Nitrate removal from wastewater generated in wet Flue Gas Desulphurisation Systems (FGD) in coal-fired power generation using the heterotrophic denitrification method
, , y | 30 sept 2020
Publicado en línea: 30 sept 2020
Páginas: 27 - 34
DOI: https://doi.org/10.2478/oszn-2020-0012
Palabras clave
© 2020 Krzysztof Iskra et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
The most commonly used method of flue gas desulphurization (FGD) in power plants and cogeneration (CHP) plants in Poland is the wet lime based technology, which has been described by the European Union as one of the best available technology (BAT) [Brinkmann et al. 2017]. The flue gas desulphurisation installations basically generate large amounts of wastewater with particular characteristics, that is, very high salinity (even up to 40,000 g/m3 of chlorides and sulphates in total) and high temperature up to 50°C [Vredenbreght et al. 1997]. This wastewater also contains a whole array of heavy metals [Huang et al. 2013]. It often has relatively high chemical oxygen demand (COD) (above 300 gO2/m3), with a very low proportion of readily biodegradable fraction expressed as biochemical oxygen demand (BOD5). The essential aspect, from the point of view of this paper, is that such wastewater has a significant content of nitrogen compounds in all forms, with a predominance of nitrates, which usually constitute 70% of the total nitrogen content. There are also cases of very high ammonium nitrogen levels of 600 gN-NH4/m3 and nitrate nitrogen up to 200 gN-NO3/m3. Before further management or disposing to a reservoir, it is purified in physico-chemical processes in one- or two-stage systems that ensure reaching normative values, such as pH value, temperature, concentration of suspended solids, most of heavy metals and, to a partial extent, of compounds expressed as COD and total organic carbon (TOC); however, they are not adapted for removing nitrogen forms [Litwinowicz 2014]. From the point of view of efficiency of the heterotrophic denitrification process, especially in industrial wastewater, the supply of easily digestible substrate is necessary [Hirata, Nakamura 2001]. Alternatively, some studies are being conducted on autotrophic denitrification that has aroused the interest of many researchers [Park, Yoo 2009]. An attempt to remove nitrates from industrial wastewater of particular characteristics was undertaken by, among other Gabaldón [Gabaldón et al. 2007]. He received for wastewater from metalworking industry (high salinity and low BOD5 concentration) the nitrate removal efficiency exceeding 90%.
The need to maintain allowable nitrogen concentrations in treated wastewater is regulated by the Act of Minister of Marine Economy and Inland Navigation of 12 July 2019 on substances that are particularly harmful to the aquatic environment and the conditions to be met when disposing wastewater into waters or into the ground, as well as when disposing rainwater or snowmelt into waters or into water facilities [J. of L. 2019, item 1311]. The limit for total nitrogen has been set at 30 gN/m3, which for the energy sector plants is determined individually on the basis of integrated permits, however, often at the level of several hundred gN/m3. According to the authors of this publication, the legal solution to the problem of wastewater disposal containing large amounts of nitrogen into the aquatic environment is insufficient, and over time, it may pose a significant contribution to the inland water quality deterioration and the Baltic Sea consequently. The previously mentioned document [Lecomte et al. 2017] and BREF summary for waste water and waste gas treatment and management in the chemical sector [Brinkmann et al. 2016] indicate that it is required to look for opportunities to improve the situation, with particular regard to streams with a higher concentration, charge, potential threat and impact on the water reservoir, to which it is disposed. According to available information, the global technology providers for FGD systems have no technological solutions in their offers allowing to reduce the wastewater nitrate concentration to the required level. This creates a well-grounded basis for seeking methods to intensify the nitrogen removal for wastewater from the wet FGD units. One of such methods is the use of heterotrophic denitrification, which involves the reduction of nitrates through nitrites to gaseous products under anoxic conditions with a respectively available amount of readily biodegradable organic compounds [Zhu et al. 2008]. Table 1 presents the methods used for wastewater treatment from wet FGD systems given their major advantages and disadvantages. The methods for which the nitrate removal is possible are additionally highlighted.
The summary of methods used for wet FGD wastewater treatment
| 1 | Simplified methods | Operation does not require qualified service Resilience to changes in the FGD technology | Large area required for use of lagoons for wastewater / sediment disposal | No |
| 2 | Chemical methods | Small installation volume (short time chemical reaction) Technology with a wide range of both customization and control strategy options | High demand for chemical reagents | No |
| Biological methods | ||||
| 3 | Treatment reactors | Lower maintenance costs compared to chemical methods Consuming less chemical reagents | Operation requires qualified staff Small resilience to changes in wastewater composition | Yes |
| 4 | Hydrophyte wastewater treatment plants | Possibility of achieving parameters similar to those for treatment reactors | Large area required Significant reduction in the efficiency of wastewater treatment at low temperatures | Yes |
| Physical methods | ||||
| 5 | Evaporative methods | A significant reduction in the amount of generated wastewater | High maintenance costs of concentrating wastewater on evaporators | No |
| 6 | Membrane methods | Lower maintenance costs compared to evaporative methods | The need for frequent flushing of membranes High risk of precipitation of mineral salts on the surface of membranes (fouling) High investment costs | Yes1 |
| 7 | Ion exchange methods | In the case of an existing ion exchange installation at the facility, the extension will take lower investment costs | Need for frequent ionites regeneration Higher efficiency of sulphate ion exchange than nitrate ion | Yes |
The aim of this study was to evaluate the possibilities for the application of heterotrophic denitrification process in the nitrates removal from wastewater from wet flue gas desulphurization installations and optimization of the process in the scope of basic technological parameters. The object of the studies was particular wastewater from wet FGD installations where the flue gas desulphurisation process is carried out with the wet lime method.
The pilot experimental installation consisted of one technological line, and its diagram is presented in Fig. 1. Before the wastewater was disposed into a reactor in the form of a biological bed, it had been collected in a balancing tank (ZŚ) with a capacity of 1 m3. The tank was equipped with an
Figure 1
Diagram of the installation for conducting the denitrification process: ZŚ - raw wastewater tank, ZO - treated wastewater tank, ZD - denitrification bed, Q - wastewater flow measurement, α - denitrified effluent recirculation pump, pH - pH measurement, NO3 - nitrate measurement, T - temperature measurement, RedOx - measurement of oxidoreductive potential
The material was the wastewater from one of the country wet FGD installations after physicochemical treatment. Table 2 provides detailed characteristics of the material from the studies. Nitrate concentrations varied, and they changed within quite a wide range from 53 to 185 gN-NO3/m3, constituting a significant part of total nitrogen. It was also found that the experimental material had a very low concentration of phosphorus in absorbable form and BOD5 and high concentration of dissolved compounds and suspended solids. This wastewater differs radically from typical municipal wastewater, being simultaneously an unfavourable material for conducting biological processes. Fig. 2 presents the changes in the wet FGD wastewater composition (wastewater from an actual energy facility). These data indicate that the wastewater composition has changed quite significantly. The nitrate nitrogen concentrations reached the highest value in first days and in the 120th day of the studies. In periods when its concentration was relatively low, before being transferred to the model reactor, it would be increased to approx. 100 gN-NO3/m3 by dosing the sodium nitrate solution. The COD mean value was 222 gO2/m3, however, it was a substratum inassimilable for microorganisms leading the denitrification process. The value of BOD5 determining the amount of easily biodegradable organic compounds did not exceed 5% of the COD value. There were also no assimilable forms of phosphorus recorded. The experimental bacteria inoculum was taken from the activated sludge system from a municipal wastewater treatment plant. The composition bacteria biomass was not tested.
Figure 2
Changes in nitrate and COD concentration in wastewater from wet FGD unit
The wet FGD wastewater characteristics during the studies
| 1 | pH value | - | 8.84 | 8.87 | 8.57 | 7.47 | - |
| 2 | BOD5 | gO2/m3 | 0.73 | 0.87 | 1.11 | 1.39 | 1.0 |
| 3 | COD | gO2/m3 | 209 | 213 | 246 | 222 | 222.5 |
| 4 | Total nitrogen (Ntotal) | gN/m3 | 233 | 68.2 | 87.7 | 101.9 | 122.7 |
| 5 | Kjeldahl nitrogen (TKN) | gN/m3 | 43.2 | 21.9 | 35.9 | 0.97 | 25.5 |
| 6 | Ammonium nitrogen (N-NH4) | gN/m3 | 30.5 | 14.9 | 19.6 | 0.62 | 16.4 |
| 7 | Nitrate nitrogen (N-NO3) | gN/m3 | 185.2 | 53 | 58 | 100.9 | 99.3 |
| 8 | Nitrite nitrogen (N-NO2) | gN/m3 | 4.61 | 0,84 | 1.1 | 0.03 | 1.6 |
| 9 | Total phosphorus (P) | gP/m3 | 7.5 | 5.73 | 1.7 | - | 5.0 |
| 10 | Orthophosphates (P-PO4) | gP/m3 | 0.043 | 0.18 | 0.07 | 0.03 | 0.1 |
| 11 | Total suspended solids (TSS) | g/m3 | 422 | 392 | 212 | 69 | 273.8 |
| 12 | Chlorides (Cl−) | g/m3 | 35400 | 19200 | 18500 | 16000 | 22275.0 |
| 13 | Sulphates (SO42−) | g/m3 | 1186 | 1100 | 1227 | - | 1171.0 |
Before the wastewater was disposed into the model reactor, it was subjected to preliminary treatment by raising its temperature to approx. 35°C, pH value correction to 8.3 and, according to needs, correction of nitrate concentration to the value of approx. 100 gN-NO3/m3. The wastewater was disposed to the process reactor by peristaltic pumps dosing in the amount resulting from the studies stage. A separate stream provided a source of organic carbon and phosphorus in the amount of 130% of stoichiometric consumption in the denitrification process. In order to increase the degree of even spread of the specific surface loading rate on the denitrification bed, recirculation of treated wastewater in the amount of 200% of the inflow value was used. The operational control of the denitrification bed included daily sampling of treated wastewater and determination of the most important parameters from the point of view of the denitrification process – COD, Ntotal, N-NO3, P-PO4. An additional tool for controlling the outflow quality was the use of an optical probe monitoring the nitrate nitrogen concentration
The effects of the biological bed performance are shown in Figure 3. In the initial period of the studies involving the adaptation of biomass to the composition of wastewater from wet FGD, quite big fluctuations of obtained results were observed. Drastic decreases in efficiency were observed especially in periods corresponding to the increase in the share of wet FGD wastewater inflow to the reactor. However, the most important element during this period was not obtaining a satisfactory wastewater quality but avoiding permanent inhibition of the biological bed biomass caused by the unfavourable for the microorganisms composition of the wet FGD wastewater. Therefore, the reactor was operated at a low nitrates loading rate. After about 3 months of developing the reactor, the municipal wastewater supply was turned off and a systematic increase of nitrate loading rate began. Not counting periodic performance decreases, a very high process efficiency was observed over the next three months. The nitrate nitrogen removal rate was above 95% and considering its initial concentration at the level of approx. 100 g/m3, it was found that the target effluent quality was successfully achieved, that is, below 30 gN/m3 at the wastewater reaction time being 180 minutes during the denitrification rate test (NUR). It was also determined that the optimal loading range of the active surface of the bed about 300 m2/m3 should be within 1.5–2.5 gN-NO3/m2·d, which is illustrated by the results presented in Figure 4. During the period of optimal work of the reactor, a test was also carried out to determine the loading rate limits and resistance of organisms to the abrupt changes in loading rate. To this end, the inflow loading rate was reduced to approx. 5 gN-NO3/d, followed by a sudden increase to value approx. 10 gN-NO3/d, which corresponded to the loading rate on the active surface of the bed being 3 gN-NO3/m2·d. The basis for this experiment was the fact of large irregularities on the inflow and in the composition of the wastewater from wet FGD. The sudden change in the bed loading rate inhibited the efficiency of nitrate reduction, which fell from 99 to barely 20%. However, this continued for a fairly short time, because in the next days of the experiment, the reactor operation was improving, namely, the denitrification biomass worked with increasing efficiency, eventually reaching the nitrate reduction level of 93–97% after approx. 17 days. This experiment has shown that the biological system does not tolerate rapid changes conditions of its operation. It has been defined that the work regime changes should be introduced gradually, and they must not exceed 10–15% per day in relation to the initial values.
Figure 3
Nitrate nitrogen removal rate apposed its loading rate at the reactor inflow during the studies
Figure 4
Effect of active surface loading rate on efficiency (% removal)
Removal of nitrate nitrogen from wastewater from wet FGD installation is a relatively new issue. This specific wastewater group is characterized by significantly high concentrations of chlorides and sulphates, the presence of heavy metals and suspensions (mainly gypsum) and significant content of forms of nitrogen compounds. So far, the main attention has been paid mainly to the removal of heavy metals, which are quite strictly limited in BAT conclusions established for large combustion plants (LCP). Depending on the type of metal, emission standards range from 0.2 μg/L for cadmium to 200 μg/L for zinc. Controlling these requirements implies the use of highly effective precipitation methods, often requiring several steps of its treatment. However, such treatment plants have a minor impact on the content of nitrogen compounds. This studies showed a great potential in removing nitrate nitrogen using denitrification bacteria. Cheng et al. [2011] considered that artificial ecosystem (wetlands) can be a cost-effective and low-energy method of removing biogenic pollutants. Biological reactors constructed in this way can remove and degrade nitrogen, phosphorus and some metals. In relation to nitrogen compounds, these can be both nitrification and denitrification processes. In turn, Zhu and Liu [2017] believed that elevated levels of salinity in industrial wastewater can have a major impact on denitrification, but few studies have been done so far. To remove nitrates from such kind of wastewater, it is important to understand the effect of salinity on process kinetics, selection of carbon sources, periodic fluctuations in salinity and microorganism communities in process selection and engineering design.
The problem of the presence of nitrogen forms in wastewater from wet FGD installations has also been noticed by researchers from the Electric Power Research Institute (EPRI). In a report on pollution from the energy sector [EPRI, 2007], they suggested that the presence of nitrates may require additional purification steps due to high concentrations. As a possible way, they indicate the ABMet® system, which can effectively remove metal, metalloid, non-metal and inorganic compounds such as nitrates. The purification system consists of a series of connected chambers (in the form of fluidized beds), each of which contains specialized bacteria. Problem of nitrate nitrogen removal from high salinity wastewater was also taken up by Yang et al. [1995]. In their studies, they analysed the process of immobilizing microbial cells trapped in a mixed culture (EMCI) to remove nitrates from brine waste resulting from the regeneration of used ion exchange resins. With salinity up to 20 g/L, they achieved nitrate nitrogen concentration in the outflow of 0.1–0.2 g/m3.
The wet FGF installations are provided with wastewater treatment plants, the technological level of which does not allow to meet all the legal requirements for the treated wastewater from the energy sector, which in particular relates to nitrogen compounds. In this context, the existing wastewater treatment plants can be considered as a two-stage physico-chemical pre-treatment plants providing correction of pH value, temperature, concentration of suspensions, COD and heavy metals. One of the promising methods for their efficiency improvement is the extension by a biological facility, enabling the removal of nitrogen compounds by the method of heterotrophic denitrification. The studies show that after providing right conditions for the development of microorganisms, it is possible to develop and adapt the denitrification biomass completely to the adverse composition of the wastewater from wet FGD installations. While maintaining the optimal bed loading rate in the range of 1.5–2.5 gN-NO3/m2·d, the over 90% reduction level of nitrate nitrogen has been repeatedly achieved. Considering that this parameter constituted for approx. 70% of total nitrogen in influent wastewater, it gives a strong chance of reduction to the level required by law. The developed technology has been submitted to the Patent Office of the Republic of Poland and currently pending at the application stage.