Several Methods for Removing Nitrates from Water

The pollution caused by nitrogen-containing wastewater discharged during industrial production, nitrogen fertilizers applied in agriculture washed into rivers and lakes by rainwater, and domestic sewage discharged into receiving water bodies is becoming increasingly serious and has attracted widespread attention. This is because NO3- is harmful to human health. After entering the human body, NO3- is reduced to NO2-, which has carcinogenic effects. Furthermore, NO3- inhaled by infants and young children enters the bloodstream and reacts with hemoglobin, oxidizing Fe(II) to Fe(III), leading to the formation of methemoglobin. Methemoglobin irreversibly binds with oxygen, causing methemoglobinemia. The World Health Organization (WHO) has set the maximum allowable concentration of NO3-N in drinking water standards at 10 mg/L, while the NO3-N content in groundwater in some provinces and cities in my country is as high as 20-50 mg/L.

Nitrates have high solubility and good stability in water, making them difficult to co-precipitate or adsorb. Therefore, traditional simple water treatment technologies, such as lime softening and filtration, are insufficient to remove nitrates from water.

Currently, methods for removing nitrates from water include chemical denitrification, catalytic denitrification, reverse osmosis, electrodialysis, ion exchange, and biological denitrification. This article will briefly introduce these methods, focusing on the principle, method, and application status of ion exchange technology for removing nitrates from water, and comparing it with other methods.

Methods for Removing Nitrate

Chemical Denitrification

Under alkaline pH conditions, nitrates in water can be reduced to ammonia by chemical methods. The reaction equation can be expressed as:

NO3- + 8Fe(OH)2 + 6H2O → NH3 + 8 F(OH)3 + OH-

This reaction is carried out under the action of the catalyst Cu. The ratio of Fe/NO3- is 15:1. This process will produce a large amount of iron sludge, and the ammonia formed needs to be removed by air stripping.

Sorg[1] studied the removal of nitrates using ferrous compounds. The results showed that due to the high cost, this process is difficult to apply in practice. Murphy et al.[2] used powdered aluminum to remove nitrates. The main product of the reaction was ammonia, accounting for 60-95%, which could be removed by air stripping. The optimal pH for the reaction was 10.25, and the reaction equations were:

3NO3- + 2Al + 3H2O → 3NO2- + 2Al(OH)3

NO2- + 2Al + 5H2O → 3NH3 + 2Al(OH)3 + OH-

2NO2- + 2Al + 4H2O → N2 + 2Al(OH)3 + 2OH-

This process can be effectively used in water treatment plants that use lime as a softening agent because lime can usually raise the pH value to 9.1 or above. Therefore, the cost of adjusting the pH value is low, and the reaction of aluminum with water can be expressed as:

Al + 6H2O → 2Al(OH)3 + 3H2

When the pH value is 9.1-9.3, the amount of aluminum lost due to the above reaction is less than 2%. Experimental results show that 1.16g of aluminum is needed to reduce 1g of nitrate.

Reverse Osmosis

Commonly used reverse osmosis membranes include: cellulose acetate membranes, polyamide membranes, and composite membranes. The pressure range is 2070～10350kPa. These membranes are generally non-selective.

Guter[3] used a cellulose acetate membrane reverse osmosis system to remove nitrates. When the nitrate concentration in the feed water was 18～25mg/L, the nitrate removal rate reached 65% after 1000h of continuous operation.

Clifford et al.[4] studied the removal of nitrates using a reverse osmosis system with polyamide and triacetate membranes. Adding sulfuric acid and sodium hexamethyl phosphate to the feed water can prevent membrane scaling. The results show that polyamide membranes are more effective than triacetate membranes. Compared with ion exchange and electrodialysis, reverse osmosis systems are more expensive.

Rautenbach et al.[5] conducted a pilot-scale study using a composite membrane reverse osmosis system with an operating pressure of 14Pa and a treatment capacity of 2m3/h.

Electrodialysis

Miquel et al. developed a method for selectively removing nitrates using electrodialysis. This method can reduce the nitrate concentration from 50 mg/L to below 25 mg/L without the need for any chemical reagents. Rautenbach et al. [6] studied the removal of nitrates by electrodialysis and compared it with reverse osmosis. They believed that the cost of the two methods was roughly equivalent in reducing nitrates from 100 mg/L to 50 mg/L.

Catalytic Denitrification

Horold et al. [7] developed a method for removing nitrite and nitrate from drinking water. The results showed that in the presence of hydrogen, Pd-Al alloys can effectively reduce nitrite to nitrogen (98%) and ammonia. The Pb(5%)-Cu(1.25%)-Al2O3 catalyst can completely remove nitrates with an initial concentration of 100 mg/L within 50 minutes. The catalyst's nitrate removal capacity is 3.13 mg NO3-/min·g catalyst, which is about 30 times that of microbial denitrification activity. This method can be carried out at a temperature of 10°C and a pH of 6–8. The process is easy to automate and suitable for small-scale water treatment systems. This technology is currently in the research stage, and many factors, such as kinetic parameters and the long-term stability of the catalyst, require further investigation.

Biological Denitrification

Biological denitrification, also known as biological denitrification, refers to the process by which microorganisms, under anoxic conditions, utilize NO3- as an electron acceptor to perform anaerobic respiration, oxidizing organic matter and reducing nitrates to nitrogen gas. It can be represented as:

NO3- → NO2- → NO → N2O → N2

Many microorganisms exist in nature, such as Pseudomonas, Micrococcus, Denitrifying Bacteria, Achromobacter, Aerobacterium, Alcaligenes, Spirulina, Proteus, and Thiobacillus, which can grow under anaerobic conditions and reduce NO3- to N2. In this process, NO3- or NO2- replaces oxygen as the terminal electron acceptor and produces ATP. When electrons are transferred from the donor to the acceptor, the microorganisms obtain energy, which is used to synthesize new cellular material and maintain the life activities of existing cells.

Biological denitrification can be divided into heterotrophic denitrification and autotrophic denitrification, depending on the carbon source for microbial growth.

Ion Exchange Method

The principle of ion exchange for nitrate removal is that NO3- in the solution is removed by exchanging with Cl- or HCO3- on the ion exchange resin. After the resin becomes saturated, it is regenerated with NaCl or NaHCO3 solution.

Generally, the selectivity of anion exchange resins for several anions is in the following order:

HCO3- < Cl- < NO3- < SO42-

Therefore, it is difficult to treat nitrates in sulfate-containing brine using conventional ion exchange resins. This is because the resin exchanges almost all the sulfates in the water before exchanging with nitrates. In other words, the presence of sulfates reduces the resin's ability to remove nitrates. Using resins with preferential selectivity for nitrates can better solve this problem. Such resins preferentially exchange nitrates, and their nitrate exchange capacity is not affected by the amount of sulfate in the water.

Increasing the number of carbon sources around the N atom in the resin functional group NR3+ can improve the resin's selectivity for nitrates. The selectivity order of this type of resin for nitrates is as follows:

HCO3- < Cl- < SO42- < NO3-

When the methyl group around the nitrogen atom in the resin NR3+ is changed to an ethyl group, the selectivity coefficient KSN of the resin for nitrates and sulfates increases from 100 to 1000.

The results of Clifford et al. [8] show that increasing the distance between ion exchange sites can reduce the selectivity for sulfates, while increasing the hydrophobicity of the resin group and functional groups can increase the selectivity for nitrates. This increase in the selectivity of the resin for nitrates can be attributed to the fact that as the number of alkyl carbon sources increases, its volume increases, requiring more space, thus increasing the steric strain of the resin. NO3- has a stronger ability than SO42- to reduce this steric strain. The results of Guter et al. [3] showed that compared with trimethylamine resin, triethylamine resin could extend the life of the resin bed by 62% and reduce the amount of regenerator by 25-50% when treating influent containing 1.5 meq/L NO3- and 6.5 meq/L SO42-. Therefore, the operating cost of the ion exchange process was reduced. The results of Korngold et al. [9] showed that seawater can be used as a regenerator for the resin. The results of Eliassen et al. [10] showed that using strong basic anion exchange resin can reduce the NO3- concentration in the effluent of activated sludge treatment plants from 18 mg/L to 6.8 mg/L, and the treatment volume can reach 200 BV (bed volume). The presence of organic matter in the influent can easily cause resin blockage. Adding bentonite clay to the backwash water helps the resin recover. The resin is regenerated with 1N NaOH and 1N HCl. The results of Viraraghavan et al. [11] showed that the presence of silica and iron deposits in the water reduces the resin's ability to remove nitrates. Gaunlett[12] studied the continuous removal of nitrates in a closed-loop ion exchange system. Guter[3] studied the removal of NO3--N from groundwater using ion exchange technology, with a concentration range of 16–23 mg/L. Lauch et al.[13] investigated the actual operation of ion exchange resin technology for NO3- removal. The selected resin was a non-nitrate selective resin with a treatment capacity of 155 m3/h. The resin was regenerated with saturated brine. The waste brine was discharged into the aeration pond of the municipal wastewater treatment plant. The total treatment cost (including investment, operation and maintenance costs) was approximately RMB 0.53/ton. Investment costs included equipment and infrastructure costs, while operating costs included labor, electricity consumption, resin and regenerator costs. The energy consumption of the treatment plant was 0.064 kW·h/ton. The wastewater generated per ton of influent treated was: 1.4 liters of brine, 6.6 liters of resin water, and 10.3 liters of backwash water. Clifford et al. [4] conducted a 15-month pilot-scale study to compare and evaluate the removal of NO3- from drinking water using ion exchange, reverse osmosis, and electrodialysis. The influent contained 18–25 mg/L NO3-, 43 mg/L SO42-, and 530 mg/L total dissolved solids (TDS). The results showed that all three processes could reduce the NO3- concentration to below 10 mg/L. The ion exchange process resulted in higher TDS in the effluent, reaching 500 mg/L. Nitrate breakthrough always preceded sulfate breakthrough and was usually accompanied by an increase in pH. When the SO42- concentration in the influent increased from 42.5 mg/L to 310 mg/L, the nitrate breakthrough time decreased from 400 BV to 180 BV. Resin regeneration and the disposal of regenerant are important factors in the application of ion exchange processes. Guter et al. [14] reported on the operation of a plant with a treatment capacity of 155 m3/h using ion exchange to remove NO3-, and the results showed that the annual salt consumption reached 250 tons. Therefore, the disposal of large amounts of waste brine will be a very important issue. For a plant that has been operating for 20 years, its regeneration cost is more than twice the initial equipment cost. Clifford et al. [15] studied the regeneration methods of resin and believed that partial regeneration of ion exchange columns (such as eluting 60% NO3-) is more economical than complete regeneration (such as eluting more than 95% NO3-). The research results of Lanch et al. [13] showed that ion exchange is about 5 times more economical than reverse osmosis. Richard [16]'s research showed that compared with biological denitrification, the investment in ion exchange treatment plants is 2.5 to 3 times cheaper, and its operating cost is also slightly cheaper than that of biological denitrification. Richard[16] reported that in 1985, six plants in France were operating using ion exchange technology to treat NO3-, with a total treatment capacity of 576 m3/h. Woodword[17] reported that in 1990, the first ion exchange treatment plant in the UK officially started operation, with a treatment capacity of 288 m3/h. It used nitrate-selective resin with a capacity of 170 meq/L. The NO3- concentration in the influent was greater than 18 mg/L, and the daily salt required for resin regeneration reached 1000 kg. Calcium carbonate precipitation was observed in the ion exchange column and in the plant's pipelines. Acid addition could control the formation of CaCO3 precipitation. Philipot[18] et al. developed a new process in which exchange and regeneration proceed in the same direction. The nitrate concentration could be reduced from 15.8 mg/L to 5.7 mg/L, and the system could control NO3- leakage to less than 3.4 mg/L. The regenerant dosage was 90 g NaCl/L resin. The lack of in-depth research on the leaching of organic components during the reuse of synthetic resins has hindered the application of ion exchange technology in removing NO3- from drinking water.

Dore et al.[19] studied the effluent quality after NO3- removal using a strongly basic anion exchange resin regenerated with brine. The results showed that the monomer components leached from the resin included styrene, divinylbenzene, trimethylamine and its derivatives. Pretreatment of the resin with NaOH, distilled water, and HCl solution revealed that distilled water could eliminate most of the leached monomer components, and the resin did not increase the organic components in the effluent. On the contrary, the resin could adsorb some micro-pollutants in the influent, such as aromatic compounds, chlorinated organic solvents, pesticides, and nitrosamines. Therefore, the ion exchange process does not increase toxic organic pollutants in the treated effluent.

The increased Cl- concentration and decreased alkalinity in the effluent treated by the ion exchange process lead to an increased potential for selective leaching of zinc from the water pipes. This property is called the "dezincification potential" of water. When the ratio of Cl- concentration (mg/L) to alkalinity (expressed as mg/L CaCO3) in water is greater than 0.5, the water can be considered as zinc-removed water. The zinc removal potential of ion exchange process effluent can be controlled by the following measures: ① Installing a large mixing tank before water distribution; ② Regenerating the resin with brine and then rinsing it with NaHCO3 solution (two-stage regeneration system); ③ Achieving a higher NO3- breakthrough concentration in the resin.

The study by Croll et al. [20,21] found that the chloride/alkalinity ratio in the effluent from nitrate-selective resins was lower than that from effluent from general resins.

Based on the principle of ion exchange, saturated resins after the removal of NO3-, SO42-, and hardness by ion exchange processes can be regenerated with CO2 [22-24]. The process can be represented as follows:

The saturated ion exchange resin is regenerated by contacting a CO2 solution. The ion exchange resin removes neutral salts from the solution and releases an equivalent amount of carbon dioxide. Compared with conventional ion exchange processes, this process does not increase the salt content in the regenerated effluent. Therefore, only the salts removed during the exchange process need to be discharged. Based on laboratory and pilot-scale studies, a treatment plant using the above-mentioned ion exchange process was built in Germany with a treatment capacity of 170 m3/h. This process can reduce the NO3- concentration from 9 mg/L to 5.7 mg/L, and the CO2 consumption is 0.35 kg/m3 of treated water. Due to the low CO2 regeneration efficiency, nitrate-selective resins can be selected to improve the nitrate removal efficiency. The results of Wenli et al. [24] show that CO2 dissolved in water can be used as an effective regenerator at a pressure of 5–5.5 Pa. It can be seen that the driving force of this process is the partial pressure of carbon dioxide in the system. Under high pressure, the concentration of dissolved carbon dioxide in the water is high, and the reaction proceeds to the left, regenerating the resin. When the concentration of carbon dioxide in the water is low, the reaction proceeds to the right, removing sulfates, nitrates, and hardness from the water.

The advantage of ion exchange processes using carbon dioxide as a regenerator is that it does not produce excessive regeneration wastewater. Furthermore, carbon dioxide can be reused, saving on regenerator usage. The disadvantages of this process are its complexity and difficulty in operation and management. In addition, carbonates are weak acids, and the concentrations of protons and bicarbonate ions released are very low, resulting in incomplete resin regeneration.

Combined Ion Exchange/Biological Denitrification Process

Ion exchange processes require a large amount of NaCl solution (50–100 g/L) for resin regeneration. The regeneration wastewater typically contains high concentrations of NO3-, SO42-, and Cl-, which require further treatment, increasing operating costs. The effluent from biological denitrification processes requires further treatment to remove microorganisms and organic pollutants. Combining ion exchange and biological denitrification processes can overcome some of the problems of the individual processes mentioned above. A schematic diagram of the combined process flow is shown in Figure 1. In the combined ion exchange/biological denitrification process, the ion exchange process is used to remove NO3- from the water, and the biological denitrification process is used to treat the waste liquid generated during the regeneration of the resin, which contains a large amount of NO3- and Cl-. The combined process avoids direct contact between the denitrifying microorganisms and the raw water. The bioreactor can denitrify under high salinity (25-30 g/L) conditions. This process unifies the nitrate removal process into a closed-loop system, and compared with the traditional ion exchange process, this combined process can reduce the amount of waste brine generated by 95%.

Clifford[25] developed a process combining ion exchange and sequencing batch reactor (SBR) for biological denitrification. The regenerated solution contains 30 g/L NaCl and 835 mg/L NO3--N. When methanol is added to make the methanol/NO3--N ratio 2.2, NO3--N can be completely removed after 20 h. When the methanol/NO3--N ratio is 2.7, the NO3--N removal rate can reach 95% within 8 h. This combined process can reduce the consumption of regenerator by 50% and the discharge of waste brine by 90% [20].

Comparison of various methods

Ion exchange, biological denitrification and reverse osmosis are common methods for removing NO3--N from water and have been put into practical use. Ion exchange technology is suitable for treating groundwater with low dissolved organic matter. The presence of organic matter will pollute the ion exchange resin and reverse osmosis membrane. When the total dissolved solids (TDS) in the water is <500 mg/L and SO42- is <300 mg/L, the ion exchange process can be selected. When the TDS in the water is >1000 mg/L, reverse osmosis or electrodialysis can be selected. The most significant challenge with ion exchange technology is the treatment of waste regenerators, which contain NO3-, SO42-, and NaCl. Furthermore, the effluent is prone to causing pipe corrosion. Despite this, ion exchange technology is considered a viable option due to its simplicity, durability, effectiveness, and relatively low cost. Several plants in the United States have already implemented this process. Biological denitrification technology has received considerable research and application in Europe. Data indicates that heterotrophic biological denitrification is more widely used than autotrophic biological denitrification. This is because heterotrophic denitrification has a higher specific volumetric nitrogen removal rate, ranging from 0.4–24 kg NO3- -N/m3·d to 0.5–1.3 kg NO3- -N/m3·d. The techno-economic feasibility of heterotrophic biological denitrification technology has been confirmed in several European countries. Autotrophic biological denitrification processes, due to their low reaction rate, require longer hydraulic retention times, resulting in large reactor volumes and increased investment costs. Heterotrophic biological denitrification can also remove trace organic pollutants from water, such as trichloroethylene and carbon tetrachloride. In biological denitrification, fluidized bed reactors are superior to packed bed reactors. Compared to packed beds, fluidized beds prevent clogging and channeling, and offer higher nitrate removal rates.

Influent water quality, such as trace organic pollutants and SO42-, has a greater impact on ion exchange processes than on biological denitrification. Therefore, biological denitrification is suitable for surface water, while ion exchange is more suitable for groundwater. Reverse osmosis and electrodialysis processes have high energy consumption and operating costs. Reverse osmosis membranes have high selectivity for inorganic salts, and the treated water is essentially free of inorganic salts. Therefore, only a portion of the water needs to be treated before mixing with the untreated water. Electrodialysis, on the other hand, requires the treatment of all water. If wastewater discharge costs are disregarded and water loss is negligible, the water treatment costs for both methods are almost the same. Compared to electrodialysis, reverse osmosis is simpler to manage, making it particularly suitable for small treatment plants. However, the concentration effect of reverse osmosis can lead to scaling of silica, calcium carbonate, and calcium sulfate, affecting the normal operation of the treatment process.

The ammonia produced by chemical denitrification using aluminum can react with chlorine to form chloramine, which can improve the stability of residual chlorine disinfection in the water supply system. However, care must be taken when treating residual aluminum in the water.

Table 1 compares the methods of nitrate removal using ion exchange, biological denitrification, and reverse osmosis.

Table 1 Comparison of Nitrate Removal Using Ion Exchange, Biological Denitrification, and Reverse Osmosis

In summary, each of the above methods has its advantages and disadvantages. With increasing environmental awareness, the removal of nitrates from water and the prevention of their harmful effects are receiving growing attention in my country. Developing efficient and low-consumption technologies for removing NO3- from water, suitable for my country's national conditions, is an urgent task.

References

[1] Sorg T J. Treatment technology to meet the interim primary drinking water regulations for inorganics. J. AWWA, 1978, 70(2): 105.

[2] Murphy A P. Chemical removal of nitrate from water. Nature, 1991, 350: 223-225.
[3] Guter G A. Removal of nitrate from contaminated water supplies for public use. EPA/600/S2-82-042, U.S.EPA, Cincinnati,Ohio,1982.
[4] Clifford D, Lin C and Boegel J. Nitrate removal from drinking water in Glendale, Arizona. EPA/600/52-86-107, U.S.EPA, Cincinnati,Ohio,1987.
[5] Rautenbach R, KoppW and Hellekes R et al. Separation of nitrate from well water by membrane processes (reverse osmosis/electrodialysis reversal). Aqua, 1986, 5: 279-282.
[6] Rautenbach R, Kopp W and van Opbergen G et al. Nitrate reduction of well water by reverse osmosis and electrodialysis-studies on plant performance and cost. Desalination, 1987, 65: 241-258.
[7] Horold S, Tacke T and Vorlop K D. Catalytic removal of nitrate and nitrite from drinking water: 1. Screening for hydrogenation catalysts and influence of reaction conditions on activity and selectivity. Environ. Tech., 1993, 14: 931-939.
[8] Clifford D and Weber Jr. W J. Nitrate removal from water supplies. EPA/600/2-78-052, U.S.EPA, Cincinnati,Ohio,1978.
[9] Korngold E. Removal of nitrates from potable water by ion exchange. Water Air Soil Pollution, 1973, 2: 15-22.
[10] Eliassen R, Wyckoof B M and Tonkin C D. Ion exchange for reclamation of reusable supplies. J.AWWA. 1965, 57(9): 1113-1122.
[11] Viraraghavan T and Rao G A K. Treatment processes for nitrate removal from water. Proc., 42^nd Annu. Convention of the western Canada Water and Wastewater Assn., Regina, Sask., Canada, 1990, 137-160.
[12] Gantlett R B. Nitrate removal from water by ion exchange. Water Treatment &Examination, 1975, 24: 172-193.
[13] Lauch R P and Guter G A. Ion exchange for the removal of nitrate from well water. J.AWWA., 1986, 78(5): 83-88.
[14] Guter G A. Nitrate removal from contaminated water supplies. EPA/600/S2-87-034, U.S.EPA, Cincinnati,Ohio,1982.
[15] Clifford D and Liu X. Ion exchange for nitrate removal. J.AWWA. 1993, 85(4): 135-143.
[16] Richard Y R. Operating experiences of full scale biological and ion-exchange denitrification plants in France. J. Inst. Water Environ. Mgmt., 1989, 3: 154-167.
[17] Woodword A J. Removing nitrates from potable water. Institution of water and environment management yearbook, 1994. London, England, 39-45.
[18] Philpot J M and de Larminant G. Nitrate removal by ion exchange: the ecodenit process and industrial scale facility at Binic (France). Water Supply, 1988, 6: 45-50.
[19] Dore M, Simon P and Deguin A et al. Removl of nitrate from drinking water by ion exchange _impact on the chemical quality of treated water. Water Res. 1986, 20(2): 221-232.
[20] Croll B T and Hayes C R. Nitrate and water supplies in the United Kingdom. Environ. Pollution, 1988, 50: 163-187.
[21] Andrews D A and Harward C. Isleham ion -exchange nitrate removal plant. J. Inst. Water Environ. Mgmt. 1994, 8: 120-127.
[22] Hagen K, Holl W and Kretzschmar W. The CARIX process for removing nitrate, sulfate and hardness from water. Aqua. 1986, 5: 275-278.
[23] Holl W and Kretzschmar W. Combined nitrate and hardness elimination by the CARIX ion exchange process. Water Supply, 1988, 6: 51-55.
[24] Wenli G, Wensheng H and Holl WH. Combined nitrate and hardness elimination from drinking water by the CARIX process. Aqua. 1994, 43(2): 95-101.
[25] Clifford D and Liu X. Biological denitrification of spent regenerant brine using a sequencing batch reactor. Water Res. 1993, 27(9): 1477-1484.