REMEDIATION OF WATER CONTAMINATED WITH NITRATE. DEVELOPMENT OF A CONTINUOUS ADSORPTION/CATALYSIS SYSTEM

M.B. PERONI^†‡, L. MICHELENA^†‡, A. MARTÍNEZ^§, M. FERNANDEZ^*,
B. BARBERO^§, M. CASELLA^† and M. JAWORSKI^†‡

† Centro de Investigación y Desarrollo en Procesos Catalíticos (CINDECA), UNLP (Facultad de Ciencias Exactas)-CONICETCCT-CICPBA 47 Nº 257, 1900, La Plata. Argentina.

‡ Facultad de Ingeniería, UNLP, 50 y 116, 1900, La Plata. Argentina

§ Instituto de Investigaciones en Tecnología Química (INTEQUI), UNSL-CONICET, Almirante Brown 1455, D5700HGC San Luis, Argentina.

* Centro de Tecnología de Recursos Minerales y Cerámica (CETMIC), UNLP-CIC-CONICET, Camino Parque Centenario 2499-2599, Gonnet, Buenos Aires, Argentina

Corresponding Author Email: belen.peroni@ing.unlp.edu.ar

Cite this article as:

Peroni, M.B., Michelena, L., Martínez, A., Fernandez, M., Barbero, B., Casella, M., Jaworski, M. (2023) “Remediation of water contaminated with nitrate. Development of a continuous adsorption/catalysis system”, Latin American Applied Research, 53(1) pp 65-70.

Abstract-- The elimination of NO₃^- from water was studied using PdCu structured catalysts prepared on cordierite monoliths coated with Al₂O₃. The influence of the source of alumina to coat the nude monoliths and the method of preparation of PdCu catalyst was evaluated. The adherence of the Al₂O₃ on the monoliths was carried out by washcoating using two different suspensions 1) an alumina suspension prepared in the laboratory, and 2) a commercial bohemite colloidal suspension. With bohemite suspension, higher alumina mass gain and better adherence were obtained. No difference was found in the reduction of NO₃^- with the catalyst preparation method (PdCu or CuPd). The adsorption of NO₃^- was studied using the montmorillonite clay modified with hexadecyltetraethylammonium mixed with sea sand in a column filtration system. The NO₃^- was desorbed from the column and the concentrated NO₃^- was satisfactory removed using PdCu structured catalyst. The adsorption/catalysis combination system used in this work is a promising technology for the treatment of water contaminated with NO₃^- because it allows to regenerate the adsorption column and to eliminate NO₃^-.

Keywords-- nitrate, cordierite monoliths Al₂O₃ support, PdCu catalyst, montmorillonite clay.

I. INTRODUCTION

Nowadays, there is concern about water quality. One of the most worrying problems is the high levels of nitrate concentration found in some groundwater deposits (Xu et al., Zhai et al., 2017).  Excessive NO₃^- intake is harmful, because it is easily reduced to nitrite in the mouth and intestines, causing serious health problems, particularly in children (blue baby syndrome). Also, NO₂^- are precursor of carcinogenic nitrosamines (Zhai et al., 2017). The maximum limit of nitrate allowed by the World Health Organization is 50mg/L in the population’s water supply (Fan, 2019). In our country, the Argentine Food Code establishes a maximum concentration of NO₃^- of 45 ppm (ANMAT, 2021).

From an environmental point of view, the best technique to remove NO₃^- is one that converts them to innocuous products. One of the most promising processes is the reduction of NO₃^- to N₂ using heterogeneous catalysts in the presence of H₂ as the reducing agent (Kim et al., 2016; Zoppas et al., 2016). Generally, these catalysts contain a precious metal as base metal (Pd, Rh, Ru or Pt) and a promoter metal (Cu, Ag, Fe, Hg, Ni, Cu, Zn, Sn or In).  Kim et al. (2016) and Zoppas et al. (2016) being PdCu active phase the most active and selective. In these catalysts, the bimetallic sites allow the reduction of NO₃^- to NO₂^-, which are the reduced to N₂ or NH₄⁺ in the monometallic sites, an inconvenient that these systems present (Ding et al., 2017).

However, the use of a suspended powder catalytic formulation is not technologically feasible for the use of this type of catalyst can cause problems in its recovery, leading to contamination of treated water by metallic particles. One of the ways to avoid the loss of catalytic material is to use pellets or extrudates (Yuranova, et al., 2012). Structured catalysts based on monoliths with different chemical composition are widely used in environmental catalytic applications (Landi et al., 2016).

Additionally, the development of catalytic reactors would require large centralized treatment units to treat large volumes of water. That is why it is necessary to develop a system in which the NO₃^- are concentrated using adsorbents and subsequently eliminated (Kim et al., 2016).

As exposed, the adsorption and catalytc process play a central role in the development of technology for eliminating NO₃^- in water. The adsorption of this anion has been investigated using various materials (clays, resins, active carbon) (Wu et al., 2016). Clays are not only abundant and inexpensive, but their ability to accept various functional groups (for example, by cation exchange reactions, pillaring, etc.) modifies their characteristics, expanding their applications as powerful adsorbents of NO₃^-, CrO₄ ^-2, AsO₄^-3, which are toxic to health and the environment (Gammoudi et al., 2013; Xi et al., 2010).

Figure1. Scheme of the cordierite monoliths used. Left: monolith, Right: channels of the monolith used

Montmorillonite clay (Mt), of Argentine origin, is an aluminium silicate composed of hydrated and interchangeable cations (Azaro et al., 2021). In previous work, Mt clay was modified with the surfactant hexadecyltetratehylammonium (Mt-HDTMA) to increase the adsorption capacity of NO₃^- and thus design a system that allows adsorbing and concentrating NO₃^- to be disposed in a catalytic reactor containing the structured catalysts already prepared.

In this way, this work proposes the development of different formulations and procedures that allow a better adherence of the active phase, PdCu, on structured supports based on Al₂O₃ to be tested in the elimination of NO₃^- from water. In order to design an integrated system, the NO₃^- were first adsorbed and concentrated in a montmorillonite clay from Argentina. The NO₃^- desorbed from the column will be catalytically removed in the reactor containing the structured catalysts

II. EXPERIMENTAL

A. Preparation of alumina structured supports

Structured support: monolith

The monolithic support used in this work was obtained from a commercial cordierite support provided by Corning® (2MgO₂.Al₂O₃.5SiO₂). The structured supports were cut as cylinders with 10mm of diameter and 30 mm of large (Fig. 1) and washed with distilled water in an ultrasonic device to remove impurities.

To prepare the structured support, the washcoating method was used and different alumina suspensions were prepared. In order to study the effect of the Al₂O₃ support nature and to improve the coating of Al₂O₃ on the monoliths, two suspensions were used: a γ-Al₂O₃ suspension prepared in the laboratory (C-γAl) and Nyacol^® AL20 bohemite colloidal suspension (C-Ny).

For the preparation of C-γAl catalysts, a commercial γ-Al₂O₃ (Air Products, S_BET: 279 cm²/g, V_pore: 0.63 cm³/g) was finely milled in a ball mill instrument (Fritsch, Pulverisette model 6) during 5 h. The particles size was determined using the analyzer MASTERSIZER 2000 (Malvern Instruments).

With the obtained solid, a suspension containing 12 wt. % of γ-Al₂O₃ in water was prepared at pH 4.5. This suspension was placed in an ultrasound device for 30 min to obtain a stable suspension.

For the preparation of C-γAl and C-Ny, the nude monoliths were immersed in the respective suspension during 2 min, then they were extracted and centrifuged (600 rpm, 2 min) to remove the excess suspension retained in the channels, dried and calcined in air at 500°C for 2 h. For each suspension four monoliths were made. The extraction-centrifugation-drying-calcination-weighing process was repeated tree times for all the structured supports prepared.

To evaluate the adherence of Al₂O₃ on C-Ny and C-γAl supports, the prepared supports were immersed in distilled water and subjected to ultrasonic agitation for 30 min. The monoliths were removed from water, centrifuged, dried, calcined at 500°C and weighed. The adhesion of the alumina layer was calculated by weight difference. This weight difference before and after the ultrasound treatment was used to determine the coating adherence (Jaworski et al., 2014):

where  is the weight before ultrasonic treatment and  is the weight after ultrasonic treatment

B. Synthesis and characterization of PdCu structured catalysts

The PdCu active phase was deposited on the monoliths wrapped with alumina by simple impregnation. As source of Pd, H₂PdCl₄ solution was used and for Cu a solution containing Cu(NO₃)₂.2.5H₂O was prepared. A preliminary study carried out in our laboratory optimized a powder PdCu/γ-Al₂O₃ catalyst, which presented good catalytic performance (at/at Cu/Pd ratio 0.5) (Jaworski et al., 2014) so in this work we used the same relationship.

To study the influence on the addition order of the active phase in bimetallic catalysts, the monoliths were first contacted with the Pd solution and then with the Cu solution and vice versa. The catalysts were named PdCu (first Pd and second Cu) or CuPd (first Cu and second Pd), depending on the addition of the active phase.

To prepare the bimetallic catalysts, the prepared structured supports (monoliths with alumina) were placed in a suitable container with one of the active phase solution and then inside an ultrasonic equipment for 30 min. Then, they were dried after each impregnation in microwave for 2 min at 200W. The process was repeated until obtaining the desirable amount of Pd and Cu on the structured supports. The content of Pd and Cu on supports before and after impregnation was determined by atomic absorption (AAS) in a Varian Spectra AA 55 spectrophotometer. Finally, the catalysts were calcined in air for 2 h at 400 °C of temperature before being used in reaction.

The nomenclatures of the prepared catalysts are named in Table 1.

The surface morphology and composition of the catalyst prepared were analyzed by SEM-EDS, using an Oxford SDD X-Act model FEI ESEM Quanta 200.

Table.1. Nomenclature of the structured catalysts prepared

Nomenclature	Active Phase	Support
PdCu/C-Ny	PdCu	Nyacol^® AL20
CuPd/C-Ny	CuPd	Nyacol^® AL20
PdCu/γ-Al	PdCu	γ-Al2O3

Figure 2. Diagram of the desorption-catalysis system used

C. Catalytic reaction

Before to start the reaction, the catalysts were previously reduced using H₂ gas (2 h at 400°C) before being used in reaction and were evaluated in a fixed bed reactor with recirculation (Fig 2). The reactor used was described in a previous work and consisted of a tubular glass reactor which contains the structured catalysts inside [15]. To start the reaction, 50mL of previously degassed aqueous solution containing 100ppm NO₃^- prepared in the laboratory was placed in the system and the reducing agent H₂ was bubbled continuously to the reactor.

The NO₃^- solution was re-circulated with a peristaltic pump at a flow rate of 0.31 mL/min.

The reaction was carried out for 6 h at room temperature and atmospheric pressure. Samples were taken periodically from the system to determinate NO₃^-, NO₂^- y NH₄⁺ (by ion chromatography, Metrohm 790 Personal IC equipment). The initial NO₃^- conversion rate at 15 min of reaction (V₀), the NO₃^- conversion (X_NO3- %) and the selectivity towards NO₂^- (S_NO2- %), NH₄⁺ (S_NH4+ %) and N₂ (S_N2 %), after 6 h of reaction were calculated according to the following equations:

                                  (1)

                                  (2)

                                  (3)

D. Synthesis of adsorbent

In a previous work we had used abundant clay from Argentina, montmorillonite (Mt) in the retention of NO₃^- from water (Azaro et al., 2021). The clay was modified with a positive surfactant, hexadecyl trimethyl-ammonium (Mt-HDTMA) with the corresponding concentration of 400% of HDTMA with respect to the Mt CEC (cation exchange capacity). The modified clay was used in batch and column systems obtaining good results in terms of NO₃^- removal.

When the column filtration system was used, 2 wt. % of the Mt-HDTMA adsorbent was mixed with commercial quartz sand (Cicarelli, particle size 0.106-0.850 mm). As was previously described (Jaworski et al., 2019), the column (see Fig. 2) was first conditioned by the slow flow of deionized water and then a solution containing 100 mg/L NO₃^- was passed through the column. After that, in order to regenerate the column, the NO₃^- retained on the column was eluted with 1M NaCl solution until NO₃^- was not detected in the eluted water. Then, the eluted solution with the recovered NO₃^- (100ppm NO₃^- in 50 mL solution) was eliminated using the structured PdCu catalysts. A description of the absorption/desorption/catalysis reactor is depicted in Fig. 2.

III. RESULTS

A. Catalyst characterization

Table 2 shows the results of adherence efficiency of Al₂O₃ on the monoliths C-Ny and C-γAl. In column 2, it is shown the weight increase after 3 immersions for each structured support prepared. The fourth column lists the mass gain of Al₂O₃ on the monoliths.

The particle size obtained for the milled Al₂O₃ used was 1-100 μm with an average value of 1-3 μm (see Supplementary Material) which is a suitable particle size to obtain a good coating of the monoliths with alumina.

It was observed that the percentage of Al₂O₃ retained after adhesion test is 98% when the alumina deposited came from a suspension of colloidal boehmite and 65% when finely ground Al₂O₃ is used.

Regarding the mass of alumina retained after three impregnation cycles (Table 2, column 4), the results showed that the C-Ny support retained about 150 mg of Al₂O₃, and C-γAl only retained around 50 mg of Al₂O₃. In addition, we analyzed the stability of the γ-Al₂O₃ suspension and found that this last one is not stable in the time. We observed a solid at the bottom of the flask containing the suspension.

In conclusion, with the boehmite solution a better adherence of alumina was obtained in comparison to the structured support prepared using grounded Al₂O₃. In addition, the mass of alumina retained is superior in C-Ny. Different photographs of the structured supports and the structured catalysts prepared were taken (Fig. 3). When the structured supports C-Ny and C-Al are compared, Fig. 3 b) and c) respectively, it is observed that the Al₂O₃ coverage on C-γAl is inhomogeneous. There are some rough and dark zones cover by Al₂O₃ and other light zones corresponding to the monolith without covering. On the other hand, the C-Ny support shows the complete coating of Al₂O₃ on the monoliths. The same trend is observed on the structured catalysts prepared. A homogeneous deposit of the active PdCu is obtained on PdCu/C-Ny and on PdCu/γ-Al there are some light and darks zones, which indicates that Pd and Cu metals are not uniformly, distributed on the support surface.  

The structured support and catalysts prepared were analyzed by SEM-EDX analysis (Fig.4). In Fig 4.a) it is clearly observed in C-γAl the inhomogeneous covering of Al₂O₃ on the monolith. There is a thicker layer of alumina in some sectors which generates a bad distribution on the surface of the monolith. On the contrary, on C-Ny (Fig 4.b) all the surface of the monolith contained a thin and homogeneous layer of Al₂O₃ on its surface.

The semi-quantitative analysis made by EDX showed the presence mainly of Al, O, Si, Mg and C on the structured supports prepared.

For both structured catalysts prepared (Fig. 4 c) and d)), the active phases Pd and Cu were found on the entire monoliths surface.

Table 2. Adhesion test results on the monoliths. Column 1: structured support, Column 2: weight increase of Al₂O₃ support Column 3: mass of Al₂O₃ retained on monoliths after the adherence test. Column 4: suspension stability

Figure 3: Photos of the structured supports and catalysts prepared. a) nude monolith: b) C-γAl; c) C-Ny; d) PdCu/C-γAl and e) PdCu/C-Ny

Figure 4: Photos of results of Scanning Electron Microscopy technique a) C-γAl; b) C-Ny; c) PdCu/γ-Al and d) PdCu/Ny

Table 3. Elimination of NO₃^- (X_NO3-) and selectivities towards NO₂^-, NH₄⁺ and N₂ in a fixed bed reactor after 6 h of reaction for the catalysts prepared

Catalyst	X% NO3-	SNO2- %	SNH4+ %	SN2 %
PdCu/C-γAl	19	0.2	0.04	99.76
PdCu/C-Ny	50	0.1	0.03	99.87
CuPd/C-Ny	48	0.11	0.03	99.86

Table 4. Results of conversion of NO₃^- (X_NO3-) and selectivities towards NO₂^-, NH₄⁺ and N₂ after 6 h of reaction

B. Catalytic reaction

Nitrate elimination

The results for NO₃^- reduction using the structured catalysts are presented in Table 3.

Particularly, the boehmite colloidal suspension gave better results than the alumina suspension. The catalysts prepared on C-Ny monoliths were more active than those prepared on Al₂O₃, probably because in these catalysts a greater amount of Al₂O₃ was retained and the distribution of the active phase was homogeneous. It was observed that the selectivity towards the product of interest N₂ was similar in all the analyzed reaction. Besides, an increase in pH values was detected during the reaction studied (from pH 5.5 to 8).

Additionally, we calculated the Reynolds number (see Supplementary Material) in order to determinate the type of flow that circulates through the channels of the monolith. The value obtained was 5.76 which implies a laminar flow inside the monoliths channels.

Nitrate removal

It is well known that the adsorption process plays center role in the development of a technology to remove oxyanions from water, like NO₃^-.

For this reason, we have studied the removal of NO₃^- using different adsorbents based on highly available Argentine clay (montmorillonite, Mt) modified with different positive surfactants. These adsorbents were evaluated in batch and adsorption column systems.

In a previous work, column filtration systems were filled with sea sand and 2 wt.% of a montmorillonite clay modified with the cationic surfactant hexadecyltrimethylammonium (Mt-H) (Azaro et al., 2021).

The column was saturated with NO₃^- and the NO₃^- retained in the column (100 ppm) was eluted with a 1M NaCl solution until no NO₃^- was detected in the eluted water. Next, the NO₃^- eluted was removed using the PdCu/C-Ny catalyst. The results obtained are detailed in Table 4.  The results obtained from Table 4 show a successful removal of NO₃^- using the combined adsorption/catalysis system. Although the conversion of NO₃^- and the selectivity to N₂ were lower than the same reaction using NO₃^- prepared in distilled water, the results are satisfactory for the development of an adsorption/catalysis technology for the removal of oxyanions in polluted water.

IV. CONCLUSIONS

The procedure for preparing monolithic catalysts has a great influence on both their mechanical stability and their performance in reaction tests. In this work it was shown that the different suspensions used for the coating of monolithic supports generated different adherence, homogeneity and catalytic activity in removing nitrates from water. The incorporation of the support Al₂O₃ on the monolithic structure was studied using two different methods: preparing a suspension of finely milled Al₂O₃ and using a colloidal suspension of boehmite. Better results in terms of mass gain and adherence were obtained by employing bohemite suspension. No differences were observed in the catalytic performance if the catalyst used was PdCu or CuPd.

In this work, modified montmorillonite clay with hexadicyltrimethylammonium surfactant was used in the adsorption of NO₃^- using column filtration system. To regenerate the column, the NO₃^- was desorbed from the column and the concentrated NO₃^- solution was passed through the catalytic reactor containing the structured PdCu catalysts in order to eliminate this anion. This NO₃^- was satisfactory removed, indicating that the adsorption/catalysis combination is a promising technology for the treatment of water contaminated with oxyanions such as NO₃^-.

REFERENCES

ANMAT (2021) Código Alimentario Argentino. www.argentina.gob.ar/sites/ default/files/anmat_ caa_capitulo_xiii.pdf

Azaro, M. Flores, F. M. Casella, M. Peroni, B., Rodríguez, C., Torres Sánchez, R. M. and Jaworski M. (2021) Synthesis and characterization of adsorbents for the elimination of nitrates and bromates from water aiming to develop a continuous oxyanion water elimination system. WaterSupply. 21, 1243–1252.

Ding, Y., Suna, W., Yang, W. and Li, Q. (2017) Formic acid as the in-situ hydrogen source for catalytic reduction of nitrate in water by PdAg alloy nanoparticles supported on amine-functionalized SiO₂. Appl. Catal. B-Environ. 203, 372–380.

Fan, A.M. (2019) Health in Exposure and Regulatory Implications of Nitrate and Nitrite in Drinking Water. Reference Module in Earth Systems and Environmental Sciences Encyclopedia of Environmental Health, (2nd ed) Elsevier. 417−435.

Gammoudi, S., Frini-Srasra, N. and Srasra, E. (2013) Preparation, characterization of organosmectites and fluoride ion removal. Int. J. Miner. Process. 125, 10–17.

Jaworski, M., Lick, I. Siri, G. and Casella, M. (2014) ZrO2-modified Al2O3-supported PdCu catalysts for the water denitrification reaction. Appl. Catal. B: Environmental 156–157, 53-61.

Jaworski, M., Barbero, B. and Casella M. (2019) Removal of Nitrate from Drinking Water by Using PdCu Structured Catalysts Based On Cordierite Monoliths. 36, 705-715.

Kim,Y-N., Kim, M.Y. and Choi, M. (2016) Synergistic integration of catalysis and ion-exchange for highly selective reduction of nitrate into N₂. Chem. Eng. J. 289, 423–432.

Landi, G., Barbato, P.S., Di Benedetto, A. and Lisi, L. (2016) Optimization of the preparation method of CuO/CeO2 structured catalytic monolith for CO preferential oxidation in H2-rich streamsAppl. Catal. B-Environ. 181, 727–737.

Wu, Y., Wang, Y. J., Wang, S., Xu, L., Yu, P., Corvini, W. and Thomas, J. (2016) Nitrate removal from water by new polymeric adsorbent modified with amino and quaternary ammonium groups: Batch and column adsorption study. Taiwan Inst. Chem. Eng. 66, 191–199.

Xi, Y., Mallavarapu, M. and Naidu, R. (2010) Preparation, Characterization of Surfactants Modified Clay Minerals and Nitrate Adsorption. Appl. Clay Sci. 48, 92–96.

Xu, J., Qi, Y. P. W-K., Yang, X., Tang, Y., Wa, P. and Fisher, A. (2017) Evaluation of South African Sand/Zero-Valent Iron Combinations for the Treatment of Nitrate-Contaminated Water: Kinetic and Effect of Competitive Ions. Chemosphere. 166, 197-202.

Yuranova,T., Franch, C., Palomares, A.E., Garcia-Bordejé, E. and Kiwi-Minsker, L. (2012) Structured fibrous carbon-based catalysts for continuous nitrate removal from natural water. Appl. Catal. B-Environ. 123-124, 221-228.

Zhai, Y., Zhao, X., Teng, Y., Li, X., Zhang, J., Wu, J. and Zuo, R. (2017) Groundwater nitrate pollution and human health risk assessment by using HHRA model in an agricultural area, NE China. Ecotox. Environ. Safe. 137, 130–142.

Zoppas, F.M., Marchesini, F.A., Devard, A., Bernardes, A.M. and Miró, E.E. (2016) Controlled deposition of Pd and In on carbon fibers by sequential electroless plating for the catalytic reduction of nitrate in water.  Cat. Commun. 78, 59-63.

SUPLEMENTARY MATERIAL

Calculation of the Reynolds number to determine the flow inside the monoliths

where

For calculations

With

Data:

This result indicates that there is laminar flow in the monolith channels.

GRAPH OF PARTICLE DISTRIBUTION OF γ-ALUMINA GROUND

Received: September 5, 2022

Sent to Subject Editor: September 5, 2022

Accepted: October 3, 2022

Recommended by Subject Editor Laura Briand