REMOVAL OF NITRATE FROM DRINKING WATER BY USING PdCu STRUCTURED CATALYSTS BASED ON CORDIERITE MONOLITHS

Abstract

Structured catalysts were prepared, characterized and evaluated in NO₃ ^- removal from drinking water. Different suspensions containing a previously optimized PdCu/5wt% ZrO₂-Al₂O₃ powder catalyst (hereinafter PdCu/5ZA_p) were prepared and deposited on cordierite monoliths by washcoating. The effect of suspension concentration, the particle size, the immersion number, the use of suspension stabilizer agent, and an alumina pre-coating on the coating adherence and catalytic performance were studied. All the prepared structured catalysts were active for the elimination of NO₃ ^- and presented good selectivity to N₂ (> 93%) in synthetic water samples. The catalyst performance was related to the amount of deposited catalyst. The highest activity and the best coating adherence were obtained with the structured catalyst prepared by a single immersion in a 14 wt% concentration suspension, which was obtained from the powder catalyst milled for 10 h and stabilized with colloidal alumina. This catalyst also showed good activity when it was reused for four cycles and when it was evaluated with real water samples.

Keywords:
Water pollution; Nitrate; Cordierite monoliths; PdCu catalyst; Structured catalyst

INTRODUCTION

Industrial and agricultural activities lead to the discharge of pesticides, fertilizers and organic wastes of animals and plants that contaminate the water in a diffuse but very remarkable way, in particular, groundwater. One of the most serious problems is the high levels of nitrate (NO₃ ^-) found in some groundwater reservoirs (Faridullah et al., 2017 ; Xu et al., 2017). When NO₃ ^- levels in soil are greater than the ones plants can assimilate, irrigation and rainwater cause NO₃ ^- to percolate and reach the groundwater. This fact affects the quality of life and health of a significant number of people, especially in areas dominated by agricultural activities (Zhai et al., 2017). Drinking water with excess NO₃ ^- ion can be harmful because NO₃ ^- is easily reduced to nitrite (NO₂ ^-) in the mouth and intestines, causing methemoglobinemia, a conversion of hemoglobin to methemoglobin that depletes oxygen levels, particularly in children under 6 months (blue-baby syndrome). In addition, NO₃ ^- is a precursor of carcinogenic nitrosamines (Zhai et al., 2017). The maximum contaminant level permitted by the US Environmental Protection Agency (EPA) is 10 ppm for N-NO₃ ^- (expressed in terms of nitrogen N in NO₃ ^-) in the water supply for the population (water.epa.gov/drink/contaminants/basicinformation/nitrate.cfm) .While for the World Health Organization (WHO) it is 50 ppm for NO₃ ^- and 1 ppm for NO₂ ^- (www.who.int/water_sanitation_health/dwq/chemicals/nitratenitrite2ndadd.pdf).

In Argentina, the Argentine Food Code establishes a maximum concentration of 45 ppm for NO₃ ^- and 0.1 ppm for NO₂ ^-(www.anmat.gov.ar/alimentos/normativas_alimentos_caa.asp).

There are different treatments for NO₃ ^- removal from water such as electrodialysis, distillation, ion exchange, reverse osmosis, catalytic or electrochemical reduction, biological degradation, among others (Li et al., 2017; Huo et al., 2017). For environmental reasons, the best technique to remove NO₃ ^- should convert it to gaseous N₂. One of the most promising processes is the reduction of NO₃ ^- to N₂ using heterogeneous catalysts in the presence of H₂ as a reducing agent (Huo et al., 2017; Mirabi et al., 2017). These catalysts generally contain a noble metal (Pd, Rh, Ru or Pt) and a promoter metal (Cu, Ag, Fe, Hg, Ni, Cu, Zn, Sn or In) (Ding et al., 2017; Zoppas et al., 2016; Kim et al., 2016; Martínez et al., 2017). In these catalysts, the bimetallic sites permit the reduction of NO₃ ^- to NO₂ ^-, which is then reduced to N₂ or over-reduced to NH₄ ⁺ over the monometallic sites. This latter product is an inconvenient aspect of these systems (Hamid et al., 2018; Martínez et al., 2017).

The powder catalysts containing PdCu or RhMo as active phase supported on Al₂O₃ or ZrO₂, both pure and mixtures of these supports, were previously evaluated in our group. These catalysts showed good catalytic performance in a batch reactor (Jaworski, et al., 2014; 2018). Particularly, it was found that the PdCu catalyst supported on ZrO₂-Al₂O₃ showed good catalytic activity and N₂ selectivity (Jaworski, et al., 2014). Despite the good activity and selectivity results obtained with these catalysts in a batch reactor, the use of a catalyst in a powder form is not technologically viable. The use of these catalysts could cause problems in their recovery, leading to the contamination of the treated water by metal particles. To avoid the loss of catalytic material, the use of pellets or extruded supports is proposed. A disadvantage of the use of pellets is that the diffusion of reagents and/or products into the pores of the catalysts would decrease the selectivity to N₂ (Durkin et al., 2018; Zoppas et al., 2018).

The use of structured catalysts such as monoliths, foams and grids could overcome the limitations caused by the use of powder formulations (Durkin et al., 2018; Li et al. 2019; Muñoz-Murillo et al., 2018). Structured catalysts based on monoliths with different chemical composition are widely used, especially in environmental catalytic applications. For example, Ni-based catalysts (Ni-CeO₂) prepared on ceramic monoliths have been obtained by dip-coating the support in an acid-free stable catalyst dispersion. The structured catalysts obtained were evaluated towards the biogas oxy-steam reforming reaction. The use of structured system instead of packed bed reactors allows one to obtain stable performances, in term of CH₄ (97%) and CO₂ (87%) conversion (Balzarotti et al., 2018).

Some of the structured catalysts used are supported on ceramic monoliths. These monoliths contain parallel channels of an inert oxide such as cordierite (2MgO₂.Al₂O₃.5SiO₂) and are coated with a thin layer of the catalyst (Özyönüm et al., 2016; Şem et al., 2016). Cordierite has numerous important applications due to its low cost, its excellent properties, such as low coefficient of thermal expansion, good thermal resistance, low dielectric constant, high chemical inertness, refractoriness and adequate mechanical properties (Labhsetwar et al., 2012; Yuan et al., 2016).

A series of copper/ceria washcoated cordierite monoliths was prepared and tested in the carbon monoxide preferential oxidation reaction. These catalysts were active in the CO-PROX reaction and CO-PROX stage, and could be used in the case of non-stationary applications (Landi et al., 2016). Volatile organic compounds (VOCs) that are air pollutants released from many industrial operations were eliminated using a honeycomb monolith washcoated with Mn-M mixed oxide (M= Cu, Ni or Co) (Zhang and Wu, 2016) or by using Mn, Ce and Ce/Mn supported ceramic monoliths with natural clays as raw materials (Colman-Lerner et al., 2016). Catalysts of LaCoO₃/ɤ-Al₂O₃/cordierite monolith, LaCoO₃/SiO₂/cordierite monolith and LaCoO₃/TiO₂/cordierite monolith were used to reduce diesel particulate emissions that cause serious problems in human health (Tang et al., 2017).

Considering the advantages of the structured catalysts, in this work the preparation of cordierite monoliths coated with a powder catalyst previously optimized by our group is reported. The powder catalyst is PdCu supported on Al₂O₃ modified with 5 wt% ZrO₂ (PdCu/5ZA_P), which showed good catalytic performance in the NO₃ ^- removal reaction (Jaworski, et al., 2014).

It has been shown that the adherence of the active phase to the monolith is essential in the development of structured catalysts (Ashraf et al., 2018; Sedjame et al., 2018). Different factors affect this adherence, such as the characteristics of the suspension containing the active phase, the particle size of the solid to be deposited on the monolith, the nature of the dispersion medium, the concentration of solids and pH of the suspension (Landi et al., 2016). For this reason, different suspensions of PdCu/5ZA_P were used for the washcoating of cordierite monoliths. The effects of the particle size, the solid concentration in the suspension, the addition of a suspension stabilizer, the number of immersions and the effect of a precoating on the adherence were studied. Furthermore, the catalytic performance for nitrate reduction in water was evaluated.

MATERIALS AND METHODS

Preparation of PdCu/5ZA catalyst

The synthesis and characterization of the PdCu/5ZA_P catalyst employed in this work were previously described (Jaworski et al., 2014). Briefly, in order to prepare the Al₂O₃ support modified with 5 wt% ZrO₂, ZrOCl₂.6H₂O (Fluka) was added to the γ-Al₂O₃ support (γ-Al₂O₃ Air Products, surface area 190 m²g⁻¹; pore volume 0.50 m³g⁻¹) suspended in distilled water. NH₄OH was added drop by drop under constant stirring until a gel formed. This system was aged for 8 days. The solid obtained was washed until no chloride ion was determined by AgNO₃ solution, dried at 105°C and calcined at 400ºC for 2 h. The support obtained was called 5ZA. The Pd-based catalyst (1 wt%) was prepared by impregnating the 5ZA support with a solution of H₂PdCl₄ prepared from PdCl₂ (Sigma-Aldrich) in HCl (pH=1). The solid obtained was dried at 105ºC and calcined in air at 400°C. After that, this monometallic catalyst was impregnated with a solution of Cu(NO₃)₂ (Merck), so as to obtain 0.3 wt% Cu on the support. The bimetallic catalyst was first dried at 105ºC and then calcined at 400°C.

Preparation of cordierite monoliths

The structured support used was ceramic cordierite monoliths (Corning). They were sectioned in a suitable size (length: 3 cm, cross section: 0.785 cm²), placed in an ultrasonic water bath to remove possible impurities, dried, weighed and identified. The scheme of the monoliths used is shown in Figure 1.

Figure 1
Scheme of the monoliths used.

The catalyst was incorporated into the monolithic structure using the washcoating method. This method consists of immersing the monolith in a suspension of the finely milled catalyst. In order to prepare a stable suspension, it was necessary to determine the zeta potential (pZ) of the PdCu/5ZA_P catalyst. For this determination, the PdCu/5ZA_P catalyst was milled to a particle size between 120 and 140 mesh. The determination of the pZ was carried out with solutions of KCl, KOH and HCl using Zeta-Meter System 3.0 equipment.

For the preparation of the PdCu/5ZA_P suspension, the catalyst was finely milled for 5 or 10 h in a planetary ball mill (Fritsch Pulverisette 6). The particle size of the PdCu/5ZrAl_P catalyst after milling was determined using a Fritsch A-20 photosedimentograph and 2% Triton X-10 as surfactant.

Preparation of the structured catalysts

In order to obtain a stable coating of the PdCu/5ZA_P catalyst on the monolith structure, different preparation strategies were evaluated. For example, the effect of the milling time for the PdCu/5ZA_P catalyst (5 or 10 h), the catalyst concentration in the suspension (14 or 20 wt%), the stabilizer agent (colloidal alumina Nyacol® Al 20 or polyvinyl alcohol (PVA)) and the precoating of monoliths with colloidal alumina Nyacol® Al 20 as primer were studied.

To prepare the structured catalyst, an aqueous suspension of the finely milled PdCu/5ZA_P catalyst was prepared and the pH was adjusted to 9.5 with a concentrated NaOH solution. The suspension obtained was placed in an ultrasound device for 30 min and its stability was observed. The cordierite monoliths were immersed in this suspension for 1 min. The excess suspension was eliminated by centrifugation at 650 rpm for 1 min. Finally, the monoliths containing the deposited catalyst were dried at 105ºC, calcined at 400°C for 2 h and weighed.

The monolithic catalysts prepared are listed in Table 1, where N or P indicates the Nyacol® Al 20 or PVA stabilizer, respectively, 14 or 20 is the weight percentage of the catalyst in the suspension, 5 or 10 indicates the powder catalyst milling time, Pr indicates the precoating with colloidal alumina Nyacol® Al 20 as primer, and (1) or (2) is the number of monolith dips in the suspension. Four pieces were prepared for each condition to verify the reproducibility of the preparation method.

Table 1
Nomenclature and methodologies employed to deposit the PdCu/5ZA_P catalyst on monoliths.

Monolith	Milling time (h)	Pre-coating	Stabilizer	Suspension concentration (wt.%)	Number of immersions
N20-5(1)	5	No	Nyacol	20	1
N14-5(1)	5	No	Nyacol	14	1
N14-5(2)	5	No	Nyacol	14	2
N14-10(1)	10	No	Nyacol	14	1
N14-10 Pr(1)	10	Yes	Nyacol	14	1
P14-10(1)	10	No	PVA	14	1
P14-10 Pr(1)	10	Yes	PVA	14	1

Adherence test

In order to evaluate the coating adherence of the PdCu/5ZA_P catalyst on the monoliths, an adaptation of a previously reported method (Barbero, et al., 2008) was used. The structured catalysts were immersed in distilled water inside a sealed beaker and subjected to ultrasonic agitation at approximately 30 °C for 30 min. Then, the structured catalysts were dried at 105 °C for 6 h and calcined at 400 °C for 2 h. These conditions are identical to that employed during the washcoating of PdCu/5ZA_P catalyst. Therefore, the weight difference before and after the ultrasound treatment allows one to determine the coating adherence as following:

$$\text{Adherence}\left(\text{\%}\right)=\left[1-\frac{\left({\text{W}}_{\text{bu}}-{\text{W}}_{\text{au}}\right)}{{\text{W}}_{\text{cat}}}\right]\times 10$$ (1)

where W_bu is the weight before ultrasonic treatment, W_au the weight after ultrasonic treatment, and W_ap the weight of PdCu/5ZA_P catalyst retained by washcoating.

Characterization of PdCu/5ZA structured catalysts

Monolithic catalysts were characterized by SEM-EDX microscopy. SEM-EDX measurements were carried out using a FEI Quanta 200 scanning electron microscope equipped with an energy dispersive X-ray spectroscopy facility (EDX SDD Apollo 40). In order to draw conclusions about the distribution of the components in the samples, backscattered electron (BSE) images were taken.

The temperature-programmed reduction (TPR) experiments were carried out in a conventional flow system using a reactive gas stream (5% H₂ balanced with Ar) flowing at 25 mL min⁻¹ over 50 mg of sample. The temperature was raised from room temperature to 800 °C at 10 °C min⁻¹. Hydrogen uptake during the reduction was analysed on-line by a Shimadzu GC-8A gas chromatograph equipped with a thermal conductivity detector (TCD).

The textural properties of the supports were measured by N₂ adsorption-desorption at −196 °C using a Micromeritics Accusorb 2100E apparatus.

Analysis of the catalytic properties

A tubular glass reactor (internal diameter: 1.1 cm; length: 3.1 cm) containing the structured catalysts (Figure 2) was used to study NO₃ ^- removal from water. Fifty mL of degassed aqueous solution containing 100 ppm of NO₃ ^- was loaded into the system and continuously bubbled with H₂. The solution was recirculated with a peristaltic pump at a flow rate of 5 mL/min. The reaction was carried out for 20 h at room temperature and atmospheric pressure.

Figure 2
Schematic diagram of the experimental equipment.

Samples were periodically taken to determine NO₃ ^- and NO₂ ^- concentration with an ion chromatograph (Metrohm 790 Personal IC). NH₄ ⁺ was determined at the end of the reaction by a colorimetric method using a UV-Vis spectrophotometer (UV-Vis Thermo Spectronic Helios Gamma), following a modified Berthelot method (Marchesini et al., 2010). The conversion values of NO₃ ^- (X_NO3-%) as well as selectivities towards NO₂ ^- (S_NO2-%), NH₄ ⁺ (S_NH4+%) and N₂ (S_N2%) after 20 h of reaction for all the catalysts studied were calculated according to:

$$X_{N{O}_3^{-}}=\frac{{\left[N{O}_3^{-}\right]}_{initial}-{\left[N{O}_3^{-}\right]}_t}{{\left[N{O}_3^{-}\right]}_{initial}}\times 100\%$$ (2)

$$S_{N{O}_2^{-}}={\frac{{\left[N{O}_2^{-}\right]}_t}{{\left[N{O}_3^{-}\right]}_{initial}-\left[N{O}_3^{-}\right]}}_t\times 100\%$$ (3)

$$S_{N{H}_4^{+}}={\frac{{\left[N{H}_4^{+}\right]}_t}{{\left[N{O}_3^{-}\right]}_{initial}-\left[N{O}_3^{-}\right]}}_t\times 100\%$$ (4)

$$S_{N_2}=100-S_{N{O}_2^{-}}(\%)-S_{N{H}_4^{+}}(\%)$$ (5)

RESULTS

Characterization of the structured catalysts

The uncoated monolith and the structured catalysts were analysed by SEM-EDX microscopy. A representative SEM-EDX analysis of a portion of the catalyst N14-10(1) is shown in Figure 3. The arrows indicate a progressive increase of zoom for the images. The circles indicate the points where the elements were analysed by EDX.

Figure 3
SEM image and EDX analysis for the structured catalyst N14-10(1).

The areas of the monolith covered with catalyst were clearly distinguished in all the analysed monoliths. The EDX analysis of the uncovered areas showed the presence of O, Mg, Al and Si, representative of cordierite, while the areas with the catalyst presented O, Al and Zr, which indicates the effective coating of the PdCu/5ZA_P catalyst on the monolith surface. As expected, it was not possible to differentiate the metals Pd and Cu because they are in concentrations below the detection limit of EDX.

To determine if the catalyst particles were modified during the washcoating process, the textural properties (Figure 4) and the TPR profile (Figure 5) of the original PdCu/5ZA_P catalyst were compared with those of the solid recovered by drying of the suspension used to prepare monolith N14-10(1) (hereinafter called N14-10(1)_P).

Figure 4
N₂ adsorption/desorption isotherms at for a) PdCu/5ZA_P, b) N14-10(1)_P.

Figure 5
TPR profile of the catalyst PdCu/5ZA_P and N14-10(1)_P .

The N₂ adsorption/desorption isotherms for the powder catalyst PdCu/5ZA_P and N14-10(1)_P are shown in Figure 4 a) and b), respectively. The catalytic materials analysed have similar isotherms corresponding to a type IV isotherm according to the IUPAC classification, characteristic of mesoporous solids (Muttakin et al., 2018).

The TPR profile for the catalyst PdCu/5ZA_P was described in a previous paper (Jaworski, et al., 2014) and was similar for the solid N14-10(1)_P, Figure 5. In both profiles it is possible to distinguish two different peaks: a main peak centred around 150ºC and a small peak which extends between 300 and 400ºC. The first peak is assigned to the reduction of PdO to Pd⁰, and the reduction of copper oxides promoted by the presence of the noble metal Pd (Resende et al., 2018, Zhuang, et al., 2010).The Pd crystals favor the spillover of H₂ that causes the Cu oxides to be reduced at lower temperature when compared to a monometallic Cu catalyst (300-400°C) (Lei et al. 2018; Yuan et al., 2017). The second peak corresponds to the reduction of CuO to Cuº (Lei et al., 2018).

The similarity of the TPR profiles for the catalysts PdCu/5ZA_P and N14-10(1)_P show that the Pd-Cu interaction is not modified when the catalyst is finely milled and used for preparing the suspension.

Determination of the isoelectric point

As previously mentioned, it was necessary to determine the pZ of the PdCu/5ZA_P catalyst prior to preparing the suspensions. In Figure 6 it can be seen that the isoelectric point for the PdCu/5ZA_P catalyst was approximately 4.5. Then, the pH selected to prepare the PdCu/5ZA_P catalyst suspension was 9.5, which is far enough away from pZ. In this way, the agglomeration of particles was avoided.

Figure 6
Zeta potential as a function of pH for the PdCu/5ZA_P catalyst.

Determination of particle size

Figure 7 shows the curves corresponding to the determination of the particle size distribution after the milling process (5 h and 10 h).

Figure 7
Determination of the particle size after the milling process. Left, 5 h milling. Right, 10 h milling.

The continuous line with square symbols represents the particle size distribution, while the dashed line with triangle symbols represents the accumulative distribution. From these curves it is possible to read the d₅₀ or d₉₀ (50% or 90% of the particles of the suspension are smaller than that value, respectively). For the milling time of 5 h, d₉₀ of 25.5 μm was obtained. For the10 h milling time, the d₅₀ was 7 μm and the d₉₀ 13.5 μm approximately. Evidently, the particle sizes for 5 h milling time are larger than for 10 h milling time.

Results of coated monoliths

Table 2 presents the catalyst mass gained by washcoating and the coating adherence for all prepared monoliths. The second column shows the weight increase observed after washcoating of the cordierite monolith, which correspond to the content of PdCu/5ZA_P catalyst on the cordierite support expressed as weight percentage. The values reported are an average of the four pieces prepared. The third column shows the percentage of PdCu/5ZA catalyst retained after the adherence test with respect to the catalyst retained by the washcoating process. The net content of catalyst after the adherence test is shown in the last column. These results are discussed below.

Table 2
Weight percentage of PdCu/5ZA_P catalysts retained after the washcoating (second column), coating adherence (third column) and net content of PdCu/5ZA_P catalyst after the adherence test (fourth column).

Monolith	Weight increase (wt%)	Adherence (%)	Catalyst net content (wt%)
N20-5(1)	4.79±0.03	48.57	2.33
N14-5(1)	4.35±0.02	66.67	2.90
N14-5(2)	8.45±0.01	40.00	3.38
N14-10(1)	3.33±0.02	90.00	3.00
N14-10 Pr(1)	4.55±0.02	61.67	2.80
P14-10(1)	2.05±0.01	83.33	1.71
P14-10 Pr(1)	2.88±0.02	75.00	2.16

Effect of suspension concentration

The structured catalyst prepared using the 20 wt% suspension (N20-5(1)) showed a non-uniform coverage and channels blocked by the active phase.

To avoid these problems and to obtain a homogeneous layer of the catalyst on the monolith, a more dilute suspension, 14 wt%, was used (monolith N14-5(1)). The monoliths were coated with this suspension and Table 2 (second column) shows that the mass gain of catalyst is smaller for N20-5(1) than for N14-5(1) monolith. A better adherence (third column) is achieved when the 14 wt% suspension in used. In this way, higher catalyst net content is obtained after the adherence test (fourth column).

Effect of the number of immersions

To increase the amount of catalyst in the monoliths, two immersions were made with the 14 wt% suspension, obtaining N14-5(2) monolith. From Table 2 it is noted that, with two immersions, about the double of catalyst was retained on the monolith (second column). However, a significant amount of catalyst was not retained after the adherence test for this monolith. As was previously reported, it is possible that, with a greater number of immersions, an unstable coating will be formed on the monolith (Bosko et al., 2014). The attraction forces between the first and the second layers deposited on the monolith are weak. Therefore, the coating adherence decreases for subsequent immersions.

Effect of catalyst particle size

In order to analyse the effect of the catalyst particle size, the PdCu/ZA_P catalyst was milled for 5 or 10 h (Figure 7) and two monoliths were prepared (monolith N14-5(1) and N14-10(1)). In Table 2, it can be seen that the milling time affected the final adherence of the catalyst. The monolith coated with the powder milled for 10 h was more stable than that coated with the powder milled for 5 h. This implies that the small particle size favours the adherence of the coating.

Effect of the stabilizer agent

Alternatively to the use of colloidal alumina Nyacol® Al 20 as the stabilizer agent of the suspension, PVA was used and the monolith P14-10(1) was prepared. In comparison with the monolith N14-10(1), the monolith prepared with PVA retained less catalyst amount and presented lower coating adherence (Table 2).

Effect of the use of a primer

The adherence of the catalyst to the monolith is a crucial factor. For this reason, to improve adherence, usually the monolith is firstly covered with a layer of alumina or silica oxides. The function of these oxides is to provide a bond between the catalyst particles and the monolith, without affecting the catalytic performance (Yuan et al., 2016). In this work, the cordierite monoliths were immersed in a suspension of colloidal alumina Nyacol® Al20 for 1 min. The excess of the suspension was removed by centrifugation at 650 rpm for 1 min. The monoliths containing the primer were dried, calcined at 500 °C for 2 h and weighed. The PdCu/5ZA_P powder catalyst (14 wt% suspension) milled for 10 h was then deposited on these monoliths. Two suspensions were made using the stabilizers previously evaluated, colloidal alumina or PVA, and the monoliths obtained were N14-10 Pr(1) and P14-10 Pr(1).

By comparing the monoliths without primer (N14-10(1) and P14-10(1)) with the monoliths with primer (N14-10 Pr(1) and P14-10 Pr(1)), it can be seen that the use of a primer was not favourable (Table 2). The amount of catalyst retained was higher compared to the same monoliths containing no primer (Table 2, second column) but the adherence of these coatings was lower (Table 2, third column). It was suggested that the presence of a primer could cause the loss of the surface roughness of the monolithic support and, for that reason, the adherence decreases (Yuan et al., 2016).

Catalytic reaction

Before testing the monolithic catalysts in the NO₃ ^- removal reaction in water, blank tests were performed. The first one consisted of the use of 50 mL of a 100 ppm NO₃ ^- solution in the presence of H₂ bubbling but in the absence of catalyst. The second one consisted in the use of the structured support covered by powder Al₂O₃ or 5ZA supports in the presence of 100 ppm of NO₃ ^- and H₂. In both cases, it was found that NO₃ ^- removal was null.

All PdCu/5ZA monolithic catalysts were evaluated in the NO₃ ^- removal reaction. The PdCu/5ZA_P powder catalyst was included as a reference. The results are shown in Table 3.

Table 3
Conversion of NO₃^- and concentration of NO₂^- and NH₄⁺ for the monoliths prepared after 20 hours of reaction.

Monolith	XNO3- (%)	ppm NO2-	ppm NH4+	SNO2- (%)	SNH4+ (%)	SN2 (%)	XNO3-/gcat *
N20-5(1)	45	0.8	2	1.78	4.44	93.78	13.24
N14-5(1)	75	0.55	0.4	0.73	0.53	98.73	18.75
N14-5(2)	78	0.75	0.6	0.96	0.77	98.27	16.25
N14-10(1)	83	0.4	0.55	0.48	0.66	98.86	18.44
N14-10 Pr(1)	61	1.2	0.7	1.97	1.15	96.89	16.49
P14-10(1)	42	1.1	0.4	2.62	0.95	96.43	16.80
P14-10 Pr(1)	53	0.9	0.6	1.70	1.13	97.17	18.93
PdCu/5ZA poder	100	0.5	0.3	0.50	0.30	99.20	20.00

• *Nitrate specific conversion calculated as nitrate conversion per catalyst gram.

This study showed that the structured catalysts prepared were active in the removal of NO₃ ^-. A maximum conversion of 83% after 20 h of reaction was found for the monolith N14-10(1). This monolith also had the best N₂ selectivity and the highest nitrate specific conversion (calculated as (nitrate conversion)/(catalyst mass) ratio, Table 3, eighth column). For all the monoliths studied, the selectivity to N₂ was higher than 96%, except for the monolith N20-5(1). These results make these systems promising candidates to be used as catalysts for NO₃ ^- removal from water, although some operation conditions could still be improved.

In order to evaluate the existence of diffusional effects on NO₃ ^- removal, another test was carried out using an aliquot of the PdCu/5ZA_P powder catalyst obtained from the suspension used to prepare the catalyst N14-10(1). The reaction was carried out in a semibatch system using 45 mg of this powder, which is equal to the catalyst mass retained on the monolith N14-10(1). The reaction conditions were the same as in the previous tests. The final concentrations detected for NO₃ ^-, NO₂ ^- and NH₄ ⁺ were 0 ppm, 0.5 ppm and 0.3 ppm, respectively. The difference in activity between the powder and the monolith N14-10(1) could be assigned to diffusion problems of the reactants and/or products in the structured system.

In summary, from the results shown in Tables 2 and 3, the preparation of structured catalysts using the catalyst PdCu/5ZA_P milled for 10 h in the absence of a primer is favourable since the best results in terms of catalyst retention and catalytic activity were obtained using the N14-10(1) monolith.

As previously mentioned, in the blank tests using the cordierite monolith covered with the supports Al₂O₃ or 5ZA, NO₃ ^- removal was null. Since the support itself is not active, the NO₃ ^- removal is attributed to the PdCu active phase. In this sense, the same reaction mechanism described by Jaworski et al. (2014) in a previous work is postulated. In this mechanism, the NO₃ ^- ions are adsorbed on the oxygen vacancy sites (Lewis acid sites) generated by the presence of ZrO₂ in the support.

In order to evaluate the potential reuse of the monolith N14-10(1), after reaction, the monolith was washed with distilled water, dried, calcined at 400°C for 1 h and reduced at 400°C for 1 h. The reaction was carried out using the original NO₃ ^- solution (100 ppm of NO₃ ^-). This process was repeated for each cycle. The results obtained are listed in Table 4.

Table 4
Conversion of NO₃^- and concentration of NO₂^- and NH₄⁺ after re-use of the N14-10(1) monolith.

Cycle	XNO3-%	ppm NO2-	ppm NH4+	SNO2- (%)	SNH4+ (%)	SN2 (%)
1°	83	0.4	0.55	0.48	0.66	98.86
2°	75	0.9	0.3	1.20	0.40	98.40
3°	68	0.75	0.3	1.10	0.44	98.46
4°	61	0.4	0.25	0.66	0.41	98.93

The results from Table 4 show that there is a certain deactivation in the removal of NO₃ ^- after each reuse, while the selectivity to NO₂ ^- and NH₄ ⁺ is maintained. The mass of the catalyst remains constant before and after each use. Thus, the decrease in activity for NO₃ ^- removal could be assigned to the loss of some of the active sites.

For these catalysts to have a practical application, it is necessary to evaluate them in water intended for human consumption. In this sense, the monolith with the best catalytic performance, N14-10(1), was evaluated using water samples taken from the Puelche Aquifer. Two samples of groundwater from this aquifer were extracted and analysed, and were designated C and P, as shown in Table 5. The nitrate content of these samples was higher than that allowed by current regulations. The Puelche Aquifer is one of the most exploited water reservoirs in Argentina, since it supplies drinking water to the Buenos Aires Metropolitan Region (Greater Buenos Aires), one of the most densely populated areas of Argentina. The Puelche Aquifer has low salinity (approximately 585 mg/L), being of the sodium bicarbonate type. Its potability is only affected by the NO₃ ^- content, especially in urbanized areas (Armengol et al., 2017; Zabala et al. 2016).

Table 5
Elimination of NO₃^- from water for human consumption after 20 h of reaction using N14-10(1) monolith*.

Sample	ppmi(1) NO3-	XNO3-%	ppmf(2) NO2-	ppmf(2) NH4+	SNO2- (%)	SNH4+ (%)	SN2%
C	63	49	3.3	0.8	10.64	2.58	86.78
P	58	48	2.9	0.65	10.71	2.32	86.97

• * For experimental conditions, see the text. (1) ppm_i: initial concentration (mg/L). (2) ppm_f: final concentration (mg/L).

The denitrification reaction was carried out for 20 h using the N14-10(1) monolith and H₂ as reducing agent, following the same procedure described in the experimental section for the synthetic samples. The results obtained are presented in Table 5.

The evaluated N14-10(1) monolith was active in the removal of NO₃ ^- from real water samples, although a high production of NO₂ ^- and NH₄ ⁺ was observed, compared to the results obtained when analysing synthetic samples (Table 3). It is suggested that the decrease in the activity and the selectivity towards N₂ is due to the masking of the active phase caused by the rest of the ions present in the real water samples, especially to the abundant presence of bicarbonate ions present in the Puelche Aquifer. It is well-established that bicarbonate ions present similar chemical properties to nitrate ions, and thus it can be expected that both anions would be adsorbed on the same catalytic sites. Therefore, the catalytic surface available for nitrate ions will be lower, and so will be the activity (Mendow et al., 2017).

CONCLUSIONS

Unsupported catalysts are used in numerous industrial processes; however, most of the catalysts used in environmental processes must be supported. In this work structured catalysts were satisfactorily prepared and evaluated in the NO₃ ^- removal reaction in water in the presence of H₂ as a reducing agent. The powder catalyst PdCu/5ZA_P was deposited on cordierite monoliths by washcoating. The influence of different variables in the preparation of the structured catalysts was analysed: suspension concentration, particle size, number of immersions, nature of the stabilizer and the use of primer. It was found that the use of a 14 wt% catalyst suspension is more favourable than a higher suspension concentration (20 wt%) in order to prevent blocking of the monolith channels. A milling time of 10 h (reduction of the particle size of the powder catalyst) improved the amount of catalyst deposited and the adherence of the coating.

The use of colloidal alumina Nyacol as the stabilizer agent was more favourable than the use of PVA to improve the adherence of the catalyst to the monolith. One second immersion increased the retained catalyst mass, but the adherence was low.

The monolith N14-10(1) showed the best catalytic performance among all structured catalysts tested. Though there would be diffusional limitations in the catalytic coating, it should be noted that the selectivity to nitrogen was higher than 97% and this value remained after the reuse of the catalyst, making it a very promising catalytic system.

This structured catalyst eliminated 50% of the NO₃ ^- present in samples of water for human consumption, but high NO₂ ^- and NH₄ ⁺ generation was detected. Probably the ions present in real water masked the NO₃ ^- adsorption sites.

ACKNOWLEDGEMENTS

We are grateful to Ms Mariela Theiller for her contribution and technical support. The authors would like to thank the following institutions for funding this work: CONICET (PIP 0276 and 1178), UNLP (Project X633 and X700, Subsidio Jóvenes Investigadores, Subsidio de Viajes) and ANCYT (PICT 2016/1391).

REFERENCES

• Armengol, S., Manzano, M., Bea, S., Martínez, S. Identifying and quantifying geochemical and mixing processes in the Matanza-Riachuelo Aquifer System, Argentina. Science of the Total Environment, 599-600, 1417-1432 (2017). https://doi.org/10.1016/j.scitotenv.2017.05.046
  » https://doi.org/10.1016/j.scitotenv.2017.05.046
• Ashraf, A., Sanz, O., Montes, M., Specchia, S. Insights into the effect of catalyst loading on methane steam reforming and controlling regime for metallic catalytic monoliths. International Journal Of Hidrogen Energy, 43, 11778-11792 (2018). https://doi.org/10.1016/j.ijhydene.2018.04.126
  » https://doi.org/10.1016/j.ijhydene.2018.04.126
• Barbero, B. P., Costa-Almeida, L., Sanz, O., Morales, M. R., Cadus, L. E., Montes, M. Washcoating of metallic monoliths with a MnCu catalyst for catalytic combustion of volatile organic compounds. Chemical Engineering Journal, 139, 430-435 (2008). https://doi.org/10.1016/j.cej.2007.12.033
  » https://doi.org/10.1016/j.cej.2007.12.033
• Colman-Lerner, E., Peluso, M., Sambeth, J., Thomas, H. Cerium, manganese and cerium/manganese ceramic monolithic catalysts. Study of VOCs and PM removal. Journal of Rare Earths, 34, 675-682 (2016). https://doi.org/10.1016/S1002-0721(16)60078-9
  » https://doi.org/10.1016/S1002-0721(16)60078-9
• Ding, Y., Sun, W., Yang, W., Li, Q. Formic acid as the in-situ hydrogen source for catalytic reduction of nitrate in water by PdAg alloy nanoparticles supported on amine functionalized SiO₂ Applied Catalysis B: Environmental, 203, 372-380 (2017). https://doi.org/10.1016/j.apcatb.2016.10.048
  » https://doi.org/10.1016/j.apcatb.2016.10.048
• Durkin, D., Ye, T., Choi, J., Livi, K., De Long, H., Trulove, P., Fairbrother, D., Haverhals, L., Shuai, D. Sustainable and scalable natural fiber welded palladium indium catalysts for nitrate reduction. Applied Catalysis B: Environmental , 221, 290-301 (2018). https://doi.org/10.1016/j.apcatb.2017.09.029
  » https://doi.org/10.1016/j.apcatb.2017.09.029
• Faridullah, Malik, N., Fareed, I., Irshad, M. Reducing the leachability of nitrate phosphorous and heavy metals from soil using waste material. Brazilian Journal of Chemical Engineering, 34, 715-726 (2017). https://doi.org/10.1590/0104-6632.20170343s20150617
  » https://doi.org/10.1590/0104-6632.20170343s20150617
• Hamid, S., Bae, S., Lee, W. Novel bimetallic catalyst supported by red mud for enhanced nitrate reduction. Chemical Engineering Journal , 348, 877-887 (2018). https://doi.org/10.1016/j.cej.2018.05.016
  » https://doi.org/10.1016/j.cej.2018.05.016
• Huo, X., van Hoomissen, D. J., Liu, J., Vyas, S., Strathmanna, T. J. Hydrogenation of aqueous nitrate and nitrite with ruthenium catalysts. Applied Catalysis B: Environmental , 211, 188-198 (2017). https://doi.org/10.1016/j.apcatb.2017.04.045
  » https://doi.org/10.1016/j.apcatb.2017.04.045
• Jaworski, M. A., Lick, I. D., Siri, G. J., Casella, M. L. ZrO₂-modified Al₂O₃-supported PdCu catalysts for the water denitrification reaction. Applied Catalysis B: Environmental , 156-157, 53-61 (2014). https://doi.org/10.1016/j.apcatb.2014.02.048
  » https://doi.org/10.1016/j.apcatb.2014.02.048
• Jaworski, M., Bertolini, G., Cabello, C., Siri, G., Casella, M. Use of Rh (III)-heteropolymolybdate as potential catalysts for the removal of nitrates in human drinking water: synthesis, characterisation and catalytic performance. Water, Air, & Soil Pollution, 229, 309-315 (2016). https://doi.org/10.1007/s11270-018-3962-3
  » https://doi.org/10.1007/s11270-018-3962-3
• Kim, Y.-N., Kim, M. Y., Choi, M. Synergistic integration of catalysis and ion-exchange for highly selective reduction of nitrate into N₂ Chemical Engineering Journal , 289, 423-432 (2016). https://doi.org/10.1016/j.cej.2016.01.002
  » https://doi.org/10.1016/j.cej.2016.01.002
• Labhsetwar, N., Doggali, P., Rayalu, S., Yadad, R., Mistuhashi, T., Haneda, H. Ceramics in Environmental Catalysis: Applications and Possibilities. Chinese Journal of Catalysis, 33, 1611-1621 (2012). https://doi.org/10.1016/S1872-2067(11)60440-3
  » https://doi.org/10.1016/S1872-2067(11)60440-3
• Landi, G., Barbato, P. S., Di Benedetto, A., Lisi, L. Optimization of the preparation method of CuO/CeO₂ structured catalytic monolith for CO preferential oxidation in H₂-rich streams. Applied Catalysis B: Environmental , 181, 727-737 (2016). https://doi.org/10.1016/j.apcatb.2015.08.040
  » https://doi.org/10.1016/j.apcatb.2015.08.040
• Lei, Y., Luo, Y., Li, X., Lu, J., Mei, Z., Peng, W., Chen, R., Chen, K., Chen, D., He, D. The role of samarium on Cu/Al₂O₃ catalyst in the methanol steam reforming for hydrogen production. Catalysis Today, 307, 162-168 (2018). https://doi.org/10.1016/j.cattod.2017.05.072
  » https://doi.org/10.1016/j.cattod.2017.05.072
• Li, R., Feng, C., Xi, B., Chen, N., Jiang, Y., Zhao, Y., Li, M., Dang, Q., Zhao, B. Nitrate removal efficiency of a mixotrophic denitrification wall for nitrate-polluted groundwater in situ remediation. Ecological Engineering, 106, 523-531 (2017). https://doi.org/10.1016/j.ecoleng.2017.06.010
  » https://doi.org/10.1016/j.ecoleng.2017.06.010
• Li, K., Jiao, Y., Yang, Z., Zhang, J. Comparative study of Ni/Al₂O₃-SiC foam catalysts and powdercatalysts for the liquid-phase hydrogenation of benzaldehyde. Journal of Materials Science & Technology, 35, 159-167 (2019). https://doi.org/10.1016/j.jmst.2018.09.018
  » https://doi.org/10.1016/j.jmst.2018.09.018
• Marchesini, F. A., Gutierrez, L. B., Querini, C. A., Miró, E. E. Pt,In and Pd,In catalysts for the hydrogenation of nitrates and nitrites in water. FTIR characterization and reaction studies. Chemical Engineering Journal , 159, 203-211 (2010). https://doi.org/10.1016/j.cej.2010.02.056
  » https://doi.org/10.1016/j.cej.2010.02.056
• Martínez, J., Ortiz, A., Ortiz, I. State-of-the-art and perspectives of the catalytic and electrocatalytic reduction of aqueous nitrates. Applied Catalysis B: Environmental , 207, 42-59 (2017). https://doi.org/10.1016/j.apcatb.2017.02.016
  » https://doi.org/10.1016/j.apcatb.2017.02.016
• Mendow, G., Sánchez, A., Grosso, C., Querini, C. A novel process for nitrate reduction in water using bimetallic Pd-Cu catalysts supported on ion exchange resin. Journal of Environmental Chemical Engineering, 5, 1404-1414 (2017). https://doi.org/10.1016/j.jece.2017.01.033
  » https://doi.org/10.1016/j.jece.2017.01.033
• Mirabi, M., Ghaderi, E., Sadabad, H. R. Nitrate reduction using hybrid system consisting of zero valent magnesium powder/activated carbon (Mg⁰/AC) from water. Process Safety and Environmental Protection, 111, 627-634 (2017). https://doi.org/10.1016/j.psep.2017.08.035
  » https://doi.org/10.1016/j.psep.2017.08.035
• Muñoz-Murillo, A., Martínez, L., Domínguez, M., Odriozola, J., Centeno, M., Selective CO methanation with structured RuO₂/Al₂O₃ catalysts. Applied Catalysis B: Environmental 236, 420-427 (2018). https://doi.org/10.1016/j.apcatb.2018.05.020
  » https://doi.org/10.1016/j.apcatb.2018.05.020
• Muttakin, M., Mitra, S., Thu, K., Ito, K., Bidyut Saha, B. Theoretical framework to evaluate minimum desorption temperature for IUPAC classified adsorption isotherms. International Journal of Heat and Mass Transfer, 122, 795-805 (2018). https://doi.org/10.1016/j.ijheatmasstransfer.2018.01.107
  » https://doi.org/10.1016/j.ijheatmasstransfer.2018.01.107
• Özyönüm, G., Yildirim, R. Water gas shift activity of AueRe catalyst over microstructured cordierite monolith wash-coated by ceria. International Journal Of Hidrogen Energy , 41, 5513-5521 (2016). https://doi.org/10.1016/j.ijhydene.2016.02.025
  » https://doi.org/10.1016/j.ijhydene.2016.02.025
• Resende, K., Teles, C., Jacobs, G., Davis, B., Cronauer, D., Kropf, A., Marshall, C., Hori, C., Noronha, F. Hydrodeoxygenation of phenol over zirconia supported Pd bimetallic catalysts. The effect of second metal on catalyst performance. Applied Catalysis B: Environmental , 232, 213-231 (2018). https://doi.org/10.1016/j.apcatb.2018.03.041
  » https://doi.org/10.1016/j.apcatb.2018.03.041
• Sedjame, H.-J., Rachid Brahmi, R., Lafaye, G., Barbier Jr., J., Fontaine, C. Influence of the formulation of catalysts deposited on cordierite monoliths for acetic acid oxidation. Comptes Rendus Chimie, 21, 182-193 (2018). https://doi.org/10.1016/j.crci.2017.08.004
  » https://doi.org/10.1016/j.crci.2017.08.004
• Şem, M., Emiroğlu, A., Çelik, M., CO and C₃H₈ oxidation activity of Pd/ZnO nanowires/cordierite catalyst. Applied Thermal Engineering, 99, 841-845 (2016). https://doi.org/10.1016/j.applthermaleng.2016.01.134
  » https://doi.org/10.1016/j.applthermaleng.2016.01.134
• Tang, L., Zhao, Z., We, Y., Liu, J., Peng, Y., Li, K. Study on the coating of nano-particle and 3DOM LaCoO₃ perovskite-type complex oxide on cordierite monolith and the catalytic performances for soot oxidation: The effect of washcoat materials of alumina, silica and titania. Catalysis Today, 297, 131-142 (2017). https://doi.org/10.1016/j.cattod.2017.06.016
  » https://doi.org/10.1016/j.cattod.2017.06.016
• Xu, Y. P., Qi, W.-K., Yang, X., Tang, Y., Wa, P., Fisher, A. Chemical removal of nitrate from water by aluminum-iron alloys. Chemosphere, 166, 197-202 (2017). https://doi.org/10.1016/j.chemosphere.2016.09.102
  » https://doi.org/10.1016/j.chemosphere.2016.09.102
• Yuan, E., Wu, C., Hou, X., Dou, M., Liu, G., Li, G., Wanga, L. Synergistic effects of second metals on performance of (Co, Ag, Cu)-doped Pd/Al₂O₃ catalysts for 2-ethyl-anthraquinone hydrogenation. Journal of Catalysis, 34, 779-788 (2017). https://doi.org/10.1016/j.jcat.2017.01.003
  » https://doi.org/10.1016/j.jcat.2017.01.003
• Yuan, J., Zhao, K., Cai, T., Gao, Z., Yang, L., He, D. One-step dip-coating of uniform γ-Al₂O₃ layers on cordierite honeycombs and its environmental applications. Ceramics International, 42, 14384-14390 (2016). https://doi.org/10.1016/j.ceramint.2016.06.003
  » https://doi.org/10.1016/j.ceramint.2016.06.003
• Zabala, M., Martínez, S., Manzano, M., Vives, L. Groundwater chemical baseline values to assess the Recovery Plan in the Matanza-Riachuelo River basin, Argentina. Science of the Total Environment 541, 1516-1530 (2016). https://doi.org/10.1016/j.scitotenv.2015.10.006
  » https://doi.org/10.1016/j.scitotenv.2015.10.006
• Zhang, X., Wu, D., Ceramic monolith supported Mn-Ce-M ternary mixed-oxide (M=Cu, Ni or Co) catalyst for VOCs catalytic oxidation. Ceramics International , 42, 16563-16570 (2016). https://doi.org/10.1016/j.ceramint.2016.07.076
  » https://doi.org/10.1016/j.ceramint.2016.07.076
• Zhai, Y., Zhao, X., Teng, Y., Li, X., Zhang, J., Wu, J., Zuo, R. Groundwater nitrate pollution and human health risk assessment by using HHRA model in an agricultural area. Ecotoxicology and Environmental Safety, 137, 130-142 (2017). https://doi.org/10.1016/j.ecoenv.2016.11.010
  » https://doi.org/10.1016/j.ecoenv.2016.11.010
• Zhuang, H.-D., Bai, S.-F., Liu, X.-M., Yam, Z.-F. Structure and performance of Cu/ZrO₂ catalyst for the synthesis of methanol from CO₂ hydrogenation. Journal of Fuel Chemistry and Technology, 38, 462-467 (2010). https://doi.org/10.1016/S1872-5813(10)60041-2
  » https://doi.org/10.1016/S1872-5813(10)60041-2
• Zoppas, F. M., Marchesini, F. A., Devard, A., Bernardes, A. M., Miró, E. E. Controlled deposition of Pd and In on carbon fibers by sequential electroless plating for the catalytic reduction of nitrate in water. Catalysis Communications, 78, 59-63 (2016). https://doi.org/10.1016/j.catcom.2016.02.012
  » https://doi.org/10.1016/j.catcom.2016.02.012
• Zoppas, F., Bernardes, A., Miró, E., Marchesini, F. Improving selectivity to dinitrogen using Palladium-Indium coated on activated carbon fibers: Preparation, characterization and application in water-phase nitrate reduction using formic acid as an alternative reductant source. Journal of Environmental Chemical Engineering , 6, 4764-4772 (2018). https://doi.org/10.1016/j.jece.2018.07.015
  » https://doi.org/10.1016/j.jece.2018.07.015

Publication Dates

• Publication in this collection
  30 Sept 2019
• Date of issue
  Apr-Jun 2019

History

• Received
  02 July 2018
• Reviewed
  28 Oct 2018
• Accepted
  06 Nov 2018