Optimizing Washcoating Conditions for the Preparation of Zeolite

Article pubs.acs.org/IECR

Cite This: Ind. Eng. Chem. Res. 2019, 58, 11799−11810

Optimizing Washcoating Conditions for the Preparation of ZeoliteBased Cordierite Monoliths for NOx CH4‑SCR: A Required Step for Real Application M. Carmen Bacariza,*,† Acácio Nobre Mendes,†,‡ Cansu Ozhan,‡ Patrick Da Costa,‡ and Carlos Henriques†

Downloaded via GUILFORD COLG on July 31, 2019 at 13:43:09 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

CATHPRO, Centro de Química Estrutural, Instituto Superior Técnico, Universidade de Lisboa. Av. Rovisco Pais, 1049-001 Lisboa, Portugal ‡ Institut Jean le Rond d’Alembert, CNRS UMR7190, Sorbonne Université, 2 place de la gare de ceinture, 78210 Saint Cyr l’Ecole, France S Supporting Information *

ABSTRACT: This work reports an exploratory study in terms of monoliths preparation for application in NOx CH4 selective catalytic reduction (NOx CH4-SCR), a promising route for treating the exhaust gas emissions from natural gas vehicles. By using washcoating method and a zeolite precursor, the effects of milling time (1, 2, or 16 h) and zeolite concentration in the slurry (15, 20, or 30 wt %) on the coating characteristics were evaluated. Then, the optimized washcoating conditions were used in order to obtain two PdCe-HMOR zeolite-coated monoliths with different number of CPSI (400 or 600). Tests were carried out under representative conditions of real exhaust gases from a heavy duty vehicle’s engine. Results showed that the differences in the performances of both materials could not be attributed to hydrodynamic factors being thus proposed an explanation related to the catalyst surface area in contact with the inlet feed gas, significantly higher for the most active monolith.

1. INTRODUCTION Air pollution is defined as the effect caused by concentration of solids, liquids, or gases on the air that have a negative impact on the surrounding environment and people.1 The main primary air pollutants are sulfur oxides (SOx), nitrogen oxides (NOx), particulate matter (PM), carbon monoxide (CO), volatile organic compounds (VOCs), ammonia (NH3), and ground-level ozone (O3).1,2 Nitrogen oxides are toxic for human health and responsible for acid rain. They are composed by nitrogen oxide (NO) and nitrogen dioxide (NO2).1−3 In terms of emission sources and considering the information recently reported by the European Environment Agency and referring to 2016 data from the EU28 countries,2 NOx are mainly emitted by road transport (39%), energy production and distribution (17%), commercial, institutional and households (14%), energy use in industry (11%), nonroad transport (9%), agriculture (6%), industrial processes and product use (3%), and waste (1%). Con© 2019 American Chemical Society

sequently, as road transport represents the most significant contribution, the development of suitable processes for reducing nitrogen oxides emissions in mobile sources is a critical issue. Indeed, NOx emissions abatement has been the focus of many different studies in the last decades, including mainly thermal but also photo- or even plasma-activated catalytic processes.4−18 Three key strategies have been mainly developed: NOx adsorption, nonselective catalytic reduction, and selective catalytic reduction (SCR), performed with ammonia or hydrocarbons.4,5,7−9,11−16,19−27 This later technology, namely, NOx-SCR with CH4, is particularly promising Received: Revised: Accepted: Published: 11799

March 4, 2019 May 8, 2019 June 18, 2019 June 18, 2019 DOI: 10.1021/acs.iecr.9b01216 Ind. Eng. Chem. Res. 2019, 58, 11799−11810

Article

Industrial & Engineering Chemistry Research

the monolith were assessed. On the basis of these preliminary studies and the data reported in the literature, a washcoating procedure was defined using cordierite monolithic supports with 400 and 600 CPSI and a bimetallic zeolite catalyst optimized in our former works (Pd(0.3)Ce(2.0)HMOR48−52).

for natural gas vehicles since it can eliminate both main pollutants, NOx and CH4. Though the development of new catalytic formulations performed at lab-scale is commonly made considering powder formulations, commercial applications of catalysts for real emission control systems (both for automotive and stationary sources) consider the use of honeycomb monoliths.28−30 In particular, for automotive applications, ceramic monoliths made of cordierite (2MgO·2Al2O3·5SiO2) are one of the most considered structures for the majority the after-treatment systems, namely, lean NOx traps (LNT),31−34 3-ways catalysts,35,36 diesel oxidation catalysts (DOC),37−39 and selective catalytic reduction by ammonia (NH3-SCR).40,41 This material is often considered due to its mechanical strength and its low thermal expansion coefficient. Additionally, its porosity and pore size distribution is suitable to be used for washcoat application due to its favorable adherence properties.28,42 In order to deposit an active catalyst, namely, a zeolite-based one, on the walls of honeycomb monoliths, two different methods can be adopted: (i) hydrothermal synthesis, consisting of growing the zeolite particles directly in the monolith walls,43 or (ii) deposition of the zeolite slurry, also known as washcoating.44 In the case of the latter method, it is possible to first perform the deposition of the zeolite in the structure and then to introduce the active metal phase, for instance, throughout an ion-exchange procedure similar to the ones performed in the powder formulations.45 Alternatively, the washcoating procedure can be performed with the final active catalyst formulation, being this a more suitable alternative when complex catalysts formulations, containing several metals introduced in the zeolite by different methods, are intended to be used. Currently, there is no commercial application available for NOx SCR systems using methane as reductant (NOx CH4SCR) and few works reported in literature consider the use of zeolite-based washcoated monoliths for this application. Zamaro et al.28 studied the effect of the slurry concentration and the use of SiO2 binder in the preparation of washcoated cordierite monoliths (400 CPSI) considering MOR, MFI, and FER zeolites. Authors concluded that the use of MFI led to a more stable washcoat. In the same study, the authors also presented the catalytic performances of an In-MFI washcoated monolith, comparing it to the powder formulation, which exhibited similar activities. In a later study, the authors also assessed the effect of zeolite slurry concentration and the use of solvents rather than water in the preparation of MFI slurries.46 Boix et al.44 studied the catalytic performance of Co-MOR washcoated in a cordierite monolith (400 CPSI), assessing the effect of water in the catalytic performances as well the use of cab-o-sil (a commercial SiO2) as a binder. Furthermore, authors also compared the use of methane and butane as reductants. In another study conducted by these authors,47 the effect of extra-framework aluminum species on Co-MFI catalysts performances toward NOx CH4-SCR was assessed, and it was concluded that Al3+ and Co2+ species in zeolite slurry originated a mixed Co/Al-based oxide, inactive for NOx CH4-SCR, after calcination. In the present work, parameters regarding the preparation of zeolite washcoated monoliths were studied. In a first approach, the effect of the milling time in the zeolite particle size distribution and viscosity as well as the influence of the zeolite slurry concentration in the total catalyst load washcoated on

2. MATERIAL AND METHODS 2.1. Samples Preparation. 2.1.1. Pd−Ce−Zeolite Powder Catalyst. In this study, a bimetallic Pd−Ce−Zeolite was used for monoliths preparation. This sample was prepared by using as support a NH4MOR commercial zeolite (CBV21A, Si/Al = 10, Zeolyst) and following the procedure reported and optimized in previous works.48−52 In summary, a certain mass of the parent zeolite was ion-exchanged with a Pd(NH3)4(NO3)2 aqueous solution (Aldrich, 99.99%) for 24 h, at room temperature, in order to introduce 0.3 wt % Pd. Afterward, the catalyst was centrifuged and dried overnight. Then, calcination under air flow was performed at 500 °C for 1 h (1 °C/min). This sample was then impregnated with a solution of Ce(NO3)3 (Fluka, 99%), in order to introduce 2 wt % Ce. After drying overnight, the catalyst was calcined at 500 °C (5 °C/min), for 8 h. The final sample was named as Pd(0.3)Ce(2.0)-HMOR and information regarding its main properties and performances can be found in previous works.48,52 2.1.2. Washcoated Monoliths Preparation. In this work, two cordierite honeycomb monoliths (Corning, 400 CPSI, dchannel ≈ 1.09 mm, twall ≈ 0.2 mm/NGK, 600 CPSI, dchannel ≈ 0.96 mm, twall ≈ 0.1 mm), containing a cylindrical shape with height and diameter of 1 in., were used as substrates. A summary of the work proposal can be found in Figure 1.

Figure 1. Schematic summary of the work developed in the present study.

As observed, preliminary studies were carried out (section 3.1) by using the commercial NH4MOR zeolite in order to optimize the milling time (1, 2, or 16 h) and the zeolite slurry concentration (15, 20, or 30 wt % using as substrate the 400 CPSI monolith). Wet ball milling was carried out at 140 rpm in a Fritsch Planetary Micro Mill PULVERISETTE 7 premium, equipped with two grinding bowl with capacity for 80 mL each (with 28 balls of zirconia, 10 mm diameter). In a second step (section 3.2), two different Pd(0.3)Ce(2.0)-HMOR zeolite slurries of 45 g were prepared again at 140 rpm by wet ball milling using the previously optimized milling time and zeolite slurry concentration. Thus, two 11800

DOI: 10.1021/acs.iecr.9b01216 Ind. Eng. Chem. Res. 2019, 58, 11799−11810

Article


Figure 2. Monoliths preparation procedure steps.

products nr 1040525CTT, 1070525 CTT, and 1100525CTT, respectively). Viscosity of the zeolite slurries was measured using a AND SV-10 vibro viscometer, using 10 mL of suspension. Particle size distribution was measured using a laser particle sizer Fritsch Analysette 22 compact. Measurements were performed in full-range mode (0.3−300 μm). For each sample, a background was collected (with mechanical mixer and ultrasounds on). Then, the measurement was collected twice for each sample, in order to verify repeatability of the measurements. At the end of each measurement, the dispersion chamber was cleaned. Finally, XRD of the dried slurry was carried out on a Bruker D8 Advance diffractometer. Previously calcined samples were placed in the well of a glass sample holder and pressed using a microscope slide, in order to smooth the sample surface. 2.2.2. Washcoated Monoliths Characterization. Monoliths were characterized by SEM/EDS analyses in Instituto de Carboquı ́mica - ICB (CSIC, Zaragoza) using a Hitachi S-3400 N microscope. Additionally, N2 adsorption was carried out at −196 °C for the spent washcoated monoliths on an Autosorb iQ equipment from Quantachrome. Before adsorption, samples were degassed under vacuum at 90 °C for 1 h and then at 350 °C for 4 h. The total pore volume (Vtotal) was calculated from the adsorbed volume of N2 at a p/p0 of 0.97, whereas the micropores volume (Vmicro) and the external surface area (Sext) were determined using the t-plot method. The mesopores volume (Vmeso) was given by the difference Vtotal − Vmicro. Finally, XRD was also performed for spent M_400 and M_600 catalysts. 2.3. Catalytic Tests. Catalytic tests were performed in temperature-programmed conditions, using a gas hourly space velocity (GHSV) of 40 000 h−1 calculated assuming a total gas flow (F) of 8.58 L/min and a volume of catalyst (Vcatal) of 12.9 mL. First, the reaction mixture was stabilized in a reactor bypass. Once stable, the reaction mixture was fed to the reactor and a heating ramp of 10 °C/min was applied until 550 °C. Once the final temperature was reached, the mixture was put in bypass to the reactor, and the signals were recorded (second bypass point). After the bypass, the reactor was cooled down using compressed air until room temperature. Products were analyzed by three NO/NO2, CO/CO2/N2O/O2, and HC analysers (Environnment SA TOPAZE 32M chemiluminescence analyzer, Environnment SA MIR 2M infrared analyzer and Environnment SA GRAPHITE 52M flame ionization

suspensions, S_400 and S_600, were prepared for carrying out the washcoating over the 400 CPSI and 600 CPSI monolithic supports, respectively, with their pH, viscosity and particle size distribution measured. Prior to the washcoating procedure, fresh monolithic supports were calcined in a muffle (500 °C, 4 h, 2 °C/min) in order to clean their surface being their masses registered after this pretreatment. Also, a mass of colloidal silica (LudoxHS 40, Sigma-Aldrich) corresponding to 3 wt % of the mass of zeolite slurry was incorporated to the previously prepared suspensions (binder) in order to enhance the mechanical stability and adherence of the washcoat.53 The washcoating procedure started with the immersion of the monolith in the zeolite slurry for 3 min. After this, the monolith was removed from the slurry, and compressed air was blown through the channels. The monolith was then placed in an oven (110 °C) for 1 h. After this period, the monolith was removed from the oven, and 10 min was allowed to ensure proper cooling down. The mass of washcoated monolith was then registered. Commercial catalysts for environmental automotive applications usually consider a mass of catalyst per volume of monolith (m/V) ratio around 150 g/L (this value may depend on the manufacturer and the type of technology, but it will be considered as reference for this work). In this work, monolithic supports used have a cylindrical shape, with 1 in. diameter and 1 in. height, which results in a volume of ca. 12.9 cm3. Assuming the previously mentioned m/V ratio, the mass of catalyst to be deposited would be 1.93 g. Consequently, the procedures above-described were repeated until the mass of washcoat was ca. 1.9 g. Calcination was finally performed in a muffle (500 °C, 4 h, 2 °C/min) without being verified differences in the monolith masses before and after this treatment for the two prepared samples: M_400 and M_600. The preparation procedure above-described is summarized in Figure 2. 2.2. Characterization Techniques. 2.2.1. Zeolite Slurries Characterization. Measurements of pH, viscosity and particle size distribution of the prepared slurries were performed in order to support the preparation of monolith-based structured catalysts. pH was measured using a Mettler Toledo Education line pH meter, equipped with an Inlab Expert Pro probe adequate for suspensions. Before each series of measurements, the pH meter was calibrated using two buffer solutions of the three available, with different pH (10, 7, and 2, from Reagecon; 11801


Article

Industrial & Engineering Chemistry Research Table 1. Gas Feed Properties Used as Input Data for the Hydrodynamic Flow Simulations in the Studied Monoliths T (°C) 520

density (kg/m3) 0.435

NO diffusivity (m2/s) 1.243 × 10

−4

CH4 diffusivity (m2/s)

i ∂u y ρjjj + u·∇·uzzz = −∇p + μ∇2 u + S k ∂t {

∂Ci = ∇·(D∇ci) − ∇·(uci) + R i ∂t

[NO] (ppm)

[CH4] (ppm)

3.599 × 10−5

400

1600

1.265 × 10

monolith considered for the simulations are presented in Tables 1 and 2, respectively.

detector (FID), respectively). Washcoated monoliths were weighted before and after tests, without being verified any mass loss during the catalytic experiments. Also for comparison purposes, catalytic tests were run for two samples of previous mentioned cordierite honeycomb bare monoliths (without washcoating); results (not shown) revealed no CH4 or NOx conversion in the entire range of temperature tested. 2.4. Flow Simulation Inside of Monolith Channels. 2.4.1. Methodology. Single-channel 2D simulations, corresponding to the monoliths prepared with 400 and 600 CPSI, were performed by adopting the methodology followed by Ozhan54 using Gerris Flow solver, a free software program for the solution of the partial deferential equations.55 According to the methodology proposed,54 the gas feed was assumed to be an incompressible fluid, which is an assumption widely used for simulations in catalytic converters due to the fact that Mach number is smaller than 0.05, acoustic waves have a negligible impact, and the variation in pressure is lower than 10% of total absolute pressure. The implemented model considers eqs 1 and 2, where t is the time, u is the velocity, ρ is the fluid density, p is the pressure, μ is the viscosity and S is the momentum source term. Additionally, N − 1 transport equations, where N is the number of components present in the system, were also considered (eq 3, where ci is the concentration of component i, D is the diffusion coefficient, and Ri is the reaction rate). These three equations were solved by imposing proper boundary conditions. In the inlet, velocity is assumed to be known and, at the outlet, classical outflow boundary conditions are applied (Dirichlet boundary condition for pressure and Neumann boundary condition for the normal velocity). At the walls, the velocity is imposed to be zero. ∇·u = 0

viscosity (N·s/m2)

−4

Table 2. Geometrical Characteristics Used as Input Data for Hydrodynamic Flow Simulations in the Studied Monoliths monolith

dchannel (mm)

νchannel (m/s)a

M_400 M_600

1.09 0.96

1.30 1.12

Calculated assuming a flow of 22.83 L/min, corresponding to the volumetric flow of 8.58 L/min of with 400 ppm of NO, 1600 ppm of CH4, 800 ppm of CO, 7% O2, N2 balance, at 25 °C, which corresponds to the feed mixture in the catalytic test.

a

The data representing the simulation scenario considered in Tables 1 and 2 was used in order to obtain the nondimensional variables that are indeed the input data of the simulations performed (Table 3). Table 3. Non-Dimensional Variables Used as Input Data for the Hydrodynamic Flow Simulations in the Studied Monolithsa

400 CPSI 600 CPSI

u̅ uc

ν uclc

D NO uclc

DCH4

1.0 1.0

0.058 0.076

0.088 0.116

0.089 0.118

uclc

xNO

xCH4

0.0004 0.0004

0.0016 0.0016

a u̅: uniform velocity profile imposed at the inlet of the channel; uc: characteristic velocity of the problem (chosen to be the average velocity in the channel); lc: characteristic length of the problem (chosen to be the length of the channel); ν: kinematic viscosity of the gas; DNO: molecular difussity of NO in the gas; DCH4: molecular diffusivity of CH4 in the gas; xNO: molar fraction of NO in the gas and xCH4: molar fraction of CH4 in the gas.

(1)

3. RESULTS AND DISCUSSION 3.1. NH 4 MOR Slurry Properties Effect on the Preparation of Monolithic Catalysts. When preparing structured catalysts by washcoating techniques, there are several parameters related to the slurry that influence the washcoat features and that are directly related to the quality of the final monolith to be obtained. Several studies have been performed and documented in literature describing the effects of such parameters in the preparation of structured catalysts, mostly considering cordierite washcoated with metal oxides catalysts (such as Al2O356−58) but also for zeolite-based washcoated cordierite.28,46,53 These studies may guide in the preparation of structured catalysts. Nevertheless, it is important to consider that significant amounts of catalyst are required for preparing washcoated monoliths when comparing to the powder formulations. For instance, with about 1 g of starting material (NH4MOR), it is possible to obtain enough catalyst for all characterization techniques and catalytic tests, these lasts requiring ca. 200 mg.48−52 For preparing a monolith (with dimensions D × L = 1 × 1 in.2), at least ca. 9 g of catalyst are required. Hence, it is important to have an idea of the influence of, at least, some parameters such as particle size distribution, viscosity, and

(2)

(3)

The conversion process in catalytic converters is a heterogeneous reaction where the transformation of hazardous exhaust gases occurs at the monolith catalytic walls. This leads to a concentration gradient near the wall, which requires a high resolution in this region, increasing the computational cost. One of the interesting characteristics of the solver is the capability to perform dynamic adaptive mesh refinement (AMR) using quadtree meshes. Hence, this method was applied in order to capture the concentration boundary layer near the catalytic walls, by decreasing the computational cost. The accuracy and efficiency of the code and the AMR solution has been demonstrated by Ozhan.54 2.4.2. Input Data Considered in the Simulations. The simulations were performed assuming the gases temperature of 520 °C (approximately the maximum conversion temperature observed) for a channel of 1 in. (25.4 cm) long. The properties of the gas feed and the geometrical characteristics of each 11802


Article

Industrial & Engineering Chemistry Research zeolite concentration on the slurry. As previously pointed out in this work, the purpose of this study is not to perform an exhaustive study of the influence of all the parameters that affect the washcoating process for obtaining zeolite-based monoliths but, instead, giving some insights on the effect of, probably, the most relevant zeolite slurry properties that shall be considered during the preparation of washcoated monoliths. To this end, in this first approach several suspensions were prepared using the parent NH4MOR powder zeolite (Si/Al = 10) in order to study the effect of the milling time in the particle size distribution and viscosity. The quantification of the total mass load as a function of the number of immersions during the washcoating process was also determined considering different concentrations of zeolite slurry. 3.1.1. Effect of Milling Time. According to Mitra and Kunzru,53 in order to have an efficient washcoating process the zeolite particle sizes should be in the 2−3 μm range. Wet ball milling is an effective method for reducing the particle size, since it allows keeping the crystallinity and surface area of the powder. However, the required milling time may depend on the zeolite considered, namely, its particle size distribution, and the rotational speed. The particle size distribution of the NH4MOR zeolite considered in this study and the zeolite slurries with 20 wt % concentration in distilled water submitted to a wet milling process for 1, 2, and 16 h, at a rotational speed of 140 rpm, are presented in Figure 3. Additionally, information regarding the

Table 4. Average Particles Diameter ( d p ), Respective Average Deviation (σ), Diameter at Which 50% of the Sample’s Mass Is Comprised of Smaller Particles (d50), pH, and Viscosity of NH4MOR Slurries Considering Different Milling Timesa milling time (h) NH4MOR (as received) 1 2 16

dp (μm)

σ (μm)

d50 (μm)

5.6

4.3

3.3

2.8 2.9 2.0

1.4 1.5 1.4

2.4 2.4 1.6

pH

viscosity (mPa·s)

7.44 7.83 7.83

1.83 1.95 2.51

For comparison purposes, d p , σ, and d50 values of the NH4MOR zeolite as received are also presented.

a

milling time that allowed us to obtain the smallest d p (ideal range for obtaining a good quality washcoat between 2 and 3 μm)53 and is reported as the time considered for the preparation of zeolite slurries in other published works,45 it was decided to consider this time for the preparation of the zeolites slurries in order to obtain the washcoated monoliths presented hereafter. 3.1.2. Effect of Zeolite Slurry Concentration. The concentration of zeolite in the slurry will naturally have an influence in the quantity and quality of washcoat deposited in the monoliths. In this way, Mitra and Kunzru53 compared MOR-washcoated catalysts obtained from slurries with 20 and 40 wt % after 4 immersions and observed that with a higher concentration the removal of slurry by blowdown after the immersion resulted in a nonhomogeneous coating of the channels. One can also expect that when using more concentrated slurries, the number of immersions that will result in a certain mass of washcoat will be smaller. As already discussed above, the mass of catalyst to be deposited would be ca. 1.9 g since a mass of catalyst per volume of monolith (m/V) ratio of 150 g/L was considered as reference. In order to define the number of immersions required to achieve this desired mass of washcoat, depending on the zeolite slurry concentration, three slurries with zeolite concentrations of 15, 20, and 30 wt % were prepared. pH and viscosities of these slurries were registered before the washcoating procedure (Figure 4A). As already indicated, cordierite honeycomb monolith with 400 CPSI was used as substrate for this study. The total mass of the monolith was registered before and after the immersions, being the total mass load versus number of immersions on the different zeolite slurries shown in Figure 4B. As seen, the number of immersions required to obtain ca. 1.9 g of washcoat (3 for 30 wt %, 5 for 20 wt %, and 7 for 15 wt %) increases significantly with the dilution of the zeolite slurry. An adherence test consisting in submersing the washcoated monoliths in an ultrasonic water bath (37 kHz) for 30 min followed by a drying procedure at 110 °C overnight was also performed. Results showed a loss of mass of 0, 0.3, and 2.3% for the monoliths obtained from the slurries with 15, 20, and 30 wt % of zeolite concentration, respectively. Another important factor to be considered in the slurry’s preparation is the required mass of catalyst. Since all the catalysts were prepared at laboratory scale, the more concentrated the slurry, the higher the number of batches required to obtain the powder catalyst. For instance, by

Figure 3. Particle size distribution and accumulated volume of NH4MOR zeolite as received (dashed lines) and NH4MOR slurries considering different milling times (1, 2, and 16 h, represented by light gray, dark gray, and black lines, respectively).

average particles’ diameter ( d p ) and deviation, the diameter at which 50% of the sample’s mass is composed by smaller particles (d50), the viscosity, and pH can be found in Table 4. The results obtained after 1 and 2 h of milling are very similar, with the average particles diameter close to 3 μm (and d50 = 2.4 μm), significantly smaller than the values for the as received zeolite. However, after 16 h milling, the average particle diameter becomes 2 μm (and d50 = 1.6 μm). As expected, viscosity increases significantly with the increase of milling time to 16 h, which is reasonable since the slurry is composed of smaller particles at the end of the process. pH does not vary significantly, and the fact that it is slightly basic can be attributed to an eventual ion-exchange of some NH4+ ions compensating Al− in the zeolite framework. Since 16 h was the 11803


Article


Figure 4. Viscosity (A, black line, right axis), pH (A, dark gray line, left axis), and total mass load vs number of immersions (B) for NH4MOR slurries considering zeolite concentrations of 15, 20, and 30 wt % represented by light gray, dark gray, and black lines, respectively.

considering a total mass of slurry of 45 g (approximately the minimum quantity required to successfully prepare a washcoated monolith, determined empirically), it would be necessary to prepare 6.75, 9, and 13.5 g of catalyst considering 15, 20, and 30 wt % concentrations, respectively. As conclusion, one can say that by using a slurry with 20 wt % zeolite concentration, a reasonable compromise can be achieved between mass of catalyst required, number of immersions (and, hence, time of preparation), and washcoat adherence. On the basis of all these factors, it has been decided to consider this concentration for the preparation of the zeolites slurries to obtain the washcoated monoliths presented hereafter. 3.2. Preparation and Characterization of PdCe-HMOR Washcoated Monoliths. As already summarized in Figure 1, two different cordierite honeycomb monoliths with 400 and 600 CPSI were used as substrates in the second part of this work. In this way, the two slurries (S_400 and S_600) prepared for the washcoated monoliths preparation using Pd(0.3)Ce(2.0)-HMOR zeolite were analyzed in terms of pH and viscosity (Table 5). Additionally, particle size distributions,

Figure 5. Particle size distribution (solid line) and accumulated volume (dashed line) of Pd(0.3)Ce(2.0)-HMOR slurries S_400 and S_600: powder before milling (black lines) and after (gray lines).

suspension, without being observed significant differences (92% in both cases). As verified, all zeolite catalysts (before and after milling) presented only the characteristic diffraction peaks of MOR zeolite, since Pd and Ce species are considerably low and well dispersed and hence not detected by this technique. Additionally, Table 6 summarizes the masses and washcoat load of catalysts M_400 and M_600. As it can be verified, the Table 6. Masses and Washcoat Loads of M_400 and M_600 Catalysts M_400 M_600

before milling after milling before milling after milling

S_400

S_600

dp (μm)

σ (μm)

d50 (μm)

8.0

7.3

3.6

3.39

2.6

1.5

1.9

n.a.

n.a.

8.8

8.1

3.9

2.61

2.80

2.6

1.6

1.9

pH

viscosity (mPa·s)

n.a.

n.a.

3.31

mwashcoated monolith (g)

washcoat load (g/L)

5.9 4.6

7.9 6.5

154 143

washcoat mass obtained in both cases was ∼1.9 g and the washcoat loads were also similar. The influence of the number of cells per square inch (CPSI) in the catalytic performances was assessed herein by comparing M_400 and M_600 monoliths, prepared using S_400 and S_600 slurries, respectively. Cordierite monoliths are known to present a rough surface, easily identified in SEM images, as shown in several works reported for the same honeycomb materials used in the present work in the literature.30,59−61 Consequently, when catalysts are successfully washcoated on monolith’s walls, their surface becomes visually smoother and more uniform compared to bare substrate. Indeed, SEM/EDS images for the monoliths along the channels (parallel cut to the gas flow, Figure 6) indicated that for both CPSI the washcoating of the active phase was achieved as revealed by the smooth surfaces verified in the channels. In Figure 6A1,B1, one can easily depict the difference in texture between the channel (containing the washcoat) and the cut walls, which do not contain washcoat and hence illustrate well what a part of bare substrate would look like. Nonetheless, the washcoat seems to be more homogeneous in M_400 rather than in

Table 5. pH, Viscosity, Average Particles Diameter (d p ), and Respective Average Deviation (σ) of Pd(0.3)Ce(2.0)HMOR Zeolite Slurries Used to Prepare the Washcoated Monoliths from the Present Section slurry

mbare substrate (g)

before and after 16 h milling were also analyzed (Table 5 and Figure 5). As seen, no significant differences can be verified when comparing the two prepared slurries. The crystallinity of the zeolite catalyst samples after wet ball milling was assessed by powder X-ray diffraction (Figure S.1) and compared with the crystallinity of the catalysts before 11804


Article


Figure 6. SEM images of washcoated monoliths M_400 (right, A.1, A.2, and A.3) and M_600 (left, B.1, B.2, and B.3) along the channels.

M_600, as in the second one more textural irregularities can be observed (Figure 6B.2,B.3). Energy-dispersive X-ray spectroscopy (EDS) analysis was also performed (Figures S.2 and S.3) in order to obtain the mapping of elements distribution (Mg, Al, Si, Pd, and Ce). Mg is only present in cordierite whereas Pd and Ce should only be present in the catalyst washcoat. For M_400 (Figure S.2), it is possible to observe that Pd and Ce are well-distributed along the image, which confirms that the washcoat is present in practically the entire area of the image recovered. Regarding the Mg mapping image, it is possible to observe that Mg is also well-distributed along the image. The fact that this element belonging exclusively to the cordierite substrate is detected by EDS analysis can be explained due to some penetration of Xray radiation through the washcoat. It should also be highlighted that in some regions, namely, near the leftdownward corner of Mg mapping image, Mg appears to be more concentrated, which can be an indicator of a segment presenting a thinner washcoat and allowing a deeper penetration of the radiation and reaching the cordierite. For M_600 (Figure S.3), similar conclusions to the ones presented for M_400 monolith can be withdrawn. Notwithstanding, it is worth highlighting that on Mg mapping image a more concentrated zone can be observed. This zone is overlapped with what seems to be a crate in the SEM image, which clearly suggests that the washcoat in this region is likely to be thinner. In summary, a qualitative comparison of the images obtained for both monoliths seems to suggest than M_600 monolith contains more irregularities in the washcoat, which may be related to the fact of being thinner for this catalyst. Moreover, Figure 7 illustrates SEM images for M_400 and M_600 monoliths, obtained by performing a transversal cut to the channels. In this way, for M_400 it is possible to observe that all the channels present washcoat, although in some of

Figure 7. SEM images of washcoated monoliths M_400 (right, A.1− A.4) and M_600 (left, B.1−B.4): transversal planes.

them the distribution seems not to be homogeneous being even identified some irregularities (Figure 7A.1) such as parts resembling fissures (Figure 7A.3, left-downward corner). It is likely that such irregularities may be related to a nonhomogenous blowdown of the channels that led to a higher accumulation of washcoat in some of them. An estimation of the maximum and minimum thickness of the washcoat was obtained considering image Figure 7A.4 (consisting of a zoom of Figure 7A.2), which are, respectively, 152 μm (at the corner of the channel) and 13 μm (at the center). In the case of M_600, it is possible to verify that all channels possess washcoat. However, even if no irregularities similar to the ones observed for M_400 are seen in the SEM images of M_600, it is possible to observe that some channels clearly exhibit different washcoat thicknesses in the corners, suggesting a nonhomogeneous washcoat. Moreover, it is also observable that the washcoat thickness is in general lower than the one observed for M_400. An estimation of the maximum and minimum thickness of the washcoat was also obtained for M_600 considering image Figure 7B.4, which are, respectively, 66 μm (at the corner of the monolith) and 12 μm (at the center). 3.3. Evaluation of PdCe-HMOR Washcoated Monoliths Performance on Synthetic Gas Bench. Monoliths were finally tested in a synthetic gas bench (temperatureprogrammed tests, from room temperature to ca. 550 °C), 11805


Article

Industrial & Engineering Chemistry Research considering an inlet gas feed mixture representative of real exhaust gases of a heavy duty vehicle’s engine. The results are presented in Figure 8, in the absence of water in the inlet gas

Figure 8. NOx conversion into N2 and N2O (A, black and gray lines, respectively) and CH4 and CO conversion (B, black and gray lines, respectively) of washcoated monoliths: M_400 (solid lines) and M_600 (dashed lines). Conditions: 400 ppm of NO, 1600 ppm of CH4, 800 ppm of CO, 7% O2, and GHSV of 40 000 h−1.

Figure 9. Simulation results for NO concentration profile over the first 10 mm of a single channel of M_400 (up) and M_600 (down) monoliths.

feed. As seen, the NOx, CH4, and CO conversions are always higher for M_600 than M_400, and the conversion starts at lower temperatures when considering 600 CPSI. One parameter that could influence the performances of both materials is their surface area. As cordierite monoliths are known to present very low external surface areas and negligible pore volumes,59,60 textural properties of the washcoated monoliths will be provided by the active phase, whose nature and mass was the same for both materials. In this way, no significant differences were expected between these materials in terms of external surface areas and micro- and mesopore volumes, as indeed confirmed by the results obtained from N2 sorption analysis and presented in Figure S.4. In addition, since both monoliths exhibit different geometries, one cannot exclude that eventual differences in the hydrodynamic flow inside the channels could explain the results observed. In order to address this issue, two simulations corresponding to a single channel of the monoliths were performed, considering the methodology followed by Ozhan described above.54 Simulation results concerning NO profiles over the channels of M_400 and M_600 catalysts are exhibited in Figure 9. For each case, profiles corresponding to three different times (2, 4, and 10 s) are presented. At the later time, the system is already in steady-state. For brevity, only the first 10 mm of the channel are exhibited, since in the rest of the channel the concentration is always zero. It is important to clarify that the implemented model does not consider the reaction that takes place in the catalyst, since it is assumed that once the reactants touch the wall their concentration becomes zero (i.e., 100% of conversion). Nevertheless, the purpose of these analyses is only to compare the hydrodynamic flow in the system in both monoliths. By comparing the representations presented in Figure 9 it is possible to observe that, in both cases, the system converges rapidly to the steady-state and no significant changes are observed between the different times. Moreover, by comparing the simulations obtained after 10 s, it is possible to see that, though some differences are observed in the very beginning of the channels (up to ca. 2 mm), the

majority of the channel presents a concentration that is zero (dark blue color) for both configurations. NO and CH4 conversions along the channel (axial distance) are presented in Figure 10; in the first 2 mm of the channel,

Figure 10. NO (squares) and CH4 (triangles) conversions along the channels of M_400 (gray lines and symbols) and M_600 (black lines and symbols) monoliths, according to the model developed by Ozhan.37

the conversions are higher for 600 CPSI (i.e., a higher percentage (conversion) of the reactants reached the wall of the catalyst). However, after 7 mm, all the reactants have reached the walls, and all the profiles become the same. It is then clear that if at the beginning some slight differences could be observed due to the hydrodynamic flow in the system then 11806


Article


conversion values would be significantly low, namely, below 7% for NOx conversion and 5% for CH4 conversion−values observed for M_400, which could jeopardize the interpretation of results. Moreover, for commercial automotive applications (e.g., heavy-duty vehicles) monoliths above 600 CPSI are not even considered for natural gas applications, due to high pressure drops that cause a loss of engine performance (power and fuel consumption). Furthermore, Figure 11 illustrates the catalytic performances of both M_400 and M_600 monoliths tested in the same conditions previously mentioned, but considering 2 vol % of water in the inlet gas feed. Results show that the catalytic performances are drastically affected by the presence of water in the inlet feed (e.g., for M_600, NOx conversion into N2 decreases from ca. 20 to 8%). Additionally, the temperature at which the monoliths start to exhibit some activity for NOx and CH4 conversion also increases. It is worth mentioning that for CO conversion the difference in temperatures at which conversion starts is lower; in both situations, CO conversion is practically complete above 520 °C. This negative effect of water in the catalytic performances can be attributed to the inhibiting role of H2O in the reaction, as widely reported in the literature.48,62−74 Finally, it is important to highlight that NOx and CH4 conversion values obtained in the tests with monoliths are considerably lower than the ones obtained with the powder formulation presented elsewhere.48 Indeed, at 500 °C, in the absence of water in the inlet gas feed, NOx conversion into N2 was 19 against 70%. However, it is important to keep in mind that the geometry of both systems is completely different. Indeed, in the powder tests, GHSV is calculated by dividing the total flow by the volume of the catalytic bed since, in this geometry, the gas flow crosses the catalytic bed. However, in the monolith tests, GHSV is determined by dividing the total flow by the volume of monolith, which is a cylinder with diameter and height equal to 1 in. (this procedure is the one

these differences are mitigated and become irrelevant at the end of the channel. Thus, the differences in the hydrodynamic profiles due to the different geometries cannot explain, per se, the differences in the catalytic performances. In fact, the most reasonable explanation lays on the surface of catalyst that contacts the feed gas. An estimation of the washcoat total surface area for the two prepared monoliths is presented in Table 7, this value Table 7. Washcoat Surface Area Estimated for M_400 and M_600 Monoliths monolith

dchannel (mm)

ncellsa

twashcoat (μm)b

Asurface,total (m2)c

M_400 M_600

1.09 0.96

314 471

13 12

0.0267 0.0352

a

Considering cylindrical monolith, Dmonolith = 1 in. bCorresponding to the minimum washcoat thicknesses obtained by SEM. cConsidering circular section of each channel with effective diameter = dchannel − 2twashcoat.

being 32% higher for M_600 than M_400 (0.0352 vs 0.0267 m2). It should be mentioned that the previously presented estimations were made considering that the monolith channels are cylindrical. However, though this approximation may be roughly made for M_400 (see Figure 7A.1−A.4), in the case of M_600 monolith it is clear that the transversal geometry of the channel is somewhere between a square and a circle (see Figure 7B.1−B.4) which means that the actual washcoat surface area is likely to be even higher than the one obtained in the performed estimation. An interesting way of confirm this hypothesis would be to prepare a catalyst with lower or higher number of CPSI (e.g., 200 or 900) and verify if some proportionality (for instance linear) would exist between total washcoat surface area and NOx, CO, or CH4 conversions. Notwithstanding, it is important to highlight that below 400 CPSI the expected

Figure 11. NOx conversion into N2 and N2O (A, black and gray lines, respectively) and CH4 and CO conversion (B, black and gray lines, respectively) of Pd(0.3)Ce(2.0)-HMOR washcoated monoliths: M_400 (solid lines) and M_600 (dashed lines). Conditions: 400 ppm of NO, 1600 ppm of CH4, 800 ppm of CO, 7% O2, 2% H2O, and GHSV of 40 000 h−1. 11807


Article

Industrial & Engineering Chemistry Research adopted by industrial catalysts manufacturers and automotive constructors). Consequently, when using a certain total flow and monolithic supports with the same dimensions, the GHSV will be independent of the washcoated mass of catalyst in the supports. Moreover, in this geometry, instead of crossing the catalyst, the gas sweeps its surface along the channel without crossing it. As a result, care must be taken when performing comparisons between the different formulations (powder vs monolith), since they do not have the same physical meaning. Notwithstanding, it is interesting to see the F/mcatalyst ratio for both tests. For the powder tests, the total flow was 15 L/h for 0.188 g of catalyst, which results in ca. 80 L/(h.g). For the monolith tests, the total flow was 8.58 L/min for 1.93 g of catalysts (mass of washcoat), which results in ca. 267 L/(h·g). These numbers show that not only the geometry is different but also the total flow per mass of catalyst in the system is much higher when considering monoliths, which could also contribute to the differences in the conversion values obtained.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

M. Carmen Bacariza: 0000-0003-4236-6724 Patrick Da Costa: 0000-0002-0083-457X Carlos Henriques: 0000-0001-5878-5742 Notes

The authors declare no competing financial interest.

■

4. CONCLUSIONS In this work, the effect of key parameters on the preparation of structured catalysts, consisting in cordierite monoliths washcoated with zeolites, were studied. The effect of the milling time in the preparation of MOR zeolite slurries with 20 wt % concentration was assessed, and it was concluded that after 16 h of wet milling (at 140 rpm) the average particle size could be reduced from 5.6 to 2.0 μm which is considered within the ideal range (2−3 μm) in order to perform an effective washcoating procedure. The effect of the zeolite concentration in the slurry (15−30 wt %) was also assessed by measuring the total mass load of washcoat deposited in the cordierite monolith during the washcoating procedure, after each immersion. It was concluded that with 20 wt % a reasonable compromise could be obtained between the number of immersions required to obtain a given mass load, the loss of washcoat after adherence test and the required mass of catalyst (prepared at lab-scale) to prepare the monoliths. On the basis of these studies, it was decided to prepare the monoliths using zeolite slurries with 20 wt % concentration, obtained after wet ball milling for 16 h. In order to assess the effect of the CPSI of the monolith in the catalytic performances, two monoliths with 400 and 600 CPSI were washcoated with Pd(0.3)Ce(2.0)-HMOR catalyst. Catalytic test results considering representative conditions of real exhaust gases from a heavy-duty vehicle’s engine showed that M_600 monolith was the one that exhibited higher NOx and CH4 conversions. A simulation of the hydrodynamic flow inside the channels of both M_400 and M_600 revealed that only slight differences in the reactant’s flows are expected between both monoliths, which is why the discrepancy in the catalytic performances of these monoliths is not likely to be due to hydrodynamic factors. In fact, since the washcoat mass is practically the same in both monoliths, the most plausible explanation lays on the catalyst surface area that contacts the inlet feed gas, which is estimated to be 32% higher in M_600.

■

well as for the grinded washcoated monoliths after reaction; EDS mapping from a segment of catalytic layer (corresponding to Figure 7E) of washcoated M_400 monolith; EDS mapping from a segment of catalytic layer (Figure 7D) of washcoated M_600 monolith; textural properties obtained from N2 sorption of the spent washcoated monoliths (PDF)

ACKNOWLEDGMENTS We acknowledge “Fundaçaõ para a Ciência e a Tecnologia (FCT)” (UID/QUI/00100/2013 and grant SFRH/BD/ 78639/2011) and ENGIE (project ENGIE/IST/UPMC) for financial support.

■

REFERENCES

(1) International Energy Agency. WEO-2016 Special Report: Energy and Air Pollution; World Energy Outlook; International Energy Agency, 2016; p 266. (2) European Environmental Agency. Air Quality in Europe - 2018 Report; European Environmental Agency, 2018. (3) U.S. Environmental Protection Agency. Nitrogen Dioxide (NO2) Pollution. https://www.epa.gov/no2-pollution (accessed November 15, 2018). (4) Cheng, X.; Bi, X. T. A Review of Recent Advances in Selective Catalytic NOx Reduction Reactor Technologies. Particuology 2014, 16, 1−18. (5) Dhal, G. C.; Dey, S.; Mohan, D.; Prasad, R. Simultaneous Abatement of Diesel Soot and NOX Emissions by Effective Catalysts at Low Temperature: An Overview. Catal. Rev.: Sci. Eng. 2018, 60 (3), 437−496. (6) Ji, Y.; Xu, D.; Bai, S.; Graham, U.; Crocker, M.; Chen, B.; Shi, C.; Harris, D.; Scapens, D.; Darab, J. Pt- and Pd-Promoted CeO2−ZrO2 for Passive NOx Adsorber Applications. Ind. Eng. Chem. Res. 2017, 56 (1), 111−125. (7) Jeevahan, J.; Mageshwaran, G.; Joseph, G. B.; Raj, R. B. D.; Kannan, R. T. Various Strategies for Reducing Nox Emissions of Biodiesel Fuel Used in Conventional Diesel Engines: A Review. Chem. Eng. Commun. 2017, 204 (10), 1202−1223. (8) Roy, S.; Madras, G. Photocatalytic NOx Abatement: A Short Review. Curr. Org. Chem. 2015, 19 (21), 2122−2131. (9) Leng, X.; Zhang, Z.; Li, Y.; Zhang, T.; Ma, S.; Yuan, F.; Niu, X.; Zhu, Y. Excellent Low Temperature NH3-SCR Activity over MnaCe0.3TiOx (a = 0.1−0.3) Oxides: Influence of Mn Addition. Fuel Process. Technol. 2018, 181, 33−43. (10) Hoekman, S. K.; Robbins, C. Review of the Effects of Biodiesel on NOx Emissions. Fuel Process. Technol. 2012, 96, 237−249. (11) Xu, J.; Chen, G.; Guo, F.; Xie, J. Development of WideTemperature Vanadium-Based Catalysts for Selective Catalytic Reducing of NOx with Ammonia: Review. Chem. Eng. J. 2018, 353, 507−518. (12) Lasek, J.; Yu, Y.-H.; Wu, J. C. S. Removal of NOx by Photocatalytic Processes. J. Photochem. Photobiol., C 2013, 14, 29−52. (13) Nguyen, D. B.; Heo, I. J.; Mok, Y. S. Enhanced Performance at an Early State of Hydrocarbon Selective Catalyst Reduction of NOx

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.iecr.9b01216. XRD diffractograms obtained for the two dried zeolite slurries (S_400 and S_600) before and after milling as 11808


Article

Industrial & Engineering Chemistry Research by Atmospheric Pressure Plasma. J. Ind. Eng. Chem. 2018, 68, 372− 379. (14) Li, W.; Guo, R.; Wang, S.; Pan, W.; Chen, Q.; Li, M.; Sun, P.; Liu, S. The Enhanced Zn Resistance of Mn/TiO2 Catalyst for NH3SCR Reaction by the Modification with Nb. Fuel Process. Technol. 2016, 154, 235−242. (15) Skalska, K.; Miller, J. S.; Ledakowicz, S. Trends in NOx Abatement: A Review. Sci. Total Environ. 2010, 408 (19), 3976− 3989. (16) Gao, C.; Shi, J.-W.; Fan, Z.; Yu, Y.; Chen, J.; Li, Z.; Niu, C. EuMn-Ti Mixed Oxides for the SCR of NOx with NH3: The Effects of Eu-Modification on Catalytic Performance and Mechanism. Fuel Process. Technol. 2017, 167, 322−333. (17) Oskooei, A. B.; Koohsorkhi, J.; Mehrpooya, M. Simulation of Plasma-Assisted Catalytic Reduction of NOx, CO, and HC from Diesel Engine Exhaust with COMSOL. Chem. Eng. Sci. 2019, 197, 135−149. (18) Zhao, D.; Gao, Z.; Xian, H.; Xing, L.; Yang, Y.; Tian, Y.; Ding, T.; Jiang, Z.; Zhang, J.; Zheng, L.; et al. Addition of Pd on La0.7Sr0.3CoO3 Perovskite To Enhance Catalytic Removal of NOx. Ind. Eng. Chem. Res. 2018, 57 (2), 521−531. (19) Yin, D.; Li, J.; Wang, J.; Ling, L.; Qiao, W. Low-Temperature Selective Catalytic Reduction of NOx with Urea Supported on Carbon Xerogels. Ind. Eng. Chem. Res. 2018, 57 (20), 6842−6852. (20) Metkar, P. S.; Harold, M. P.; Balakotaiah, V. Experimental and Kinetic Modeling Study of NH3-SCR of NOx on Fe-ZSM-5, CuChabazite and Combined Fe- and Cu-Zeolite Monolithic Catalysts. Chem. Eng. Sci. 2013, 87, 51−66. (21) Karamitros, D.; Koltsakis, G. Model-Based Optimization of Catalyst Zoning on SCR-Coated Particulate Filters. Chem. Eng. Sci. 2017, 173, 514−524. (22) Park, J.-H.; Ahn, J.-W.; Kim, K.-H.; Son, Y.-S. Historic and Futuristic Review of Electron Beam Technology for the Treatment of SO2 and NOx in Flue Gas. Chem. Eng. J. 2019, 355, 351−366. (23) Rodriguez-Rivas, F.; Pastor, A.; Barriga, C.; Cruz-Yusta, M.; Sánchez, L.; Pavlovic, I. Zn-Al Layered Double Hydroxides as Efficient Photocatalysts for NOx Abatement. Chem. Eng. J. 2018, 346, 151−158. (24) Moon Lee, S.; Su Kim, S.; Chang Hong, S. Systematic Mechanism Study of the High Temperature SCR of NOX by NH3 over a W/TiO2 Catalyst. Chem. Eng. Sci. 2012, 79, 177−185. (25) Wang, T.; Sun, B. Effect of Temperature and Relative Humidity on NOX Removal by Dielectric Barrier Discharge with Acetylene. Fuel Process. Technol. 2016, 144, 109−114. (26) Zhao, Y.; Choi, B.; Kim, D. Effects of Ce and Nb Additives on the De-NOx Performance of SCR/CDPF System Based on Cu-Beta Zeolite for Diesel Vehicles. Chem. Eng. Sci. 2017, 164, 258−269. (27) Ting, A. W.-L.; Harold, M. P.; Balakotaiah, V. Elucidating the Mechanism of Fast Cycling NOx Storage and Reduction Using C3H6 and H2 as Reductants. Chem. Eng. Sci. 2018, 189, 413−421. (28) Zamaro, J. M.; Ulla, M. A.; Miró, E. E. Zeolite Washcoating onto Cordierite Honeycomb Reactors for Environmental Applications. Chem. Eng. J. 2005, 106 (1), 25−33. (29) Millet, C.-N.; Chédotal, R.; Da Costa, P. Synthetic Gas Bench Study of a 4-Way Catalytic Converter: Catalytic Oxidation, NOx Storage/Reduction and Impact of Soot Loading and Regeneration. Appl. Catal., B 2009, 90 (3), 339−346. (30) Govender, S.; Friedrich, H. B. Monoliths: A Review of the Basics, Preparation Methods and Their Relevance to Oxidation. Catalysts 2017, 7 (2), 62. (31) Pereda-Ayo, B.; González-Velasco, J. R. NOx Storage and Reduction for Diesel Engine Exhaust Aftertreatment. In Diesel Engine Combustion, Emissions and Condition Monitoring; IntechOpen, 2013. (32) Watling, T. C.; Bolton, P. D.; Swallow, D. The Effect of NO and O2 Concentration on NOX Storage over a Lean NOX Trap: An Experimental and Modelling Study. Chem. Eng. Sci. 2018, 178, 312− 323.

(33) Balaji, N.; Aghalayam, P.; Kaisare, N. S. Global Kinetic Modeling and Analysis of Lean NOx Traps (LNT) Catalysts. Ind. Eng. Chem. Res. 2018, 57 (20), 6853−6862. (34) Millo, F.; Rafigh, M.; Sapio, F.; Wahiduzzaman, S.; Dudgeon, R.; Ferreri, P.; Barrientos, E. Modeling NOx Storage and Reduction for a Diesel Automotive Catalyst Based on Synthetic Gas Bench Experiments. Ind. Eng. Chem. Res. 2018, 57 (37), 12335−12351. (35) Bruehlmann, S.; Forss, A.-M.; Steffen, D.; Heeb, N. V. Benzene: A Secondary Pollutant Formed in the Three-Way Catalyst. Environ. Sci. Technol. 2005, 39 (1), 331−338. (36) Kwon, H. J.; Baik, J. H.; Kwon, Y. T.; Nam, I.-S.; Oh, S. H. Enhancement Effect of Water on Oxidation Reactions over Commercial Three-Way Catalyst. Chem. Eng. J. 2008, 141 (1), 194−203. (37) Lizarraga, L.; Souentie, S.; Boreave, A.; George, C.; D’Anna, B.; Vernoux, P. Effect of Diesel Oxidation Catalysts on the Diesel Particulate Filter Regeneration Process. Environ. Sci. Technol. 2011, 45 (24), 10591−10597. (38) Herreros, J. M.; George, P.; Umar, M.; Tsolakis, A. Enhancing Selective Catalytic Reduction of NOx with Alternative Reactants/ Promoters. Chem. Eng. J. 2014, 252, 47−54. (39) Václavík, M.; Kočí, P.; Novák, V.; Thompsett, D. NOx Conversion and Selectivity in Multi-Layer and Sequential DOC-LNT Automotive Exhaust Catalysts: Influence of Internal Transport. Chem. Eng. J. 2017, 329, 128−134. (40) Colombo, M.; Koltsakis, G.; Nova, I.; Tronconi, E. Modelling the Ammonia Adsorption−Desorption Process over an Fe−Zeolite Catalyst for SCR Automotive Applications. Catal. Today 2012, 188 (1), 42−52. (41) Wang, J.; Peng, Z.; Chen, Y.; Bao, W.; Chang, L.; Feng, G. InSitu Hydrothermal Synthesis of Cu-SSZ-13/Cordierite for the Catalytic Removal of NOx from Diesel Vehicles by NH3. Chem. Eng. J. 2015, 263, 9−19. (42) Williams, J. L. Monolith Structures, Materials, Properties and Uses. Catal. Today 2001, 69 (1), 3−9. (43) Zamaro, J. M.; Ulla, M. A.; Miró, E. E. Growth of Mordenite on Monoliths by Secondary Synthesis: Effects of the Substrate on the Coating Structure and Catalytic Activity. Applied. Appl. Catal., A 2006, 314 (1), 101−113. (44) Boix, A. V.; Aspromonte, S. G.; Miró, E. E. Deactivation Studies of the SCR of NOx with Hydrocarbons on Co-Mordenite Monolithic Catalysts. Appl. Catal., A 2008, 341 (1), 26−34. (45) Pereda-Ayo, B.; De La Torre, U.; Romero-Sáez, M.; Aranzabal, A.; González-Marcos, J. A.; González-Velasco, J. R. Influence of the Washcoat Characteristics on NH3-SCR Behavior of Cu-Zeolite Monoliths. Catal. Today 2013, 216, 82−89. (46) Zamaro, J. M.; Ulla, M. A.; Miró, E. E. The Effect of Different Slurry Compositions and Solvents upon the Properties of ZSM5Washcoated Cordierite Honeycombs for the SCR of NOx with Methane. Catal. Today 2005, 107−108, 86−93. (47) Boix, A. V.; Miró, E. E.; Lombardo, E. A.; Fierro, J. L. G. The Inhibiting Effect of Extra-Framework Al on Monolithic Co-ZSM5 Catalysts Used for NOx SCR. Catal. Today 2008, 133−135, 428−434. (48) Mendes, A. N.; Lauga, V.; Capela, S.; Ribeiro, M. F.; Da Costa, P.; Henriques, C. Application of PdCe-HMOR Catalyst as NOx CH4SCR System for Heavy-Duty Vehicles Moved by Natural Gas. Top. Catal. 2016, 59 (10), 982−986. (49) Mendes, A. N.; Ribeiro, M. F.; Henriques, C.; Da Costa, P. On the Effect of Preparation Methods of PdCe-MOR Catalysts as NOx CH4-SCR System for Natural Gas Vehicles Application. Catalysts 2015, 5 (4), 1815−1830. (50) Mendes, A. N.; Zholobenko, V. L.; Thibault-Starzyk, F.; Costa, P. D.; Henriques, C. On the Enhancing Effect of Ce in Pd-MOR Catalysts for NOx CH4-SCR: A Structure-Reactivity Study. Appl. Catal., B 2016, 195, 121−131. (51) Mendes, A. N.; Matynia, A.; Toullec, A.; Capela, S.; Ribeiro, M. F.; Henriques, C.; Da Costa, P. Potential Synergic Effect between MOR and BEA Zeolites in NOx SCR with Methane: A Dual Bed Design Approach. Appl. Catal., A 2015, 506, 246−253. 11809


Article

Industrial & Engineering Chemistry Research (52) Mendes, A. N. Development of an Innovative System for Pollution Abatement in the New Natural Gas Vehicles; Universidade de Lisboa, Université Pierre et Marie Curie: Lisboa, Portugal, 2015. (53) Mitra, B.; Kunzru, D. Washcoating of Different Zeolites on Cordierite Monoliths. J. Am. Ceram. Soc. 2008, 91 (1), 64−70. (54) Ozhan, C. Multi-Scale Simulation of Automotive Catalytic Converters; Université Pierre et Marie Curie - Paris VI: Paris, 2014. (55) Popinet, S. Gerris: A Tree-Based Adaptive Solver for the Incompressible Euler Equations in Complex Geometries. J. Comput. Phys. 2003, 190 (2), 572−600. (56) Blachou, V.; Goula, D.; Philippopoulos, C. Wet Milling of Alumina and Preparation of Slurries for Monolithic Structures Impregnation. Ind. Eng. Chem. Res. 1992, 31 (1), 364−369. (57) Agrafiotis, C.; Tsetsekou, A. The Effect of Processing Parameters on the Properties of γ-Alumina Washcoats Deposited on Ceramic Honeycombs. J. Mater. Sci. 2000, 35 (4), 951−960. (58) Agrafiotis, C.; Tsetsekou, A. The Effect of Powder Characteristics on Washcoat Quality. Part II: Zirconia, Titania Washcoats  Multilayered Structures. J. Eur. Ceram. Soc. 2000, 20 (7), 825−834. (59) Varela-Gandía, F. J.; Berenguer-Murcia, Á .; Lozano-Castelló, D.; Cazorla-Amorós, D.; Sellick, D. R.; Taylor, S. H. Total Oxidation of Naphthalene at Low Temperatures Using Palladium Nanoparticles Supported on Inorganic Oxide-Coated Cordierite Honeycomb Monoliths. Catal. Sci. Technol. 2013, 3 (10), 2708−2716. (60) Rezaei, F.; Lawson, S.; Hosseini, H.; Thakkar, H.; Hajari, A.; Monjezi, S.; Rownaghi, A. A. MOF-74 and UTSA-16 Film Growth on Monolithic Structures and Their CO2 Adsorption Performance. Chem. Eng. J. 2017, 313, 1346−1353. (61) Baharudin, L.; Watson, M. J. Monolithic Substrate Support Catalyst Design Considerations for Steam Methane Reforming Operation. Rev. Chem. Eng. 2018, 34 (4), 481−501. (62) Kikuchi, E.; Yogo, K. Selective Catalytic Reduction of Nitrogen Monoxide by Methane on Zeolite Catalysts in an Oxygen-Rich Atmosphere. Catal. Today 1994, 22 (1), 73−86. (63) Kikuchi, E.; Ogura, M.; Aratani, N.; Sugiura, Y.; Hiromoto, S.; Yogo, K. Promotive Effect of Additives to In/H-ZSM-5 Catalyst for Selective Reduction of Nitric Oxide with Methane in the Presence of Water Vapor. Catal. Today 1996, 27 (1), 35−40. (64) Lónyi, F.; Solt, H. E.; Valyon, J.; Decolatti, H.; Gutierrez, L. B.; Miró, E. An Operando DRIFTS Study of the Active Sites and the Active Intermediates of the NO-SCR Reaction by Methane over In,Hand In,Pd,H-Zeolite Catalysts. Appl. Catal., B 2010, 100 (1), 133− 142. (65) Decolatti, H.; Solt, H.; Lónyi, F.; Valyon, J.; Miró, E.; Gutierrez, L. The Role of Pd−In Interactions on the Performance of PdInHmordenite in the SCR of NOx with CH4. Catal. Today 2011, 172 (1), 124−131. (66) Serra, R.; Vecchietti, M. J.; Miró, E.; Boix, A. In,Fe-Zeolites: Active and Stable Catalysts for the SCR of NOxKinetics, Characterization and Deactivation Studies. Catal. Today 2008, 133−135, 480−486. (67) Pieterse, J. A. Z.; van den Brink, R. W.; Booneveld, S.; de Bruijn, F. A. Durability of ZSM5-Supported Co-Pd Catalysts in the Reduction of NOx with Methane. Appl. Catal., B 2002, 39 (2), 167− 179. (68) Ogura, M.; Hayashi, M.; Kikuchi, E. Intrapore Catalysis in Reduction of Nitric Oxide with Methane. Catal. Today 1998, 42 (1), 159−166. (69) Wang, X.; Zhang, T.; Sun, X.; Guan, W.; Liang, D.; Lin, L. Enhanced Activity of an In−Fe2O3/H-ZSM-5 Catalyst for NO Reduction with Methane. Appl. Catal., B 2000, 24 (3), 169−173. (70) Ren, L.; Zhang, T.; Liang, D.; Xu, C.; Tang, J.; Lin, L. Effect of Addition of Zn on the Catalytic Activity of a Co/HZSM-5 Catalyst for the SCR of NOx with CH4. Appl. Catal., B 2002, 35 (4), 317− 321. (71) Li, Z.; Flytzani-Stephanopoulos, M. Selective Catalytic Reduction of Nitric Oxide by Methane over Cerium and Silver IonExchanged ZSM-5 Zeolites. Appl. Catal., A 1997, 165 (1), 15−34.

(72) Ferreira, A. P.; Capela, S.; Da Costa, P.; Henriques, C.; Ribeiro, M. F.; Ribeiro, F. R. CH4-SCR of NO over Co and Pd Ferrierite Catalysts: Effect of Preparation on Catalytic Performance. Catal. Today 2007, 119 (1), 156−165. (73) Ramallo-López, J. M.; Requejo, F. G.; Gutierrez, L. B.; Miró, E. E. EXAFS, TDPAC and TPR Characterization of PtInFerrierite: The Role of Surface Species in the SCR of NOx with CH4. Appl. Catal., B 2001, 29 (1), 35−46. (74) Kubacka, A.; Janas, J.; Sulikowski, B. In/Co-Ferrierite: A Highly Active Catalyst for the CH4-SCR NO Process under Presence of Steam. Appl. Catal., B 2006, 69 (1), 43−48.

11810


Optimizing Washcoating Conditions for the Preparation of Zeolite

Recommend Documents