Investigation of Iron Vanadates for Simultaneous Carbon Soot Abatement and NH₃-SCR

Abstract

FeVO₄ and Fe_0.5Er_0.5VO₄ were prepared and loaded over standard Selective Catalytic Reduction (SCR) supports based on TiO₂-WO₃-SiO₂ (TWS) and redox active supports like CeO₂ and CeZrO₂ with the aim of finding a suitable formulation for simultaneous soot abatement and NH₃-SCR and to understand the level of interaction between the two reactions. A suitable bi-functional material was identified in the composition FeVO₄/CeZrO₂ where an SCR active component is added over a redox active support, to increase carbon oxidation properties. The influence of the presence of ammonia in soot oxidation and the effect of the presence of soot on SCR reaction have been addressed. It is found that the addition of NO and NO/NH₃ mixtures decreases at different levels the oxidation temperature of carbon soot, while the presence of carbon adversely affects the NH₃-SCR reaction by increasing the oxidation of NH₃ to NO, thus lowering the NO removal efficiency.

1. Introduction

The study of Diesel exhaust after-treatment technologies is focusing on one side in finding novel formulations for active soot oxidation [1] and de-NO_x catalysts [2,3,4] and on the other side in looking for suitable strategies to limit the number of different process units and combine more functionalities in one single step [5]. Nowadays, the mobile exhaust gas after-treatment is comprised of a series of stand-alone devices, starting from an oxidation catalyst (DOC), a diesel particulate filter (DPF) and a de-NO_x reactor which includes a NO_x storage/reduction system or, alternatively, a NH₃-SCR (Selective Catalytic Reduction) unit. Among the NO_x removal technologies, Selective Catalytic Reduction with NH₃ (NH₃-SCR) has been successfully employed both in stationary and heavy duty mobile applications [6].

Increasingly restrictive limitations concerning emissions of nitrogen oxides and soot particulates from vehicle engines will demand the simultaneous application of a particulate filter and a SCR de-NO_x system. The motivation for the integration of the two reactions is to obtain a reduction of cost and volume of the engine after-treatment system via inclusion of SCR and DPF functionalities into a single entity. Different bi-functional configurations are possible with DPF and SCR catalyst placed in the same housing [7], or with a DPF filter coated with an SCR catalyst formulation [8,9,10].

The aim of this work is to study the behaviour of a suitable catalyst with acceptable performances both in soot oxidation and in the NH₃-SCR reaction conducted simultaneously. For this purpose, an active SCR catalyst formulation has been loaded over supports showing different soot oxidation functionalities. SCR catalyst materials were selected among metal/rare earth vanadates that have been shown to act as efficient catalyst precursors for NH₃-SCR reaction [11,12,13,14,15,16], while among the different supports, standard TiO₂-WO₃-SiO₂ (TWS) was used and compared with more efficient carbon oxidation carriers like CeO₂ and Ce_0.75Zr_0.25O₂ [17]. In particular, this work investigates FeVO₄ and Fe_0.5Er_0.5VO₄ supported on TWS, CeO₂, Ce_0.75Zr_0.25O₂ (CeZrO₂), and Al₂O₃ in the simultaneous soot oxidation and NH₃-SCR reactions. Specific attention is given to the influence of all the components of the complex atmosphere under which these catalysts operate and to the study of the interactions between the two reactions. A promising catalytic system is proposed as a suitable candidate for both reactions.

2. Results and Discussion

2.1. Catalysts Characterization

All the samples were characterized by B.E.T. (Brunauer-Emmett-Teller) surface area measurements (reported in

Table 1. B.E.T. surface areas, hydrogen consumption in TPR, H/V atomic ratio and onset temperature of reduction for each sample.

Catalyst	B.E.T. (m2/g)	H2 Consumption (mmol/gcat)	H/V Atomic Ratio 1	Onset of Reduction (°C)
TWS	89	0.953	/	350
8.4% FeVO4/TWS	68	2.184	8.88	344
50% FeVO4/TWS	21	5.786	3.95	361
CeO2	26	0.997	/	190
8.4% FeVO4/CeO2	12	1.926	7.83	285
50% FeVO4/CeO2	6	6.379	4.36	338
Ce0.5Zr0.5O2 (CZ)	52	1.404	/	186
8.4% FeVO4/CZ	24	2.448	9.95	320
50% FeVO4/CZ	8	6.694	4.57	348
Al2O3	168	0.080	/	/
8.4% FeVO4/Al2O3	166	1.022	4.15	320
50% FeVO4/Al2O3	89	5.240	3.58	320

¹ Calculated on the basis of H₂ consumption from TPR profiles and total vanadium content of the catalyst.

and Table S1) and powder XRD analysis. As it can be observed, the specific surface area of bare supports is negatively affected by the loading of iron vanadate (B.E.T. = 4 m²/g) and iron erbium vanadate (13 m²/g). XRD patterns of samples made with FeVO₄ are reported in Figure 1A, whereas XRD diffraction patterns of supported Fe_0.5Er_0.5VO₄ are reported in Figure 1B. In the diffraction profile of 8.4 wt % FeVO₄/TWS only diffraction lines due to anatase-TiO₂ and Fe₂O₃ were observed. This is due to partial decomposition of FeVO₄ on TiO₂-based supports following thermal treatments [14,18]. This decomposition generates Fe₂O₃ plus VO_x species which at low coverage are not detected by XRD, while the remaining FeVO₄ could not be clearly identified because of superposition of its main diffraction line to the anatase main diffraction peak. In the XRD pattern of 8.4 wt % FeVO₄/CeO₂ instead, diffraction lines due to CeO₂, Fe₂O₃, and CeVO₄ are identified. In this case, the precursor FeVO₄ partially decomposes releasing Fe₂O₃ and VO_x species, which then react with CeO₂ already at 650 °C, resulting in the formation of CeVO₄ [19]. A very similar situation was detected for the 8.4 wt % FeVO₄/Ce_0.75Zr_0.25O₂ catalyst. The formation of surface VO_x species was confirmed by Raman spectroscopy on a representative 8.4 wt % FeVO₄/TWS sample calcined at 650 °C (Figure 2). We observed Raman shifts at ca. 1000 cm⁻¹ that have been attributed to monomeric or dimeric species formed upon decomposition of FeVO₄ [20,21,22]. By carefully looking at different spots of the sample, also on the remaining FeVO₄, that were not observed by XRD, was easily detected by Raman as highlighted in Figure 2. FeVO₄ decomposition takes place also on alumina as revealed by Fe₂O₃ peaks in 8.4 wt % FeVO₄/Al₂O₃ diffraction pattern. Also in this case this is accompanied by the release of non-detectable VO_x species. Residual FeVO₄ species, not detectable by XRD, could be identified by Raman spectroscopy also on 8.4 wt % FeVO₄/Al₂O₃, as shown in Figure S1.

XRD spectra of supported Fe_0.5Er_0.5VO₄ (Figure 1B) are characterized by the presence of ErVO₄ and Fe₂O₃. It was previously highlighted that Fe_0.5Er_0.5VO₄ does not form a solid solution and the two vanadates (FeVO₄ and ErVO₄) are present as an intimate physical mixture [13]. The FeVO₄ fraction of the mixed vanadate partially decomposes releasing Fe₂O₃ and VO_x, while ErVO₄, which is more thermally stable than FeVO₄ [23], can still be detected on the various supports.

For a better understanding of the effect of the vanadate-support interaction on catalytic activity, some additional investigations were conducted also on a family of catalysts made by loading 50 wt % of vanadate on the support. Their XRD diffraction profiles are reported in Figure S2 of the supplementary material; as already noticed for 8.4 wt % FeVO₄ loading, the presence of Fe₂O₃ in the XRD patterns and of VO_x in Raman spectrum (Figure S3) indicates that part of FeVO₄ is decomposed.

Catalyst reducibility was measured by temperature-programmed reduction under H₂ atmosphere (TPR). Since many reducible species are involved, complex reduction profiles were obtained, as shown in Figure 3. From XRD results it is clear that FeVO₄ undergoes decomposition over all supports, with the formation of surface Fe₂O₃, VO_x and, on ceria-containing supports, CeVO₄. On Al₂O₃, which is a non-reducible support, all contributions to reduction profile are related to Fe₂O₃, VO_x, and FeVO₄ species. The attribution of the single TPR peaks is not straightforward, but similarly to what reported for FeVO₄ supported on TWS [24], the low temperature feature can be related mostly to the reduction of Fe₂O₃ and VO_x species, whereas at increasing temperature H₂ consumption due to FeVO₄ reduction is detected. At higher loading, an increase in hydrogen consumption is observed, which has been tentatively attributed to the residual part of FeVO₄, based on quantitative analysis obtained also from XRD spectra. The onset reduction temperature is independent from FeVO₄ initial loading (Table 1). TWS reduction profile (Figure 3A) is characterized by three reduction steps of WO₃ with the main hydrogen consumption taking place in the 800–950 °C range [25]. This feature is maintained also in the supported catalyst. Hydrogen consumption at lower temperature for 8.4 wt % FeVO₄/TWS (400–700 °C) is due to reduction of the supported species VO_x, Fe₂O₃, and FeVO₄. As for Al₂O₃ samples, it is noteworthy that the onset temperature of reduction is not related to the vanadate loading (Table 1).

TPR profile of CeO₂ shows a low-T reduction feature due to surface species and a higher-T peak due to bulk reduction, as expected for pure ceria [26]. For Ce_0.75Zr_0.25O₂ one single hydrogen consumption peak is detected in the medium temperature range, in agreement with reported TPR profiles of ceria-zirconia solid solutions [27]. In the case of CeO₂-containing supports, a difference in the onset of the reduction can be detected at increasing FeVO₄ loading, in contrast to what observed for TWS and Al₂O₃-supported vanadates, indicating a strong influence of vanadate loading on sample reducibility. The temperatures of the onset of reduction are reported in Table 1 together with the overall hydrogen consumption and the H/V atomic ratio. As it can be observed from Table 1 and Figure 3, when increasing the vanadate loading up to 50 wt % the reduction peaks become more intense, the overall H₂ consumption increases but with a lower H/V ratio, indicating a better reducibility of the low-loaded samples.

2.2. Soot Oxidation Activity

Table 2. Soot oxidation characteristic temperatures of investigated catalysts.

Catalyst	T50 (°C) 1 air	Tp (°C) 2 NO/O2/N2	Tp (°C) 3 NO/O2/N2/NH3
8.4% FeVO4/TWS	520	490	480
8.4% Fe0.5Er0.5VO4/TWS	567	548	541
8.4% FeVO4/CeO2	473	459	462
8.4% Fe0.5Er0.5VO4/CeO2	460	448	440
8.4% FeVO4/CZ	459	459	447
8.4% Fe0.5Er0.5VO4/CZ	423	419	411
8.4% FeVO4/Al2O3	602	-	-
8.4% Fe0.5Er0.5VO4/Al2O3	603	-	-
FeVO4	450	-	-
Fe0.5Er0.5VO4	450	-	-
TWS	580	-	-
50% FeVO4/TWS	474	-	-
50% Fe0.5Er0.5VO4/TWS	449	-	-
CeO2	390	-	-
50% FeVO4/CeO2	493	-	-
50% Fe0.5Er0.5VO4/CeO2	493	-	-
CeVO4	495	-	-
Ce0.75Zr0.25O2 (CZ)	385	-	-
50% FeVO4/CZ	486	-	-
50% Fe0.5Er0.5VO4/CZ	488	-	-
Al2O3	612	-	-
50% FeVO4/Al2O3	546	-	-
50% Fe0.5Er0.5VO4/Al2O3	505	-	-

¹ Temperature of 50% weight loss measured in thermogravimetric temperature programmed ramps from RT up to 800 °C on catalyst samples mixed in tight contact with soot. ² Peak temperature calculated from CO₂ profiles during TPO (Temperature Programmed Oxidation) runs in a flow-trough microreactor under 500 ppm NO, 10% O₂, (balance N₂) atmosphere. ³ Peak temperature calculated from CO₂ profiles during TPO runs in a flow-trough microreactor under 500 ppm NO, 500 ppm NH₃, 10% O₂, 5% H₂O, (balance N₂) atmosphere.

shows soot oxidation efficiency of the samples investigated under air atmosphere. For 8.4 wt % loaded catalysts activity under NO/N₂/O₂ and NO/N₂/O₂/NH₃ mixtures is also reported.

Soot oxidation values (T50) of the vanadates supported on TWS indicate the low efficiency of TiO₂-based materials in this reaction, as reported in the literature [17]. Compared to the oxidation potential of pure vanadates (FeVO₄ and Fe_0.5Er_0.5VO₄) supported catalysts show an increase of T50. These formulations were compared with those obtained by loading vanadates on supports that are known to be more active for soot oxidation like CeO₂ and Ce_0.75Zr_0.25O₂ [28].

As expected, CeO₂-based materials show lower values of T50. This can be explained through the so called “active oxygen mechanism” where, thanks to the redox ability of ceria, oxygen can be activated over a vacancy and spill over the carbon soot for oxidation [17,29,30]. Conversely, the use of redox inactive supports like Al₂O₃, results in catalyst with the highest T50. A comparison of the TPR profiles reported in Figure 3 with the corresponding onset temperatures of reduction (Table 1) evidences the correlation between sample reducibility and soot oxidation behavior. Bare CeO₂ or Ce_0.75Zr_0.25O₂ are the most active materials thanks to their low-temperature surface reducibility. Even with 8.4 wt % FeVO₄ a change in the onset temperature of reducibility of ceria is observed and, consequently, soot oxidation activity is lowered. For 50 wt % FeVO₄/CeO₂, the reduction is further shifted towards higher temperatures resulting in a higher soot oxidation temperature (Table 2). The opposite is obtained for TWS materials, for which an increase in FeVO₄ loading favours activity as no detrimental effect is displayed on reducibility in the temperature range of soot oxidation.

These results clearly show that the mechanisms involved in soot oxidation on ceria and ceria-zirconia are different than those observed over TWS and alumina. This is better understood by looking at Figure 4 which shows the correlation between soot oxidation activity of pure and supported vanadates against loading. Similar results were obtained with FeErVO₄ (Supplementary material Figure S4).

For FeVO₄ supported on either Al₂O₃ or TWS the addition of FeVO₄ progressively increases activity. The effect is much more pronounced on TWS, where a strong influence is already observed at low loading. On Al₂O₃, a simple dilution effect seems to prevail, with a linear decrease in T50 going towards pure FeVO₄. From XRD analyses of these compounds (Figure 1 and Figure S2) it can be observed that both in Al₂O₃ and TWS samples the decomposition of FeVO₄ is confirmed by the presence of Fe₂O₃ diffraction lines (corresponding VO_x were detected by Raman spectroscopy on TWS support—Figure 2 and Figure S3 in supplementary material). If on TWS the role of VO_x formed is essential for soot oxidation enhancement, FeVO₄ decomposition on Al₂O₃ does not have the same beneficial effect and activity is governed simply by the dilution of the catalysts components. An opposite trend is observed with CeO₂ and Ce_0.75Zr_0.25O₂-based materials, where the highest activity is obtained with the two bare supports; the activity then decreases reaching a minimum for the 50 wt % mixtures to increase again in correspondence of the pure vanadates. This catalytic behaviour is due to formation of CeVO₄ which is less active than either CeO₂ or FeVO₄ (Table 2). For the catalyst with a higher vanadate loading, the higher amount of CeVO₄ formed (Figure S2A in supplementary material), contributes to a further decrease in activity.

The effect of addition of NO and NO/NH₃ to the gas feed stream for soot oxidation was also evaluated (Table 2). In comparison to O₂/N₂ atmosphere, the presence of NO enhances soot oxidation performances of the catalysts by lowering the peak temperature due to the formation of NO₂ with its better oxidation properties compared to O₂ [31]. In a subsequent step, 500 ppm of ammonia were added to the feed gas mixture with the overall effect of slightly lowering Tp temperature for almost all catalysts (Table 2). The positive effect of the presence of NH₃ on soot combustion efficiency using NO₂ was observed by Tronconi et al. for Cu-Zeolite catalysts [32].

2.3. NH₃-SCR Activity

NO conversion (defined as {[NO_in] − [NO_out]/[NO_in]}) in the NH₃-SCR reaction was investigated for the vanadates on the different supports. Figure 5 compares their activity in the range of temperature 200–500 °C. The NO reduction profile shows a typical volcano-type shape with NO conversion increasing with temperature up to a maximum value which is reached at ca. 300–450 °C depending on the catalyst composition. At this point, the activity starts to decrease due to the ammonia oxidation reaction which is prevailing over NO reduction to N₂. The activity of supported FeVO₄ is higher than that of Fe_0.5Er_0.5VO₄ irrespective of the support used, in agreement with our earlier findings [13]. Among the different carriers, the highest activity is found with TWS, for which maximum conversions in the range 80–100% are reported. The best performing de-soot catalysts show lower activity in the standard NH₃-SCR reaction with NO conversion values in the range 45–60% at medium SCR temperatures. Al₂O₃-based catalysts are almost inactive up to 250 °C. Overall, the NO_x conversion values coupled with the better efficiency in carbon soot oxidation at lower temperature makes vanadates supported on CeO₂ and Ce_0.75Zr_0.25O₂ as promising catalysts to investigate the interaction between the two reactions occurring simultaneously.

2.4. Simultaneous NH₃-SCR and Soot Oxidation Activity

To study the influence of soot oxidation on SCR reaction, a different set up configuration was used employing 20 mg of powder catalyst mixed in tight contact with soot. Under these conditions, the SCR reaction that is carried out at a GHSV of ca. 900,000 h⁻¹ results in a lower NO_x conversion than that observed previously. This downscale of the catalytic bed was necessary to reduce pressure drop issues originating by mixing a higher amount of catalyst with soot in powder form. Figure 6 shows that for 8.4 wt % catalysts, the comparison of NO conversion obtained under the conditions described above in SCR and SCR/de-soot reaction. In addition, the CO₂ concentration profile due to soot combustion is also reported. In agreement with the results obtained with lower space velocity (Figure 5), SCR only reaction (blue curve in Figure 6), shows a typical volcano shape with a maximum conversion in the range 350–400 °C. Iron vanadates over TWS support are the best performing catalytic materials with a maximum NO conversion in the range 40–60% while a drop to lower conversion values is observed using CeO₂ and CeZrO₂ as supports (16–18%). In correspondence of CO₂ evolution peak, a decrease of NO conversion is observed, which is much more evident for CZ based supports. Similar results are obtained with CeO₂ (supplementary material Figure S5). NH₃ evolution profiles (not shown) reveal a decrease in NH₃ concentration in correspondence to soot oxidation that indicates a large consumption of ammonia to form NO.

On pure FeVO₄ and Ce_0.75Zr_0.25O₂, negative NO conversion values are obtained (Figure 7). This indicates that NO can be generated through ammonia oxidation, and this occurs in correspondence to carbon soot oxidation. Monitoring catalyst temperature during heating ramp allowed to exclude that a local increase of temperature during carbon oxidation might affect reaction mechanism (Figure S6, Supplementary). A similar result, that better highlights the decrease of NO conversion in SCR, is found when running the fast-SCR with NO₂ present in the feed gas on bare Ce_0.75Zr_0.25O₂ mixed with soot. In this case, the high NO conversion observed already at 150 °C, drops in correspondence of soot oxidation and CO₂ release (Figure 7D).

Interestingly, a small drop in NO conversion in correspondence to soot oxidation is observed also on pure FeVO₄ and TWS (Figure 7A,B) which indicates that soot oxidation slightly affects NH₃-SCR also on these compounds. For this reason, ammonia oxidation reaction was studied for the single catalyst components. In Figure 8, NH₃ oxidation results over FeVO₄, TWS, CeO₂, Ce_0.75Zr_0.25O₂ are shown. The four different profiles compare reaction carried out with empty reactor, and with reactor filled with soot, catalyst and soot/catalyst mixtures. Compared to the empty reactor, soot alone has a minimal effect towards ammonia oxidation, as already observed by Mehring et al. [33]. FeVO₄ is the most active in the oxidation of NH₃, followed by ceria and ceria-zirconia. For FeVO₄, there is no large difference in NH₃ oxidation activity with or without the presence of soot. For CeO₂ and Ce_0.75Zr_0.25O₂ instead, the tight contact with soot (Figure 8C,D—red lines) leads to an additional NH₃ consumption in correspondence of CO₂ release due to soot combustion. This consumption is higher than the sum of contributions of catalyst and soot alone, showing a synergistic effect of the two components in NH₃ oxidation reaction and indicating that the presence of carbon soot on CeO₂ and CeZrO₂ produces a synergic effect increasing NH₃ oxidation to NO. No significant effect is displayed in NH₃ oxidation reaction by CeVO₄ which is formed on these catalysts when FeVO₄ is supported on CeO₂ as highlighted in Figure S7 in supplementary material. This effect is the responsible of the large decrease in NO conversion observed over ceria-containing materials in NH₃-SCR reaction.

3. Experimental Section

3.1. Catalyst Preparation

FeVO₄ and Fe_0.5Er_0.5VO₄ were prepared by co-precipitation of metal nitrates (Fe(NO₃)₃·9H₂O and Er(NO₃)₃·6H₂O, Treibacher Industrie AG, Althofen, Austria)) and ammonium metavanadate (99+%, Sigma-Aldrich, St. Louis, MO, USA). Aqueous solutions of the precursors were separately prepared at 80 °C. After unification, the pH was adjusted by means of ammonia (28% aqueous solution, Sigma-Aldrich) to obtain the respective precipitate which was filtered, washed and dried at 120 °C overnight. These vanadates were subsequently used to make the supported catalysts. All support used in this study were commercial materials (CeO₂ from Treibacher Industrie AG, CeZrO₂ from W.R. Grace (Columbia, MD, USA), TWS from Cristal (Hong Kong, China), Al₂O₃ from Sasol Germany GmbH (Hamburg, Germany)). Supported catalysts were prepared by mixing two separate aqueous slurries containing, respectively, the vanadate and the support. Vanadate/support weight ratio was 8.4/91.6 or 50/50. Combined slurries were evaporated to dryness. The resulting solids were dried at 120 °C overnight followed by calcination at 650 °C/2 h in a muffle furnace to obtain the final catalyst.

3.2. Catalytic Tests

Soot combustion efficiency under air was measured by means of thermogravimetry using soot (Printex U-Degussa)/catalyst mixtures under tight contact conditions obtained by mixing the two components in an agate mortar for 10 min. The soot/catalyst weight ratio was 1:20. The samples were placed on Pt crucibles licked by air flow (60 mL/min) and heated at a constant rate of 10 °C/min from RT up to 800 °C. T50 values, corresponding to the temperature at which 50 wt % of carbon is lost, was used as a measure of soot oxidation activity. Soot oxidation under NO/O₂/N₂ and NO/O₂/N₂/NH₃ was carried out in a laboratory set-up equipped with a quartz microreactor for powder testing. In these experiments 20 mg of sample in powder form (soot/catalyst ratio of 20:1, mixed in tight contact) were exposed to a 300 mL/min feed stream of composition 500 ppm NO, 500 ppm NH₃ (when present), 10% O₂, 5% H₂O, balance N₂ with a GHSV of 900,000 h⁻¹.

NH₃-SCR activity was measured in the same laboratory set-up described above on 100 mg of catalyst pressed and sieved (350 < Ø < 425 μm) under a 300 mL/min feed stream of composition 500 ppm NO, 500 ppm NH₃, 10% O₂, 5% H₂O, balance N₂ with a GHSV of 180,000 h⁻¹.

NH₃ oxidation experiments were conducted on 20 mg catalytic beds under a reactive atmosphere of composition 500 ppm NH₃, 10% O₂, balance N₂ with a GHSV of 900,000 h⁻¹. Ammonia concentration was monitored during a temperature ramp from 150 °C up to 650 °C. When soot was present, also CO₂ release due to carbon oxidation was monitored.

For all experiment conducted in the gas bench, the composition of the feed stream (reagents and products) was analyzed with an on line MKS 2030 MultiGas Analyzer FT-IR instrument.

3.3. Characterization of the Catalysts

Specific surface areas of catalysts were measured by nitrogen adsorption at -196 °C according to the B.E.T. method on a Tristar 3000 gas analyzer (Micromeritics, Norcross, GA, USA). Prior to the measurement all samples were outgassed at 150 °C for 1.5 h. Powder X-ray diffraction patterns were collected on a Philips X’Pert diffractometer (PANalytical B.V., Almelo, The Netherlands) operated at 40 kV and 40 mA using Ni filtered Cu Kα radiation. Data were recorded from 10 to 100° 2θ using a step size of 0.02° and a counting time of 40 s per angular abscissa. Phase identification was conducted with the Philips X’Pert Highscore software (PANalytical B.V., Almelo, The Netherlands, 2002). Reducibility of the samples was measured by temperature programmed reduction (TPR) runs; approximately 50 mg of samples were heated at 10 °C/min in a U-shaped quartz reactor from room temperature up to 960 °C under a 4.5 vol % H₂/N₂ flow (35 mL/min). The samples were pretreated at 500 °C for 1 h under air flow. Hydrogen consumption was monitored with a thermal conductivity detector (TCD) in an Autochem II 2920 instrument (Autochem II 2920, Micromeritics, Norcross, GA, USA). Raman spectra were collected with a Xplora Plus Micro-Raman system (Horiba, Kyoto, Japan) equipped with a cooled CCD detector (−60 °C) and Edge filter. The samples were excited with the 512 nm radiation in a Linkam CR1000 in situ cell at 200 °C under flowing air. The spectral resolution was 1 cm⁻¹ and the spectra acquisition was of 5 accumulations of 10 or 20 s with a 50× LWD objective. The optical images were collected with an integrated microscope Olympus BX43 (Olympus, Tokyo, Japan) with a 10× objective.

4. Conclusions

A systematic screening to evaluate the interaction between soot oxidation and NH₃-SCR has been carried out investigating the behavior of FeVO₄ and Fe_0.5Er_0.5VO₄ loaded on standard SCR supports like TiO₂-WO₃-SiO₂, redox active carriers like CeO₂ or Ce_0.75Zr_0.25O₂ and Al₂O₃.

It is shown by combined XRD, TPR, and Raman spectroscopy that vanadates partially decompose over TiO₂-based materials and Al₂O₃ with formation of Fe₂O₃ and surface VO_x species. In the presence of CeO₂/CeZrO₂, formation of crystalline CeVO₄ along with Fe₂O₃ is reported.

Al₂O₃-based materials are almost inactive either in SCR and soot oxidation while, as expected, TiO₂-based materials are highly active in NH₃-SCR and CeO₂/CeZrO₂ are active in carbon soot oxidation. The addition of ammonia is found to slightly influence soot oxidation by reducing its combustion temperature by a few degrees while the presence of soot is found to affect SCR reaction, especially in the presence of CeO₂ and CeZrO₂. This is due to the combined effect of the catalyst and carbon soot that results in an enhancement of NH₃ oxidation reaction.