Methanol Oxidation over Co- and/or Ag-Based Catalysts: Effect of Impurities (H₂O and CO)

1. Introduction

Methanol (MeOH or CH₃OH) has received considerable attention in different fields worldwide, due to its low production cost, safety and properties [1,2]. Industrially, methanol is produced by synthesis gas and from other carbon-based feedstocks and processes, such as coal gasification, biomass gasification and direct synthesis from CO₂ [3,4].

In petrochemical industry, methanol is used as a feedstock for the production of chemicals, such as formaldehyde [5], acetic acid [6], plastics and resins, like poly(ethene) and poly(propene) [7,8]. Furthermore, CH₃OH is considered an excellent candidate for hydrogen generation and has received global attention as an alternative environmentally friendly fuel in industries, transportation, shipping and fuel cell applications, due to its lower flammability and its toxicity, which is comparable to or better than gasoline [3,9,10,11,12,13]. Some of the properties and characteristics that make methanol an attractive molecule is that it is liquid at standard temperature and pressure (STP), which means it can be easily stored and transported, with low carbon-to-hydrogen molar ratio, low combustion temperature, high flame speed and elevated knock resistance [14].

From an environmental perspective, methanol has many benefits, such as lower greenhouse gas emissions compared to gasoline and diesel and reduced air pollution. CH₃OH combustion produces less particulate matter, soot, SOx and NOx [15]. However, it is worth mentioning that partial methanol combustion can lead to the emission of other harmful products, such as formaldehyde (HCHO), carbon monoxide (CO) and unburned methanol vapors [16]. The complete methanol oxidation reaction (MOR) produces CO₂ and H₂O, so it is a great challenge to develop catalysts resistant to water vapor, as it can affect the active sites of the catalysts [17]. Hence, the composition of the exhaust gases can be complex and may contain impurities, like unburnt methanol, CO, H₂O and CO₂ [18,19,20,21]. Therefore, it is very important to develop a catalyst with the appropriate properties that will allow the complete oxidation of methanol even in the presence of these impurities.

Transition metals (Co, Fe, Mn, Au, Ag, etc.) based on a support (Al₂O₃, CeO₂, LiO₂, etc.) seem to be promising catalysts for the oxidation of volatile organic compounds (VOCs) [6,21,22,23,24]. The redox properties, the metal-support interaction and the promotion with noble metals (Pd, Pt) play a crucial role for the catalyst performance [20,21,25,26]. Cobalt (Co) due to its electronic and magnetic properties is used in a variety of reactions in energy and environmental sectors [27]. In methanol oxidation reaction, the mechanism of the reaction depends on the oxidation state of cobalt (Co₃O₄, CoO) and the CH₃OH to O₂ ratio. Zafeiratos et al. [24] revealed that over Co₃O₄ species, methanol is oxidized to CO₂ and H₂O, over CoO partial oxidation to HCHO takes place, while metallic cobalt is oxidized to CoO in methanol rich environments. Silver (Ag) on the other side is widely used in various selective catalytic oxidation reactions carried out in the chemical industry, like catalytic oxidation of methanol to HCHO [28,29]. Jablonska et al. [30] tested different metals, like Cu, Mn, Ag and bimetallic systems on γ-Al₂O₃ support for methanol incineration and concluded that the samples loaded with 1wt.% Ag presented the highest activity due to the dispersed Ag⁺ species on alumina.

Among the studied catalytic supports, CeO₂ has received considerable attention due to its unique characteristics, such as high oxygen storage capacity, oxygen mobility and redox properties [20,31,32]. On the other side, γ-Al₂O₃ is cheaper, with high surface area, while the synergy potential of metal active sites - support, in combination with the dispersion of metal species on alumina tend to enhance the final catalytic efficiency [33]. Noble metals, especially Pt and Pd have also shown high activity for oxidation reactions at low temperature, due to the synergy of noble metal with the support, which increased the reducibility of the catalyst [21,25,34,35]. For the combustion of VOCs, Pd-based catalysts are preferred, since they present also resistant to moisture and sintering [35,36,37,38].

In a recent study [39], we have investigated the catalytic oxidation of CO and CH₃OH over Pd/Co-alumina catalysts. In detail, the catalysts were prepared using different cobalt loadings (1wt.% and 5wt.%) and synthesis methods (the conventional wet impregnation WI and the advanced spray impregnation SI technique). The superiority of spray impregnation method was clear for both oxidation reactions, due to the formation of core-shell catalytic nanostructures and the deposition of Co₃O₄ on the outer surface of the γ-Al₂O₃ support, securing the chemical/thermal stability of active sites on the catalyst carrier. Palladium (0.5wt.%) incorporation on 5 wt.% Co/γ-Al₂O₃ prepared by spray impregnation boosted the MOR, achieving a light-off temperature of 48 °C (T₅₀) and a complete conversion temperature of 64 °C (T₉₀).

The present work aims to investigate further the performance of doped Co-based catalysts on MOR, via incorporation of noble metals (0.1wt.%, 0.5wt.% loading), like Pt or Pd and transition metals, such as Ag (2wt.%). Additionally, Ag-based alumina catalysts were prepared applying either wet and/or spray impregnation methods, to compare them with bare Co-based catalysts. The optimum catalyst was further tested concerning its stability in the presence of impurities, such as H₂O and CO, and time-on-stream. The catalysts were fully characterized fresh (as prepared) and used (after reaction) in order to evaluate the effect of the synthesis method, reaction conditions and impurities on the catalytic properties.

2. Results

2.1. Textural and Structural Properties of Fresh Catalysts

Table 1 presents the textural properties of all as received/prepared catalytic materials (γ-Al₂O₃ support, Ag- and/or Co-based alumina catalysts, doped with Pt or Pd), such as metal content, surface area, pore volume, pore size and the crystallite size of the formed Co₃O₄, calculated via Scherrer equation from the peak at 2θ: 35-38° from the XRD patterns (Figure 1a). The metal content measured by ICP, revealed a Co content of approximately ~ 5.5 wt.%, while the high and low Pd-doped catalysts contained ~ 0.58 wt.% and 0.12 wt.% respectively and the corresponding Pt loading was ~0.6 wt.% and 0.17 wt.%. The silver content ranged between ~ 5.5 - 5.9 wt.%, with ~ 0.47 wt.% Pd. It is worth mentioning the difference between the Pd loadings in the 0.5Pd-5Co/Al-SI and 0.5Pd-5Ag/Al-SI catalysts (0.58 wt.% and 0.47 wt.%, respectively). Regarding surface area, as expected, the addition of metals over alumina reduced the surface area of the carrier and increased the pore size.

In the case of Co-based catalysts, the crystallite size of Co₃O₄ was estimated from the XRD diffractograms (Figure 1a). In addition to the classic peaks of γ-alumina carrier (at 2θ: 37.2°, 45.6° and 67.2° based on JCPDS reference no. 00-010-0425 for γ-Al₂O₃), the Co-based catalysts exhibit intense peaks due to the formation of Co₃O₄ at 2θ: 19°, 31.3°, 36.8°, 44.8°, 55.6°, 59.3° and 65.2° [40]. For the undoped Co/Al catalyst prepared via the SI method, the crystallite size was calculated to be 36.7 nm, while the addition of Pt decreased the crystallite size to 35.1 nm and 31.8 nm, for the catalysts with 0.1 wt.% and 0.5 wt.% Pt, respectively. Further reduction in the size of Co₃O₄ was revealed for the Pd-doped cobalt catalysts, with crystallite sizes 31.5 nm for the 0.1Pd-5Co/Al-SI and 30 nm for the 0.5Pd-5Co/Al-SI. One the other hand, the XRD patterns of the Ag-based catalyst (Figure 1b) did not reveal any peak due to any silver phase (Ag₂O, Ag⁰), possibly due to the efficient dispersion of Ag over the alumina carrier or the formation of small Ag-based particles. Even the different synthesis method did not affect the metal deposition on γ-Al₂O₃. Furthermore, the XRD pattern of the Ag-doped Co catalyst (2Ag-5Co/Al-SI), did not show any Ag peak, while the Co₃O₄ crystallite size was the highest (38 nm) measured. For the noble metal - doped catalysts (Pt or Pd), only one low intensity peak was observed for the catalysts with a Pd content of 0.5 wt.%, at 2θ: 33.8° attributed to the fromation of PdO phase [41].

So far, the synthesis method (WI vs. SI) has not shown any effect on the properties of the derived silver-based catalysts. As well known, SEM images provide a better understanding of the metals deposition on alumina carrier. Indeed, in our previous study [39], the SEM image of 5Co/Al-SI revealed the formation of a core–shell nanostructure, where Co₃O₄ was deposited on the outer surface of alumina particle. Besides Pd addition at different loadings, the catalyst, in the current study, was also doped with Ag (2 wt.%) and Pt (0.1 wt.% and 0.5 wt.%Pt).

In the case of Ag-based catalyst prepared by WI and SI methods, no significant differences were observed in SEM images concerning Ag deposition on the alumina carrier (Figure 2a and 2b, respectively). The silver deposition occurred homogeneously on the alumina particle, without creating the core-shell structure formed via spray impregnation method, as in the case of 5Co/Al-SI catalyst (Figure 2c). This fact is probably related to the mobility of Ag over the alumina particle during synthesis and more specifically during the thermal treatment (calcination at 500°C under air flow) of the material. Keijzer et al. [42] showed an increase in Ag particle size during oxidative thermal treatment up to 400°C. Even during a reaction/process, such as ethylene epoxidation [42] or methanol oxidation to formaldehyde [43], changes in the morphology of the silver active sites are possible due to the mobility of Ag species. SEM images of the 2Ag-5Co/Al-SI catalyst revealed the formation of a core-shell structure of Co deposition on alumina carrier, while Ag species (via dry impregnation method) were widely and uniformly dispersed on alumina carrier.

2.2. Reducibility of Fresh Catalysts

Structural characterization revealed the formation of Co₃O₄ species over cobalt-based catalysts, while in the case of silver-based catalysts the oxidation state of Ag is not clear. Thus, the reducibility of the Ag-based catalysts was investigated via TPR-H₂ and the reduction profiles are presented in Figure 3. The corresponding reduction profiles for the cobalt catalysts were presented in Iliopoulou et al. [39], where two main reduction peaks were observed, related to the reduction of Co₃O₄ to CoO (low temperature peak) and the reduction of CoO to Co⁰ (high temperature peak) respectively. Moreover, the addition of Pd shifted the reduction profiles to lower temperatures, indicating a significant increase in the reducibility of the catalysts.

As shown in Figure 3, the 5Ag/Al-WI catalysts exhibit two low intensity peaks at 110°C and 170°C, and two broad peaks at 300°C and 600°C. The low temperature peaks are related to the reduction of large Ag₂O clusters, while at higher temperatures (> 300°C) the reduction of small Ag₂O clusters takes place. The silver catalyst prepared with SI method (5Ag/Al-SI) and the Pd-doped catalyst (0.5Pd-5Ag/Al-SI) present similar reduction profiles, with some differences in the intensity of the peaks. A main, low temperature, broad peak with A maximum at 100-130°C and two broad peaks centered at 250-300°C and at temperatures above 500°C are observed, attributed to large Ag₂O clusters and well-dispersed smaller Ag₂O [44,45].

2.3. Methanol Oxidation Reaction

2.3.1. Effect of Noble and Transition Metals on Co-Based Catalysts

Methanol oxidation performance of doped cobalt-based catalysts is shown in Figure 4. The superiority of the 0.5Pd-5Co/Al-SI catalyst is obvious, achieving 50% methanol conversion at 43°C (T₅₀), while the decrease of the Pd content in the Pd-Co catalyst to 0.1wt.% (0.1Pd-5Co/Al-SI) shifted the conversion plot to higher temperatures with T₅₀: 95°C, maintaining the superiority even compared to the Pt-doped Co/Al catalyst (Pt content of 0.5wt.%) presenting T₅₀: 119°C. Decreasing the Pt content to 0.1wt.% did not affect the catalyst performance, presenting almost similar plots (T₅₀: 126°C). On the other hand, the incorporation of 2 wt.% Ag did not improve the methanol oxidation activity of 5Co/Al-SI, but remained the same or slightly reduced it, shifting T₅₀ from 191°C for 5Co/Al-SI to 198°C for 2Ag-5Co/Al-SI. In the literature, the combination of silver and cobalt on alumina carrier resulted in synergistic interaction of these metals, demonstrating high activity for CO oxidation reaction, which was affected by catalyst synthesis method [46]. In the methanol oxidation reaction, the combination of Ag and Co did not have the same effect. Probably the deposition of Co on the outer surface of the alumina particles (when applying spray impregnation) and then the incorporation of 2wt.% Ag using the dry impregnation method, dispersed silver all over the alumina particles as revealed SEM images (Figure 2c and d), without creating the expected synergy between Co-Ag.

Figure 5 present the MOR products for 0.5Pt-5Co/Al-SI, 2Ag-5Co/Al-SI and 0.1Pt-5Co/Al-SI (Fig. 5a, 5b and 5c, respectively). Different reaction pathways appear to be followed by each catalyst. The Pt-doped cobalt catalysts (same products despite different Pt loading) at high reaction temperatures of 225-300°C oxidized methanol to CO₂ and H₂O, while at lower temperatures (T < 225°C) dimethyl ether and formaldehyde were produced from MOR. The addition of Ag on Co/Al catalyst presented similar products with bare 5Co/Al-SI catalyst [39], but with lower concentration of DME and similar concentration of HCHO. This implies that the reaction follows the same mechanism with 5Co/Al-SI catalyst. The catalyst with low Pd content of 0.1wt.% (Fig. 5c) mainly produce CO₂ and H₂O during methanol oxidation reaction at the temperature range of 115-300°C, while at lower temperatures the formation of HCHO (maximum concentration of 55 ppm) was observed in higher concentrations than the corresponding catalyst with 0.5wt.% Pd, where traces of formaldehyde were detected [39]. The different products revealed different reaction mechanisms for each catalyst and for this reason bare 5wt.% Ag on γ-Al₂O₃ was prepared (applying both wet and spray impregnation methods), in order to study silver performance in MOR.

2.3.2. Effect of Synthesis Method and Pd-Doping on Ag-Based Catalysts

Silver is widely used for methanol oxidation to formaldehyde, but also has studied for the oxidation of volatiles organic compounds to CO₂ and H₂O, like formic acid, acetic acid, ethanol, etc. [47]. Thus, instead of using silver as a dopant for Co/Al catalyst, Ag-based alumina catalysts were prepared applying both wet and spray impregnation methods. In Figure 6, it is obvious that there is no significant difference in the methanol oxidation performance of the differently prepared silver catalysts, as the 5Ag/Al-WI and -SI catalysts present similar conversion plots. These results agree with their structural characterization (XRD patterns in Figure 1b and SEM images in Figure 2a, b), where the synthesis method did not affect catalysts’ properties, presenting similar structures. In the case of Co-based catalysts, important differences were observed between the catalysts and the difference in T₅₀ was ~ 23°C between the two differently synthesized Co-based samples [39], while in silver-based catalysts is only 10°C. However, the silver-based catalysts present higher methanol oxidation performance than cobalt-based catalysts, with 50% MeOH conversion at 137°C for 5Ag/Al-SI and at 191°C for 5Co/Al-SI.

Incorporation of 0.5wt.% Pd on 5Ag/Al-SI catalyst shifted the conversion plots to lower temperatures, enhancing the oxidation performance of bare silver catalysts, achieving T₅₀ at 81°C. As compared with the doped-cobalt catalyst, the 0.5Pd-5Co/Al-SI (T₅₀: 43°C) presented the highest activity, followed by the 0.5Pd-5Ag/Al-SI (T₅₀: 81°C) and the 0.1Pd-5Co/Al-SI (T₅₀: 95°C). It seems that Pd (0.5 wt.%) improved the performance of the Co-based catalyst more than the Ag/alumina catalyst, despite the higher Pd content in the catalysts as shown in Table 1 (0.58 wt.% Pd for Co/Al and 0.47 wt.% Pd for Ag/Al), as the difference in ΔT₅₀ between the 0.1Pd-5Co/Al-SI catalyst with 0.12 wt.% Pd content and the 0.5Pd-5Ag/Al-SI catalyst is 14°C, while between 0.5Pd-Co/Al-SI and 0.5Pd-5Ag/Al-SI is 38°C.

In the case of Ag-based catalysts different reaction mechanism seemed to take place based on the products of MOR (Figures 7). The production of CO₂ and H₂O was observed at the high temperature area (175-300°C), while at lower temperatures formaldehyde was detected, without the production of DME. The incorporation of Pd (Figure 7c) into the 5Ag/Al-SI catalyst changed the reaction mechanism and appears to be similar to the corresponding 0.5Pd-5Co/Al-SI catalyst, where when the methanol conversion started to decrease (T < 100°C) formaldehyde production was observed during MOR. So, it seems that the increase in catalytic activity of 5Ag/Al-SI catalyst was likely due to Pd, with Ag promoting formaldehyde formation.

2.4. Effect of Impurities (H₂O and CO) and Catalysts Stability with Time-on-Stream

2.4.1. Effect of CO and/or H₂O on Optimum 0.5Pd-5Co/Al-SI Catalyst Activity

The catalyst with the highest activity 0.5Pd-5Co/Al-SI was further tested for methanol oxidation reaction in the presence of CO and/or H₂O. Figure 8 presents the effect of the impurities on catalyst activity during light-off experiments. The addition of 2 vol.% H₂O in the feed (Feed 2) shifted conversion plots at higher temperatures and increased T₅₀ from 43°C to 97°C, possibly due to the competitive adsorption of methanol and water on catalyst active sites, resulting in partial loss of activity [33]. The Feed 3, except CH₃OH, O₂ and H₂O contains 0.1vol.% CO and the catalytic performance of 0.5Pd-5Co/Al-SI is further diminished, presenting 50% methanol conversion at 203°C. In the presence of CO, it seems that CO and methanol are competing further for the catalyst’s active sites, and probably the adsorption of methanol is poorer than CO, based on the CO conversion results (Figure 8). At T < 300°C, methanol conversion started to decrease, while CO conversion was increased achieving almost complete oxidation (~96%) at 220°C. This catalyst has been also tested for CO oxidation, showing high activity [39]. The coexistence of CO and MeOH (along with O₂ and H₂O) indicates a weaker sorption of methanol on the active sites of the 0.5Pd-5Co/Al-SI catalyst, as compared to the CO adsorption, which seems to be stronger, hindering the methanol adsorption.

2.4.2. Catalyst Stability with Time-on-Stream

Figure 9 shows the stability performance of the 0.5Pd-5Co/Al-SI catalyst for 12h TOS, under (Feed 3) 0.1 vol.% CH₃OH-0.1 vol.% O₂-2 vol.% H₂O-0.1 vol.% CO and at 230°C. The reaction temperature was chosen based on the catalyst performance in Figure 8 (under Feed 3), where the methanol conversion was approximately 75%. In the first 1h of the reaction, the methanol conversion was 80% and with TOS the conversion increased, reaching 94% after 9h MOR. In the last 3h of the reaction (t: 9-12h), the methanol conversion was increased further, achieving complete methanol conversion (100%). CO conversion ranged between 97-100% during the experiment and the main reaction products were CO₂ and H₂O, while traces (less than 3 ppm) of DME and formaldehyde were detected (Figure 10).

2.5. Physicochemical Characteristics of Used Catalysts

The improvement of the activity of the 0.5Pd-5Co/Al-SI catalyst during the stability experiment was investigated by characterizing the used catalyst via TPR-H₂ and XRD methods. The reduction profiles of fresh and used (after stability experiment) Pd-Co/Al-SI catalysts are shown in Figure 11. The fresh catalyst presents three peaks, with maximum at 160°C, 310°C and at 800°C. The first two peaks are assigned to the reduction of Co₃O₄ to CoO and then to metallic Co⁰, and are shifted at lower temperatures compared to bare Co₃O₄ [32] or Co/γ-Al₂O₃ reduction profiles [39], due to the incorporation of Pd which enhance catalyst reducibility. The high temperature peak is assigned to the reduction of Co-Al species, like CoAl₂O₄, where there is a strong interaction of cobalt with alumina carrier [48]. The reduction profile of used catalyst presents the same peaks with the fresh one, but with differences in the intensities of H₂ consumption peaks. The intensity of the first peak (reduction of Co₃O₄ → CoO) is similar with the fresh one, while the intensity of the second peak (reduction of CoO → Co⁰) is higher, which means that more CoO species existed in the used sample than the fresh one, which were formed in situ during the stability experiment.

The XRD patterns of fresh and used are compared in Figure 12 and confirm the conclusions from the reduction profiles. Significant differences are identified in the XRD diffractograms related to the oxidation state of both Co and Pd. The fresh catalyst, in addition to the peaks of γ-Al₂O₃, also showed peaks due to the formation of Co₃O₄ and PdO. The used catalyst present lower intensity Co₃O₄ peaks and no any PdO peak, while new peaks were formed due to CoO phase, at 2θ: 42.4°, 61.7°, 73.5° and 77.5° [49]. It seemed that under the reaction conditions (0.1 vol.% CH₃OH - 0.1 vol.% O₂ - 2 vol.% H₂O - 0.1 vol.% CO) and with TOS, significant structural changes occurred. Thus, it is likely that the active phase of the 0.5Pd-5Co/Al-SI catalyst for MOR is a mixture of Co₃O₄ and CoO phases. Zafeiratos et al. [24] studied two important parameters for methanol oxidation reaction, (i) the active phase of cobalt and (ii) the CH₃OH/O₂ mixing ratio. The methanol to O₂ ratio strongly affected the oxidation state of Co, which is a mixture of CoO and Co₃O₄ (CoOx) and the reaction mechanism.

3. Discussion

The doping of Co/γ-Al₂O₃ catalyst, prepared with spray impregnation method, with noble metals (Pd or Pt) and/or a transition metal (Ag) revealed that each “combination” results in a different reaction pathway. Methanol desorption study offered an insight into the MOR mechanism, where the results are in full agreement with the results from catalysts’ evaluation performance. Figure 13 (a-d) shows the MS signals of CO₂ (a-b), DME (c) and HCHO (d) desorbed from the saturated with methanol catalysts, under 0.1 vol.% O₂/He. The CO₂ desorption profiles of Co- (Figure 13a) and Ag-based (Figure 13b) catalysts present higher intensities as compared with DME and HCHO profiles. Bare 5Co/Al-SI catalyst exhibits one, low intensity, CO₂ peak at high temperatures (T > 220°C), similar with the 2Ag-5Co/Al-SI catalyst. The Pt- and Pd-doped cobalt catalysts present CO₂ desorption during the whole temperature range (RT – 300°C). The Pt-Co/Al catalyst present two temperature areas of CO₂ desorption, at 50 – 140°C and at 160 – 300°C, while Pd-Co/Al desorbed CO₂ during the whole range, with different peak intensities.

Bare 5Ag/Al catalyst prepared either with WI or SI method presents similar desorption profiles of CO₂ (T > 200°C), while the addition of Pd, desorbed CO₂ at lower temperatures starting from 100°C. The signal of DME (Figure 13c) was detected for 5Co/Al-SI, 2Ag-5Co/Al-SI and 0.5Pt-5Co/Al-SI catalysts, with similar intensities and temperature area (RT – 200°C), while formaldehyde (Figure 13d) signal was detected for all the catalysts, with higher intensities for Ag-based catalysts and 0.5Pt-5Co/Al-SI.

Based on the literature, the initial step (Ι) during MOR is the formation of methoxy species (CH₃O-) through CH₃OH adsorption and dissociation on catalyst [24,33,50], while the next step and the different intermediates of MOR strongly depend on the metal and metal oxidation state [24]. Thus, three different reaction pathways were observed during the current study (Figure 14). The first one (A), is the complete oxidation of methanol to CO₂ and H₂O through dehydrogenation of methoxy species (complete oxidation), which takes place on Pd-doped catalysts (0.5Pd-5Co/Al-SI and 0.5Pd-5Ag/Al-SI), where the main reaction products were CO₂ and H₂O [43]. A second pathway (B), is the reaction of methoxy species with another methanol molecule (CH₃OH dehydration) producing DME, which was observed for 5Co/Al-SI, 2Ag-5Co/Al-SI and 0.5Pt-5Co/Al-SI catalysts [33,51]. The last one (C) pathway that may take place during MOR is the dehydrogenation of methoxy species to intermediates like formaldehyde, which can be followed by dehydrogenation/oxidation of intermediates to CO₂ and H₂O, was observed for Ag/Al and Pt-Co/Al catalysts [33,43].

It seems that Pd enhances the initial mechanism step, the formation of more methoxy species and in the case of the Ag catalyst it promotes pathway (C). On the other side, the characterization of the Pd-Co/Al catalyst revealed an interaction between the two metals, where Pd also improved the reducibility of the catalyst and reduced the size of the Co₃O₄ crystallites to 30 nm (Table 1), which leads to complete oxidation of methanol without the formation of intermediates. However, stability experiments revealed that the active cobalt phase is a mixture of Co₃O₄ and CoO, as revealed by the characterization of the used catalyst.

4. Materials and Methods

4.1. Catalyst Preparation

Different synthesis methods were used to prepare catalytic materials. Wet impregnation (WI) and spray impregnation (SI) were used to impregnate 5wt.% of the transition metal Co or Ag on alumina support (supplied by Saint Gobain NorPro), while incipient wetness impregnation method (dry impregnation, DI) was used for the incorporation of the transition metal Ag (2wt.%) and of low amounts (either 0.1wt.% or 0.5wt.%) of noble metals, such as Pd and Pt. The precursor salts were Co(CH₃CO₂)₂∙4H₂O (purchased from Sigma-Aldrich with purity > 99%), AgNO₃ (supplied by Honeywell-Fluka and purity ≥ 99.8%), Pd(NO₃)₂∙2H₂O (from Sigma-Aldrich) and Pt(NH₃)₂(NO₂)₂ (from Sigma-Aldrich). A detailed description of the synthesis methods of Co-alumina based catalysts doped or not with Pd, is presented in Iliopoulou et al. [39]. The same methods and conditions were used for the synthesis of bare Ag- alumina-based catalysts, doped or not with Pd. The catalysts are labeled as xN-5M/Al-z, where x is the loading of the noble/transition metal (0.1wt.%, 0.5wt.% or 2wt.%), N the noble/transition metal (Pd, Pt or Ag), 5M the transition metal (5wt.% Co or Ag) and z the preparation method (WI or SI). The 5Co/Al-SI and 0.5Pd-5Co/Al-SI catalysts used in the present study constitute the second batch of catalysts, which were characterized, evaluated for MOR reaction and compared with the results of the first batch presented in the previous publication, showing great repeatability during the methanol oxidation reaction [39].

4.2. Catalyst Characterization

All catalytic materials were fully characterized regarding their textural and structural properties, via Inductively Coupled Plasma–Atomic Emission Spectroscopy (ICP-AES) in order to measure the metal content (wt.% Co, Ag, Pd and Pt). The surface area, the volume and size of pores of the catalysts were measured in a 300 DV PerkinElmer Optima spectrometer, via N₂ adsorption/desorption experiments (BET method) at −196°C, using an Automatic Volumetric Sorption Analyzer (Nova2200e Quantachrome flow apparatus) . The samples were outgassed overnight at 250°C under vacuum. The X-Ray Diffraction (XRD) diffractograms were accumulated in the range of 10-80° 2θ, with a counting time of 2 sec per step, using a SIEMENS D-500 diffractometer, employing CuKa1 radiation (λ = 0.15405 nm) and operating at 40 kV and 30 mA, in order to determine the structure of the catalysts and the formed metal oxide species (fresh and used catalysts). Scherrer equation (1) was used to calculate the crystallite size of Co₃O₄ phase (D_XRD), based on the most intense diffraction peak of Co₃O₄ at 2θ: 35-38°.

$$D_{XRD}={\displaystyle \frac{K\lambda }{\beta cos\theta }}$$

(1)

Where K is Scherrer constant, λ is the wavelength of the X-ray in nm, β is the line broadening and θ is the Bragg angle [52].

Morphological characterization was performed via Scanning Electron Microscope (SEM) imaging, using a JEOL JSM-IT500 microscope. EDS spectra and mapping were obtained using an Oxford Instruments x-Act detector. The samples were embedded in resin, which was grinded and polished, while each resin was gold-plated. The analysis was carried out by applying a voltage of 20 kV.

The reaction mechanism was investigated via temperature-programmed desorption of methanol (TPD-CH₃OH) under 0.1vol.% O₂/He. Initially, the catalysts were saturated with liquid methanol and then the desorption was performed in a fixed bed reactor connected with a mass spectrometer (MS), starting from room temperature (RT) up to 300°C (with a heating rate 10°C/min). The recorded MS signals were: m/z = 31 for CH₃OH, m/z = 45 for DME, m/z = 29 for HCHO and m/z = 44 for CO₂.

Temperature-programmed reduction with H₂ (TPR-H₂) was also used to investigate the reducibility of the catalysts and identity the oxidation states of the metal phases in the fresh and used (after reaction) catalysts. The experiments were performed in the same unit with the desorption studies, while the detailed methodology is described in [39].

4.3. Performance Evaluation

The evaluation of catalysts performance for the methanol oxidation was performed in a bench-scale unit, equipped with a quartz fixed-bed reactor. The reactor was loaded with 0.4 g, using a total gas flow of 600 cm³/min, corresponding to 40,000 h^-1 Gas Hour Space Velocity (GHSV). All catalysts were initially screened for methanol oxidation with feed composition: 0.1 vol.% CH₃OH and 0.1 vol.% O₂ balanced with He (Feed 1), in the temperature range: 30-300°C (decreasing temperature mode). The composition of the effluent gases was analyzed using an MKS FT-IR gas analyzer (MG2030, Germany), monitoring continuously CH₃OH and possible reaction products, such as CO₂, H₂O, DME, HCHO, CO, etc. Methanol conversion (X _MeOH, %) was calculated by the following equation:

$$X_{MeOH} = {\displaystyle \frac{{\left[MeOH\right]}_{in }- {\left[MeOH\right]}_{out}}{{\left[MeOH\right]}_{in}}} \times 100$$

(2)

Where [MeOH]_in and [MeOH]_out are the methanol concentration (ppm) in the inlet and outlet gas streams, respectively.

The optimum catalysts were further evaluated to test their stability in the presence of impurities (CO and/or H₂O) and time-on-stream (TOS). The feed compositions were: 0.1 vol.% CH₃OH- 0.1 vol.% O₂-2 vol.% H₂O (Feed 2) and 0.1 vol.% CH₃OH- 0.1 vol.% O₂-2 vol.% H₂O-0.1 vol.% CO (Feed 3) balanced with He. The light-off experiments were performed at the same unit and reaction conditions, while the TOS experiments were performed at a constant temperature (T: 230°C) for several hours (TOS: 12h). Water was introduced to the blend of the other gases using a saturation bath at controlled temperature conditions of He flow.

5. Conclusions

The present study moved forward the investigation of Pd-Co/alumina catalysts for the catalytic oxidation of atmospheric pollutants [39], focusing on the methanol oxidation reaction and exploring/improving further the performance and stability of the catalyst. The effect of the noble metal (Pd vs. Pt) and its content (0.5 wt.% vs. 0.1 wt.%) was investigated and the results revealed the superior performance of Pd versus Pt, as even at 0.1 wt.%, Pd had a higher catalytic activity than 0.5wt.% Pt on 5Co/Al-SI catalyst. Different reaction pathways were also followed for each noble metal. The Pd-Co/Al-SI catalyst promoted complete oxidation of methanol to CO₂ and H₂O, while the Pt-Co/Al-SI promoted the partial oxidation to formaldehyde and dehydration to dimethyl ether.

The addition of 2 wt.% Ag on 5Co/Al-SI did not improve its catalyst activity, while the bare 5Ag/Al catalyst showed higher activity than the 5Co/Al catalyst. The synthesis method did not seem to play a significant role in the case of the Ag-based catalyst, as Ag exhibits mobility during thermal treatments. The incorporation of Pd into the Ag/Al-SI catalyst further increased the oxidation activity of the catalyst, but did not achieve better performance than that of Pd-Co/Al-SI. This is probably related to the interaction and synergistic effect between Pd-metal according to the reduction profiles of the Co-based catalyst, while in the case of Ag-based catalyst palladium seems to enhance the formation of methoxy species, but Ag promote the formation of formaldehyde.

Therefore, the stability of the optimum 0.5Pd-5Co/Al-SI catalyst was finally tested in the presence of CO and/or H₂O. The catalyst performance decreased during light-off experiments, but over time the catalyst activity improved, achieving complete methanol conversion (100%) after 12 h of MOR at T: 230°C. The characterization of the used catalyst revealed that during the stability experiment, in addition to Co₃O₄, the CoO phase was also formed, revealing that the active phase of the catalyst is a mixture of Co₃O₄ and CoO.