Exploring Synthesis Approaches of Co-based Catalysts for the Efficient Oxidation of CH₄ and CO

Abstract

Co-based catalysts were synthesized and studied as novel oxidation catalysts, exploring and optimizing the effect of synthesis method on the redox behavior, the oxygen storage ability and thus the catalytic performance of the derived Co₃O₄ materials in the complete CH₄ and/or CO oxidation reactions. Thus, a series of Co-based catalysts were synthesized applying either the precipitation and/or the hydrothermal method, using different precipitating agents (Na₂CO₃, NH₃, urea or NaOH), Co precursor salt (nitrate or acetate) and finally varying the Co/Na ratio. In addition, the reaction time (6 or 24 h aging) was also investigated for the hydrothermally prepared Co₃O₄. The best catalysts for the CH₄ oxidation are the precipitated Co₃O₄, using cobalt acetate as precursor salt and NaOH as precipitating agent, presenting the highest surface areas and the lowest Co₃O₄ particle sizes. On the other side, hydrothermally prepared cobalt oxides reveal higher performance for CO oxidation, with Co₃O₄ prepared with cobalt acetate, NaOH and low aging time shown as the optimum materials. The best catalysts were further promoted with incorporation of Pd (0.5wt.%) and explored for both reactions. The addition of Pd enhanced the activity of Co₃O₄ for CH₄ oxidation, while Pd did not improve any further the catalyst performance for CO oxidation, presenting thus the same activity with pure cobalt oxides.

1 Introduction

Compressed natural gas (CNG), composed of 85–95% of methane (CH₄) has recently received increasing attention as an attractive alternative fuel for motor vehicles, as it has many advantages over traditional fossil fuels, such as gasoline and diesel; it is cheaper, has lower sulphur and nitrogen content (less SOx and NOx emissions), produces smaller amount of carbon dioxide (CO₂) due to the high H/C ratio and much lower number of particulates as compared to diesel engines [1]. Today’s heavy-duty (HD) natural gas-fueled vehicles is estimated to represent less than 2% of the total fleet. However, over the next couple of decades, predictions are that the percentage could grow to as much as 50% [2]. CNGVs under lean burn conditions present increased fuel efficiency as compared to stoichiometric conditions. Unfortunately, even with efficient lean conditions, CO and unburned HC are always released in the exhaust; about 90% of the unburned fuel is CH₄, typically equivalent to 500–1,000 ppmv. Both CH₄ and CO gases are harmful to the living beings and environment [1]. CO is a colorless, odorless, tasteless, non-irritating, poisonous gas, while CH₄ is a substantially more potent greenhouse gas (GHG) than CO₂, due to its 25 times higher global warming potential. Hence, leakage of only a small percentage of CH₄ from the supply chain could alter the net climate benefits of NG as a HD vehicle fuel [2]. Therefore, it is imperative to eliminate CO and CH₄ emissions and thus catalytic oxidation converters are needed for their abatement [3]. In contrast to thermal oxidation, catalytic oxidation is considered as green and cost-effective way to reduce the GHG emissions, as it is performed at much lower temperatures, thus being economically favorable and also preventing the formation of nitrous oxides [4].

Up to date, the catalytic combustion of CH₄ has been extensively investigated over noble metals, transition metal oxides, and mixed metal oxides. Among the noble metals, Pd and Pt catalysts belong to the most active for CH₄ combustion at low temperatures. However, despite of being expensive, they also usually deteriorate due to sintering. Alternative oxide materials such as spinels, hexaalumimnates, and perovskites have been considered as promising catalysts for total oxidation of CH₄ due to their high activity, abundant resources and distinctly lower costs when compared to the noble metals [5]. Among the spinel-type oxide catalysts, cobalt oxide (Co₃O₄) usually stands out as an earth-abundant material and possible alternative to expensive noble metals as it exhibits high activity for the catalytic combustion of CH₄. This is possibly attributed to the structure and electronic properties of spinel-type cobalt oxides with variable valence states of transition metal cations in, either tetrahedron (Co²⁺), or octahedron (Co³⁺) interstices, and multifold surface lattice oxygen anions, as well as the lower bonding energy of Co–O bond and the high turnover frequency [4]. In addition, Co₃O₄ holds several other potential advantages such as a lower tendency to sinter at high temperature compared to other transition metal oxides (e.g., Cu or Mn), a larger thermodynamic driving force for reaction with hydrocarbon fuels, and greater rates of reduction and (re)oxidation. Reduction of Co₃O₄ to metal Co provides the highest amount of available oxygen among common metal oxides used for CH₄ combustion, as Co₃O₄ offers the highest ratio of available oxygen moles/mole of metal. Activity and selectivity of Co₃O₄ catalysts depend greatly on their morphology, possibly related to the electro transfer properties of different Co₃O₄ materials, thus enormous efforts have been devoted to develop Co₃O₄ based materials of different dimensions and morphologies (cubes, sheets, belts, wires, tubes, plates, flowers, rods, boxes and nanoparticles) to enhance their catalytic activity [6].

In general, the synthesis method (e.g., type and parameters during the preparation process and/or type of catalysts precursor, etc.) has an important effect on the structural properties of the catalysts, such as surface area, the surface cobalt valence states, etc. Hydrothermal synthesis is reported with significant advantages over others methods, while hydrothermal conditions, such as precursor nature and concentration, pH value, temperature, and time, have critical impacts on the crystal structure and morphology of the product material [7, 8]. However, only few pioneering studies have been reported to explore the influence of different cobalt precursors on the catalytic combustion of CH₄ over Co₃O₄ catalysts [9]. Recently [10], we have investigated the preparation of novel Co-Ce catalytic materials targeting in the complete combustion of CH₄. Co incorporation on an optimum selected CeO₂ support aimed to promote the catalytic activity for the oxidation reactions, via an attempted improvement of the redox property and high oxygen storage capacity induced by the synergistic effect between CeO₂ and CoO_x. We purposely explored the effect of both the synthesis method of the CeO₂ support (following either the hydrothermal or precipitation technique) and the metal loading (0, 2, 5 and 15 wt. % Co) on the catalytic performance and durability for complete CH₄ oxidation. Moreover, the addition of water vapor was studied in the catalytic activity of the optimum selected 15Co/CeO₂ prepared with hydrothermal method. The corresponding material was very promising for methane oxidation purposes, as it presented good water resistance properties in wet CH₄ combustion [10]. Effect of water presence (either as gas humidity or as a combustion product) is suggested to cause deactivation mainly for Pd-based catalysts [11], due to the formation of OH groups, which block the interaction between methane and active sites [12], while a promotional effect is even reported for metal oxide materials (e.g., cobalt-manganese mixed oxides studied for methane combustion [13]. On the other hand, when testing water tolerance of CeO₂–MOx catalysts (M = Cu, Mn, Fe, Co, and Ni) tested for CH₄ or CO oxidation, it seems that catalytic performance in the absence or presence of water depends on the mixed oxide [14]. In another study [12], water effect during methane oxidation seems to also depends not only on active site (Cu or Co catalyst) but also on oxygen deficiency (O₂/CH₄ ratios). More recently, investigation of reaction kinetics of wet lean methane combustion over Co₃O₄- and Pd/Co₃O₄ catalysts outlined the beneficial contribution of Co₃O₄ support towards tolerance to water poisoning [15].

In the present work, we focus on pure cobalt oxide catalysts systematically investigating the effect of synthesis method on the physicochemical properties and catalytic efficiency of the derived cobalt oxide materials. We varied step by step several synthesis parameters such as the method (precipitation or hydrothermal procedure), the type of cobalt precursor (nitrate or acetate), the type of precipitating agent (Na₂CO₃, NH₃, urea or NaOH) and the Co precursor/precipitant molar ratio (0.2 or 0.4), trying to optimize their effect on the catalytic properties and thus oxidation performance of the derived materials. In addition, selected Co₃O₄ samples were also promoted with low amount of Pd (0.5 wt.%), in order to enhance further their catalytic performance in CH₄ and CO oxidation reactions. Last but not least catalytic performance of all materials is explored under the same reaction conditions, focusing further on the effect of differently synthesized catalysts.

2 Experimental

2.1 Materials

A series of Co-based catalysts were synthesized with two different methods; the precipitation method (Co₃O₄–P samples) and the hydrothermal method (Co₃O₄–H samples). The Co precursor salt and the precipitating agents during catalyst synthesis were also varied, using either cobalt acetate tetrahydrate (Co(CH₃CO₂)₂.4H₂O, Fluka, 99%) or cobalt nitrate hexahydrate (Co(NO₃)₂.6H₂O, Merck, 99%), with Na₂CO₃, NH₃, urea and NaOH. For the preparation of Co₃O₄–P samples an aqueous solution of cobalt precursor, acetate (ace) or nitrate (nit) was initially heated to 70 °C and then the precipitating agent (1 M solution of Na₂CO₃, NH₃ or NaOH) was rapidly added, kee** the resulting mixture at 70 °C for 1 h under continuous stirring. The resulting precipitate was filtered, washed several times with double distilled water and ethanol, dried overnight at 110 °C and then calcined at 600 °C for 3 h under air flow. The samples prepared with precipitation method are labeled herein as P-ace or nit-Na₂CO₃–, NH₃– or NaOH–x, where x is the investigated molar ratios of Co/Na (0.2 and 0.4).

For the preparation of Co₃O₄–H samples, cobalt precursor (acetate or nitrate), was dissolved in double distilled water. Then, urea or NaOH solution (1 M) was rapidly added, under vigorous stirring at room temperature. The mixed solution was placed in a 1L Teflon bottle, which was sealed and heated at 110 °C for 6 or 24 h (aging time). The as-obtained material was washed several times with double distilled water and ethanol, and then dried and calcined at 600 °C for 3 h under air flow. The hydrothermally prepared samples are labeled herein as H-ace or nit-urea or NaOH–x–t_aging, where x is the investigated molar ratios of Co/Na (0.2 and 0.4) and t_aging the aging time (6 and 24 h).

The addition of 0.5 wt.% Pd over selected Co₃O₄ samples was performed via incipient wetness impregnation method, using Pd nitrate (Pd(NO₃)₂.2H₂O, Sigma-Aldrich) as the precursor salt. The derived samples were dried at 110 °C overnight and finally calcined under air flow at 500 °C for 5 h. All the code names of the as synthesized Co₃O₄ samples and the synthesis parameters (precursor salt, Co/Na ratio, aging time, precipitating agent) are presented in Table 1.

Table 1 Catalysts synthesis parameters

2.2 Catalysts Characterization

2.2.1 Textural and Structural Characterization (BET Method, XRD, ICP)

The physicochemical characteristics of the as prepared catalysts were determined by the adsorption–desorption isotherms at −196 °C, using the Nova 2200e (Quantachrome) flow apparatus. BET surface areas were obtained according to the Brunauer–Emmett–Teller (BET) method, at the relative pressure in the range of 0.05–0.30. The total pore volume was calculated based on nitrogen volume at the highest relative pressure, whereas the average pore size diameter was determined by the Barrett-Joyner-Halenda (BJH) model, using the desorption branch of the isotherm. Prior to measurements the samples were degassed at 250 °C under vacuum.

The crystalline structure of the catalysts was determined by X-ray powder diffraction (XRD) on a Siemens D 500 diffractometer at 40 kV and 30 mA, with Cu Kα radiation (λ = 0.154 nm). Diffractograms were recorded in the 2–80° 2θ range at a scanning rate of 0.02° over 2 s. The Scherrer equation was employed to determine the particle size of Co₃O₄, based on their most intense diffraction peaks (~ 36.9° and ~ 59.4°).

Inductive coupled plasma-atomic emission spectroscopy (ICP-AES) was performed for the determination of Pd content on Co₃O₄ samples, using a 4300 DV PerkinElmer Optima spectrometer.

2.2.2 Oxygen Storage Ability and Mobility (O₂–TPD and CO–TPR)

Oxygen temperature-programmed desorption (O₂–TPD) experiments were performed in a bench-scale unit, with a fixed bed reactor (0.1 g of sample loading). The temperature was increased to 400 °C (heating rate 5 °C/min) under 20 vol.% O₂/N₂ (50 cm³/min), for 30 min and was cooled to room temperature to adsorb oxygen for 30 min. After that, the samples were purged under He flow for 1 h, in order to remove physically adsorbed oxygen. The desorption step was conducted under He flow (50 cm³/min) up to 600 °C at a rate of 10 °C/min. The reactor exit was coupled with a mass spectrometer (MS), detecting the signal of O₂ (m/z = 32).

Temperature-programmed reduction with CO (CO–TPR) was performed in the same unit. The reactor was loaded with the sample (0.1 g) and a pre-treatment was initially performed at 400 °C for 1 h under He flow, in order to remove the humidity and any other physiosorbed species from the sample surface. The reduction step was performed with 5 vol.% CO/He (50 cm³/min), from room temperature up to 600 °C (heating rate: 5 °C/min). The effluent gases were analyzed by mass spectrometry (MS). The main m/z fragments registered were CO: 28 and CO₂: 44.

2.3 Catalytic Activity Measurements

Catalytic performance experiments were carried out in a quartz fixed-bed reactor, loaded with 0.6 g of catalyst. The total gas flow feed rate was 900 cm³/min corresponding to a gas hourly space velocity (GHSV) of 40,000 h⁻¹. For the CH₄ combustion experiments, the composition of the feed was 0.5 vol.% CH₄ and 10 vol.% O₂, balanced with He. The CH₄ conversion was monitored in the range 300 – 600 °C (in a decreasing temperature mode). For the CO oxidation, the feed composition was 1 vol.% CO, 1 or 10 vol.% O₂, balanced with He and the CO conversion was monitored in the 80–300 °C range (in a decreasing temperature mode). The composition of the effluent gas for both reactions, was analyzed using a FT-IR gas analyzer from MKS instruments (MKS-MG2030Germany).

The conversion of CH₄ and CO was calculated using the Eqs. (1) and (2) respectively, where the C_{CH4, in} and C_{CO, in} represent the CH₄ and CO concentration of inlet and C_{CH4, out} and C_{CO, out} the concentration of the outlet stream.

$$CH_{4}\ Conversion \left( \% \right) = \frac{{C_{CH4, in} - C_{CH4, out} }}{{C_{CH4,in} }} \times 100$$

(1)

$${\varvec{O}}\ Conversion \left( \% \right) = \frac{{C_{CO, in} - C_{CO, out} }}{{C_{CO,in} }} \times 100$$

(2)

3 Results and Discussion

3.1 Catalytic Activity of Co₃O₄ Catalytic Materials for CH₄ Oxidation

3.1.1 Effect of Synthesis Method and Precipitating Agent

The catalytic performances of Co₃O₄ materials, prepared by different synthesis methods and parameters, in complete oxidation of CH₄ were investigated using a fixed-bed reactor [10]. Figure 1 presents the CH₄ oxidation performance of Co₃O₄ catalysts prepared with precipitation and hydrothermal methods, using different precipitating agents, like Na₂CO₃, NH₃, urea and NaOH.

Fig. 1

Effect of precipitating agent and synthesis method of Co₃O₄ catalysts for CH₄ conversion (GHSV: 40,000 h.⁻¹, Feed: 0.5 vol.% CH₄, 10 vol.% O₂, balanced with He)

Co₃O₄ prepared with the precipitation method from Co acetate and using Na₂CO₃ as the precipitating agent (P-ace-Na₂CO₃) presented the worse performance among the different solvents used with T_50% ~ 597 °C, followed by P-ace-NH₃, with T_50% ~ 492 °C. When NaOH was used for the synthesis, cobalt oxide catalytic performance was improved and the light-off curve shifted to lower temperatures, achieving 50% CH₄ conversion at 434 °C for P-ace-NaOH-0.4. The same was observed for the hydrothermally prepared cobalt oxides (NaOH vs urea), where H-ace-NaOH-0.4-24 h present higher activity than the urea catalyst (T_50% ~ 458 °C and T_50% ~ 478 °C, respectively).

Tables 2 and 3 present the textural properties of the as prepared catalysts and it is worth to mention that their physicochemical properties seem to follow their catalytic performance. Thus, the limited activity of the samples prepared with Na₂CO₃, NH₃ and urea is attributed to the low surface area and mainly to the high mean particle size, which is 57.5 nm, 66.7 nm and 65.3 nm, respectively. On the other side, when NaOH was used the size of the formed crystallites was smaller, ranging between 47 and 49 nm. The mean particle size of all the samples were calculated applying Scherrer equation at 2θ: 36.9° and 59.4° from the XRD diffractograms.

Table 2 Textural properties of Co₃O₄ catalysts, using different precipitating agents

Table 3 Physicochemical properties of synthesized Co₃O₄ catalysts

XRD patterns (not shown here) of all Co₃O₄ samples prepared with either the precipitation or the hydrothermal method exhibits characteristic diffraction peaks at 2θ: 19.0°, 31.3°, 36.9°, 38.5°, 44.8°, 55.9°, 59.4° and 65.2°, which are attributed to the Co₃O₄ and suggest that the cobalt precursor was completely turned into a face centered cubic unit cell of Co₃O₄ (space group Fd3m) with a spinel type structure after calcination [10].

Comparing the two synthesis methods with NaOH as precipitating agent, it seems that the precipitation method led to higher surface areas and smaller crystallites, enhancing the CH₄ oxidation performance of P-ace-NaOH-0.4. Thus, NaOH was selected as the precipitating agent to further investigate the effect of the synthesis parameters on the CH₄ oxidation performance and the physicochemical properties of Co₃O₄ for both synthesis methods.

3.1.2 Effect of Synthesis Parameters: Cobalt Precursor, Co/Na Molar Ratio and Aging Time

The effect of cobalt precursor, Co/Na ratio and the aging time (for hydrothermal Co₃O₄ catalysts) for the CH₄ oxidation is presented in Figs. 2 and 3, respectively. The superiority of the Co acetate precursor during precipitation synthesis is evidenced from the whole CH₄ conversion vs temperature curves (Fig. 2a). The curve is shifted to lower temperatures, as compared to the catalyst prepared with cobalt nitrate and the difference in T_50% is > 20 °C (T_50% ~ 434 °C for P-ace-0.4 and 458 °C for P-nit-0.4). More significantly, Co/Na molar ratio (0.2 vs 0.4) does not seem to affect the CH₄ oxidation performance of Co₃O₄ prepared by precipitation method (Fig. 3a). It should be also mentioned that when attempting to use Co/Na molar ratios higher 0.5, complete precipitation was not achieved, as evidenced by significant Co leaching in the filtrate. On the other hand, lower surface areas are observed when applying the hydrothermal method (Co₃O₄–H samples), thus leading to bigger crystallites. In this case, the aging time (6 or 24 h) seems to be more crucial than the type of Co precursor salt (Fig. 2b) and Co/Na molar ratio (Fig. 3b). As expected, longer aging (24 h) and higher Co/Na ratio (0.4), leads to larger crystallites, directly affecting the surface area of the derived materials. Use of rather low Co/Na ratio (0.2 instead of 0.4) interestingly results in similar surface area and crystallite size.

Fig. 2

Effect of Co precursor salt for CH₄ conversion, during a precipitation and b hydrothermal synthesis (GHSV: 40,000 h.⁻¹, Feed: 0.5 vol.% CH₄, 10 vol.% O₂, balanced with He)

Fig. 3

Effect of synthesis parameters of Co₃O₄ catalysts for CH₄ conversion, prepared with: a precipitation and b hydrothermal method (GHSV: 40,000 h.⁻¹, Feed: 0.5 vol.% CH₄, 10 vol.% O₂, balanced with He)

Among all the Co₃O₄ samples, the best catalysts for the CH₄ oxidation are P-ace with Co/Na ratio 0.2 and 0.4 and H-ace-0.4-6 h presenting the highest surface areas (~ 16—18 m²/g) and the lowest Co₃O₄ particle sizes (~ 33—43 nm). High surface areas and small crystallite sizes enhance the catalytic activity, providing more active sites (Co³⁺) [16]. All three catalysts present more or less the same CH₄ oxidation performance with T_50% ~ 430 °C. Different morphology (crystallite shape and size) is generally suggested, when varying the synthesis parameters; a property expected to affect the catalytic activity [6] for both CH₄ and CO oxidation reactions and currently explored via HR-TEM.

3.2 Oxygen Storage Mobility (O₂-TPD) and Reducibility (CO-TPR) of Co₃O₄

Catalyst efficiency is directly associated with the oxygen availability/mobility of the catalysts and possibly related with their morphology. In addition, besides the interaction of Co₃O₄ catalytic materials with O₂, further investigation was also performed with CO (CO-TPR) to study the reducibility of the catalysts.

The oxygen mobility of Co₃O₄ catalysts was investigated with O₂-TPD technique and the results are presented in Fig. 4. Selected samples, with low (P-ace-Na₂CO₃, H-ace-urea-24 h) and high (P-ace-0.2, P-ace-0.4 and H-ace-NaOH-0.2-6 h) CH₄ oxidation performance were studied and the corresponding desorption curves revealed two desorption areas; the low (< 250 °C) and the high (250 – 600 °C) temperature area. The peaks at low temperatures can be assigned to the desorption of surface adsorbed oxygen, while at high temperatures the desorption of surface lattice oxygen takes place [17,18,19].

Fig. 4

O₂ desorption profiles of Co₃O₄ catalysts

It is obvious that the desorption ability of the adsorbed surface oxygen species is better when NaOH is used as the precipitating agent, instead of Na₂CO₃ and/or urea. P-ace-Na₂CO₃ presented two peaks at temperatures higher than 250 °C, with maximum at 320 °C and 450 °C, while H-ace-urea-24 h presented very low intensity peaks at 155 °C and 346 °C (with a shoulder at 430 °C).

On the other side, the use of NaOH as a precipitating agent during Co₃O₄ synthesis improves the mobility of the oxygen species, shifting the peaks at lower temperatures and increasing the intensity of the desorption profiles, indicating the desorption of higher amount of oxygen. Between the precipitation and hydrothermal method, the cobalt oxides prepared via precipitation present higher oxygen mobility, with the desorption peak of P-ace-NaOH-0.2 starting at 70 °C, with a maximum at 150 °C and 430 °C.

The oxygen mobility of P-ace-NaOH-0.2 enhances the formation of oxygen vacancies on the surface of the catalyst, which reflects the amount of O_ads that could be formed. This oxygen species reacts with CO adsorbed on Co³⁺ ions and produces CO₂. Thus, Co³⁺ species are the active sites for CH₄ and CO oxidation, while Co²⁺ is almost inactive. In combination with the oxidation state of cobalt, the morphology of Co₃O₄ is strongly related to its catalytic performance [20,21,22].

The formation of oxygen vacancies enhances the redox properties of the catalysts, improving the reducibility of Co₃O₄ [23]. To further investigate the reducibility of the Co₃O₄ catalysts, CO–TPR experiments were performed. Figure 5(a) presents the intensity of CO signal (28), while Fig. 5(b) the CO₂ (44) during CO–TPR. In all cases, it is obvious that the consumed CO was converted to CO₂ and both signal profiles are perfectly matched. All the CO profiles present two main peaks, centered at 315 °C the first one and a broad peak, with shoulders in the temperature range of 335 – 550 °C. The reduction of Co₃O₄ is a two-step reduction process; Co₃O₄ is reduced to CoO and then, to metallic Co [19, 24]. The first peak at 315 °C may be assigned to the reduction of Co³⁺ to Co²⁺. In addition, all the CO profiles of Co₃O₄ present one low intensity peak at 120 °C, probably related to the reduction of bulk Co₃O₄ and the reaction of CO with the catalysts surface lattice oxygen species [25]. The broad peak starts at 335 °C and the reduction of CoO to metallic cobalt under CO was completed at 500 – 540 °C.

Fig. 5

CO-TPR profiles of Co₃O₄ catalytic materials: a CO signal (m/z: 28) and b CO₂ signal (m/z: 44)

It is obvious that in the case of the samples prepared with urea and Na₂CO₃ as precipitating agents (H-ace-urea-24 h and P-ace-Na₂CO₃), the reduction profiles are shifted to higher temperatures, from 315 °C to 330 °C for H-ace-urea-24 h and 322 °C for P-ace-Na₂CO₃, for the first peak, while the second broad peaks started at temperatures higher than 350 °C, compared with the samples prepared with NaOH, indicating that the precipitating agent affect the reducibility of the cobalt catalysts. It is probably related with the size of cobalt crystallites; as the size increases, i.e., from 43 nm for P-ace-NaOH-0.2 to 57 nm for P-ace-Na₂CO₃, the peaks are shifted to higher temperatures. Li et al. [26] studied the size effect of Co₃O₄ nanoparticles and tested them for CO oxidation. The results revealed that the catalyst activity is depending on the size of the Co₃O₄ nanoparticles, with the smallest being highly active. In addition, the areas of the reduction peaks are smaller, indicating that these catalysts have less active oxygen species [19].

The Co₃O₄ prepared with NaOH present peaks at the same temperature area, with different reduction areas. P-ace-NaOH-0.2 present the highest CO consumption (and CO₂ formation), followed by P-ace-NaOH-0.4 and H-ace-NaOH-0.2-6 h. Especially the first peak at 315 °C due to the reduction of Co³⁺, revealed that P-ace-NaOH-0.2 present higher amount of Co³⁺ active sites, which is in line with the O₂-TPD profiles. The combination of oxygen adsorption capacity, mobility and cobalt’s oxidation state of P-ace-NaOH-0.2 enhances the efficiency of the catalyst for the CH₄ oxidation reaction.

3.3 Catalytic Activity of Optimum Co₃O₄ Catalysts for CO Oxidation

Selected Co₃O₄ catalysts were also tested for the oxidation of CO using two different reaction feeds: 1 vol.% CO, 1 vol.% O₂ and 1 vol.% CO, 10 vol.% O_2, balanced with He and the results are presented in Fig. 6. The profiles are shifted to lower temperatures, when 10% vol. O₂ was used in the feed, with a simultaneous important decrease of T_50%. The order of the catalyst’s performance is the same under both feeds, with H-ace-NaOH-0.2-6 h (T_50% ~ 131 °C under 1 vol.% O₂ and 114 °C under 10 vol.% O₂ in the feed) shown as the optimum catalysts, followed by P-ace-NaOH-0.2 and P-ace-NaOH-0.4. In addition, P-ace-Na₂CO₃ was tested for CO oxidation reaction. The lower catalyst performance is also evident for the oxidation of CO, with T_50% ~ 190 °C. The low oxidation performance (as presented above) is related to the large particle size (57 nm), while the same trend was observed when urea was used as precipitating agent (resulting in particle sizes of 65 nm).

Fig. 6

CO oxidation performance of Co₃O₄ catalysts, with feed: 1 vol.% CO-1 vol.% O₂ (solid line-closed marker) and 1 vol.% CO-10 vol.% O₂ (dashed line-open marker), balanced with He (GHSV: 40,000 h.⁻¹)

3.4 Pd-Promoted Co₃O₄ Catalysts

As already mentioned, selected Co₃O₄ catalysts were further promoted with the addition of 0.5 wt.% Pd. Among the precipitated catalysts, the impregnation of Pd was performed over the samples originally synthesized with cobalt acetate and Co/Na ratio 0.2 and 0.4, while H-ace-NaOH-0.2-6 h was selected among the hydrothermally prepared cobalt oxides. The catalysts were tested for both CH₄ and CO oxidation (1% vol. and 10% vol. O₂ in feed) reactions and the results are presented in Fig. 7a and Fig. 7b, c, respectively.

Fig. 7

Effect of 0.5 wt.% Pd on Co₃O₄ catalysts for: a CH₄ oxidation (GHSV: 40,000 h⁻¹, Feed: 0.5 vol.% CH₄, 10 vol.% O₂, balanced with He), b CO oxidation, with feed: 1 vol.% CO-1 vol.% O₂ (solid line) and 1 vol.% CO-10 vol.% O₂ (dashed line), balanced with He (GHSV: 40,000 h.⁻¹)

The incorporation of Pd over the optimum pure cobalt oxides improved further their CH₄ oxidation performance (Fig. 7a). The curves were shifted to lower temperatures and the T_50% was decreased up to 60 °C for Pd/P-ace-NaOH-0.2 (T_50% ~ 373 °C), 67 °C for 0.5Pd/H-ace-NaOH-0.2-6 h (T_50% ~ 387 °C) and 54 °C for Pd/P-ace-NaOH-0.2 (T_50% ~ 388 °C). The 0.5Pd/P-ace-NaOH-0.2 sample is shown as the best catalyst for CH₄ oxidation, while H-ace-NaOH-0.2-6 h exhibited the highest improvement with the addition of Pd.

On the other side, CO oxidation efficiency of Pd-doped Co₃O₄ catalysts (Fig. 7b) did not follow the same activity sequence as CH₄ oxidation. A similar trend is observed in the literature [27], where the effect of Pd was different for CH₄ and CO oxidation; Pd enhances the activity of Co₃O₄ for CH₄ oxidation, while a significant decline was observed for the CO oxidation. In our case, Pd enhanced the activity of P-ace-NaOH-0.4 (under both feeds), shifting light-off curves to lower temperatures and decreasing T_50% ~ 17 °C (from 153 °C for P-ace-NaOH-0.4 under 1% vol. O₂ to 137 °C with Pd impregnation and from 137 °C to 118 °C, under 10% vol. O₂). The 0.5Pd/P-ace-NaOH-0.2 cobalt oxide presents more or less the same catalytic performance for CO oxidation with pure cobalt oxide (P-ace-NaOH-0.2), especially under 1% vol. O₂ in the feed, while under 10% vol. O₂ the curve is slightly shifted to lower temperatures, with up to 5 °C lower T_50%.

The best catalyst for CO oxidation, with or without Pd, is Co₃O₄ prepared by hydrothermal method and with Co/Na ratio 0.2 and aging time 6 h (H-ace-NaOH-0.2-6 h sample). The addition of Pd did not improve further its catalytic performance, presenting the same activity with pure cobalt oxide, with T_50% ~ 134 °C and 114 °C under 1 vol.% and 10 vol.% O₂ in the feed, respectively.

The addition of Pd on Co₃O₄ catalysts prepared with precipitation method (P-ace-NaOH-0.2 and -0.4) slightly decreased the surface area and increased the pore diameter, as shown in Table 4. The opposite was observed for 0.5Pd/H-ace-NaHO-0.2-6 h, where the incorporation of Pd did not affect the surface area or slightly increased it (from 11.4 nm to 11.7 nm). The XRD patterns (not presented here) revealed peaks of Co₃O₄, without the existence of PdO phase, probably due to the low Pd loading (0.5% wt.) and the good dispersion of the metal over cobalt oxide. The calculated crystal size of Co₃O₄ (applying the Scherrer equation) was slightly increased for the Pd on Co₃O₄-P, while for the hydrothermally prepared cobalt oxide, the impregnation of Pd decreased the crystallites size from 53.7 nm to 48.5 nm.

Table 4 Physicochemical properties of Pd-based Co₃O₄ catalysts

No important changes were observed on the reduction profiles of Co₃O₄ (CO-TPR) after the incorporation of Pd (Fig. 8). The same double peak, at the same temperature area was observed for Pd-based Co₃O₄ catalysts. The main differences observed relates with the intensity of the peaks (CO consumption), where 0.5Pd/P-ace-NaOH-0.4 and 0.5Pd/H-ace-NaOH-0.2-6 h present more or less the same intensity and higher than 0.5Pd/P-ace-NaOH-0.2. These two catalysts present the same catalytic performance for both CH₄ and CO oxidation, with higher performance in CO oxidation, while 0.5Pd/P-ace-NaOH-0.2 in CH₄ oxidation. This is probably related to the existence of more Co³⁺ active sites according the intensity of the first peak in CO-TPR profiles of Pd-based cobalt oxides P-ace-NaOH-0.4 and H-ace-NaOH-0.2-6 h.

Fig. 8

CO-TPR profiles of Pd/Co₃O₄ catalysts: a CO signal (m/z: 28) and b CO₂ signal (m/z: 44)

In addition, O₂-TPD profiles of Pd-based cobalt oxides are presented in Fig. 9. For all Pd/Co₃O₄ catalysts the first peak is the same with the pure cobalt oxides, with the peak of 0.5Pd/P-ace-NaOH-0.2 shifted to 167 °C, compared to P-ace-NaOH-0.2, which was centered at 150 °C. On the other side, the addition of Pd over H-ace-NaOH-0.2-6 h shifted the first peak to 168 °C (from 176 °C for pure hydrothermally prepared Co₃O₄). The high temperature area (250 – 500 °C) of the desorption of surface lattice oxygen presented higher oxygen adsorption with the incorporation of palladium.

Fig. 9

O₂ desorption profiles of Pd/Co₃O₄ catalysts

Table 5 presents several Co₃O₄ catalysts reported in the literature, prepared either with hydrothermal or precipitation method, showing different morphologies and similarly tested for CH₄ and/or CO oxidation. As obvious from Table 5, the metal oxide morphology, actually configuring the surface topology is suggested in the literature as the most crucial property, mainly defining the catalytic efficiency. On the other hand, catalytic performance of Co-based materials synthesized during the current study may not reach the low T_50% efficiency of the catalysts reported in the literature, but this trend is possibly related not only with the morphology (currently explored for catalysts of the present study) but also with the Co₃O₄ crystallite size. In conclusion, both from literature data and our current investigation not only morphology but crystallite size (depending on synthesis parameters) is suggested to be very important for both CH₄ and CO oxidation reactions. Interestingly, when incorporating Pd on cobalt oxide, the differences in catalytic performance between our catalysts and those reported in the literature are not so big for both reactions. This is an important observation, if we consider that materials of the present study are loaded with much lower Pd content than other catalysts in Table 5. In this case, that is for Pd/Co₃O₄ catalysts, optimization of cobalt oxide support should focus more on catalyst stability (e.g. durability, water tolerance, etc.) and not only on the efficiency of the materials studied. Nevertheless optimally synthesized cobalt oxide support may minimize if not replace the expensive noble metal.

Table 5 Crucial characteristics and catalytic performance of Co₃O₄, prepared by different synthesis methods for CH₄ and/or CO oxidation (T_50%: Temperatures of 50% conversion)

4 Conclusions

The present study systemically investigates the effect of synthesis method (type and synthesis parameters) on the physicochemical properties and catalytic efficiency of the derived Co₃O₄ for oxidation of CH₄ and CO. Differentiating synthesis parameters directly affects the porosity and morphological characteristics (surface area and crystallite size) and other crucial properties (oxygen storage/redox performance) of the derived catalysts, thus influencing their catalytic performance in both oxidation reactions. Among the different precipitating agents, when NaOH was used, the size of the formed cobalt oxide crystallites was smaller, presenting higher catalytic activity for CH₄ oxidation. Between the precipitation and the hydrothermal method, the precipitation led to higher surface areas and smaller crystallites than hydrothermal, enhancing CH₄ oxidation performance of Co₃O₄.

The investigation of other synthesis parameters revealed the superiority of cobalt acetate as a precursor salt, while for the hydrothermally prepared cobalt oxides the aging time (6 or 24 h) seem to be more crucial than the type of Co precursor salt and Co/Na molar ratio. Longer aging (24 h) led to larger crystallites, directly affecting the surface area of the derived materials. The best Co₃O₄ for CH₄ oxidation was prepared with precipitation method, using NaOH and cobalt acetate (P-ace-NaOH-0.2). The combination of oxygen adsorption capacity, mobility and cobalt’s oxidation state of P-ace-NaOH-0.2 enhanced the efficiency of the catalyst for the CH₄ oxidation reaction.

Concerning CO oxidation on the other side, the hydrothermally prepared cobalt oxide H-ace-NaOH-0.2-6 h presented higher catalytic performance than the precipitated samples. The effect of Pd (0.5wt.%) was different for CH₄ and CO oxidation reactions. Pd enhance the activity of Co₃O₄ for CH₄ oxidation, while the same catalytic performance was observed for the CO oxidation.

Concerning Co₃O₄ materials it seems that besides morphology (currently explored) crystal size is a very crucial parameter affecting surface topology and thus catalytic performance. On the other hand, for Pd/Co₃O₄ catalysts, optimization of cobalt oxide support should focus more on catalyst stability (e.g. durability, water tolerance, etc.) and not only efficiency of the materials studied, simultaneously aiming to minimize noble metal loading.