Water–Gas Shift Activity over Ni/Al₂O₃ Composites

Abstract

The water–gas shift (WGS) performance of 10%Ni/Al₂O₃, 20%Ni/Al₂O₃ and 10%Ni/CaO-Al₂O₃ catalysts was studied to reduce CO concentration and produce extra hydrogen. Ni was added onto the Al₂O₃ support by an impregnation method. The physicochemical properties of nickel catalysts that influence their catalytic activity were examined. The most influential factors in increasing the CO conversion for the water–gas shift reaction are Ni dispersion and surface acidity. Ni metal sites were identified as the active sites for CO adsorption. The main effect of nickel metal was reducing the adsorbed CO amount by reducing the active site concentration. The 10%Ni/Al₂O₃ catalyst was more active for the WGS reaction than other catalysts. This catalyst presents a high CO conversion rate (75% CO conversion at 800 °C), which is due to its high Ni dispersion at the surface (6.74%) and surface acidity, thereby favoring CO adsorption. A high Ni dispersion for more surface-active sites is exposed to a CO reactant. In addition, favored CO adsorption is related to the acidity on the catalyst surface because CO reactant in the WGS reaction is a weak base. The total acidity can be evaluated by integrating the NH₃-Temperature-Programmed Desorption curves. Therefore, an enhancement of surface acidity is identified as the favored CO adsorption.

1. Introduction

A WGS reaction involves the reaction of steam and CO to produce H₂ and CO₂ according to the following equation:

CO + H₂O ⇆ H₂ + CO₂           ΔH₂₉₈ = −41.2 kJ/mol

A WGS reaction is an important industrial reaction used in the production of NH₃ to balance the H₂/CO ratio and provide pure hydrogen through CO loss. It is also an important side reaction combined with the main synthesis reaction in conjunction with CH₄ steam reforming and CH₃OH synthesis from the mixtures of CO, CO₂ and H₂. The water–gas shift reaction is also an important component in reducing CO amount from feed gas in a Proton-Exchange Membrane Fuel Cell (PEMFC). A trace amount of carbon monoxide poisons the anode used in PEMFC; therefore, CO content in the inlet gas must be kept below 10 ppm [1]. A purification stage is needed to reduce the carbon monoxide content to be lower than the cell tolerance level. A WGS reaction can decrease the amount of CO while producing more H₂ as a fuel for the hydrogen fuel cell. Many metallic elements such as Rh, Pt, Co and Ni [2,3] are suitable for use in a WGS reaction. It is well known that noble metal catalysts exhibit good WGS performance, but large-scale commercialization has been restricted because of their high cost. Nickel catalysts are promising as excellent commercial catalysts with relatively low costs, high WGS activity, and high methane selectivity [4,5,6,7,8]. Different catalysts have been investigated so far. Precious metals provide the best catalytic performance, but their high price makes them an unattractive choice compared with the Ni catalyst supported on alumina. However, a key challenge for Ni catalysts is deactivation due to Ni sintering and coke formation.

It is well known that Ni catalysts are familiar for CH₄ steam reforming. However, Ni also plays an important role for the WGS reaction. In previous studies, Ni/Al₂O₃ and Ni/ZnO-Al₂O₃ were studied for the WGS reaction [9]. The TPR technique revealed the types of NiO particles in the materials, as well as the degree of interaction between NiO and the support based on Al₂O₃ and ZnO-Al₂O₃. The degree of interaction rises when materials are prepared at higher temperatures, with NiAl₂O₄ particles being detected. The presence of zinc led to increased selectivity but decreased specific surface area and activity. Furthermore, Ni/CeO₂-Al₂O₃ powder was studied for the water–gas shift reaction at a high temperature (450 °C) [10]. The effect of Ni loading (ranging from 1 to 8 wt.%) on the WGS performance was investigated. It was found that by increasing the Ni content, it results in an increase in the catalytic activity of Ni/CeO₂-Al₂O₃. WGS activity increases from 24%CO conversion (for 1 wt.% Ni loading) to 54%CO conversion (for 4 wt.% Ni loading) and to 76%CO conversion (for 8 wt.% Ni loading). Moreover, the 5%Ni/Al₂O₃ catalyst presented high CO₂ conversion and CO selectivity for a reverse water–gas shift reaction [11]. The high activity and CO selectivity of Ni/Al₂O₃ catalysts for the reverse water–gas shift reaction were the results of an increase in nickel dispersion (or smaller nickel particles), the amount of surface-active sites and CO₂ adsorption by the basic nature of the catalyst. An addition of a promoter increased the dispersion of nickel species on the support, in addition to enhancing CO₂ adsorption on the catalyst surface. An addition of 1%La, 1%K and 2%K into 5%Ni/Al₂O₃ enhanced the CO₂ conversion at 700 °C from 40% to 43%, 42% and 43%, respectively.

In this work, the effect of the different supports and Ni loading on the WGS activity were studied. In addition, the key factors in enhancing the WGS activity were investigated using the following techniques: BET surface area, NH₃-temperature-programmed desorption, H₂-temperature-programmed reduction and H₂ chemisorption. The CeO₂ support has received much interest for water–gas shift reactions due to its excellent redox properties and high catalytic activity, but its high price has impeded its commercialization. Therefore, developing water–gas shift catalysts could be made cheaper by using an Al₂O₃ support. This study was expected to aid an improvement of cost-effective catalysts with high WGS activity and low carbon deposition by confirming physicochemical characteristics (the carbon deposition on the used catalysts, the acidity, the reduction behavior, the metal dispersion and average particle size) relating to the WGS performance. In addition, the reduction of NiO to generate Ni⁰ in WGS catalysis leads to an advancement of the water–gas shift reaction. The type of interaction that NiO particles develop with the support (Al₂O₃ and CaO-Al₂O₃) causes reduction peaks at different temperatures; therefore, the types of NiO particles were observed, and how they influence the catalytic activity for the water–gas shift reaction was studied.

2. Experimental Procedure

2.1. Catalysts Preparation

Al₂O₃ powder (99.99%; 85–115 m²/g, BET; pH valve 5 (100 g/L, H₂O, 20 °C); Merck (Rahway, NJ, USA)) was used as support with Ni loadings of 10 wt.% and 20 wt.%. Ni(NO₃)₂.6H₂O (99.8%; Alfa Aesar (Haverhill, MA, USA)) with a mass of 0.55 g (for 10 wt.% Ni loading) and 1.24 g (for 20 wt.% Ni loading) was dissolved in DI water and then added into Al₂O₃ powder (1 g). All catalysts were dried in an oven at 110 °C for 12 h and calcined at 800 °C for 2 h.

The catalytic activity of commercial 10%Ni/CaO-Al₂O₃ was also studied to compare with the 10%Ni/Al₂O₃ and 20%Ni/Al₂O₃ catalysts.

2.2. Catalyst Characterization

A BELSORP-MAX instrument was used to measure the specific surface areas of all samples by N₂ adsorption–desorption isotherms at −196 °C. Before the analysis, the sample was outgassed at 300 °C for 3 h. The Brunauer Emmett Teller (BET) method was used to examine the surface area of the catalysts.

The coke formation on the used catalysts was examined by thermo-gravimetric analysis (TGA) with a PerkinElmer TGA/DTA 6300 instrument. TGA was operated under an air flow rate of 100 cm³ min⁻¹. The weight loss of the catalysts was investigated as a function of temperature from 50 to 800 °C with a ramp rate of 20 °C min⁻¹.

A catalyst analyzer BELCAT-B instrument was utilized to operate the H₂ chemisorption, NH₃-Temperature-Programmed Desorption (NH₃-TPD) and H₂-Temperature-Programmed Reduction (H₂-TPR). The reduction behavior of Ni catalysts was determined by H₂-TPR. The catalyst was pretreated under a helium flow at 110 °C for 30 min and then cooled down to 50 °C. H₂-TPR analysis was carried out under 5%H₂/Ar flow from 50 to 950 °C with a ramp rate of 10 °C min⁻¹. The surface acidity of the catalyst was investigated using NH₃-TPD. The sample was first heated up to 500 °C in Ar flow and then cooled down to 50 °C. After that, the sample was exposed to NH₃ in Ar flow with a loop of a known volume until saturation. Finally, NH₃-TPD analysis was performed in the range of 50–800 °C with a ramp rate of 10 °C min⁻¹ under Ar flow.

The metal dispersion and average particle size were determined by a H₂ chemisorption technique. The catalyst was calcined in air at the rate of 100 cm³ min⁻¹ at 800 °C for 2 h and purged under He flow for 30 min, followed by reduction in 5%H₂/Ar flow at the rate of 100 cm³ min⁻¹ at 800 °C for 2 h. After that, the reduced catalyst was cooled down to 50 °C in He flow and followed by hydrogen chemisorption with pure H₂. The nickel dispersion and average particle size were calculated on the assumption that one Ni atom adsorbed one H atom using Equations (1)−(3).

D = (SF X_hydrogen F_w/% g) × 100

(1)

S (m² g⁻¹) = SF × X_hydrogen × σ_m × N_A/% g

(2)

d = 6000/S × ρ_m

(3)

where D is % metal dispersion, SF is the stoichiometric factor, X_hydrogen is chemisorbed hydrogen on the metal particles (mol), F_w is the formula weight of the particle, % is the weight percent of nickel, g is the sample weight, S is specific surface area (m² g⁻¹), σ_m is the atomic cross-sectional area of metal, N_A is the Avogadro number, ρ_m is the metal particle density and d is average particle size (nm).

2.3. Water–Gas Shift Activity

The WGS activity was studied in the range of 400–800 °C. About 150 mg of the catalyst powder was placed inside a fixed bed flow reactor between two layers of quartz wool. The reactor was made from 310 stainless steel with an outside diameter of 0.6 cm. Before WGS activity measurement, the Ni catalyst was reduced at 300 °C for 1 h under 5%H₂/N₂ flow. The flow rate of N₂ and CO was controlled by a mass flow controller, and water was fed using a syringe pump. Mixed gas containing 10%H₂O, 5%CO and a balance of N₂ was fed into the fixed bed flow reactor with a total flow rate of 100 mL min⁻¹. The gas component at the reactor outlet was determined by an online gas chromatography with a thermal conductivity detector and a ShinCarbon ST column. The CO content at the reactor outlet was repeated at least five times for each test. %CO conversion was calculated by the following equation:

$$\%\mathrm{CO} \mathrm{conversion}=\frac{{\mathrm{CO}}_{\mathrm{in}}-{\mathrm{CO}}_{\mathrm{out}}}{{\mathrm{CO}}_{\mathrm{in}}}\times 100$$

(4)

where CO_in and CO_out are the inlet and outlet of the CO molar flow rate, respectively.

3. Results and Discussion

3.1. Catalysts Characterization

The results of H₂ chemisorption and BET surface area analysis of Ni catalysts are summarized in

Table 1. BET surface area, total pore volume, average pore diameter and Ni dispersion of Ni catalysts.

Catalysts	BET Surface Area a (m2/g)	Total Pore Volume a (cm3/g)	Average Pore Diameter a (nm)	% Ni Dispersion b	Average Particle Size b (nm)
10%Ni/Al2O3	98	0.1123	3.8928	6.74	8.7
10%Ni/CaO-Al2O3	22	0.0515	8.2550	4.01	18.6
20%Ni/Al2O3	84	0.0993	3.9071	4.38	14.2

^a Estimated from N₂ adsorption at −196 °C. ^b Estimated from H₂-chemisorption.

. It was found that the 10%Ni/Al₂O₃ catalyst has a higher BET surface area and Ni dispersion than 20%Ni/Al₂O₃ and 10%Ni/CaO-Al₂O₃. This result may be due to the strong interaction between Ni metal and Al₂O₃ support, which is beneficial for inhibiting the sintering of Ni and maintaining the small particle size of Ni [12]. Among all the catalysts, 10%Ni/Al₂O₃ exhibits the greatest BET surface area (98 m²/g) and nickel dispersion (6.74%). Generally, the high dispersion of Ni on the catalyst surface results in an increase in CO adsorption and facilitation of the redox cycle [13,14]. The average particle size of Ni was determined by H₂ chemisorption. An average Ni particle size increases with the Ni amount from 8.7 nm for 10%Ni/Al₂O₃ to 14.2 nm for 20%Ni/Al₂O₃, whereas 10%Ni/CaO-Al₂O₃ catalyst possesses a larger average size of nickel particles than other catalysts. As expected, the Ni particle size enhances from 8.7 to 18.6 nm, whereas the corresponding Ni dispersion decreases from 6.74% to 4.01%.

The reduction behavior of the Ni catalysts was analyzed by using H₂-TPR characterization (Figure 1). The H₂-TPR profiles of 10%Ni/Al₂O₃, 20%Ni/Al₂O₃ and Ni/CaO-Al₂O₃ show two reduction peaks. The reduction peaks at low temperature are ascribed partly to nickel species weakly interacting with the support. The first reduction peak shifts downward to 415 °C for 10%Ni/Al₂O₃, indicating that the reduction of Ni²⁺ to Ni⁰ becomes easier, whereas the active Ni component in the 20%Ni/Al₂O₃ and Ni/CaO-Al₂O₃ catalysts is more difficult to reduce. On the other hand, the reduction peaks at about 750–800 °C are attributed to the reduction of Ni²⁺ in the NiAl₂O₄ phase or nickel interacting strongly with the support [15,16,17]. A stable structure of the NiAl₂O₄ spinel phase results in difficulty in reducing it to Ni⁰ at low temperatures. The second reduction peak of 10%Ni/Al₂O₃ shifts to a high temperature compared to the other catalysts. The higher reduction temperature of the 10%Ni/Al₂O₃ catalyst suggests that it possessed a strong metal–support interaction (SMSI), which is beneficial to prevent the sintering of Ni and maintain the small particle size of Ni during the reduction process at a high temperature.

The surface acidity of the Ni catalysts was investigated by NH₃-TPD (Figure 2, and the total acidity was calculated from the area under the desorption peak (

Table 2. Amount of NH₃ desorbed in the NH₃-TPD experiments. Values were obtained by integrating the NH₃-TPD curves in the low and high temperature ranges.

Samples	Weak Acidity (50–400 °C) (μmol/g)	Strong Acidity (500–700 °C) (μmol/g)	Total Acidity (μmol/g)
10%Ni/Al2O3	178.2	25.7	203.9
10%Ni/CaO-Al2O3	50.7	39.5	90.2
20%Ni/Al2O3	150.7	20.6	171.3

. The effect of the acidity of the catalysts on the WGS performance was analyzed by using NH₃-TPD. The desorption peaks in a temperature range of 50–400 °C or 500–700 °C are assigned to weak or strong acid sites, respectively. An enhancement of the acidity can increase CO adsorption on the catalyst surface because a CO reactant in the WGS reaction is a weak base. Furthermore, the surface acidity of the catalyst was proved to be beneficial for CO₂ desorption, leaving behind free active sites for CO and H₂O adsorption in subsequent reaction cycles [13,14]. The total acidity can be calculated by integrating the NH₃-TPD curves. The result indicates that 10%Ni/Al₂O₃ exhibits the highest content of weak acid sites (178.2 μmol/g), which results in a higher tendency for CO adsorption and subsequently easier CO₂ desorption on the 10%Ni/Al₂O₃ catalyst; thus, the overall WGS rate of this catalyst can be enhanced.

Thermo-gravimetric analysis (TGA) was used to determine the carbon deposition of the used catalyst (Figure 3). High temperature oxidation of carbon material in air leads to a weight reduction in the catalyst. Small weight loss at a temperature below 200 °C is due to the removal of moisture and volatile species [18]. The weight reduction in the range of 200–400 °C is related to the thermal decomposition of physisorbed carbonaceous species. A major mass loss of a range of 400–600 °C is ascribed to the bulky carbonaceous products on the used catalysts. The mass loss of the bulky carbonaceous species on Ni catalysts on different supports (Al₂O₃ and CaO-Al₂O₃) and Ni loadings (10% and 20%) were 22%, 20% and 15.5% for Ni/CaO-Al₂O₃, 20%Ni/Al₂O₃ and 10%Ni/Al₂O₃, respectively. Thus, the 10%Ni/Al₂O₃ catalyst after the WGS reaction test gives lower carbon deposition than other catalysts, thus obtaining higher efficiency and a higher WGS rate. For the spent 10%Ni/Al₂O₃ catalyst, the mass loss was significantly decreased. The better anti-carbon deposition performance of the 10%Ni/Al₂O₃ catalyst is attributed to the smaller size of nickel particles, as well as strong metal–support interaction.

3.2. Water–Gas Shift Performance

Figure 4 illustrates the %CO conversion of Ni/CaO-Al₂O₃, 10%Ni/Al₂O₃ and 20%Ni/Al₂O₃. For Ni/CaO-Al₂O₃, the CO conversion ascended slowly to reach the maximum of 31%CO conversion at 800 °C. As shown, the highest WGS activity was attained over the 10%Ni/Al₂O₃ catalyst, reaching 75% of CO conversion at 800 °C, which was due to its high Ni dispersion (6.74%) and surface acidity. A higher Ni dispersion with more surface-active sites are exposed to the CO reactant. The surface acidity is related to the amount of CO adsorption because a CO reactant in the WGS reaction is a weak base. Generally, the redox mechanism is accepted for the water–gas shift reaction at high temperatures where a carbon monoxide molecule adsorbs on the catalyst surface and abstracts one O from the metal oxide support to generate CO₂. A depletion of oxygen from metal oxide is fulfilled by dissociating a H₂O molecule and then desorbing as H₂ gas, while O atom is captured by the oxygen deficient metal oxide [19]. Therefore, higher CO adsorption ability and easier reducibility result in a high WGS performance.

4. Conclusions

The WGS activity of Ni/CaO-Al₂O₃, 10%Ni/Al₂O₃ and 20%Ni/Al₂O₃ was studied. The results revealed that the 10%Ni/Al₂O₃ catalyst presents the highest catalytic activity for the WGS reaction and the lowest carbon deposition. The WGS activity at 800 °C follows the order 10%Ni/Al₂O₃ (75% CO conversion) > 20%Ni/Al₂O₃ (70% CO conversion) > Ni/CaO-Al₂O₃ (31% CO conversion). The highest WGS activity of the 10%Ni/Al₂O₃ catalyst was due to its greater specific surface area (98 m²/g), as well as the acidity and Ni dispersion (6.74%) on the catalyst surface, which cause an increase in CO adsorption on the surface. The Ni metal site was the active site for CO adsorption. A high Ni dispersion provides more surface-active sites that are exposed to the CO reactant. In addition, the CO reactant in the WGS reaction is a weak base; therefore, the higher levels of acidity on the catalyst leads to favored CO adsorption. Moreover, acidic property in the 10%Ni/Al₂O₃ catalyst can alert CO₂ desorption. Therefore, the increased CO adsorption ability and subsequently easier CO₂ desorption on the 10%Ni/Al₂O₃ catalyst facilitate the redox processes at the catalyst surface, thus increasing the WGS rate.