Ammonia (NH₃) is an indispensable raw material for the chemical industry and agriculture fertilizers, which has played a crucial role in developing the human society^[1-2]. Over the past century, ammonia synthesis is carried out at high temperatures (400~600 ℃) and high pressures (20~40 MPa) using the Fe-based catalysts in Haber-Bosch process, which consume 1%~2% of annual global energy production^[3-4]. Compared to conventional Fe-based catalysts, Ru-based catalysts exhibits higher activity under mild conditions, namely low temperature (< 500 ℃) and pressure (< 10 MPa), which have attracted extensive interest in the last two decades^[5-6].

Previous studies indicated thatthe support affect the catalytic activity of Ru-based catalysts by modifying the electronic properties of the active metal Ru^[7-8]. A series of materials including active carbon^[9], zeolites^[10], boron nitride^[11] and electronic compound^[12] have been investigated as support of the Ru-based catalysts for ammonia synthesis. The lanthanide metal oxide CeO₂ has superior catalytic performance over other supports under mild conditions^[13-14]. The Ce³⁺/Ce⁴⁺ redox ability and oxygen storage capacity of CeO₂ support are key factors in increasing the electron density of Ru. For example, Ma et al. found that the amount of surface adsorption oxygen species and Ce³⁺ concentration of Ru/CeO₂ catalysts had greatly affected the ammonia synthesis activity^[15]. Ogura et al. discovered that CeO₂ with surface hydroxyls groups had increased the activity of ammonia synthesis^[16].

It has been proven that the rate-determining step in ammonia synthesis isthe cleavage of the N≡N bond (945 kJ·mol^-1)^[17-18]. The basic supports significantly enhanced the ammonia synthesis activity by weakening of the N≡N bond. Wang et al. have successfully prepared Ru/ZrO₂-KOH catalyst by the active metal Ru-loaded the superbasic ZrO₂-KOH for ammonia synthesis^[19]. As a result, the strong basicity Ru/ZrO₂-KOH catalyst after KOH loading exhibits excellent activity due to high electron density. Moreover, some researchers have found that alkali metal and alkaline earth metal promoters are essential for obtaining Ru-based catalysts of high activity in ammonia synthesis^[20-21]. It also found that the promotion effect of alkali metal and alkaline earth metal on the activity of Ru-based catalysts is inversely proportional to the electronegativity of alkali metals and alkaline earth metals in the order of Cs >Ba>K>Na^[22].

Herein, different basicity CeO₂ supports were used to study the surface properties of CeO₂ for the catalytic performance under mild conditions. The alkali metal and alkaline earth metal nitrates (KNO₃, CsNO₃, Ba(NO₃)₂) were used as promoters to enhance the catalytic performance of ammonia synthesis. And the powder X-ray diffraction (PXRD), scanning electron microscopy (SEM), N₂ adsorption-desorption, H₂-temperature programmed reduction (H₂-TPR), X-ray photoelectron spectroscopy (XPS), CO₂ temperature-programmed desorption (CO₂-TPD) were used to further analyze the effect of Ru/CeO₂ catalysts on the catalytic reaction of ammonia synthesis. The basicity of the support enhanced the ammonia synthesis performance, which could be related to the basic site, oxygen vacancy and surface composition of support.

1 Experimental Section 1.1 Materials

All chemicals were directly used without any further purification. Ruthenium (Ⅲ)-2, 4-pentanedionate (C₁₅H₂₁O₆Ru, 97.0%) and Cerium (Ⅲ) nitrate hexahydrate (Ce(NO₃)₃·6H₂O, 99.5%) were purchased from Aladdin Industrial Corporation. Potassium hydroxide G.R. (KOH ≥ 85.0%) was obtained from Beijing Chemical Works Co., Ltd. Cesium nitrate (CsNO₃, 99.7%) was bought from Beijing Xinhua Chemical Reagent Co., Ltd. Barium nitrate (Ba(NO₃)₂, 99.5%) and Potassium nitrate (KNO₃, 99.5%) were purchased from Tianjin Hengxing Chemical Reagent Co., Ltd. The purity of nitrogen and hydrogen were 99.99%.

1.2 Preparation of CeO₂ supports

The basicity CeO₂ supports were prepared by hydrothermal synthesis method. 1.736 g Ce(NO₃)₃·6H₂O was dissolved in 40 mL deionized water under ultrasound, and then 1 mol/L KOH solution was added dropwise to the above solution until the obtained solution was different pH values (pH=10, 11, 12). The resulting solution was stirred at 95 ℃ for 24 h and then cooled to room temperature. The precipitate was washed several times with deionized water and absolute ethanol. And dried at 85 ℃ for 12 h. Finally, the basic support was obtained by calcining in air at 500 ℃ for 3 h.

1.3 Preparation of Ru/CeO₂ catalyst

In a typical synthesis procedure, the Ru/CeO₂ catalysts were prepared by the incipient wetness impregnation method. Briefly, the different basicity CeO₂ supports were impregnated with a Ruthenium (Ⅲ)-2, 4-pentanedionate solution at 25 ℃ for 12 h, and then dried at 120 ℃ for 12 h. The dried sample was reduced in H₂ (45 mL·min^-1) at 450 ℃ for 4 h and finally cooled to room temperature in H₂ atmosphere. Furthermore, the promoters were introduced the 1.25%Ru/CeO₂-11 catalyst using the incipient wetness impregnation with an aqueous solution of nitrates (Ba(NO₃)₂, KNO₃ and CsNO₃) at 25 ℃ for 24 h. The resulting samples were dried at 120 ℃ for 12 h. The obtained catalysts were denoted as X%M-1.25%Ru/CeO₂-11 (M = Ba, K or Cs).

1.4 Characterization of the catalysts

The samples were analyzed on a Rigaku Mini Flex Ⅱ benchtop X-ray diffractometer using Cu-Kα radiation (30 kV, 15 mA, λ = 0.154 18 nm). The morphology of CeO₂ supports was obtained by a scanning electron microscopy (Hitachi, SU8010) equipped with energy dispersive spectrometry (EDS). The Brunauer-Emmett-Teller (BET) surface areas and pore size distributions of the catalysts were measured with a Micromeritics TriStar Ⅱ 3020 instrument using adsorption of N₂ at -196 ℃. Before experiment, the catalyst was heated at 150 ℃ for 3 h under vacuum. The X-ray photoelectron spectroscopy (XPS) test was performed on an ESCALAB 220i-XL spectrometer by using Al Kα (1486.6 eV) as the X-ray source. The equipment base pressure was 3×10^-5 Pa, and the samples were characterized at room temperature. Detailed spectra were recorded for the region of Ru 3d, Ce 3d and O 1s photoelectrons with a 0.1 eV step. The H₂-temperature programmed reduction (H₂-TPR) and CO₂-temperature programmed desorption (CO₂-TPD) test were analyzed using a Micromeritics AutoChem Ⅱ 2920 chemisorption instrument with a sample loading of 50 mg. Prior to the H₂-TPR experiment, 100 mg catalyst was pretreated in Ar at 150 ℃ for 60 min, (30 mL·min^-1) and then in Ar flow cooled to 50 ℃. The reduction process was carried out in the temperature range of 50~800 ℃ in 10% H₂/Ar (30 mL·min^-1). Before CO₂-TPD experiment, the temperature was raised to 150 ℃ at a heating rate of 10 ℃·min^-1 under Ar atmosphere for 1 h to remove water molecules and other adsorbates, and then it was reduced at 450 ℃ for 2 h in H₂ atmosphere (30 mL·min^-1). After cooling to 50 ℃ in He atmosphere (30 mL·min^-1), the mixture 10% CO₂/He was then passed through for 1.5 h. And was purged 1.5 h under He atmosphere (30 mL·min^-1). The CO₂-TPD experiment was performed under He (30 mL·min^-1) heating from 50 to 900 ℃ at a rate of 10 ℃·min^-1.

1.5 Measurements of catalytic activity

The catalytic activity for ammonia synthesis was carried out in a conventional flow system using a fixed-bed tubular micro-reactor (length =700 mm, i.d. = 8 mm). Before measurements, the 150 mg catalysts were activated in a flowing gas (mixture of N₂ and H₂ at volume ratio is 1:3, flow rate of 60 mL·min^-1) under atmospheric pressure at different temperatures (200, 300, 400, and 500 ℃ for 2 h, respectively) following a linear heating ramp of 5 ℃·min^-1. Data were obtained when the ammonia synthesis reaction was stabilized for more than 2 h. The ammonia synthesis rates in the effluent were determined via a chemical titration method with a known amount of H₂SO₄ (Congo red as the indicator)^[23].

2 Results and discussion 2.1 Structural and morphological analysis

The powder X-ray diffraction (PXRD) of the different strongly basicity CeO₂ supports were shown in Fig. 1. The characteristic peaks at 2θ = 28.6°, 33.1°, 47.5°, 56.3°, 59.1°, 69.4°, 76.7° and 79.1° are attributed to the reflections of (111), (200), (220), (311), (222), (400), (331) and (420) planes, respectively. It clearly shows that the supports can be indexed to the fluorite CeO₂ cubic structure (PDF: 43-1002), with the lattice parameters of 5.411, 5.411, and 5.411 Å. No diffraction peaks of other impurities appeared in the CeO₂, indicating the high purity of CeO₂ supports. From the Fig. 2, it can be seen that no diffraction peaks of active metal Ru and promoters were detected, indicating little loading or good dispersion of the Ru and promoter.

Fig. 3 shows the SEM images of the different basicity CeO₂ supports (pH = 10, 11, and 12) respectively. It can be observed that the basicity CeO₂ supports are composed of massive, uneven size and wide range of size distribution.

2.2 Specific surface area and pore diameter distribution

The structures of CeO₂ supports and Ru/CeO₂ catalysts were further analyzed by N₂ adsorption-desorption measurements. Fig. 4 shows N₂ adsorption-desorption isotherms and pore size distributions of CeO₂ supports and Ru/CeO₂ catalysts. From the Fig. 4(a), it can be seen that they have H₃ type hysteresis loops, which is consistent with the adsorption-desorption characteristics of mesoporous materials. Compared with the specific surface area of CeO₂=10 and CeO₂=12, the CeO₂=11 support shows the highest specific surface area. It can be seen from Fig. 4(b) that the pore size distribution of the supports is as follows: CeO₂=11 (3.8 nm) < CeO₂=10 (4.2 nm) < CeO₂=12 (4.7 nm). The large specific surface area and small mesoporous pore size of CeO₂=11 support are favorable to the catalytic reaction. The BET surface area, pore volume and pore diameter of CeO₂ supports and catalysts are listed in Table 1. The specific surface area of CeO₂ is slightly decreased after the loading of promoters (Cs/K/Ba), which may be the promoters clog pores of the CeO₂ surface^[24].

2.3 CO₂ temperature-programmed desorption(CO₂- TPD)

The CO₂-TPD measurements were used to analyze the effect of the basicity of the supports for ammonia synthesis. It can be seen from the Fig. 5 that the order of the amount of CO₂ adsorption is: 1.25%Ru/CeO₂-11(1.52 cm³/g)> 1.25%Ru/CeO₂-10 (1.22 cm³/g) > 1.25%Ru/CeO₂-12 (1.01 cm³/g), which was concerned with the surface area of supports. Catalytic results show that the catalytic activity of 1.25%Ru/CeO₂-11 catalyst was far away than that of 1.25%Ru/CeO₂-10 and 1.25%Ru/CeO₂-12, which shows that the appropriate basic site density of the catalyst was conducive to the transfer of electrons to Ru.

2.4 H₂ temperature-programmed reduction (H₂-TPR)

The reducibility of 1.25%Ru/CeO₂-11 and 2%M-1.25%Ru/CeO₂-11 catalysts were investigated by H₂-TPR characterization. It is observed from Fig. 6 that the CeO₂-11 support exhibits strong reduction peak at 475 ℃, which could be attributed to the reduction of surface oxygen of cerium oxide. And the Ru/CeO₂ catalysts have two reduction peaks. The first reduction stage focuses on the reduction temperature around 100 ℃, which belongs to the reduction of RuO₂ → Ru. The second reduction stage mainly occurs in the range of reduction temperature of 290~300 ℃, which can be attributed to the reduction peaks of ceria surface oxygen shift to lower temperatures. Specifically, the H₂ consumption amount of ceria surface oxygen in the Ru/CeO₂ catalysts is far more than that in the CeO₂-11 supports. The metal-support interaction is a key factor for interfacial charge redistribution and mass transport, which affects the catalytic property. The Table 2 shows that the order of H₂ consumption for Ru phase reduction is: 2%Cs-1.25%Ru/CeO₂-11(24.40 mL/g) > 1.25%Ru/CeO₂-11> (21.95 mL/g) > 2%Ba-1.25%Ru/CeO₂-11(16.60 mL/g) > 2%K-1.25%Ru/CeO₂-11(10.18 mL/g), which is consistent with the order of ammonia synthesis activity. The activity of Ru-based catalyst increases with the degree of reduction of active metal Ru. Therefore, the ammonia synthesis activity of the catalysts is related to the reduction of the Ru phase.

2.5 X-ray photoelectron spectroscopy (XPS)

The surface chemical state of the 1.25%Ru/CeO₂-11 and 2%M-Ru/CeO₂-11 catalysts were revealed by the XPS analysis. The Ce 3d_5/2 and Ce 3d_3/2 orbitals are fitted into 10 peaks for the surface chemical state of Ce from the Fig. 7(a). The peaks near v₀ (882.0 eV), v′ (885.7 eV), u₀ (899.3 eV), u′ (902.1 eV) are characteristic peaks of the Ce³⁺ species, and the peaks near v (883.4 eV), v″ (889.1 eV), v''' (898.1 eV), u (900.7 eV), u″ (907.7 eV), u''' (916.1 eV) belongs to the characteristic peak of the Ce⁴⁺ species^[25]. The Ce³⁺ content reflects the degree of oxygen vacancy in the sample. As shown in Table 3, the 2%Cs-1.25% Ru/CeO₂-11 catalysts had the highest Ce³⁺/Ce⁴⁺ ratio than other catalysts. The order of Ce³⁺/Ce⁴⁺ values for the samples was: 2%Cs-1.25%Ru/CeO₂-11 (0.56) > 1.25%Ru/CeO₂-11 (0.52) > 2%Ba-1.25%Ru/CeO₂-11 (0.51) > 2%K-1.25%Ru/CeO₂-11 (0.36), which is in accordance with the results obtained from the H₂-TPR spectra. The conversion of Ce⁴⁺ to Ce³⁺ led to enhancement of the amount of oxygen vacancies.

The Fig. 7(b) shows the O 1s XPS spectra. There are two characteristic peaks of O at 532.0 eV (O₂^-) and 531.0 eV (O^-), which are mainly attributed to the surface adsorption oxygen species (O_ads) of the catalysts. The peak at 529.4 eV (O^2-) was the lattice oxygen species (O_lat) of the catalysts^[26]. The resulting ratio of O^-/O^2- for the catalysts follow the order of 2%Cs-1.25%Ru/CeO₂-11 (0.58) > 1.25%Ru/CeO₂-11 (0.45) > 2%Ba-1.25%Ru/CeO₂-11 (0.39) > 2%K-1.25%Ru/CeO₂-11 (0.37), which was consistent with the conclusion of the catalytic reaction.

The Ru 3d XPS spectra are shown in Fig. 7(c). The characteristic of metallic ruthenium (Ru⁰) appears near the binding energy of 280.2 eV, and a peak of Ru⁴⁺ appears around the binding energy of 280.7 eV. Above the standard RuO₂ binding energy at Ru 3d_5/2 (280.6 eV) indicates that the Ru phase enters the CeO₂ lattice in the Ce_1-xRu_xO₂ catalyst system^[27]. The binding energy of Ru 3d_5/2 decreases with the addition of promoters, indicating that the promoter can lower the work function of Ru and transfer more efficient charges to the reverse bond orbitals of N₂. More detailed data on XPS analysis of 1.25%Ru/CeO₂-11 and 2%M-Ru/CeO₂-11 catalysts are shown in Table 3.

2.6 Catalytic performance for Ammonia synthesis

The catalytic activity results of different basicity 1.25%Ru/CeO₂-X (X = 10/11/12) catalysts in the range of temperature (375~450 ℃) under 3.8MPa are shown in Fig. 8. The catalytic activity increased with increasing temperature from 375 to 450 ℃. It was worthy noting that the catalytic performance of 1.25%Ru/CeO₂-11 (7040 μmol·g^-1·h^-1) for ammonia synthesis was far away than that of 1.25%Ru/CeO₂-10 (4225 μmol·g^-1·h^-1) and 1.25%Ru/CeO₂-12 (2115 μmol·g^-1·h^-1). Additionally, the characterizations including CO₂-TPD, XPS and H₂-TPR demonstrate that the Ru particles supported on CeO₂-11 exhibits higher electron density compared to that of Ru/CeO₂-10 and Ru/CeO₂-12 catalyst.

It was worthy noting that only the introduction of the Cs promoter increases the catalytic activity, because the Cs promoter accelerates the electron transfer and promotes the reduction of Ru in Fig. 9. The K and Ba promoters inhibit the catalytic activity of the 1.25%Ru/CeO₂-11 catalyst. It shows that the Ce³⁺/Ce⁴⁺, O^-/O^2- ratio and reduction degree of Ru phase were the dominant factor in the catalytic performance. The catalytic performance of the promoter for ammonia synthesis was further investigated by increasing the amount of Cs additive. It can see from the Fig. 10, the highest reaction rate of 4%Cs-1.25%Ru/CeO₂-11 catalyst is 12 000 μmol·g^-1·h^-1 at 450 ℃ and 3.8 MPa, which is much higher than that of Ru catalysts supported on other supports. Excessive Cs reduces the catalytic activity, which may be due to the Cs covering the active site of the Ru.

3 Conclusions

In summary, the effect of the basic surface of the supports on the ammonia synthesis activity was investigated by synthesizing different basic CeO₂ supports. After the loading of Ru is 1.25%, the appropriately strengthened surface basic support significantly enhanced the catalytic activity of the Ru-based catalyst. By adjusting the surface properties of CeO₂, the electron transport ability of the support can be improved. It was found the CeO₂-11 displayed the best ammonia synthesis activity, and the addition of Cs could evidently improve the performance. The introduction of alkali metal and alkaline earth metal increased the reducing ability of the active metal Ru. The strong interaction between the metal and the support promotes the reduction of Ru and weakens the adsorption competition of hydrogen and nitrogen. Therefore, modulating surface property of cerium may be a promising catalyst for ammonia synthesis.

Acknowledgements: The authors acknowledge the financial support of the Natural Science Foundation of China (Grant No. 21671147), Scientific and Technological Innovation Programs of Higher Education Institutions in Shanxi, State Key Laboratory of Coal and CBM Co-mining (Grant No. 2016012004).