Elsevier

Applied Catalysis B: Environment and Energy
《应用催化 B 辑:环境与能源》

Volume 366, 5 June 2025, 125075
第 366 卷,2025 年 6 月 5 日,125075 页
Applied Catalysis B: Environment and Energy

Electrically driven gaseous ammonia decomposition on Co-based SiC composite catalysts for low-temperature H2 production
钴基碳化硅复合催化剂电驱动气态氨分解用于低温制氢

https://doi.org/10.1016/j.apcatb.2025.125075Get rights and content  获取权限与内容
Full text access  全文访问

Highlights  研究亮点

  • An electrically driven gaseous ammonia decomposition strategy for low-temperature hydrogen production was proposed.
    提出了一种电驱动气态氨分解策略用于低温制氢。
  • The ammonia decomposition reaction was carried out in the absence of external heating using Co-based SiC composite catalysts.
    在无外部加热条件下,采用钴基碳化硅复合催化剂实现了氨分解反应。
  • An NH3 conversion of 73 % was achieved at ca. 200℃ over SiC-supported catalyst (Co/SiC-Al2O3).
    在约 200℃条件下,碳化硅负载催化剂(Co/SiC-Al 2 O 3 )实现了 73%的氨转化率。
  • Hydrogen species migration is important for NHx dehydrogenation and N-N recombination.
    氢物种迁移对于 NH x 的脱氢和 N-N 重组至关重要。

Abstract  摘要

Ammonia (NH3) is regarded as a promising carbon-free hydrogen (H2) carrier due to its ease of storage. However, due to the slow kinetics, highly efficient H2 production from NH3 decomposition below 300℃ is hard to achieve, which hinders its practical multi-scenario application. Here, a promising electrically driven NH3 decomposition strategy using Co-based SiC composite catalysts is proposed. The electrical strategy was conducted without external heating over gaseous NH3, and the joule heating keeps the system at low temperatures. The differences in catalytic activity and mechanism of two main SiC composite catalysts, including SiC-supported catalyst (Co/SiC-Al2O3) and SiC-mediated catalyst (Co/Al2O3-SiC), are compared. Co/SiC-Al2O3 has a more pronounced synergistic effect with the electric field, on which an NH3 conversion of 73 % was achieved at ca. 200℃, outperforming state-of-the-art catalysts in conventional thermal NH3 decomposition. The systematic characterizations demonstrated that the charge migration stimulated by the electric field was important for intermediates evolution. This work shows the potential for highly efficient H2 production from NH3 with earth-abundant transition metals at low temperatures.
氨(NH₃)因其易于储存的特性被视为一种极具前景的无碳氢(H₂)载体。然而,由于反应动力学缓慢,在 300℃以下实现 NH₃高效分解制氢难以实现,这阻碍了其多场景实际应用。本文提出了一种采用钴基碳化硅复合催化剂的电驱动 NH₃分解新策略。该电驱动策略无需外部加热即可在气态 NH₃中实施,焦耳热效应使系统维持在低温状态。研究对比了两种主要碳化硅复合催化剂(包括碳化硅负载型催化剂 Co/SiC-Al₂O₃和碳化硅介导型催化剂 Co/Al₂O₃-SiC)的催化活性与反应机制差异。Co/SiC-Al₂O₃与电场具有更显著的协同效应,在约 200℃条件下可实现 73%的 NH₃转化率,性能优于传统热催化 NH₃分解的尖端催化剂。系统表征表明,电场激发的电荷迁移对反应中间体的演变具有重要作用。 本研究表明,利用地球储量丰富的过渡金属在低温条件下从氨气中高效制氢具有巨大潜力。

Keywords  关键词

Hydrogen
Ammonia decomposition
Electric field
Co-based catalysts
SiC

氢气 | 氨分解 | 电场 | 钴基催化剂 | 碳化硅

1. Introduction  1. 引言

In the past two hundred years, the use of carbon-based fossil fuels has emitted a large amount of carbon dioxide into the atmosphere, which has severely affected the development of human society due to its greenhouse effect. Therefore, researchers are in search of carbon-free fuels to replace fossil fuels [1]. Green hydrogen (H2), produced through renewable resources, is considered the ultimate energy with zero carbon emissions [2], [3]. However, traditional high-pressure hydrogen storage is challenging with low mass hydrogen density, high energy consumption, and the risk of leakage [4]. Alternatively, H2 can be stored in materials and released at high temperatures on demand. One of the promising materials is ammonia (NH3), which has a high gravimetric H2 storage capacity (17.6 wt%) and can be easily liquefied under mild conditions (20℃ and 0.8 MPa) [5]. Additionally, H2 produced from NH3 decomposition via wind and solar power is C-free and the unreacted NH3 is easily removed from the H2 stream by using adsorbents [6], which meets the needs of fuel cell vehicles (FCVs) [7]. All these features make NH3 an ideal hydrogen storage medium.
在过去两百年间,碳基化石燃料的使用向大气排放了大量二氧化碳,其温室效应严重影响了人类社会发展。为此,研究者们正寻求用零碳燃料替代化石燃料[1]。通过可再生能源制备的绿氢(H 2 )被视为零碳排放的终极能源[2][3]。然而传统高压储氢存在质量储氢密度低、能耗高及泄漏风险等技术瓶颈[4]。作为替代方案,H 2 可通过材料储存在高温条件下按需释放。其中极具前景的储氢材料氨(NH 3 )具有 17.6 wt%的高重量储氢密度,且能在温和条件(20℃、0.8 MPa)下液化储存[5]。此外,利用风能/太阳能驱动 NH 3 分解制取的 H 2 全程零碳排,未反应的 NH 3 可通过吸附剂从 H 2 流中高效脱除[6],完全满足燃料电池汽车(FCVs)的用氢需求[7]。这些特性使 NH 3 成为理想的储氢介质。
NH3 decomposition reaction requires high temperatures (> 400℃) because the reaction is endothermic (NH3 → 1/2N2 + 3/2H2, ΔH = +46 kJ mol−1). High-temperature environment makes onsite H2 storage less energy efficient, limits the usage scenarios, and decreases the service life of catalysts. It is generally believed that Ru is the most active metal for NH3 decomposition [8]. NH3 decomposition on Ru/CNTs promoted by K has been reported to approach thermodynamic equilibrium at temperatures as low as 450°C [9]. However, the scarcity and high cost of Ru drive efforts to develop more earth-abundant catalysts for a large-scale application. Recently, Hassina et al. [10] reported a highly efficient CoNi bimetal catalyst that approached complete NH3 conversion at 450℃, showing the prospects of Co-based catalysts. Unfortunately, when the temperature was further reduced to 400℃, the NH3 conversion was only about 40 %. Further improvements in low-temperature performance by complex catalyst design need to overcome the uncontrollable costs. Accordingly, the researchers tried to incorporate external field to enhance catalytic activity [11], [12], such as plasma catalysis [13], [14] and electro-catalysis [15]. Among these, NH3 decomposition assisted by the adequate DC electric field has shown great potential to enhance the catalytic activity at low temperatures [16], [17]. Ofuchi Y, et al. achieved high NH3 conversion at markedly lower temperatures using an applied electric field on Ru/CeO2 catalyst and found the rate-determined step changed by surface protonics [18]. The electric field-assisted NH3 decomposition is a non-Faradaic process with lower electrical input power, which is more efficient than the electrolysis of NH3 [15] and plasma catalysis [13].
氨分解反应需要高温(>400℃),因为该反应为吸热过程(NH₃ → 1/2N₂ + 3/2H₂,ΔH = +46 kJ mol⁻¹)。高温环境导致现场制氢能效降低、应用场景受限,同时会缩短催化剂使用寿命。学界普遍认为钌(Ru)是氨分解活性最高的金属催化剂[8]。研究表明,钾(K)促进的 Ru/CNTs 催化剂在低至 450℃时即可接近热力学平衡转化率[9]。然而钌的稀缺性和高昂成本促使研究者致力于开发更廉价的地球丰产元素催化剂。近期 Hassina 等人[10]报道的 CoNi 双金属催化剂在 450℃下可实现氨的完全转化,展现了钴基催化剂的潜力。但温度进一步降至 400℃时,其氨转化率仅约 40%。若通过复杂催化剂设计进一步提升低温性能,则需面临成本不可控的难题。 为此,研究人员尝试引入外场以增强催化活性[11][12],例如等离子体催化[13][14]和电催化[15]。其中,在适当直流电场辅助下的 NH3 分解在低温条件下展现出显著提升催化活性的潜力[16][17]。Ofuchi Y 等学者通过在 Ru/CeO2 催化剂上施加电场,实现了显著低温条件下的高 NH3 转化率,并发现表面质子传导改变了速率决定步骤[18]。这种电场辅助的 NH3 分解是一种非法拉第过程,其电能输入需求低于 NH3 电解[15]和等离子体催化[13]。
Common oxide supports such as Al2O3, SiO2, ZrO2, and MgO with high resistance would severely hinder current transmission (Table. S1), whereas other oxide semiconductors, such as CeO2, that often used to transport the electric field in previous studies [12], [16], suffered from conductive instability and serious discharge without external heating (Fig. S1). Additional external heating limits the further reduction of reaction temperature, which makes the high catalytic synergistic effect of the electric field with catalyst cannot fully develop. SiC is a widely used semiconductor in the electronics industry, which has an excellent electric field transport capability due to its wide bandgap and high thermal conductivity [19]. Therefore, it can be used to enhance the catalyst’s electric field transportation ability at low temperatures. Moreover, SiC has also been used in conventional thermocatalysis due to strong heat durability [20], [21]. Here, the electrically driven NH3 decomposition strategy without external heating using Co-based SiC composite catalysts is proposed. Due to stable electric activation and efficient thermal activation for the catalytic process, highly efficient NH3 decomposition was realized at ca. 200℃. The differences in catalytic activity and mechanism of two main SiC composite catalysts, including SiC-supported catalyst (Co/SiC-Al2O3) and SiC-mediated catalyst (Co/Al2O3-SiC), are compared.
常见的氧化物载体如 Al₂O₃、SiO₂、ZrO₂和 MgO 因具有高电阻会严重阻碍电流传输(表 S1),而其他氧化物半导体(如先前研究中常用于传输电场的 CeO₂[12,16])则存在导电不稳定性和无外部加热时严重放电的问题(图 S1)。额外的外部加热限制了反应温度的进一步降低,使得电场与催化剂的高效催化协同效应无法充分发挥。SiC 作为电子工业中广泛应用的半导体材料,凭借其宽带隙和高导热性[19]展现出优异的电场传输能力,因此可用于增强催化剂在低温下的电场传输性能。此外,由于 SiC 具有出色的耐热性[20,21],也常被用于传统热催化领域。本研究提出采用钴基 SiC 复合催化剂实现无需外部加热的电驱动 NH₃分解策略。 由于稳定的电激活和高效的热激活催化过程,在约 200℃下实现了高效的 NH3 分解。比较了两种主要碳化硅复合催化剂(包括碳化硅负载型催化剂 Co/SiC-Al2O3 和碳化硅介导型催化剂 Co/Al2O3-SiC)在催化活性与反应机理上的差异。

2. Experimental  2. 实验部分

2.1. Catalyst preparation
2.1. 催化剂制备

Co-based SiC composite electric field catalysts were prepared using two facile steps: deposition-precipitation (DP) [22] and high-pressure grinding. Briefly, 2.0 g γ-Al2O3 (analytical reagent (AR) grade, Aladdin) was suspended in 100 mL of D.I. water and 1.975 g Co (NO3)2·6H2O (99.99 % metals basis, Macklin) was dissolved in 50 mL of D.I. water. Then, the Co precursor solution was added dropwise into the suspension with a nominal Co loading of 15.7 wt%. After 3 h of vigorously stirring, the pH was controlled to 9.5 by ammonia solution dropwise. The resulting suspension was stirred at room temperature overnight. The obtained precipitation was separated by filtration, washed with deionized water, and dried in an air oven at 70℃ for 8 h. Then, the sample was ground and calcined in air at 600℃ for 4 h with a ramp rate of 5℃/min. The obtained sample was denoted Co/Al2O3. Co/SiC was synthesized using SiC (40 nm, Macklin) and Co (NO3)2·6H2O in the same way as Co/Al2O3, except that, Ar was used to protect it during calcination at 600℃ for 4 h with ramp rate of 5℃/min, because SiC is easily oxidized at high temperatures, resulting in poor conductivity. For comparison, Co/(Al2O3-SiC) was synthesized by supporting Co on a mixture of Al2O3 and SiC (mass ratio = 1: 1) in the same way as Co/SiC.
采用沉积-沉淀法(DP)[22]与高压研磨两步简易工艺制备钴基碳化硅复合电场催化剂。具体步骤为:将 2.0 克γ-Al₂O₃(分析纯级,阿拉丁试剂)悬浮于 100 毫升去离子水中,同时将 1.975 克 Co(NO₃)₂·6H₂O(金属基纯度 99.99%,麦克林试剂)溶于 50 毫升去离子水。随后将钴前驱体溶液以 15.7 wt%的标称载钴量逐滴加入悬浮液,剧烈搅拌 3 小时后通过氨水逐滴调节 pH 至 9.5。所得悬浮液室温下持续搅拌过夜,经抽滤分离沉淀物,去离子水洗涤后于 70℃空气烘箱中干燥 8 小时。研磨后的样品以 5℃/分钟升温速率在 600℃空气氛围中煅烧 4 小时,最终产物标记为 Co/Al₂O₃。 采用与 Co/Al₂O₃相同的制备方法,以 SiC(40 纳米,麦克林)和 Co(NO₃)₂·6H₂O 为原料合成 Co/SiC 催化剂。不同之处在于,由于 SiC 在高温下易被氧化导致导电性下降,煅烧过程需在氩气保护下以 5℃/分钟的升温速率升至 600℃并保持 4 小时。为进行对比,按照 Co/SiC 相同方法,将 Co 负载于 Al₂O₃与 SiC(质量比 1:1)的混合物上,制得 Co/(Al₂O₃-SiC)复合催化剂。
Co/Al2O3 with high resistance severely hinders current transmission, whereas, Co/SiC with too high conductivity, limits the increase of electric field intensity (Table S1) and shows a low activity (Fig. S2). To obtain the electric field catalysts with moderate conductivity, the catalyst resistance is controlled by high-pressure grinding to introduce the second component of the catalyst. The determination of the two components’ ratio is described in Table. S1. For SiC-mediated catalyst, briefly, fully ground Co/Al2O3 and SiC (mass ratio = 1: 1) were mixed by a vortex oscillator for 30 min. Then, the mixture was vigorously ground in a stainless-steel mortar for another 30 min. The obtained sample was pressed at a pressure of 15 MPa for 20 minutes, and then it was crushed and sieved into particles of 178–425μm. The obtained SiC-mediated catalysts were denoted Co/Al2O3-SiC. Similarly, SiC-supported catalyst (Co/SiC-Al2O3) was prepared in the same way using fully ground Co/SiC and Al2O3 (mass ratio = 2: 1). Since Al2O3 and SiC alone have negligible catalytic activity for NH3 decomposition (Fig. S3), the activity tests and characterizations conducted in this work were based on the same mass of Co/Al2O3, Co/SiC and Co/(Al2O3-SiC).
高电阻的 Co/Al₂O₃严重阻碍电流传输,而导电性过高的 Co/SiC 则限制了电场强度的提升(表 S1),并表现出较低的活性(图 S2)。为获得具有适中导电性的电场催化剂,通过高压研磨引入催化剂第二组分来控制电阻。两组分比例的确定方法详见表 S1。对于 SiC 介导的催化剂,简要步骤如下:将充分研磨的 Co/Al₂O₃与 SiC(质量比=1:1)通过涡旋振荡器混合 30 分钟,随后在不锈钢研钵中强力研磨 30 分钟。所得样品在 15 MPa 压力下压制 20 分钟,经破碎筛分获得 178-425μm 颗粒。制得的 SiC 介导催化剂标记为 Co/Al₂O₃-SiC。同理,采用相同方法使用充分研磨的 Co/SiC 与 Al₂O₃(质量比=2:1)制备了 SiC 负载型催化剂(Co/SiC-Al₂O₃)。由于单独的 Al₂O₃和 SiC 对 NH₃分解的催化活性可忽略不计(图 S3),本研究中进行的活性测试与表征均基于相同质量的 Co/Al 2 O 3 、Co/SiC 及 Co/(Al 2 O 3 -SiC)催化剂。

2.2. Catalyst characterization
2.2. 催化剂表征

Co content of each catalyst was determined by inductively coupled plasma optical emission spectroscopy (ICP-OES) on an Avio 500 instrument. XRD of the catalysts was performed in an X-ray diffractometer (D8 ADVANCE DaVinci) employing Cu Kα radiation (λ = 1.5406 Å). Raman spectra were collected using a Renishaw inVia Qontor spectrometer equipped with an excitation laser beam of 532 nm. Average pore size, pore volume, and BET were determined by N2 adsorption-desorption measurements at liquid nitrogen temperature using a Quanta chrome NOVA 2000e automated gas sorption instrument. XPS was obtained with a multifunctional X-ray photoelectron spectrometer *ESXCALAB Xi+ (Thermo Fisher) with an Al Kα radiation source, and corrected using a C1s signal at 284.8 eV. TEM and HRTEM were performed on a JEOL JEM-2100 analytical transmission electron microscope operating at an accelerating voltage of 200 kV.
采用 Avio 500 型电感耦合等离子体发射光谱仪(ICP-OES)测定各催化剂钴含量。催化剂 X 射线衍射(XRD)测试在配备 Cu Kα辐射源(λ=1.5406 Å)的 D8 ADVANCE DaVinci 型 X 射线衍射仪上进行。拉曼光谱通过配备 532 nm 激发激光源的 Renishaw inVia Qontor 共聚焦拉曼光谱仪采集。采用 Quantachrome NOVA 2000e 全自动气体吸附仪在液氮温度下进行 N₂吸附-脱附测试,测定平均孔径、孔容及 BET 比表面积。X 射线光电子能谱(XPS)测试使用配备 Al Kα辐射源的 Thermo Fisher ESCALAB Xi+多功能光电子能谱仪完成,并以 284.8 eV 的 C1s 峰位进行荷电校正。透射电子显微镜(TEM)及高分辨透射电子显微镜(HRTEM)分析在加速电压 200 kV 的 JEOL JEM-2100 型分析型透射电镜上完成。
H2 temperature-programmed reduction (H2-TPR) was carried out on BUILDER PCA-1200 equipped with a TCD. First, the sample (100 mg) was dehydrated at 200℃ for 30 min in Ar. After cooling to 30℃, the gas was switched to 10 vol% H2/Ar (30 mL/min), and the temperature was increased to 800℃ (10℃/min).
采用配备 TCD 检测器的 BUILDER PCA-1200 仪器进行氢气程序升温还原(H2-TPR)测试。首先将 100 mg 样品在氩气氛围下 200℃脱水 30 分钟,冷却至 30℃后切换为 10 vol% H2/Ar 混合气(流速 30 mL/min),并以 10℃/min 的升温速率加热至 800℃。
CO2 temperature-programmed desorption (CO2-TPD) was performed with the same instrument. The sample (100 mg) was placed in the reactor and reduced at 600℃ for 1 h with 10 vol% H2/Ar (30 mL/min) followed by a purge process with He (30 mL/min) to remove residual H2. Pure CO2 (30 mL/min) was then introduced for pre-adsorption for 1 h at 30℃. It was then purged again with He (30 mL/min) to remove physisorbed CO2 for another 30 min. TPD was then performed at 10℃/min in He flows (30 mL/min). To prepare the electric field-treated samples, the catalysts were electrified for 3 h at 8 W before the experiment.
采用相同仪器进行一氧化碳程序升温脱附(CO-TPD)测试。将 100 mg 样品置于反应器中,在 10 vol% H₂/Ar(30 mL/min)气氛下 600℃还原 1 小时,随后用 He(30 mL/min)吹扫去除残留氢气。接着在 30℃下通入纯 CO(30 mL/min)进行 1 小时预吸附,再用 He(30 mL/min)吹扫 30 分钟去除物理吸附的 CO。最后以 10℃/min 升温速率在 He 气流(30 mL/min)中进行 TPD 测试。电场处理样品制备方法为:实验前对催化剂施加 8 W 功率通电处理 3 小时。
NH3 temperature-programmed desorption (NH3-TPD) was carried out the same way as CO2-TPD, but the pre-adsorption gas was 10 vol% NH3/Ar. To prepare the electric field-treated samples, the catalysts were electrified for 3 h at 8 W before the experiment.
氨气程序升温脱附(NH₃-TPD)测试流程与 CO-TPD 相同,但预吸附气体改为 10 vol% NH₃/Ar 混合气。电场处理样品制备方法为:实验前对催化剂施加 8 W 功率通电处理 3 小时。
H2 temperature-programmed desorption (H2-TPD) was also carried out the same way as CO2-TPD, but the pre-adsorption gas was 10 vol% H2/Ar, and the carrier gas during temperature desorption was changed to Ar.
氢气程序升温脱附(H₂-TPD)测试流程同样参照 CO-TPD 方法,但预吸附气体采用 10 vol% H₂/Ar 混合气,且程序升温脱附阶段载气更换为 Ar 气。
H2 electrically driven desorption (H2-EDD) was carried out with an in-house designed electric field reactor equipped with mass spectrometry (Hiden Analytical, HPR-20 EGA). The pretreatment is the same as H2-TPR. After cooling to 30℃, 1 vol% H2/Ar was introduced into the reactor. After the system stabilized (20 min), the electric field was applied on the catalysts, and the desorption gas was monitored by MS. After a few minutes, the electric field was switched off. Four H2-EDD were performed at a different electrical input power of 3, 5, 7, and 9 W.
氢电驱动脱附(H-EDD)实验采用自主设计的配备质谱仪(Hiden Analytical,HPR-20 EGA)的电场反应器进行。预处理步骤与 H-TPR 相同。待系统冷却至 30℃后,向反应器通入 1 vol% H2/Ar 混合气。系统稳定 20 分钟后,对催化剂施加电场,并通过质谱监测脱附气体。数分钟后关闭电场。分别在 3、5、7 和 9 W 四种不同电输入功率下进行了四次 H-EDD 测试。
NH3 surface reaction was carried out with an in-house designed temperature-programmed reactor equipped with mass spectrometry. The pre-activation procedure is the same as CO2-TPD. Then, for the thermal surface reaction, the temperature of the catalyst was maintained at 300℃, while for the electrical surface reaction, the electric field (8 W) was applied to the catalyst and maintained until the end of the test. After that, 2 vol% NH3/Ar (100 mL/min) was introduced into the reactor. Mass spectrometry was used to monitor the NH3 surface reaction.
氨表面反应实验采用自主设计的程序升温反应器配合质谱仪完成。预活化流程与 CO-TPD 相同。对于热表面反应,催化剂温度维持在 300℃;而电表面反应则对催化剂施加 8 W 电场并持续至测试结束。随后向反应器通入 2 vol% NH3/Ar 混合气(流速 100 mL/min),利用质谱仪监测 NH3 表面反应过程。
Product desorption of the used catalysts was carried out using the same instrument as the NH3 surface reaction. First, the catalysts were tested in pure NH3 at 400℃ for 2 hours. After cooling to 30℃, the gas was changed to Ar (30 mL/min) to remove the residual NH3. Then, for temperature-programmed desorption (TPD), the temperature was increased to 600℃ (10℃/min); while for electrically powered desorption, the electric field (8 W) was applied to the catalyst in a short time. The products desorption was monitored by mass spectrometry.
对使用过的催化剂进行产物脱附时,采用了与 NH 3 表面反应相同的仪器。首先,在 400℃纯 NH 3 气氛中对催化剂进行 2 小时测试。冷却至 30℃后,切换为氩气(30 mL/min)以去除残留 NH 3 。随后进行程序升温脱附(TPD)时,以 10℃/min 速率升温至 600℃;而电驱动脱附实验则是在短时间内对催化剂施加 8W 电场。所有脱附产物均通过质谱仪进行实时监测。
In-situ Diffuse Reflectance Infrared Fourier Transform (DRIFT) experiments were taken on an FTIR spectrometer (Nicolet 6700) to gain insight into the evolving species on the catalyst during reaction. Spectra were recorded at a scanning number of 32 scans with a resolution of 4 cm−1. Prior to the experiment, each sample was reduced in 20 % H2/Ar (50 mL/min) at 400 ℃ for 1 h. The background spectra at the measurement temperature were collected in flowing pure Ar. Then, the gas was switched to 5 % NH3/Ar (50 mL/min).
采用原位漫反射红外傅里叶变换光谱(DRIFT)在傅里叶变换红外光谱仪(Nicolet 6700)上进行实验,以研究反应过程中催化剂表面物种的演变。光谱采集参数设置为扫描次数 32 次,分辨率 4 cm −1 。实验前,每个样品均在 20% H 2 /Ar(50 mL/min)气氛中于 400℃下还原 1 小时。测量温度下的背景谱图在纯 Ar 气流中采集,随后将气氛切换为 5% NH 3 /Ar(50 mL/min)。

2.3. Catalytic activity tests
2.3. 催化活性测试

As shown in Fig. S4, the electric field reactor was modified from a columnar quartz tube reactor (OD: 12 mm, ID: 7 mm). 100 mg of the catalysts (Co/support) were packed into the space between the two filters prepared from foamy copper (as current collectors) in the reaction cell. Two separate copper tubes serving as electrodes pass through both ends of the reactor and are connected to a power supply (Teslaman TD2200). Copper (Cu) is generally considered a metal with poor catalytic activity for NH3 decomposition, thus, the effects of the foamy copper and electrodes on catalytic reaction are negligible. The thermal diffusion condition of the reactor was natural convection without external heating unless otherwise specified. Joule heating brought a certain temperature rise, monitored in operando by an infrared (IR) camera (HIKMICRO, HM-TPK20–3AQF/W). For the water-cooling test, the double-layer quartz served as a reactor with a 5℃ recirculating cooling water flowing through the outer layer of the reactor. For comparison, conventional thermal NH3 decomposition was heated by a temperature-programmed tube furnace. The products were monitored by a gas chromatography (GC-9860) equipped with TCD.
如图 S4 所示,电场反应器由柱状石英管反应器(外径 12 毫米,内径 7 毫米)改造而成。将 100 毫克催化剂(Co/载体)填充于反应池内两块泡沫铜(作为集流体)制备的滤网之间。两根独立铜管作为电极贯穿反应器两端,连接至电源(Teslaman TD2200)。铜(Cu)通常被认为是氨分解催化活性较差的金属,因此泡沫铜和电极对催化反应的影响可忽略不计。除非另有说明,反应器的热扩散条件为自然对流且无外部加热。焦耳热效应导致一定温升,通过红外热像仪(HIKMICRO HM-TPK20–3AQF/W)进行原位监测。水冷测试采用双层石英反应器,5℃循环冷却水流经反应器外层。作为对比实验,传统热催化氨分解采用程序升温管式炉加热。 产物通过配备热导检测器(TCD)的气相色谱仪(GC-9860)进行监测。
The activity tests were evaluated by introducing pure gaseous NH3 (99.999 %) into the reactor at atmospheric pressure. For conventional thermal NH3 decomposition, the catalysts were pre-reduced at 600℃ in 70 vol% H2/Ar for 1 h. Then the tests were performed from 400℃ to 650℃ with a step of 50℃. For electrically driven NH3 decomposition, the catalyst was pretreated at an electrical input power of 8 W in NH3 for 15 min before changing the power from 5 to 11 W for activity tests. The NH3 conversion XNH3(%) is obtained according to Eq. (1). The H2 production rate rH2 (mmol/gcat./s) is obtained using Eq. (2).
活性测试通过向反应器中通入纯气态 NH₃(99.999%)并在常压下进行评价。对于常规热 NH₃分解,催化剂需先在 600℃、70 vol% H₂/Ar 混合气中预还原 1 小时。随后测试温度从 400℃逐步升至 650℃,升温间隔为 50℃。对于电驱动 NH₃分解,催化剂需先在 8W 电功率的 NH₃气氛中预处理 15 分钟,随后将功率调节至 5-11W 范围内进行活性测试。NH₃转化率根据公式(1)计算得出,H₂产率(mmol/g/s)则通过公式(2)求得。
(1)XNH3=VNH3,inVNH3,outVNH3,in×100(2)rH2=VNH3,in22.4XNH31.5mcat.60Where VNH3,in and VNH3,out are the NH3 flow rate for inlet and outlet gas, respectively. mcat. is the mass of Co/support.
其中 VNH3,inVNH3,out 分别代表进出口气体中 NH 3 的流量,m cat. 为钴基载体催化剂的质量。

2.4. Kinetic studies  2.4 动力学研究

The apparent activation barrier was obtained by fitting the Arrhenius plot, as expressed in Eq. (3). Under electrical conditions, the external heating was applied to construct a temperature gradient while the electrical input power was kept at 4.6 W. The NH3 conversion was controlled by less than 25 % to exclude the mass and heat transfer limitation by adjusting the flow rate and reaction temperatures.
通过拟合阿伦尼乌斯曲线(如公式(3)所示)获得表观活化能。在通电条件下,维持电输入功率为 4.6W 的同时施加外部加热以构建温度梯度。通过调节流量和反应温度,将 NH 3 转化率控制在 25%以下以排除传质和传热限制。
(3)k=k0eEaRTWhere k and k0 are reaction constant (mol/gcat./s) and frequency factor (mol/gcat./s), respectively; R is universal gas constant (kJ mol−1 K−1); T is Kelvin temperature (K).
式中 k 和 k 0 分别为反应常数(mol/g cat. /s)和频率因子(mol/g cat. /s);R 为通用气体常数(kJ mol −1 K −1 );T 为开尔文温度(K)。
Reaction orders were measured by varying mixture (NH3/H2/N2/Ar) concentrations. The reaction orders under thermal conditions were measured at 500℃. The reaction orders under electrical conditions were measured at an electrical power of 4.6 W and the temperatures were controlled at ca. 500℃ by an external heat source. The results were obtained by Eq. (4) and Eq. (5).
通过改变混合气体(NH₃/H₂/N₂/Ar)浓度测定反应级数。热条件下反应级数在 500℃下测定,电条件下反应级数在 4.6 W 电功率下测定,并通过外部热源将温度控制在约 500℃。结果通过式(4)和式(5)获得。
(4)rNH3=kpNH3αpH2βpN2γ(5)logγNH3=logk+αlog[NH3]+βlog[H2]+γlog[N2]Where rNH3 is NH3 reaction rate (mol/gcat./s); α, β, γ are the reaction orders of NH3, H2, and N2, respectively; [NH3], [H2] and [N2] are partial pressure. It is widely acknowledged that the reaction order of N2 (γ) is approximately zero and the presence of N2 does not influence the activity [23], [24], which was also found in the electric field (Fig. S5), so the N2 dependence was not involved in the reaction order analysis.
式中 rNH3 表示 NH₃反应速率(mol/g·cat/s);α、β、γ分别为 NH₃、H₂和 N₂的反应级数;[NH₃]、[H₂]和[N₂]为分压。学界普遍认为 N₂的反应级数(γ)近似为零,且 N₂的存在不影响反应活性[23,24],电场条件下也发现此现象(图 S5),因此反应级数分析未考虑 N₂依赖性。

3. Results and discussion
3. 结果与讨论

3.1. Comparison of NH3 decomposition activity
3.1. NH₃分解活性对比

The resistivity of the obtained SiC-supported catalyst (Co/SiC-Al2O3) and SiC-mediated catalyst (Co/Al2O3-SiC) is comparable (Table. S1), therefore, the synergistic effects of the electric field and catalysts under the same electric field intensity can be investigated. The electrically driven NH3 decomposition was carried out without external heating and insulation, as shown in Fig. S4, which makes the device portable. However, due to the joule heating in the electric field, the catalysts had a temperature rise. As shown in Fig. 1a-h, infrared thermograms of the reactor demonstrated that the temperatures in the electric field were below ca. 200℃, which were far lower than those in conventional thermal NH3 decomposition (usually > 400℃). The energy consumption of NH3 decomposition includes chemical energy conversion (i.e., NH3 → 1/2N2 + 3/2H2, ΔH = +46 kJ mol−1) and heat dissipation. Chemical energy conversion is constant when NH3 conversion is the same, while lower temperature in the electric field means less heat dissipation, thus, reducing the overall energy consumption. In conventional thermal NH3 decomposition, as shown in Fig. 2a, the Co-based catalysts activated at 400℃ and needed to reach ca. 600℃ for 70 % NH3 conversion and exceed 650℃ for full NH3 conversion. The activity sequence follows Co/Al2O3-SiC > Co/(Al2O3-SiC) > Co/SiC-Al2O3. The SiC-mediated catalyst (Co/Al2O3-SiC) shows the best activity, indicating Co directly loaded on Al2O3 can reach better thermal activation.
所得 SiC 载体催化剂(Co/SiC-Al 2 O 3 )与 SiC 介导催化剂(Co/Al 2 O 3 -SiC)的电阻率相当(表 S1),因此可在相同电场强度下研究电场与催化剂的协同效应。电驱动 NH 3 分解反应在无需外部加热与绝热的条件下进行(图 S4 所示),这使得装置具有便携性。然而由于电场中的焦耳热效应,催化剂存在温升现象。如图 1a-h 所示,反应器的红外热成像图表明电场中的温度始终低于约 200℃,远低于传统热催化 NH 3 分解温度(通常>400℃)。NH 3 分解的能耗包含化学能转化(即 NH 3 →1/2N 2 +3/2H 2 ,ΔH=+46 kJ mol −1 )与热耗散两部分。当 NH 3 转化率相同时,化学能转化值为定值,而电场中较低的反应温度意味着更少的热耗散,从而降低总能耗。如图示传统热催化 NH 3 分解过程中... 2a 图中显示,钴基催化剂在 400℃下活化,需升温至约 600℃才能实现 70%的氨气转化率(文献 13),而完全转化氨气(文献 14)则需要超过 650℃。催化活性顺序为:Co/Al₂O₃-SiC > Co/(Al₂O₃-SiC) > Co/SiC-Al₂O₃。其中碳化硅介导的催化剂(Co/Al₂O₃-SiC)表现出最佳活性,这表明直接将钴负载在 Al₂O₃上可获得更优的热活化性能。
Fig. 1
  1. Download: Download high-res image (312KB)
    下载:下载高分辨率图片(312KB)
  2. Download: Download full-size image
    下载:下载全尺寸图片

Fig. 1. (a) Visible infrared thermogram of the reactor at an electrical input power of 5 W. (b)-(h) Infrared rainbow thermograms of the reactor at an electrical input power of (b) 5 W, (c) 6 W, (d) 7 W, (e) 8 W, (f) 9 W, (g) 10 W, (h) 11 W.
图 1. (a) 反应器在 5W 电输入功率下的可见光-红外热成像图。(b)-(h) 反应器在不同电输入功率下的红外彩虹热成像图:(b) 5W,(c) 6W,(d) 7W,(e) 8W,(f) 9W,(g) 10W,(h) 11W。

Fig. 2
  1. Download: Download high-res image (247KB)
    下载:下载高分辨率图片(247KB)
  2. Download: Download full-size image
    下载:下载全尺寸图片

Fig. 2. (a) Catalytic activity of conventional thermal NH3 decomposition, WHSV: 12,000 mL gcat.−1 h−1; (b) Catalytic activity of electrically driven NH3 decomposition, WHSV: 12,000 mL gcat.−1 h−1. (c) Comparison of T70 with other reported catalysts: 1: 1.22 %Na-Co3O4 [25], 2: Ru/La0.03Ce0.67 [22], 3: 10Ni/Al1Ce0.05Ox [26], 4: Ru/MgO(111) [23], 5: HEA-Co25Mo45 [27], 6: K-CoNialloy-MgO-CeO2-SrO [10], 7: Cs0.018Co3Mo3N [28], 8: 1.0Ru/CeO2 [29], 9: Ru/Sm2O3 [30]. E represents the electrical condition.
图 2. (a)传统热催化 NH₃分解活性,重时空速:12,000 mL·g⁻¹·h⁻¹;(b)电驱动 NH₃分解催化活性,重时空速:12,000 mL·g⁻¹·h⁻¹。(c)与其他报道催化剂的 T₅₀对比:1: 1.22%Na-Co₃O₄[25],2: Ru/La₂Ce₂[22],3: 10Ni/Al₂Ce₃O₇[26],4: Ru/MgO(111)[23],5: HEA-Co₄Mo₃[27],6: K-CoNi-MgO-CeO₂-SrO[10],7: Cs₂Co₃Mo₂N[28],8: 1.0Ru/CeO₂[29],9: Ru/Sm₂O₃[30]。E 代表电驱动条件。

In comparison, as shown in Fig. 2b, the performance of electrically driven NH3 decomposition is excellent in spite of the low temperatures. But unlike conventional thermal NH3 decomposition, Co/SiC-Al2O3 shows the best catalytic performance in the electric field (7–11 W) while the performance of Co/Al2O3-SiC is the worst, indicating that Co directly loaded on SiC can synergize better with the electric field. As the electrical input power increases to 11 W, the activity of these catalysts gradually converges because the reaction approaches the thermodynamic equilibrium at the corresponding temperature, where an NH3 conversion of 73 % was achieved for Co/SiC-Al2O3. To enable a visual comparison, the activity is further compared with the selected state-of-the-art catalysts in conventional thermal NH3 decomposition, as shown in Fig. 2c. Due to the limitation of thermodynamic equilibrium at low temperatures, NH3 cannot be fully converted, so the activity of T70 (the temperature at which 70 % of NH3 is converted) is compared. The performance of the electrically driven strategy based on non-noble metal in this work outperforms the reported state-of-the-art catalysts including Ru-based catalysts in conventional thermal NH3 decomposition, where the T70 is reduced by at least 150 ℃.
相比之下,如图 2b 所示,电驱动氨分解在低温条件下仍表现出优异性能。但与常规热催化氨分解不同,Co/SiC-Al₂O₃在电场(7-11 W)中展现出最佳催化活性,而 Co/Al₂O₃-SiC 性能最差,这表明直接负载于 SiC 上的钴能与电场产生更好的协同效应。当电输入功率增至 11 W 时,这些催化剂的活性逐渐趋同,因为反应接近相应温度下的热力学平衡状态,此时 Co/SiC-Al₂O₃实现了 73%的氨转化率。为便于直观对比,图 2c 进一步将其活性与精选的常规热催化氨分解前沿催化剂进行对照。由于低温热力学平衡的限制,氨无法完全转化,因此采用 T₇₀(实现 70%氨转化所需的温度)作为活性比较指标。 本研究基于非贵金属的电驱动策略在性能上超越了包括钌基催化剂在内的已报道最先进催化剂,在常规热 NH3 分解中将反应温度降低了至少 150℃。

3.2. Possible mechanism  3.2. 可能的反应机理

The electric field established in the catalyst would result in migration of current carrier and joule heating, there are reasons to believe that Al2O3 and SiC play different roles in thermal activation and current carrier-enhanced activation since the activities of SiC-mediated and SiC-supported catalysts were different between conventional thermal NH3 decomposition and electrically driven NH3 decomposition. To give insight into the intrinsic activity of current carrier-enhanced activation, a homemade reactor (Fig. S6a) wrapped with a flow of constant temperature (5℃) water was constructed to eliminate the joule heating as far as possible, so that the catalyst maintained at low temperatures in the electric field (Fig. S6b). As shown in Fig. 3a, the catalytic activities decreased to some extent but still maintained a relatively high activity compared with that in conventional thermal NH3 decomposition. This result indicates that the catalytic activity in the electric field contains current carrier-enhanced activity and partial joule thermal activity. The former is the main source of intermediate activation, while the latter further thermally activates the catalyst. Since the partial joule thermal activity existed in low temperatures in the electric field (< 200℃) compared with those in the thermal NH3 decomposition (> 400℃), it is suggested that the thermal activation by joule heating is more efficient than conventional external heating for active site activation. This phenomenon is somewhat like the photothermal effect in hot carrier-driven photocatalysis, in which hot carriers stimulated by non-radiative decay of localized surface plasmons can be more energetic than those produced by direct photoexcitation [31]. On the other hand, the activity of Co/SiC-Al2O3 still outperforms Co/Al2O3-SiC under water-cooling condition, proving that SiC-supported catalyst can achieve better current carrier-enhanced activation. It is worth mentioning that Co/(Al2O3-SiC) has the best activity at 5–6 W (Fig. 2b), indicating the thermal activation of SiC-supported catalyst is improved by Al2O3 partial substitution at low electrical input power, but this limits the current carrier-enhanced activation at high electrical input power.
催化剂内部建立的电场将导致载流子迁移和焦耳热效应。有理由认为 Al₂O₃和 SiC 在热活化与载流子增强活化过程中发挥不同作用,因为 SiC 介导与 SiC 负载催化剂在传统热 NH₃分解和电驱动 NH₃分解中的活性存在差异。为探究载流子增强活化的本征活性,本研究构建了恒温(5℃)水循环包裹的自制反应器(图 S6a),以最大限度消除焦耳热效应,使催化剂在电场中保持低温状态(图 S6b)。如图 3a 所示,与传统热 NH₃分解相比,催化活性虽有所下降但仍保持较高水平。该结果表明电场中的催化活性包含载流子增强活性和部分焦耳热活性,前者是中间体活化的主要来源,后者则进一步对催化剂进行热活化。 与热 NH3 分解(>400℃)相比,电场中(<200℃)存在的部分焦耳热活性表明,焦耳加热产生的热活化比传统外部加热对活性位点激活更为高效。这一现象类似于热载流子驱动光催化中的光热效应——由局域表面等离子体非辐射衰变激发的热载流子,其能量可高于直接光激发产生的载流子[31]。另一方面,在水冷条件下 Co/SiC-Al2O3 的活性仍优于 Co/Al2O3-SiC,证明碳化硅载体催化剂能实现更优异的载流子增强活化。值得注意的是,Co/(Al2O3-SiC)在 5-6W 功率下表现出最佳活性(图 2b),说明低电输入功率时 Al2O3 部分取代可改善碳化硅载体催化剂的热活化性能,但会限制高电输入功率下的载流子增强活化效应。
Fig. 3
  1. Download: Download high-res image (175KB)
  2. Download: Download full-size image

Fig. 3. Comparison of Co/Al2O3-SiC and Co/SiC-Al2O3 in (a) Electrically driven NH3 decomposition under water cooling, the temperature of cooling water was 5℃; (b) Arrhenius plots. T: thermal NH3 decomposition; E: electrically driven NH3 decomposition, the electrical input power was 4.6 W.

The apparent activation energy (Ea) can give insight into the activation barrier of the SiC-mediated and SiC-supported catalysts. Therefore, as shown in Fig. 3b, Arrhenius plots are obtained with or without the electric field by linear fitting of Ln (reaction rate) versus 1/T. The Ea value of Co/SiC-Al2O3 (105.36 kJ/mol) is higher than that of Co/Al2O3-SiC (97.62 kJ/mol) in conventional thermal NH3 decomposition, which is in accordance with the activity tests. However, the Ea in the electrically driven NH3 decomposition is significantly reduced for both catalysts. The Ea for Co/Al2O3-SiC is 30.53 kJ mol−1 in the electric field, while that for Co/SiC-Al2O3 is only 5.86 kJ mol−1. Such low activation energy has never been reported in conventional thermal NH3 decomposition, which indicates that the reaction mechanisms are different between thermal and electrically driven NH3 decomposition. Moreover, the smaller Ea of Co/SiC-Al2O3 proves that the SiC-mediated catalyst is more active in the electric field.
Reaction orders can help us understand the dependence of NH3 decomposition reaction. As shown in Fig. S7 and Table S2, in thermal condition, the NH3 order (α) for Co/SiC-Al2O3 and Co/Al2O3-SiC is 0.29 and 0.25, respectively. Small and positive α means retardation of nitrogen species [23], [32], which is usually considered a kinetic limited step due to the high strength of N-metal (500–630 kJ mol−1) [5]. The H2 order (γ) for Co/SiC-Al2O3 and Co/Al2O3-SiC is −0.95 and −0.99, respectively. The large but negative γ indicates that hydrogen species are strongly adsorbed onto the active sites, well-known as hydrogen poisoning, preventing further activation and dehydrogenation of NHx on these competitive active sites [22]. However, all the reaction orders increase when the electric field is applied, especially over Co/SiC-Al2O3, thus, giving a very small -γ/α (1.11). These results were consistent with the previous report that hydrogen poisoning was weakened and nitrogen desorption was accelerated in the electric field[18], [23]. Moreover, since the increase of H2 reaction order over Co/SiC-Al2O3 is the most obvious (from −0.95 to −0.40), there are reasons to believe that it is the H species migration stimulated by the electric field that promotes the stepwise dehydrogenation of NHx and products desorption, thus, making Co/SiC-Al2O3 better electrically driven performance. It was reported that the electric field can form a directional potential field in the catalyst to facilitate proton (H+) hopping [12], [33], which probably exposed the sites for NHx dehydrogenation and back donate electrons to the metal-N antibonding orbitals to promote N2 desorption [23]. These mechanisms will be demonstrated in detail in the mechanism characterization section subsequently.

3.3. Mechanism characterization

Firstly, the physicochemical properties of the catalysts were characterized. As presented in Table 1, the Co contents of the obtained nanoparticle catalysts are close to the theoretical design confirmed by ICP-OES analysis. The N2 adsorption-desorption isotherms (Fig. S8a) of the obtained catalysts show a similar Type IV isotherm with H3 hysteresis, confirming the mesoporous and slit structure in the samples. The specific surface area (SBET) and total pore volume (Vpore) of Al2O3 are much larger than that of SiC but the gap is narrowed after Co loaded (Table 1), indicating that Al2O3 can provide more sites for Co to load so that the pores were blocked while the loading of Co on SiC formed new pores. Co/Al2O3 with larger SBET and Vpore can provide more sites for NH3 adsorption as NH3-TPD (Fig. 8a) illustrated, which is one of the reasons for the better thermal activity.

Table 1. Textural properties of Co-based catalysts.

CatalystsCo content
-(wt%)a
SBET (m2 / gcat.)bVpore (cm3 / gcat.)b
SiC̸23.340.083
Al2O3̸123.650.823
Co/SiC16.0244.890.411
Co/Al2O314.15108.640.619
Co/(Al2O3-SiC)15.1379.710.327
a
Determined by ICP-OES analysis.
b
The specific area (SBET), and the total pore volume (Vpore) were measured by N2 physisorption analysis.
XRD patterns for the Co-based catalysts are displayed in Fig. 4a. The diffraction patterns show that Co species in Co/Al2O3 exist in the form of cubic Co3O4 phase (JCPDS 74–2120) with amorphous Al2O3, while in Co/SiC, cubic CoO (JCPDS 78–0431) is the main phase for Co species with highly crystalline SiC (JCPDS 29–1129), which is mainly due to the difference of calcining atmosphere. Accordingly, the coordination environment of cobalt ions is further investigated by Raman spectra. As displayed in Fig. 4b, four peaks appear at 475, 516, 613, and 682 cm−1 in Co/Al2O3, corresponding to different modes of crystalline Co3O4 [34]. The peaks at 475 and 682 cm−1 are associated with the Eg and A1 g modes of Co3O4, respectively. And the peaks at 516 and 613 cm−1 are assignable to the F2 g mode of Co3O4 [35], [36], [37]. However, Co/SiC shows the absence of the peak at 475, 516, and 613 cm−1. Compared with Co/Al2O3, the peak at 682 cm−1 is blue-shifted to 665 cm−1 in Co/SiC, and the corresponding intensities decrease. These changes indicate that Co2+ filled in octahedral is the main form of Co species in Co/SiC [34], [38], which is consistent with the XRD results. Additionally, the peak at 782 cm−1 in Co/SiC arises from the transverse optical (TO) and longitudinal optical (LO) mode of 3C-SiC crystal [39]. Both the XRD and Raman spectra show that Co/Al2O3-SiC and Co/SiC-Al2O3 maintained the basic structure of Co/Al2O3 and Co/SiC, respectively, indicating the incorporation of SiC or Al2O3 would not change the basic structures.
Fig. 4
  1. Download: Download high-res image (215KB)
  2. Download: Download full-size image

Fig. 4. (a) XRD patterns, (b) Raman results of the Co-based catalysts.

The HRTEM images of Co/Al2O3-SiC and Co/SiC-Al2O3 are presented in Fig. 5a and Fig. 6a, respectively. The lattice fringes with a spacing of 0.288 nm and 0.468 nm are assigned to the (220) and (111) planes of Co3O4 (Fig. 5a), respectively, while those with a spacing of 0.214 nm and 0.248 nm are assigned to the (200) and (111) planes of CoO (Fig. 6a), respectively. These results correspond to the XRD analysis. The overall morphology and elemental composition of Co/Al2O3-SiC and Co/SiC-Al2O3 can be seen in Figs. 5b-f and 6b-f. For Co/Al2O3-SiC (Fig. 5b-f), SiC is incorporated into Co/Al2O3 as an electric field-mediated component. For Co/SiC-Al2O3 (Fig. 6b-f), Co is directly loaded on SiC and mixed with Al2O3. These properties of both samples ensure that the active sites can be effectively charged when the electric field is established. Nevertheless, In terms of catalytic activity, Co directly loaded on highly conductive SiC may be more conducive to the transport of charge and intermediate in the electric field.
Fig. 5
  1. Download: Download high-res image (857KB)
  2. Download: Download full-size image

Fig. 5. (a) HRTEM images, (b) TEM images and (c-f) Energy-dispersive X-ray spectroscopy elemental mapping of Co/Al2O3-SiC.

Fig. 6
  1. Download: Download high-res image (927KB)
  2. Download: Download full-size image

Fig. 6. (a) HRTEM images, (b) TEM images and (c-f) Energy-dispersive X-ray spectroscopy elemental mapping of Co/SiC-Al2O3.

The electrically driven activity test shows significant differences between SiC-supported (Co/SiC-Al2O3) and SiC-mediated (Co/Al2O3-SiC) catalysts. The differences in structure lead to the change of conductive path in the electric field, which affects the surface elemental compositions and valence states. Therefore, XPS was carried out to give insight into the effects of the electric field on surface chemical state. Fig. 7a displays the Co 2p envelope of the samples, which have two main peaks of Co 2p3/2 and Co 2p1/2 with shakeup satellites. The satellite peaks of Co 2p at around 785.6 and 804.1 eV are associated with Co2+ coordinating with six oxygen atoms in the octahedral coordination [40]. The peak of Co 2p3/2 at around 781 eV can be deconvoluted into two peaks, which is attributed to the tetrahedral Co2+ (782.1–782.4 eV) and octahedral Co3+ (780.3–780.5 eV) in Co3O4 [41], [42], [43], respectively. For the pristine samples, Co/SiC-Al2O3 has a higher Co2+/(Co3++Co2+) ratio based on the peak area, which is in accordance with the previous results. However, after being treated by the electric field, although the Co2+/(Co3++Co2+) ratios of both catalysts increased, the ratio for Co/Al2O3-SiC increased more, from 51.0 % to 62.6 %. It is suggested that there was a strong electron accumulation effect stimulated by the electric field on the catalyst surface, which can help to electronically excite the Ru-N bond and reduce the N2 desorption barrier [23], and this effect seems to be more pronounced for the Co/Al2O3-SiC. Al2O3 that directly supports Co is likely to provide more O2- ions to increase the electronegativity of the metal [22], which can be further illustrated by CO2-TPD (Fig. 11b).
Fig. 7
  1. Download: Download high-res image (456KB)
  2. Download: Download full-size image

Fig. 7. XPS spectra of Co/Al2O3-SiC and Co/SiC-Al2O3 for fresh and the electric field-treated samples (termed as, E) in the electric field at 8 W in Ar: (a) Co 2p; (b) Si 2p.

The Si 2p XPS spectra of Co/SiC-Al2O3 and Co/Al2O3-SiC with or without the electric field are shown in Fig. 7b, in which the spectra were deconvoluted into two peaks. The peaks at around 100.9 and 102.0 eV are attributed to Odouble bondSi=O and Si-C, respectively [44]. The proportion of Odouble bondSi=O in pristine Co/SiC-Al2O3 is higher than that in pristine Co/Al2O3-SiC, possibly because the interaction between SiC and oxygenated species (CoxOy or Al2O3) in Co/SiC-Al2O3 is stronger. Interestingly, after being treated by the electric field in Ar, the Odouble bondSi=O ratio of both catalysts increased, especially for Co/SiC-Al2O3 (from 49.3 % to 65.7 %). This result indicates that electric field strengthened the interaction of different catalyst components, which is conducive to current carriers’ transportation. This result is also consistent with the increase in surface oxygen (OA) as illustrated by the XPS of O 1 s (Fig. S9). In conventional thermal NH3 decomposition, it is reported that the enhanced metal-support interaction is beneficial for ammonia decomposition by improving H2/N2 desorption [22], [24]. Since Co/SiC-Al2O3 shows better activity in the electric field, it is believed that the enhanced interaction is one of the mechanism by which the electric field promotes NH3 decomposition [45].
NH3 temperature-programmed desorption (NH3-TPD) was performed to investigate the NH3 adsorption properties of catalysts with and without the electric field. As shown in Fig. 8a (solid lines), pristine catalysts show obvious two desorption peaks. Generally, NH3 desorbed below 250℃ is weakly adsorbed on the catalyst surface, which is commonly associated with Brønsted acid sites, while the desorption peak above 250℃ may be attributed to moderately or strongly adsorbed NH3 on Lewis acid sites originating from unsaturated surface bonds [46], [47]. NH3 desorption peak between 100 and 250℃ is larger than that above 250℃ for both catalysts, showing the Brønsted acid sites account for a significant proportion on the catalyst surface. In comparison, NH3 desorption peak over Co/Al2O3-SiC is larger over the full temperature range, indicating Co/Al2O3-SiC had a better NH3 adsorption capacity. This is consistent with the result that Co/Al2O3-SiC has better activity in conventional thermal NH3 decomposition. NH3-TPD over the electric field-treated catalysts were also performed to investigate the electric field effect on these adsorption sites. As shown in Fig. 8a (dot lines), the NH3 desorption peaks shifted to lower temperatures while the peaks’ intensity increased, indicating that the NH3 adsorption strength was weakened while the amount of NH3 adsorption was increased. These changes are favorable for the NH3 decomposition reaction in the electric field since the electrically driven strategy greatly enhanced the sites’ activity at low temperatures, more NH3 adsorption allows more NH3 to participate in the reaction, and weaker adsorption strength mitigates the coverage of intermediates. Obviously, these changes were more pronounced on Co/SiC-Al2O3, indicating there is a stronger synergistic effect between the electric field and Co/SiC-Al2O3.
Fig. 8
  1. Download: Download high-res image (198KB)
  2. Download: Download full-size image

Fig. 8. Comparison for SiC-mediated and SiC-supported catalysts (a) NH3-TPD; (b) H2-TPR of fresh and the electric field-treated samples (identified as, E).

H2 temperature-programmed reduction (H2-TPR) was performed to evaluate the reduction behavior and interaction between Co and supports. As shown in Fig. 8b, for the pristine sample (solid line), there are two distinct reduction peaks for Co/Al2O3-SiC, showing an obvious two-step reduction process. Generally, Al2O3 and SiC are difficult to be reduced [48], and H2 consumption is mainly due to the reduction of Co species. The peak at ca.400℃ can be attributed to the reduction of Co3O4 to CoO and that at ca.672℃ can be assigned to the reduction of CoO to metal Co [10], [48]. By contrast, there is no obvious multi-peak structure for pristine Co/SiC-Al2O3 but large and overlapping reduction peak emerges at high temperatures, which indicates the interaction between Co species and SiC is stronger and more CoO there [25]. These results are in accordance with the previous analysis. It is worth noting that there is a negligible peak at 238℃, possibly originating from hydrogen spillover to the support [22]. This phenomenon is critical to reveal the difference in the promotion of both catalysts, which will be discussed subsequently. After an electrical process in Ar (Fig. 8b, dot line), the intensity of both reduction peaks for Co/Al2O3-SiC decreased, indicating that many Co species were reduced by the electric field. However, the reduction of Co/SiC-Al2O3 by the electric field is not so pronounced, instead, the peak slightly shifts to a higher temperature. As demonstrated by XPS analysis, the interaction between Co and SiC further strengthens in the electric field for Co/SiC-Al2O3, which enables Co species to steadily receive charge carriers from SiC in the electric field [45] and inhibits hydrogen adsorption during NH3 decomposition [17].
Therefore, the SiC-supported (Co/SiC-Al2O3) and SiC-mediated (Co/Al2O3-SiC) catalysts have different synergistic effects in the electric field, which can be further figured out by the feature of product desorption. Firstly, NH3 surface reactions in thermal (300℃) and electrical condition were carried out. As shown in Fig. 9a-b, NH3 was injected into the reactor after the system was stable, then, H2 and N2 were produced. However, in thermal conditions (Fig. 9a), the H2 and N2 signal quickly returned to the initial level, indicating intermediates or products retardance inhibited NH3 decomposition at low temperatures. Therefore, the NH3 conversion in the activity test was negligible at 300℃. Interestingly, the H2 signal over Co/SiC-Al2O3 was higher but the N2 signal was lower compared with Co/Al2O3-SiC. This result shows that H2 desorption is easier to happen over Co/SiC-Al2O3 while N2 desorption is easier to happen over Co/Al2O3-SiC. In comparison, the NH3 surface reaction was greatly changed in electrical conditions. As shown in Fig. 9b, N2 and H2 signals arose but only fell back slightly, then continued to rise to high levels. This result proves the promotion effect for product desorption driven by the electric field. The H2 desorption over Co/SiC-Al2O3 was still stronger than that over Co/Al2O3-SiC while the N2 desorption was close on both. Therefore, not only the H2 desorption was further promoted over Co/SiC-Al2O3, but also the difficulty of N2 desorption was well compensated by the electric field. It has been reported in previous electric field-assisted catalysis that proton hopping stimulated by the electric field could facilitate the evolution of the intermediates and thus change the RDS [12], [18]. Therefore, it is reasonable to believe that proton(H+) hopping is greatly promoted over SiC-supported catalysts in the electric field due to the highly conductive SiC and enhanced metal-support interaction. However, the proton (H+) hopping over the SiC-mediated catalyst in the electric field is not promoted so well, which makes NHx dehydrogenation still retarded.
Fig. 9
  1. Download: Download high-res image (227KB)
  2. Download: Download full-size image

Fig. 9. NH3 surface reaction for Co/SiC-Al2O3 and Co/Al2O3-SiC (a) at 300℃ without the electric field; (b)in the electric field (8 W).

Product desorption profiles of the used catalysts further prove the above conclusion. As TPD showed in Fig. 10a, the remaining species after NH3 decomposition reaction began to desorb as the temperature rose. The H2 desorption over the used Co/Al2O3-SiC was larger than that over the used Co/SiC-Al2O3 while the N2 desorption over the latter was slightly larger, indicating more H species remained on Co/Al2O3-SiC surface after the NH3 decomposition reaction. When the desorption was carried in the electric field, as shown in Fig. 10b, the catalyst temperature increased with the application of the electric field. After a short time lag, the remaining species began to desorbed. However, product desorption over the used Co/Al2O3-SiC was less than that over the used Co/SiC-Al2O3, especially for H2, which is different from the TPD result (Fig. 10a). This result indicates that the remaining H2 is not sufficiently desorbed on Co/Al2O3-SiC in the electric field, which is in good accordance with NH3 surface reaction, i.e., the proton (H+) hopping over Co/Al2O3-SiC in the electric field is not promoted so well. It is worth noting that although the desorption signal slightly delays behind the temperature signal, it can still be considered that the product desorption and the electric field application occur simultaneously, because the electrical signal of temperature changes faster than the gas desorption signal. This indicates that the synergistic effect mainly stems from the directional potential field established by the electric field rather than the joule heating.
Fig. 10
  1. Download: Download high-res image (199KB)
  2. Download: Download full-size image

Fig. 10. Product desorption of the used catalysts (a) TPD profiles; (b) Electrical desorption profiles. The fresh catalysts were tested at 400℃ for 2 hours in pure NH3 to obtain the used catalysts.

H2 temperature-programmed desorption (H2-TPD) can directly give insight into the H2 migration and desorption behavior over both catalysts. As displayed in Fig. 11a, the profiles show one H2 desorption peak for Co/Al2O3-SiC while two peaks for Co/SiC-Al2O3 in the reaction temperature range. The peak at 413℃ for Co/Al2O3-SiC is weaker than that for Co/SiC-Al2O3 at corresponding 392℃. An additional broad and strong desorption peak appears at a lower temperature (171℃) for Co/SiC-Al2O3. This result indicates that the hydrogen spillover is much stronger over Co/SiC-Al2O3 than that over Co/Al2O3-SiC [49]. The H species migrated along the catalyst surface can release the occupied active sites for further H species adsorption and desorb at lower temperatures during the temperature-programmed course, thus, increasing the overall adsorbed/desorption H2. The spillover of surface hydrogen species can be involved in the surface reaction of NH3 decomposition and dehydrogenating hydrogen species are quickly removed to expose the sites for the next stage of NHx dehydrogenation. However, this phenomenon cannot explain the activity difference in thermal condition. The rate-determined step in the classical catalytic NH3 decomposition over metal nanoparticles (NPs) consists of stepwise homolytic cleavages of N-H bonds and recombination of N [8], therefore, the latter seems to be more critical for NH3 decomposition in thermal condition.
Fig. 11
  1. Download: Download high-res image (158KB)
  2. Download: Download full-size image

Fig. 11. Comparison for Co/SiC-Al2O3 and Co/Al2O3-SiC catalysts (a) H2-TPD profiles; (b) CO2-TPD of pristine and the electric field treated samples (termed as, E).

CO2-TPD was performed to investigate the basicity of pristine and electric field-treated catalysts. The high density of basic sites can enhance the electronegativity of the metal, in other words, surface basicity can increase the potential of electron-donation to the active sites, which is beneficial to N associative desorption [22]. Generally, the desorption peak below 170℃, 170–320℃, and above 320℃ can be attributed to weak, medium, and strong basic sites, respectively [22]. As shown in Fig. 11b and Table. S3, the overall amount of CO2 desorption over pristine Co/Al2O3-SiC is greater than that over pristine Co/SiC-Al2O3, indicating a stronger intrinsic electron-donation ability for Co/Al2O3-SiC. Considering that Co/Al2O3-SiC has higher thermal activity for NH3 decomposition, the N associative desorption is likely the rate-determined step in thermal condition, which is consistent with the previous studies [8]. Interestingly, the basic sites for both catalysts increased after being treated in the electric field, indicating electrons accumulated in the electric field, thus, the N2 desorption was enhanced. The difference is that the medium basic sites mainly increased for Co/SiC-Al2O3, while the strong basic sites were markedly increased for Co/Al2O3-SiC. It was reported that weak basic sites were usually derived from OH-, while medium and strong basic sites were formed on the surface of O2- [50], [51], which was more critical for electron transfer from support to metal-N antibonding orbital to activate N-metal bond [5]. The associative desorption of N is theoretically advantageous for Co/Al2O3-SiC, but the dehydrogenation did not proceed very well in the electric field, as illustrated before, which limits the ultimate electrically driven NH3 decomposition activity. In comparison, the intrinsic ability of electron donation is not as well over Co/SiC-Al2O3, but the pronounced proton(H+) hopping as well as the residual electrons that can donate to metal-N antibonding make the electrically driven NH3 decomposition activity outstanding.
Time-resolved in-situ DRIFT spectra were also performed to further give insight into the corresponding surface change during NH3 decomposition. Fig. 12 shows the evolution of principal surface species at 200℃ and 400℃ during NH3 decomposition. The peaks at about 1626 cm−1 and 3332 cm−1 can be assigned to gaseous-phase NH3 and /or physically adsorbed NH3 [16], [25], [52]. The bands around 1598 cm−1 and 1683 cm−1 were ascribed to NH3 coordinated with Lewis acid sites and NH4+ coordinated with Brønsted acid sites [25], respectively. The bands around 1526 cm−1 and 2166 cm−1 corresponds to -NH2 species and -N2 species[16], [25], respectively. It can be seen that -NH2 species over Co/SiC-Al2O3 species increased from 200℃ to 400℃ (Fig. 12b and d). In contrast, -NH2 species almost disappeared over Co/Al2O3-SiC at 400℃ with -N2 species emerged (Fig. 12a and c). This result is consistent with the conclusion that Co/Al2O3-SiC has a better intrinsic thermal activity. However, broad OH peaks at about 3470 cm−1 exhibit significantly stronger intensity over Co/SiC-Al2O3 at 200℃ (Fig. 12b) [16]. The growth of surface [OH] at low temperatures over Co/SiC-Al2O3 indicates a better capacity of H transfer from active sites to support after N-H scission. The better capacity of H transfer at low temperatures can facilitate proton (H+) hopping around the metal-support interface in the electric field to collide with intermediates in the reaction.
Fig. 12
  1. Download: Download high-res image (649KB)
  2. Download: Download full-size image

Fig. 12. Time-resolved in situ DRIFTS over Co/Al2O3-SiC at (a) 200℃ and (c) 400℃; Co/SiC-Al2O3 at (b) 200℃ and (d) 400℃.

Therefore, additional experiment of H2 electrically driven desorption (H2-EDD) was designed to further evaluate the proton (H+) hopping capacity in the electric field. As shown in Fig. 13, H2 desorption on Co/Al2O3-SiC and Co/SiC-Al2O3 in the electric field were different. Distinct H2 desorption peaks appeared immediately when the electric field was applied on the Co/SiC-Al2O3 (middle), followed by slight catalyst reduction (see H2O signal). Besides, the H2 desorption peaks increased with the increasing input power. However, the H2 desorption was negligible when the electric field was applied on Co/Al2O3-SiC until the input power was increased to 9 W (Top), instead, obvious negative H2 peaks and H2O signal appeared during H2-EDD, indicating more pronounced catalyst reduction. These results are consistent with previous analysis, implying that the proton (H+) was more easily stimulated by the electric field and hopped around Co/SiC-Al2O3 interface in the electric field to collide with intermediates in the reaction, giving a better electrically driven NH3 decomposition activity. Whereas, adsorbed hydrogen was difficult to stimulate by the electric field on Co/Al2O3-SiC, instead, it was captured by the oxygen-terminated surface to reduce the catalyst as the temperature rose.
Fig. 13
  1. Download: Download high-res image (124KB)
  2. Download: Download full-size image

Fig. 13. H2-EDD at different electrical input power. Top: H2 and H2O signal on Co/Al2O3-SiC; Middle: H2 and H2O signal on Co/SiC-Al2O3; Bottom: Temperature evolution. Abscissa is the relative time.

3.4. Long-term stability and WHSV effect

To investigate the long-term operational reliability of the electrically driven NH3 decomposition, the stability of Co/SiC-Al2O3 in the electric field was tested for 100 hours. As shown in Fig. 14a and Fig. S10, the electric current was set at 47.6 mA at the beginning and the corresponding electrical input power was 10 W, giving an initial NH3 conversion of 67 %. As time went on, the catalyst bed was affected by thermal stress due to the joule heating of the electric field, leading to particle loosening and resistance increase (reflected in the automatic voltage increase, Fig. S10). Therefore, the electrical input power was automatically increase which led to a slight increase in the conversion rate. In order to accurately judge the stability, the current was manually turned down to bring the electrical input power back down to 10 W at 77 h, accordingly, the conversion rate reverted to the initial value and maintained until the end of the 100-hour test. Besides, stability tests at higher WHSV also showed no trend of deactivation (Fig. S11). These results show that the NH3 decomposition activity under the given electrical input power is highly stable. The high activity stability can be owed to the enhanced metal-support interaction in the electric field that can effectively prevent active metal particle aggregation and sintering [53].
Fig. 14
  1. Download: Download high-res image (151KB)
  2. Download: Download full-size image

Fig. 14. (a) Stability test for electrically driven NH3 decomposition on Co/SiC-Al2O3 at 12,000 mL gcat.−1 h−1 in the electric field. Top: activity performance; bottom: corresponding electrical input power; (b) The conversion rate of electrically driven NH3 decomposition at different WHSV on Co/SiC-Al2O3.

As shown in Fig. 14b, NH3 conversion rate decreased with the increasing of WHSV, which is conforms to the basic laws of heterogeneous catalysis, that is, higher WHSV reduced the residence time for NH3 adsorption[23]. However, the decrease in conversion rate did not exceed the increase in WHSV, resulting in a more significant H2 production rate at higher WHSV (Fig. S12), where a H2 production rate of 30.52 mmol gcat.−1 min−1 was realized at 72,000 mL gcat.−1h−1 at 11 W, outperforming the state-of-the-art Ru-free catalyst at comparable temperatures (Table. S4). This result indicating the energy efficiency of electrically driven NH3 decomposition increases with the WHSV, therefore, the electrical strategy can provide an energy-saving solution for industrial large-scale H2 production from NH3 decomposition.

4. Conclusion

This work proposes an electrical strategy to facilitate H2 production from NH3 decomposition at low temperatures using Co-based SiC composite catalysts. An NH3 conversion up to 73 % was achieved at ca. 200℃ over Co/SiC-Al2O3. The interaction of catalyst components in the electric field is enhanced, contributing to high catalytic stability for more than 100 hours. We demonstrate that Co directly loaded on highly conductive SiC (Co/SiC-Al2O3) can be more conducive to transporting proton (H+) and electron in the electric field, which can increase collision and expose the sites for NHx dehydrogenation and donate the electrons to the metal-N antibonding orbitals to promote N2 desorption, thus, decreasing the Ea along the reaction pathway. However, the lack of basic sites of Co/SiC-Al2O3 leads to the poor capacity of intrinsic electron donation, limiting the intrinsic thermal activation activity. Therefore, it will be our next work to improve the intrinsic thermal activation activity while keeping the electric activation activity of the SiC-supported catalyst. In a word, the electrical strategy for NH3 decomposition using earth-abundant transition metals with low energy consumption, quick response, and compact layout shows a multi-scenario application for H2 production, such as low-cost, rapid, and portable online H2 production in FCVs.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was funded by the National Natural Science Foundation of China (52106173).

Appendix A. Supplementary material

Data Availability

Data will be made available on request.

References

View Abstract