Brief introduction of the results
In heterogeneous catalysis, hydrocarbon-catalyzed dehydrogenation often exhibits a negative pressure dependence on H2 due to the competitive chemical adsorption of hydrocarbons and H2. However, some catalysts exhibit positive pressure dependence on propane dehydrogenation, which is an important reaction for propylene production.
Professor Gong Jinlong of Tianjin Universityet al. found that the positive activity of gallium oxide-based catalysts is dependent on the partial pressure of H2 and is mediated by metastable hydrides. Through in-situ spectroscopy, kinetics, and computational analysis, the authors demonstrated that under the reaction conditions of H2 co-feeding, H2 dissociates on the surface of partially reduced gallium oxide, resulting in the chemical bonding of H atoms to coordination unsaturated GA atoms. These metastable gallium hydrides promote the activation of c-H bonds while inhibiting deep dehydrogenation.
The authors found that the surface coverage of gallium hydride determines the catalytic performance. Therefore, the gallium oxide catalyst Gaox-IR-K Al2O3 modified by alumina-supported trace additives exhibits high activity and selectivity at high concentrations of propane under the conditions of appropriate H2 co-feeding. Related work is based on ".metastable gallium hydride mediates propane dehydrogenation on h2 co-feeding" as the title in ".nature chemistry**.
**Reading guide
Figure 1Catalytic performance of alumina-supported gallium oxide in PDH.
In this paper, alumina-supported gallium oxide (Ga2O3 Al2O3) catalysts were prepared by wet impregnation. HRTEM images confirm that gallium oxide nanoparticles with a size of 4 nm are well dispersed on the -alumina support. The PDH performance of the catalyst was evaluated at 600 °C using different H2 C3H8 ratios. As shown in Figure 1b, the initial propylene generation rate (after 5 min of reaction) of the new unhydrogenated catalyst is relatively low at 26 mmol GCAT-1 H-1 with 80% propylene selectivity. Under the pre-reduction conditions, the initial generation rate of propylene was slightly increased to 29 mmol gcat-1 h-1, and the propylene selectivity was increased to 86%.
It is worth noting that the H2 co-feed significantly affects the activity and selectivity of the prereduction catalyst during the PDH process. With the increase of the feed ratio of H2 C3H8, the initial generation rate of propylene increased significantly at first and then decreased slightly, with a ratio of 2The maximum value (53 mmol GCAT-1 H-1) was reached at 0 with a propylene selectivity of up to 92%.
In addition, the catalytic performance of constant gravity space-time velocity (WHSV) and H2 C3H8 feed ratio was used to investigate the catalytic performance at different C3H8 concentrations. The initial formation rate of propylene increased with the simultaneous increase of the concentrations of C3H8 and H2, but the selectivity decreased slightly, which verified the positive effect of the increase of the partial pressure of H2 on the dehydrogenation activity of Gaox Al2O3. To demonstrate the industrial relevance of this strategy, maintaining a constant high inlet C3H8 concentration of 40 vol% and gradually increasing the H2 concentration, the positive pressure dependence of PDH on H2 is highly consistent with that at a C3H8 concentration of 14 vol% (Figure 1B).
The results show that when the ratio of H2 C3H8 is 10. When the concentration of C3H8 is 40 vol%, the initial catalytic performance of the GAOX Al2O3 catalyst is comparable to that of the commercial Crox-K Al2O3 catalyst without H2. The formation rate of propylene on GAOX Al2O3 was 61 mmol GCAT-1 H-1 after catalyst mass normalization, which was slightly lower than that on Crox-K Al2O3 (76 mMOL GCAT-1 H-1).
H2 co-feeding can alter the intrinsic dehydrogenation activity of GAOX Al2O3. In the TPSR experiment, the PDH starting temperature (400) of the new catalyst was higher than that of the pre-reduction catalyst. The PDH starting temperature of the pre-reduction catalyst was 375 and the PDH starting temperature of the H2 co-feed pre-reduction catalyst was further reduced to 325, indicating that the catalyst's ability to activate the C-H bonds in propane was improved (Fig. 1C).
In addition, the effect of H2 co-feeding on the apparent activation energy (EA) of PDH was also studied. The EA value of the pre-reduction catalyst decreased from 140 kJ mol-1 to 120 kJ mol-1 when the feed ratio of H2 C3H8 increased from 0 to 2At 0, the EA value of the prereduction catalyst was further reduced to 79 kJ mol-1 (Figure 1D). EA is strongly dependent on the H2 concentration in the reaction system, indicating that the H2 co-feed leads to the change of the reaction mechanism. This strongly indicates that there is a novel H2-induced high-efficiency PDH site on the surface of the catalyst under the condition of H2 co-feed reaction.
Figure 2H2 induces the formation of metastable gallium hydride on defective gallium oxide.
To elucidate the role of H2, the structural transition of the H2-induced catalyst was studied (Figure 2). The oxidation state of gallium was detected by in-situ xanes during the continuous treatment of the catalyst with different gases (Figure 2A). The K-edge Xanes spectra of GA2O3 Al2O3 show that tetrahedral oxygen-coordinated Ga3+ cations (Ga3+(T)) and octahedral oxygen-coordinated Ga3+ (Ga3+(O)) have two distinct contributions. After the reduction of 600 and H2, the K-edge Xanes spectrum of Ga showed a decrease in the intensity of the white line and a lower energy of the edge (about -0.).6 ev).
The results show that during the reduction process, gallium oxide undergoes partial reduction of Ga3+(t) or Ga3+(O) cations to Gaδ+ox, resulting in an irreversible negative shift of edge energy, and at the same time produces metastable gallium hydride, resulting in a reversible low-energy shift.
The authors also used in-situ XPS to quantify the reduced gallium species (Figure 2B), and the new catalyst was exhibited in AR at 11188 EV centered on GA2P3 2 peaks, which can be attributed to the GA3+ species. After the reduction of 600 and H2, due to the contribution of reduced gallium, the GA2P32 peak became obviously asymmetrical, showing about 0Negative displacement of 3 EV (Gaδ+,0
In this paper, in-situ drifts were used to monitor the evolution of adsorbent species on the surface of the catalyst during the reduction process, and a wide peak at 3530 cm-1 was observed after exposure of fresh catalyst to H2 (Fig. 2C). This peak can be attributed to the O-H bond with the stretching mode of the GA atom ((GAO-H)), which is different from the hydroxyl peak of aluminum. As the reduction progresses, the intensity of the GAO-H peak decreases. This suggests that the removal of surface O atoms by dehydroxylation results in a decrease in the coordination number of Ga. The reduction process was completed within 30 minutes and gradually disappeared with the GAO-H peak (Fig. 2C, G).
Notably, two peaks appeared at 2013 and 1981 cm-1 and increased with the weakening of the GAO-H peak. These two peaks are assigned to the tensile mode ((GACUS-H)) of coordination unsaturated GA atoms bonded by H atoms adjacent to O vacancies on tetrahedra and octahedral, respectively. The rapid disappearance of the Gacus-H peak during AR cleanup suggests that the Gacus-H species cannot survive without H2 (Fig. 2D, F). After re-exposure to H2, the GACUS-H peak recovered rapidly and completely, while the GAO-H peak remained weak (Figure 2E).
Figure 3Dependence of PDH performance on metastable gallium hydride species.
In this study, in-situ drifts were used to detect the presence of gallium hydride under reaction conditions. No GACUS-H peaks were observed in PDH without H2 co-feed (Figure 3A), whereas during PDH co-fed with H2 with a pre-reduction catalyst, these peak strengths increased progressively and were present throughout the process (Figure 3B-E). As expected, washing with AR resulted in the rapid disappearance of the Gacus-H peak (Figure 3G-J). These results clearly show that the H2 co-feeding forms metastable gallium hydride in PDH. In addition, the pre-reduction catalyst did not undergo a dehydroxylation reaction under this reaction condition, indicating that the pre-reduction can lead to a higher degree of reduction of the catalyst (Fig. 3a-e).
With the increase of the inlet H2 C3H8 ratio, the coverage of gallium hydride on the catalyst surface, represented by the Gacus-H peak (IgA-H) intensity, increases highly consistent with the observed increase in dehydrogenation activity represented by the propylene generation rate (Fig. 3K, L). The linear relationship between dehydrogenation activity and IgA-H3 pH2 (partial pressure of H2) suggests that gallium hydride is involved in more efficient PDH (Figure 3M).
In addition, it is widely believed that carbon deposition is the main cause of PDH catalyst deactivation. The Drifts peak, centered at 1560-1545 cm-1, appears in the PDH reaction and remains stable at AR purging (Figure 3A-K), which can be attributed to coke formation on the catalyst surface. The amount of coke decreases as the GaN coverage increases (Fig. 3a-k), suggesting that GaN improves the anti-coking resistance of the catalyst.
Figure 4DFT calculations and KIES reveal metastable gallium hydride-mediated PDH mechanisms.
In this paper, the properties of gallium hydride in catalytic dehydrogenation reactions are further investigated by DFT calculations. A sequential dehydrogenation mechanism occurs on both the original surface GA2O3 (100) and the defective surface GAox (100) (Fig. 4A). The free energy barrier for the first step of dehydrogenation on these two surfaces is calculated to be 3., respectively22 and 285 EV, which is the rate-determining step. However, on the defective surface covered by gallium hydride, the free energy barrier of the first gallium hydride-mediated dehydrogenation reaction drops to 263 EV, indicating that Gacus-H and Gacus are required for the first C-H bond activation.
To explain the low adsorption energy of C3H7 on the gallium hydride-covered defective surface (+130 EV), a natural bond orbital (NBO) analysis was performed to calculate the antibond orbital occupancy of C-Ga bonds on defect surfaces with and without gallium hydride coverage with values of 0., respectively38 and 042 (Fig. 4f), indicating that GaN-induced c-ga bonding is stronger. The dehydrogenation barrier of C3H6 adsorbed on the defect surface (2.)01 EV) is significantly higher than that of the defect surface (085 EV) and defective surfaces (167 EV) (Figure 4b), which illustrates that the pre-reduction and H2 co-feed improved the propylene selectivity and anti-coking of the catalyst.
To further elucidate whether C-H bond activation or H2 desorption is a rate-determining step, the deuterium kinetic isotope effect (KIES) in H2 and C3H8 was measured and found that the rate of formation of propylene was proportional to the partial pressure height of propane performed with different feeds, suggesting that the primary mechanism of propane was independent of feed composition (Figure 4C-E). Co-feeding D2 instead of H2 led to an increase in the rate of C3H6 generation, exhibiting inverse deuterium kie (KH2 kd2;0 in Figure 4c79 and 076)。The desorption temperature of D2 is higher than that of H2, indicating that D binds to Gacus more strongly than H. Therefore, inverse KIE does not include H2 desorption as a rate-determining step. In addition, the substitution of C3H8 with C3D8 resulted in a significant reduction in the propylene formation rate (Fig. 4D), and it can be concluded that C-H bond activation is a rate-determining step.
Figure 5Positive effect of H2 co-feed on other gallium oxide-based catalysts in PDH
This facilitating effect of H2 can also be generalized to other gallium oxide-based catalysts. For the addition of trace amounts of additives (e.g., 005 wt% IR and 025 wt% K) of the Alumina supported gallium oxide catalyst, the positive effect of H2 co-feed on PDH was more significant.
In the same H2 C3H8 feed ratio (15) The in-situ drifts showed that the GaN coverage followed the following order: GAOX Al2O3 Thus, proper H2 co-feeding increased the propane conversion of GAOX-IR-K Al2O3 from 26% to 37%, propylene selectivity from 91% to 96%, and propylene generation from 63 to 96 mmol GCAT-1 H-1, exceeding the propane conversion of the commercial Crox-K Al2O3 catalyst (31%) , propylene selectivity (92%), propylene generation rate (76 mmol, gcat-1 h-1) (Figure 5D). In addition, at 600, GAOX-IR-K Al2O3 exhibited greater stability than CROX-K Al2O3 (Figure 5E).
Bibliographic information
metastable gallium hydride mediates propane dehydrogenation on h2 co-feeding，nature chemistry，2024.