English

• Hydrogen (H₂) is regarded as the most promising energy carrier to replace traditional fossil fuels in the future due to its high energy density, cleanliness, and renewability [1-4]. However, efficient and safe hydrogen storage technology is still a critical problem that needs to be solved. Formic acid (FA), which can be produced via CO₂ reduction and biomass conversion, has received extensive interest as a convenient and safe hydrogen carrier due to its wide range of sources, renewable, high hydrogen content (4.4 wt%), and convenient storage/transportation [5, 6]. Formic acid can be selectively decomposed by dehydrogenation procedure (HCOOH → H₂ + CO₂) or via a dehydration pathway (HCOOH → H₂O + CO). The latter approach should be avoided because the generated CO is toxic to the fuel cells. Recently, a variety of efficient catalysts including the homogeneous and heterogeneous catalysts have been developed toward the hydrogen evolution from the FA dehydrogenation reaction (FDR). Heterogeneous catalysts have attracted much attention due to their advantages of being easily separated and recovered from reaction mixtures [7-9]. Pd-containing catalysts [10-19], especially PdAg [14–16] and AuPd [17–19] bimetallic catalysts, were found to be very active toward FDR.

  Confining the active metal nanoparticles (NPs) in the matrix with large surface area would be an efficient method toward enhancing the catalytic activity of catalysts [20-29]. To date, several efficient supporting materials have been applied in FDR, such as graphene oxide [30, 31], porous carbon [32, 33], and metal-organic frameworks [34-36]. The supported Pd-containing catalysts showed the enhanced activity toward FDR; however, it remains a challenge to improve the catalytic performance toward the actual application in future. Mesoporous materials with high thermal and chemical stabilities have been widely used in adsorption, separation and catalysis [37-41], owing to their high specific surface area and regular pore structure. In addition, amine groups-modified mesoporous support can provide the anchoring sites for metal species to achieve small metal NPs, leading to the enhanced catalytic performance [42-45]. KIT-6 with 3D cubic structure as a carrier has more advantages in mass transfer and diffusion of reactive molecules, which can reduce hole blockage and provide more active sites. Due to its surface residual surface silicon (Si–OH), KIT-6 could be easily modified with amine groups, which can be used as the interaction site of metal NPs to obtain the ultrafine metal NPs as well as a proton scavenger to promote the breaking of O-H bond of FA.

  Herein, small and highly dispersed PdIr alloy NPs with a mean particle size of 1.8 nm immobilized by amino-functionalized 3D mesoporous silica (KIT-6) were prepared by an impregnation-reduction approach and used as an efficient catalyst for FDR (see Supporting information for detailed experimental procedures). Benefiting from bimetallic synergy between Pd and Ir, strong interaction between small PdIr alloy NPs and KIT-6-NH₂, as well as the promotion of amino-groups of KIT-6-NH₂, the as-synthesized PdIr/KIT-6-NH₂ catalyst exhibits a superior activity toward FDR with a TOF value of 3533 h⁻¹ at 303 K, which is comparable to the best heterogeneous catalysts ever reported [5–7, 10–19, 30–36].

  As shown in Fig. 1, the typical transmission electron microscopy (TEM) images of KIT-6-NH₂ (Fig. 1a) and PdIr-KIT-6-NH₂ (Fig. 1b) display the ordered channel structure, similar as that of pure KIT-6 (Fig. S1 in Supporting information), confirmed by scanning electron microscopy (SEM) images (Figs. S1a and c). The Pd, Ir, N, Si and O elements were uniform dispersed in EDS elemental mappings (Figs. 1d and e). PdIr NPs with a mean size of about 1.8 nm were well dispersed in KIT-6-NH₂ (Fig. 1b). In contrast, larger and coagulated PdIr NPs with a mean size of 7.4 nm (Fig. 1f) were observed in bare KIT-6 without amino-group. It is clearly demonstrated that amino-groups of KIT-6-NH₂ result in the formation of smaller NPs due to the fact that amino-groups can effectively anchor the metal NPs and prevent them from aggregation. The amine-groups of KIT-6-NH₂ could be verified by Fourier transform infrared (FT-IR) spectra (Fig. S2 in Supporting information). The lattice spacing of the PdIr NPs of 0.224 nm (Fig. 1c) shows the formation of PdIr alloy structure in KIT-6-NH₂, which is between the (111) planes of face-centered cubic Ir (0.222 nm) and Pd (0.229 nm). The alloy structure of PdIr NPs was also confirmed by XRD results (Fig. S3 in Supporting information), in which a wide peak between the characteristic peaks of Pd (111) and Ir (111) can be observed. After annealed in Ar atmosphere at 873 K for 4 h (Fig. S4 in Supporting information), the alloy structure of PdIr NPs could be clearly observed.

  Figure 1

  

  Figure 1.  (a) TEM of KIT-6-NH₂, (b) TEM, (c) HRTEM, (d) HAADF-STEM, (e) the corresponding EDS mapping of PdIr/KIT-6-NH₂ and (f) TEM of PdIr/KIT-6 (inset: the corresponding particle size histogram).

  DownLoad: Full-Size Img PowerPoint

  As shown in Fig. 2a, the N₂ adsorption-desorption curves of the KIT-6, KIT-6-NH₂ and PdIr/KIT-6-NH₂ samples are of type IV curves, which is indicative of the ordered mesoporous materials and consistent with their TEM and small-angle XRD results (Fig. S3). As compared to KIT-6, the decrease in BET specific surface area, pore diameter, and pore volume of PdIr/KIT-6-NH₂ (Fig. 2a, Fig. S5 and Table S1 in Supporting information) can be owing to the amino-modification and metal NPs loading.

  Figure 2

  

  Figure 2.  (a) Nitrogen sorption isotherms of KIT-6, KIT-6-NH₂, and PdIr/KIT-6- NH₂, high-resolution XPS spectra of (b) N 1s in PdIr/KIT-6-NH₂ and KIT-6-NH₂, (c) Pd 3d in PdIr/KIT-6-NH₂ and Pd/KIT-6-NH₂ and (d) Ir 4f in PdIr/KIT-6-NH₂ and Ir/KIT-6-NH₂.

  DownLoad: Full-Size Img PowerPoint

  The survey XPS spectrum of PdIr/KIT-6-NH₂ reveals the existence of Pd, Ir, N, Si and O elements (Fig. S6 in Supporting information). As shown in Fig. 2b, N 1s has a higher binding energy in PdIr/KIT-6-NH₂ than that in KIT-6-NH₂, indicating the -NH₂ group may provide some electrons to the PdIr alloy NPs, resulting in electron enrichment of metal NPs, and thus enhance the dehydrogenation rate toward FDA. Compared with the PdIr/KIT-6-NH₂ catalysts, the Pd 3d peaks of Pd/KIT-6-NH₂ are shifted to lower binding energies, and an opposite tendency is observed in the case of the Ir 4f peaks of Ir/KIT-6-NH₂ (Figs. 2c and d). This outcome is possibly attributed to the change of the electronic structure after Pd alloyed with Ir.

  The catalytic performances of PdIr/KIT-6-NH₂ and comparative samples were investigated for H₂ generation at 303 K. As shown in Figs. 3a and b, the Pd₆Ir₄-KIT-6-NH₂ catalyst display the best catalytic performance for FA dehydrogenation among a series of samples with different Pd/Ir molar ratios, providing a high TOF value of 3533 h⁻¹ (measured by the total metal content of Pd and Ir, at a FA conversion rate of 20%), which is one of the highest values for heterogeneous hydrogen so far (Table S4 in Supporting information). In contrast, physically mixed catalysts (PdIr and KIT-6-NH₂) exhibit a lower activity than that of Pd₆Ir₄/KIT-6-NH₂ under identical conditions (Fig. S7 in Supporting information), further indicating the important contribution of the strong interaction between PdIr and KIT-6-NH₂ on the activity of catalyst toward FDR.

  Figure 3

  

  Figure 3.  Volume of the generated gas versus time for FDR in FA-SF aqueous solution over (a) Pd_xIr_10-x/KIT-6-NH₂ with different Pd/Ir ratios and (b) Pd₆Ir₄ NPs confined by different supports. GC spectra for the commercial gas and the generated gas from FDR over Pd₆Ir₄/KIT-6-NH₂ using (c) TCD detector and (d) FID-methanator.

  DownLoad: Full-Size Img PowerPoint

  Surface grafted amino-groups play an important role in the catalytic activity of PdIr catalysts. We investigated the catalytic activity of PdIr catalysts confined in KIT-6-NH₂ which was functionalized with different organic amines via reflux method (Fig. S8 in Supporting information), and the results show that APTES is a good candidate as the amino-groups for FDR. Bare KIT-6-supported PdIr catalyst has almost no activity (Fig. 3b). However, PdIr NPs supported by KIT-6-NH₂ show an enhanced activity, because the amino-groups on the surface of KIT-6-NH₂ can not only serve as the Brønsted basic sites to provide a basic environment around PdIr NPs to promote the dissociation of the O-H bond of HCOOH, but also act as the anchoring sites to achieve the small PdIr NPs with high dispersity to increase the active sites toward FDR. The results of UV-vis confirmed the strong interaction between the -NH₂ groups of KIT-6-NH₂ and the metal ions (Fig. S9 in Supporting information). Compared with Pd₆Ir₄/SBA-NH₂, Pd₆Ir₄/KIT-6-NH₂ shows a higher activity owing to its 3D cubic la3D structure of KIT-6, which is more conducive to mass transfer than that of SBA-15 with 2D hexagonal P6mm structure (Fig. 3b). The gas mixtures generated from FA catalyzed by PdIr-KIT-6-NH₂ were CO₂ and H₂ analyzed by gas chromatography (GC) (Fig. 3c). There were neither CO nor other gas detected (Fig. 3d) from FDR at 303 K, suggesting that the as-prepared PdIr/KIT-6-NH₂ catalyst has a 100% selectivity for FDR. Besides, the activity of the PdIr/KIT-6-NH₂ catalyst is related to the molar ratio of FA/SF (SF is the abbreviation of sodium formate) and the optima value of FA/SF is 1/2. The presence of SF in the solution can significantly accelerate the hydrogen production rate (Fig. S10 in Supporting information). There is almost no gas production from SF alone, suggesting a negligible contribution from direct SF hydrolysis. The PdIr/KIT-6-NH₂ catalyst with a 15 wt% loading of PdIr NPs (Fig. S11 in Supporting information) shows the best activity among the investigated samples with different metal loading, which is close to the designed theoretical values as examined by ICP-AES (Table S2 in Supporting information). We also studied the activity of Pd alloyed with other noble metals (i.e., Rh, Pt, Au, Ag and Ru). As shown in Fig. S12 (Supporting information), Pd₆Ir₄/KIT-6-NH₂ shows the highest catalytic activity.

  The kinetic characteristics of FDR were also studied in Fig. 4. It could be found that the catalytic activity of the catalysts is positively correlated with reaction temperature (Figs. 4a and b). The activation energy of FDR over Pd₆Ir₄/KIT-6-NH₂ is 36.8 kJ/mol, which is lower than that of Pd/KIT-6-NH₂ (53.6 kJ/mol) and most of the reported heterogeneous catalysts (Table S4 in Supporting information). The decrease in the activation energy of FDR over PdIr/KIT-6-NH₂ can be attributed to the smaller size of metal NPs (Fig. S13 in Supporting information) and the contribution of PdIr alloy effects. Figs. 4c and d show FDR over the Pd₆Ir₄/KIT-6 catalyst is near to the first-order reaction and zero-order reaction with respect to catalyst concentrations and FA concentrations, respectively.

  Figure 4

  

  Figure 4.  Volume of generated gas versus time for FDR over (a) PdIr/KIT-6-NH₂ and (b) Pd/KIT-6-NH₂ at different temperatures, and volume of generated gas versus time for FDR over PdIr/KIT-6-NH₂ with (c) different FA concentrations and (d) catalyst concentrations (inset: the corresponding kinetic parameters).

  DownLoad: Full-Size Img PowerPoint

  As shown in Fig. S14 (Supporting information), the H₂ selectivity and productivity over PdIr/KIT-6-NH₂ toward FDR remain unchanged after four runs, but the activity shows a slightly decrease (Fig. S14). After the reusability test, the recovered catalysts were analyzed by XRD and nitrogen adsorption-desorption analyzed, and there is no obvious change in structure of KIT-6 (Figs. S15 and S16 in Supporting information). The small decrease in the activity of PdIr/KIT-6-NH₂ can be attributed to a slight increase in the size of PdIr NPs (Fig. S17 in Supporting information) and a small decrease of amino-groups amounts of KIT-6-NH₂ analyzed by EA (Table S3 in Supporting information).

  In summary, PdIr/KIT-6-NH₂ catalyst was prepared, for the first time, with a simple wet chemical method. Compared with PdIr NPs immobilized by SBA-15-NH₂ and bare KIT-6 without amino-groups, the PdIr/KIT-6-NH₂ catalyst displayed the highest catalytic performance toward FDR, achieving an initial TOF value of 3533 h⁻¹ and 100% selectivity at ambient temperature. Besides the contribution of electron-rich Pd-Ir alloy NPs with high dispersion and small size, the high catalytic performance can be attributed to the amino-groups of KIT-6-NH₂, which cannot only serve as the anchoring sites to form the ultrafine PdIr NPs and enhance the interaction between PdIr NPs and KIT-6-NH₂ but also act as a proton scavenger to promote the breaking of O-H bond of FA. This work provides an effective and simple method for preparing ultrafine metal NPs confined by porous carrier for various applications.

  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.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (Nos. 21763012, 22162014) and Scientific Research Foundation of Graduate School of Jiangxi Province (No. YC2020-B067).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2021.08.033.

  
  
• 1. [1]

     Y. Zhang, J. Liu, S. Li, et al., EnergyChem 1 (2019) 100021 doi: 10.1016/j.enchem.2019.100021

  2. [2]

     W. Hua, H. H. Sun, F. Xu, et al., Rare Met. 39 (2020) 335-351 doi: 10.1007/s12598-020-01384-7

  3. [3]

     A. Zhang, H. Du, Z. H. Lu, Acta Chim. Sin. 79 (2021) 885-902 doi: 10.6023/a21030126

  4. [4]

     J. Zhang, F. Lin, L. Yang, et al., Chin. Chem. Lett. 31 (2020) 2019-2022 doi: 10.1016/j.cclet.2019.11.042

  5. [5]

     X. Gu, Z. H. Lu, H. L. Jiang, et al, J. Am. Chem. Soc. 133 (2011) 11822-11825 doi: 10.1021/acs.accounts.7b00132

  6. [6]

     Z. J. Zhang, Y. X. Luo, S. W. Liu, et al., J. Mater. Chem. A7 (2019) 21438-21446 doi: 10.1039/c9ta06987a

  7. [7]

     Q. M. Sun, J. M. Ben, W. J. Chen, et al., Angew. Chem. Int. Ed. 59 (2020) 20183-20191 doi: 10.1002/anie.202008962

  8. [8]

     Y. Yang, Z. H. Lu, Y. Hu, et al., RSC Adv. 4 (2014) 13749-13752 doi: 10.1039/C3RA47023G

  9. [9]

     X. Cao, J Zhou, S. Li, et al., Rare Met. 39 (2020) 113-130 doi: 10.1007/s12598-019-01350-y

  10. [10]

      L. S. Jose, M. S. Cristina, I. Svetlana, et al., Appl. Catal. B Environ. 282 (2021) 119615 doi: 10.1016/j.apcatb.2020.119615

  11. [11]

      Q. Y. Bi, J. D. Lin, J. D. Liu, et al., Angew. Chem. Int. Ed. 55 (2016) 11849-11853 doi: 10.1002/anie.201605961

  12. [12]

      M. Farajzadeh, H. A. lamgholiloo, F. Nasibipour, et al., Sci. Rep. 10 (2020) 18188 doi: 10.1038/s41598-020-75369-y

  13. [13]

      Y. Luo, W. Nie, Y. Ding, et al., ACS Appl. Energy Mater. 4 (2021) 4945-4954 doi: 10.1021/acsaem.1c00520

  14. [14]

      Y. Chao, X. F. Guo, B. Shen, et al., J. Mater. Chem. A6 (2018) 23766-23772 doi: 10.1039/C8TA09342C

  15. [15]

      J. Han, Z. J. Zhang, Z. R. Hao, et al., J. Colloid Interfaces Sci. 587 (2021) 736-742 doi: 10.1016/j.jcis.2020.11.033

  16. [16]

      K. Mori, S. Y. Masuda, H. Tanaka, et al., Chem. Commun. 53 (2017) 4677-4680 doi: 10.1039/C7CC00864C

  17. [17]

      J. L. Santos, C. Leon, G. Monnier, et al., Int. J. Hydrog. Energy 45 (2020) 23056-23068 doi: 10.1016/j.ijhydene.2020.06.076

  18. [18]

      W. Hong, M. Kitta, N. Tsumori, et al., J. Mater. Chem. A 7 (2019) 18835-18839 doi: 10.1039/c9ta06014f

  19. [19]

      Z. Z. Wang, S. P. Liang, X. Y. Meng, et al., Appl. Catal. B Environ. 291 (2021) 120140 doi: 10.1016/j.apcatb.2021.120140

  20. [20]

      W. Wang, X. Hong, Q. Yao, et al. J. Mater. Chem. A8 (2020) 13694-13701 doi: 10.1039/d0ta05322h

  21. [21]

      C. G. Jessica, N. G. Miriam, S. T. David, et al., ACS Sustain. Chem. Eng. 8 (2020) 15030-15043 doi: 10.1021/acssuschemeng.0c05906

  22. [22]

      T. Wang, J. Du, Y. Sun, et al., Chin. Chem. Lett. 32 (2021) 1186-1190 doi: 10.1016/j.cclet.2020.07.044

  23. [23]

      Y. Yan, H. Li, Z. H. Lu, et al., Chin. Chem. Lett. 30 (2019) 1153-1156

  24. [24]

      J. Cheng, X. Gu, X. L. Sheng, et al., J. Mater. Chem. A4 (2016) 1887 184

  25. [25]

      K. Wang, Q. Yao, Q. Qing, et al., J. Mater. Chem. A7 (2019) 9903-9911 doi: 10.1039/c9ta01066a

  26. [26]

      I. Choi, H. Lee, G. Lee, et al., Rare Met. 39 (2020) 767-783 doi: 10.1007/s12598-019-01203-8

  27. [27]

      Z. Pan, Z. Tang, Y. Zhang, et al., Tungsten 2 (2020) 390-402 doi: 10.1007/s42864-020-00065-3

  28. [28]

      Z. Zhang, S. Zhang, Q. Yao, et al., Inorg. Chem. Front. 5 (2018) 370-377 doi: 10.1039/C7QI00555E

  29. [29]

      K. Zhan, P. Liu, J. Dong, et al., Chin. Chem. Lett. 31 (2020) 1630-1634

  30. [30]

      X. Ji, K. T. Lee, R. Holden, et al., Nat. Chem. 2 (2010) 286-293 doi: 10.1002/adma.201703038

  31. [31]

      F. Z. Song, Q. L. Zhu, X. C. Yang, et al., Adv. Energy Mater. 8 (2017) 1701416

  32. [32]

      F. Q. Tang, L. L. Li, D. Chen, Adv. Mater. 24 (2012) 1504-1534 doi: 10.1039/c9ta10196a

  33. [33]

      Q. J. Wang, N. Tsumori, M. Kitta, et al., ACS Catal. 8 (2018) 12041-12045 doi: 10.1021/acscatal.8b03444

  34. [34]

      G. X. Yang, Y. Kuwahara, S. Masuda, et al., J. Mater. Chem. A8 (2020) 4437-4446 doi: 10.1021/ja200122f

  35. [35]

      X. Li, A. E. Surkus, J. Rabeah, et al., Angew. Chem. Int. Ed. 59 (2020) 15849-15854 doi: 10.1002/anie.202004125

  36. [36]

      A. B. Redondo, F. L. Morel, M. Ranocchiari, et al., ACS Catal. 5 (2015) 7099-7103 doi: 10.1021/acscatal.5b01987

  37. [37]

      G. D. Feng, J. Y. Wang, M. Boronat, et al., J. Am. Chem. Soc. 140 (2018) 4770-4773 doi: 10.1021/jacs.8b00093

  38. [38]

      Y. D. Zou, B. T. Huang, L. H. Cao, et al., Adv. Mater. 33 (2020) 2005215

  39. [39]

      J. Liang, Z. B. Liang, R. Q. Zou, et al., Adv. Mater. 29 (2017) 1701139 doi: 10.1002/adma.201701139

  40. [40]

      S. Masuda, K. Mori, Y. Futamura, et al., ACS Catal. 8 (2018) 2277-2285 doi: 10.1021/acscatal.7b04099

  41. [41]

      M. Huang, Q. Yao, G. Feng, et al., Inorg. Chem. 59 (2020) 5781-5790 doi: 10.1021/acs.inorgchem.0c00600

  42. [42]

      K. Koh, M. Jeon, C. W. Yoon, et al., J. Mater. Chem. A5 (2017) 16150-16161 doi: 10.1039/C7TA02040F

  43. [43]

      Y. X. Luo, Q. F. Yang, W. D. Nie, et al., ACS Appl. Mater. Interfaces 12 (2020) 8082-8090 doi: 10.1021/acsami.9b16981

  44. [44]

      W. D. Nie, Y. X. Luo, Q. F. Yang, et al., Inorg. Chem. Front. 7 (2020) 709-717 doi: 10.1039/c9qi01375j

  45. [45]

      S. P. Wang, S. H. Hou, C. Wu, et al., Chin. Chem. Lett. 30 (2019) 398-402

• 

  Figure 1  (a) TEM of KIT-6-NH₂, (b) TEM, (c) HRTEM, (d) HAADF-STEM, (e) the corresponding EDS mapping of PdIr/KIT-6-NH₂ and (f) TEM of PdIr/KIT-6 (inset: the corresponding particle size histogram).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Nitrogen sorption isotherms of KIT-6, KIT-6-NH₂, and PdIr/KIT-6- NH₂, high-resolution XPS spectra of (b) N 1s in PdIr/KIT-6-NH₂ and KIT-6-NH₂, (c) Pd 3d in PdIr/KIT-6-NH₂ and Pd/KIT-6-NH₂ and (d) Ir 4f in PdIr/KIT-6-NH₂ and Ir/KIT-6-NH₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Volume of the generated gas versus time for FDR in FA-SF aqueous solution over (a) Pd_xIr_10-x/KIT-6-NH₂ with different Pd/Ir ratios and (b) Pd₆Ir₄ NPs confined by different supports. GC spectra for the commercial gas and the generated gas from FDR over Pd₆Ir₄/KIT-6-NH₂ using (c) TCD detector and (d) FID-methanator.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Volume of generated gas versus time for FDR over (a) PdIr/KIT-6-NH₂ and (b) Pd/KIT-6-NH₂ at different temperatures, and volume of generated gas versus time for FDR over PdIr/KIT-6-NH₂ with (c) different FA concentrations and (d) catalyst concentrations (inset: the corresponding kinetic parameters).

  下载: 全尺寸图片 幻灯片
•