Elsevier

Journal of Catalysis

Volume 318, October 2014, Pages 128-142
Journal of Catalysis

Stability and hydrogen adsorption of metal–organic frameworks prepared via different catalyst doping methods

https://doi.org/10.1016/j.jcat.2014.07.010Get rights and content

Highlights

  • We synthesize MOF-catalyst by various doping methods and compare structural quality.

  • Pre-bridge catalyst addition method avoids MOF structural degradation in high-temperature H2.

  • Pt/AC in Cu-TDPAT via pre-bridge has the highest catalytic capability.

  • Connectivity and mesoporosity between Pt/AC and MOF facilitate H migration.

Abstract

The stability of three metal–organic frameworks (MOFs), namely IRMOF-8, Cu-TDPAT, and Cu-BTC, was tested after incorporation of Pt. Stability was assessed with powder X-ray diffraction (PXRD), physical (N2 at 77 K) and chemical (H2 at 300 K) adsorption, and thermogravimetric analysis in H2 and N2. Introduction of Pt via wet precipitation led to MOF degradation during the H2 reduction step. Addition of pre-reduced Pt supported on activated carbon (Pt/AC) to MOFs via physical mixing also led to structural degradation. However, addition of Pt/AC via a ‘pre-bridge’ (PB) technique led to high MOF stability, with the retention of surface area, porosity, crystallinity, and thermal stability. The catalytically active surface area was assessed by hydrogen adsorption, and demonstrated extension of the catalytically active surface area to the MOF surface. High hydrogen uptake correlated with MOF particle size, due to the connectivity between Pt/AC and MOF, and the interpenetration of Pt/AC into the MOF crystal.

Introduction

The long-range structural order, high surface area (SA), adjustable pore size, and tunable functionality of metal–organic frameworks (MOFs) [1] have led to extensive study of these materials for gas storage and separation applications [2], [3], [4], [5]. High surface area generally correlates with high cryogenic adsorption: MOF-177 [6], for example, with the highest SA to date (4500 m2/g [7]), has nominally met DOE gravimetric H2 storage targets (7.5 wt% excess adsorption at 7 MPa, 77 K [8]), albeit at temperatures well below the DOE target temperature range. These same properties provide promise for application of MOFs in drug delivery [9], sensor development [10], and size-selective heterogeneous catalysis [11]. The stability of MOFs in terms of thermal and chemical stability has been tested previously [12], but to utilize MOFs as a catalyst support, the stability needs to be re-examined in the presence of the catalyst of interest.

A number of methods to incorporate catalytic transition metals into MOFs have been proposed and tested, including wet precipitation [13], [14], [15], [16], incipient wetness [17], chemical vapor deposition (CVD) [18], [19], and simple physical mixing (both with and without introduction of a carbonized sugar to provide “bridges” between the presupported catalyst and MOF) [20]. However, to realize the aforementioned benefits of MOFs in catalytic applications, it is important that the catalytic doping method retains the MOF structural integrity, without introducing substantial defects, which may lead to irreproducibilities, poorly characterized structures, and long-term instability. It is well established that PXRD characterization of MOFs can be used to characterize structural degradation: peak splitting, changes in relative peak intensity, and/or the appearance/disappearance of peaks imply loss of structure [21], [22]. Unfortunately, the structural stability of MOFs after catalyst addition has not been thoroughly examined. One possible exception is the synthesis and characterization of a Ru/MOF (MOF = Lanthanum-BTC, benzene tricarboxylate) catalyst for cyclohexene and benzene hydrogenation, where the catalyst Ru/MOF showed no obvious alteration in PXRD and had no loss in catalytic activity after five uses, with a turnover frequency (TOF) higher than 10,000 h−1 [23]. In addition, a 0.5 wt% Pd/IRMOF-1 (Zn4O(BDC)3, BDC = 1,4-benzenedicarboxylate) catalyst used for liquid phase hydrogenation of ethyl cinnamate [24], showed a higher catalytic activity than a Pd/C catalyst, and the SA and micropore volumes upon Pd introduction was retained relative to the IRMOF-1 precursor (however, the textural properties of the IRMOF-1 were reduced by a factor of 3 relative to other reports [25]). Elsewhere, catalyst addition to MOFs generally led to reduced structural integrity, reduced surface area, or both. Platinum nanoparticles supported on IRMOF-1 introduced from the metal precursor PdCl2 in liquid phase to catalyze C–C bond coupling reactions showed reasonable crystallinity in PXRD relative to undoped IRMOF-1, but N2 77 K adsorption isotherms showed a considerable reduction in SA and microporosity [16]. Despite initial reports (from multiple laboratories) that catalysts incorporated into MOFs could introduce a hydrogen dissociation source, initiate hydrogen spillover, and increase the operative hydrogen adsorption temperature to 300 K [15], [19], [20], [22], [26], [27], [28], [29], [30], [31], [32], [33], these reports have been plagued by irreproducibilities that are perhaps associated with hierarchical “mesopore” structure and/or defects [28], [34]. Close examination of the MOF characterization details after catalyst incorporation and H2 exposure showed significant changes in PXRD peak intensity and broadening, with SA reduced by one-third [19]. Previous reports show similar alterations in PXRD and SA [20].

In this paper, we test the stability of select MOF supports after various doping techniques. Introduction of a Pt cationic precursor via a “wet chemistry” technique is compared to physical mixing via ball milling/grinding (used previously by Li and Yang [26]), and a hybrid “pre-bridge” technique [35], [36], which introduces a supported Pt/AC (platinum on activated carbon) catalyst into the MOF synthesis slurry. Our primary focus was on (1) Zn-based IRMOF-8 (Zn4O(NDC)3, NDC = 2,6-naphthalenedicarboxylate) and (2) the Cu paddlewheel (Cu-PDW) based Cu-TDPAT ([Cu3(TDPAT)(H2O)3], TDPAT = 2,4,6-tris(3,5-dicarboxylphenylamino)-1,3,5-triazine). IRMOF-8 has been reported to have very high ability to incorporate H2 into its structure after catalytic doping [20], while the latter has amine functional groups, which could feasibly become positively charged to aid in traditional doping methods that require ion exchange. In select cases, we also consider (3) the Cu-BTC ([Cu3(BTC)2(H2O)3], BTC = 1,3,5-benzenetricarboxylate), as it has the same metal building unit as Cu-TDPAT (i.e., the Cu-PDW), but a purely carbon-based ligand like IRMOF-8. Structural stability is assessed with PXRD, low-pressure physical (N2 at 77 K) and chemical (H2 at 300 K) adsorption, and thermogravimetric analysis (TGA). Hydrogen adsorption is a common method to ascertain both metal dispersion and the catalytically active surface area, and also provides a screening tool for these materials for room temperature hydrogen adsorption capacity, although this specific topic will be addressed in a follow-up paper.

Section snippets

Synthesis of MOFs

The IRMOF-8 (abbreviated as I, see Fig. 1A) synthesis procedure utilized in this paper was based on a solvothermal method reported by Yaghi et al. [37] and adapted for scale-up synthesis [22]. In brief, 0.63 g Zn(NO3)2·6H2O and 0.15 g 2,6-H2NDC were dissolved in 50 ml of DEF (Diethylformamide) solvent, and ultrasonicated for better ion dispersion. The solution was then heated to 368 K for 15 h to precipitate pale white fine crystalline powders ([Zn4O(NDC)3·(DEF)6]). After filtration, the as-prepared

Direct doping, metal–organic frameworks

Cu-TDPAT (T) was directly doped (DD) with Pt via wet chemistry precipitation techniques followed by reduction of the Pt(acac)2 precursor in hydrogen at 423 K. Relative to the undoped T precursor, DD-T shows almost complete reduction of the characteristic T features as well as peaks corresponding to Cu and Pt metal (Fig. 2). It is well established that peak splitting and/or changes in relative peak intensity in MOFs imply structural degradation [21]. Although the Pt diffraction peaks at 40° (1 1 

Summary and conclusions

This paper explores three methods to introduce transition metals into MOF supports. Mechanical mixing and incipient wetness techniques lead to structural degradation of the MOF, and the resulting MOF support may bear little resemblance to the ideal crystalline structure. These results are generally consistent with previous studies that find a significant decrease in SA after metal doping of MOFs [11], [14], [15], [16], [17], [19], [30], [52], [55] as well as changes in the PXRD pattern [20],

Acknowledgments

This work was supported by the U.S. Department of Energy, Energy Efficiency and Renewable Energy program, Award DE-FG36-08GO18139. The authors thank Julie Anderson in Materials Characterization Lab at the Pennsylvania State University for the assistance in SEM and FESEM.

References (64)

  • R.J. Kuppler et al.

    Coordin. Chem. Rev.

    (2009)
  • S. Qiu et al.

    Coordin. Chem. Rev.

    (2009)
  • M.P. Suh et al.

    Coordin. Chem. Rev.

    (2008)
  • S. Gao et al.

    Appl. Catal. A – Gen.

    (2010)
  • C.Y. Wang et al.

    J. Alloy. Compd.

    (2010)
  • S. Opelt et al.

    Catal. Commun.

    (2008)
  • S.J. Park et al.

    J. Colloid Interface Sci.

    (2010)
  • S.J. Yang et al.

    Int. J. Hydrogen Energy

    (2010)
  • S.Y. Lee et al.

    Int. J. Hydrogen Energy

    (2011)
  • M. Schlesinger et al.

    Micropor. Mesopor. Mater.

    (2010)
  • J. Moellmer et al.

    Micropor. Mesopor. Mater.

    (2010)
  • D.D. La et al.

    Sensor. Actuat. B – Chem.

    (2011)
  • M. Zielinski et al.

    Catal. Commun.

    (2005)
  • A. Doménech et al.

    Electrochem. Commun.

    (2006)
  • M. Boudart et al.

    J. Catal.

    (1970)
  • M. Eddaoudi et al.

    Science

    (2002)
  • S. Kitagawa et al.

    Angew. Chem. Int. Ed.

    (2004)
  • H.L. Jiang et al.

    Chem. Commun.

    (2011)
  • H.K. Chae et al.

    Nature

    (2004)
  • D. Zhao et al.

    Energy Environ. Sci.

    (2008)
  • A.G. Wong-Foy et al.

    J. Am. Chem. Soc.

    (2006)
  • P. Horcajada et al.

    Angew. Chem. Int. Ed.

    (2006)
  • Y.K. Hwang et al.

    Angew. Chem. Int. Ed.

    (2008)
  • S.S. Kaye et al.

    J. Am. Chem. Soc.

    (2007)
  • K.L. Mulfort et al.

    Inorg. Chem.

    (2008)
  • Z. Xiang et al.

    Int. J. Hydrogen Energy

    (2011)
  • C. Zlotea et al.

    J. Am. Chem. Soc.

    (2010)
  • M. Sabo et al.

    J. Mater. Chem.

    (2007)
  • D. Esken et al.

    J. Mater. Chem.

    (2009)
  • L.F. Wang et al.

    J. Phys. Chem. C

    (2011)
  • Y. Li et al.

    J. Am. Chem. Soc.

    (2006)
  • J. Hafizovic et al.

    J. Am. Chem. Soc.

    (2007)
  • Cited by (31)

    • An insight into the electrochemical charge/discharge process of ceramic hydride electrodes for battery applications

      2022, International Journal of Hydrogen Energy
      Citation Excerpt :

      From our literature search, we have not found the use of ceramic hydrides in rechargeable battery electrodes. We could take advantage of the fact that they are light, chemically stable, with high corrosion tolerance, and have started to be used in electrochemical systems [10–12]. The study on the hydrogen storage capacity at ceramic electrodes is discussed in this work.

    • Effect of Pt nanoparticle decoration on the H<inf>2</inf> storage performance of plasma-derived nanoporous graphene

      2021, Carbon
      Citation Excerpt :

      The method is in principle straightforward, however, when additional phenomena (e.g. adsorption on the catalyst support, hydride formation and atomic hydrogen spillover) occur the dispersion cannot be calculated. In such cases, according to a previous report [64], the catalyst dispersion obtained from a H2 sorption isotherm can be expressed as pseudo-dispersion (μ), which could be higher than 100%. Moreover, in order to describe the trend of excess hydrogen that spills from a catalyst to the support, the spillover efficiency (η) is defined as the ratio of pseudo-dispersion to catalyst dispersion (η = μ/D).

    • Nanoconfinement of metal oxide MgO and ZnO in zeolitic imidazolate framework ZIF-8 for CO<inf>2</inf> adsorption and regeneration

      2020, Journal of Hazardous Materials
      Citation Excerpt :

      Metal-organic framework (MOF), on the other hand, is a new adsorbent with high SSA and microporosity in a crystalline structure constructed by metal clusters and organic ligands (Kaneti et al., 2017). MOFs have been applied for gas adsorption (Wang et al., 2010; Wang et al., 2014b), gas separation (Cheng et al., 2017; Adhikari and Lin, 2016), catalysis (Chang et al., 2019; Wen et al., 2017), etc. In addition, MOF derived materials (Wang et al., 2018; Li et al., 2018) also possess retained porosity with stable structure, and thus MOFs and the derivatives are possible candidates for CO2 adsorption (Sumida et al., 2012; Liu et al., 2012; Gadipelli and Guo, 2015).

    View all citing articles on Scopus
    View full text