Felicitas Schröder

Ruthenium Nanoparticles inside Porous [Zn4O(bdc)3] by Hydrogenolysis of Adsorbed [Ru(cod)(cot)]: A Solid-State Reference System for Surfactant-Stabilized Ruthenium Colloids

The gas-phase loading of [Zn4O(bdc)3] (MOF-5; bdc = 1,4-benzenedicarboxylate) with the volatile compound [Ru(cod)(cot)] (cod = 1,5-cyclooctadiene, cot = 1,3,5-cyclooctatriene) was followed by solid-state (13)C magic angle spinning (MAS) NMR spectroscopy. Subsequent hydrogenolysis of the adsorbed complex inside the porous structure of MOF-5 at 3 bar and 150 degrees C was performed, yielding ruthenium nanoparticles in a typical size range of 1.5-1.7 nm, embedded in the intact MOF-5 matrix, as confirmed by transmission electron microscopy (TEM), selected area electron diffraction (SAED), powder X-ray diffraction (PXRD), and X-ray absorption spectroscopy (XAS). The adsorption of CO molecules on the obtained Ru@MOF-5 nanocomposite was followed by IR spectroscopy. Solid-state (2)H NMR measurements indicated that MOF-5 was a stabilizing support with only weak interactions with the embedded particles, as deduced from the surprisingly high mobility of the surface Ru-D species in comparison to surfactant-stabilized colloidal Ru nanoparticles of similar sizes. Surprisingly, hydrogenolysis of the [Ru(cod)(cot)]3.5@MOF-5 inclusion compound at the milder condition of 25 degrees C does not lead to the quantitative formation of Ru nanoparticles. Instead, formation of a ruthenium-cyclooctadiene complex with the arene moiety of the bdc linkers of the framework takes place, as revealed by (13)C MAS NMR, PXRD, and TEM.

Loading of porous metal–organic open frameworks with organometallic CVD precursors: inclusion compounds of the type [LnM]a@MOF-5

The highly porous coordination polymer [Zn4O(bdc)3] (bdc = benzene-1,4-dicarboxylate; MOF-5 or IRMOF-1) was loaded with typical MOCVD precursor molecules 1–10 for metals such as Fe, Pt, Pd, Au, Cu, Zn, Sn. Exposure of [Zn4O(bdc)3] to the vapour of the volatile organometallic compounds, e.g. ferrocene (3), resulted in the formation of inclusion compounds of the type [LnM]a@MOF-5, where [LnM] indicates the MOCVD precursor and a denotes the effective number of molecules per cavity of the MOF-5 lattice. The obtained inclusion compounds were characterised by C/H combustion analysis, determination of the metal content by atomic absorption spectroscopy, FT-IR and solid state NMR spectroscopy and by powder X-ray diffraction. The data prove that the host lattice and the guest molecules interact only by weak van der Waals forces without any change of the framework or the chemical nature of the included molecules. Rapid desorption is observed for small and comparably volatile compounds such as pentacarbonyliron or diethyl zinc. Less labile inclusion compounds were obtained for cyclopentadienyl complexes as guest molecules, e.g. a rather high loading of six molecules of ferrocene per cavity was observed. Careful hydrolysis/calcination of [Zn(C2H5)2]2@MOF-5 resulted in the composite (ZnO)2@MOF-5 pointing to the possibility to develop a subsequent chemistry of the embedded precursor molecules to yield novel nanocomposite materials based on MOFs as host matrices and MOCVD precursors in general.

Pd@MOF-5: limitations of gas-phase infiltration and solution impregnation of [Zn4O(bdc)3] (MOF-5) with metal–organic palladium precursors for loading with Pd nanoparticles

The limitations of the loading of the porous metal–organic framework [Zn4O(bdc)3] (bdc = benzene-1,4-dicarboxylate; MOF-5 or IRMOF-1) with Pd nanoparticles was investigated. First, the volatile organometallic precursor [Pd(η5-C5H5)(η3-C3H5)] was employed to get the inclusion compound [Pd(η5-C5H5)(η3-C3H5)]x@MOF-5 via gas-phase infiltration at 10−3 mbar. A loading of four molecules of [Pd(η5-C5H5)(η3-C3H5)] per formula unit of MOF-5 (x = 4) can be reached (35 wt.% Pd). Second, the metal–organic precursor [Pd(acac)2] (acac = 2,4-pentanedionate) was used and the inclusion materials [Pd(acac)2]x@MOF-5 of different Pd loadings were obtained by incipient wetness infiltration. However, the maximum loading was lower as compared with the former case with about two precursor molecules per formula unit of MOF-5. Both loading routes are suitable for the synthesis of Pd nanoparticles inside the porous host matrix. Homogeneously distributed nanoparticles with diameter of 2.4(±0.2) nm can be achieved by photolysis of the inclusion compounds [Pd(η5-C5H5)(η3-C3H5)]x@MOF-5 (x ≤ 4), while the hydrogenolysis of [Pd(acac)2]x@MOF-5 (x ≤ 2) leads to a mixture of small particles inside the network (< 3 nm) and large Pd agglomerates (∼40 nm) on the outer surface of the MOF-5 specimens. The pure Pdx@MOF-5 materials proved to be stable under hydrogen pressure (2 bar) at 150 °C over many hours. Neither hydrogenation of the bdc linkers nor particle growth was observed. The new composite materials were characterized by 1H/13C-MAS-NMR, powder XRD, ICP-AES, FT-IR, N2 sorption measurements and high resolution TEM. Raising the Pd loading of a representative sample Pd4@MOF-5 (35 wt.% Pd) by using [Pd(η5-C5H5)(η3-C3H5)] as precursor in a second cycle of gas-phase infiltration and photolysis was accompanied by the collapse of the long-range crystalline order of the MOF.

Doping of Metal-Organic Frameworks with Functional Guest Molecules and Nanoparticles

Nanoparticle synthesis within metal-organic frameworks (MOFs) is performed by the adsorption of suitable precursor molecules for the metal component and subsequent decomposition to the composite materials nanoparticles@ MOF. This chapter will review different approaches of loading MOFs with more complex organic molecules and metal-organic precursor molecules. The related reactions inside MOFs are discussed with a focus on stabilizing reactive intermediates in the corresponding cavities. The syntheses of metal and metal oxide nanoparticles inside MOFs are reviewed, and different synthetic routes compared. Emphasis is placed on the micro structural characterization of the materials nanoparticles@MOF with a particular focus on the location of embedded nanoparticles using TEM methods. Some first examples of applications of the doped MOFs in heterogeneous catalysis and hydrogen storage are described.

Non aqueous loading of the mesoporous siliceous MCM-48 matrix with ZnO: a comparison of solution, liquid and gas-phase infiltration using diethyl zinc as organometallic precursor

Zinc oxide species hosted in the siliceous matrix MCM-48 were prepared by an organometallic route using ZnEt2 as the ZnO precursor. Gas phase as well as liquid phase infiltration of the precursor was studied in detail by ICP-AES, FT-RAIRS, 1H- and 13C-MAS-NMR, PXRD, TEM/EDX and UV–VIS. Highly loaded ZnO@MCM-48 materials with a Zn-content of up to 29.6 wt% were synthesized. A comparison of the different preparation techniques was carried out in order to find a convenient way of preparing ZnO@MCM-48 species for catalytic applications.

Loaded porous Zn4O(bdc)3 (metal@MOF-5) frameworks characterised by TEM

In recent years, metal-organic frameworks (MOFs) have received much attention because of their high specific surface areas and pore volumes, applicable in gas storage, catalysis, and photovoltaics [1]. Recently, Zn4O(bdc)3 (MOF-5; bdc = 1,4 benzenedicarboxylate) crystals have been loaded with catalytically active material like Pd, Au, Cu and Ru leading to a heightened catalytic activity in olefin hydrogenolysis and methanol synthesis [2]. These metal@MOF-5 materials can be produced by gasphase loading of metal-carrying precursors into the MOF-5 framework. This loading procedure is known to yield a MOF-5 framework loaded with catalytically active nanoparticles in the range of 1–3 nm [3]. The local distribution of these particles within the MOF-5 framework however remains unclear. Furthermore, MOFs are known to be chemically labile and the loading procedure may affect the framework structure.