Author: Grigorii Skorupskii

Corresponding author: MirceaDincă

Communication unit: Massachusetts Institute of Technology

Research highlights:

1. A series of two-dimensional MOFs based on lanthanides are reported, and it is found that effective charge transfer can occur in the vertical direction of two-dimensional flakes.

2. It is found that in layered MOFs, high conductivity does not necessarily require metal ligand bonds with high covalent properties, and the interaction between organic ligands can generate effective charge transport pathways.

Conductive MOFs

In recent years, conductive metal organic frameworks (MOFs) have received extensive attention as potential materials in the fields of supercapacitors, batteries, thermoelectric devices, chemical sensors, and electrocatalysts. A lot of effort has been invested in the development of MOFs with higher and higher conductivity.

At room temperature, the current recorded value of non-porous coordination polymers is 2500 S cm-1, and the recorded value of porous MOFs is 40 S cm-1. There are many materials close to these recorded values, such as the reported hexagonal delamination. The framework is based on the hexa-substituted triphenylene derivative and benzene connected by the first row of divalent transition metal ions such as Ni2+, Cu2+. The high conductivity values of these materials are often attributed to the strong planar conjugation between the π-system of the ligand and the metal d orbital.

Up to now, there is no research on the standard transfer of electric charge to metal-organic thin films. Although some people have tried to use metal-ligand bonding to make two-dimensional (2D) MOFs with varying degrees of covalent bonding, which can answer basic questions about charge transport in these materials, details about their structure or electronic properties There is no report yet.

Lanthanide MOFs

Lanthanide metal ions are a class of metal ions that are very suitable for systematic research on the relationship between structure and function. In the trivalent state, the 5d electron shell of the lanthanum ion (Ln3+) is empty, and all valence electrons are located on the highly shielded deep 4f orbital. Because the 4f orbital does not significantly participate in the bond formation, the ionic compounds formed by Ln3+ ions are much more than those of transition metals.

This is very attractive for MOF synthesis, because in MOF synthesis, more unstable and reversible bonds form more crystalline materials, which is essential for achieving systematic control of the electronic structure of materials. Another benefit of using Ln3+ as the targeted conductive ions for conductive MOFs is that their ionicity leads to almost the same chemical behavior. In addition, their ionic radius is very different, from La3+ to Lu3+ reduced by more than 15% .

These properties make them an ideal tool for studying the correlation of structure functions, because covalency is essentially independent of structural considerations.

Introduction

In view of this, the Mircea Dincă research group of the Massachusetts Institute of Technology and others have reported a family of MOFs composed of Ln3+ and 2,3,6,7,10,11-hexahydroxytriphenyl (H6HHTP). Although the ionic bond between Ln3+ and the catecholate ligand may reduce the efficiency of charge transfer in the two-dimensional 2D plane, when the material is measured with dual-probe polycrystalline particles, the conductivity value still reaches 0.05 S cm-1, which is comparable to the conductivity value of MOFs with the best conductivity reported so far. The high crystallinity of the material allows it to systematically control the structural parameters and reveals the direct relationship between the stacking distance, electrical conductivity and optical band gap of MOFs covering the entire 4f series of four different lanthanide materials.

Figure 1. The structure of lanthanide MOFs.

Point 1: Synthesis and structure characterization

A mixed solution of H6HHTP and hydrated Ln(NO3)3 (Ln = La, Nd, Ho, Yb), combined with N,N-dimethylimidazolinone (DMI) for solvothermal reaction, to obtain Ln1+ xHHTP(H2O)n (x = 0~ 0.2; LnHHTP for short).

Scanning electron microscopy (SEM) shows that the powder is composed of well-shaped hexagonal needles (Figure 1e), whose length varies from 1 meter to 200 meters depending on the specific synthesis conditions. The powder X-ray diffraction (PXRD) data was refined by Rietveld, and their structures were obtained, providing a good structure model for Nd3+ (Figure 1) and Yb3+ analogs. These two materials have the same structure. Since the size of Yb3+ is smaller than that of Nd3+, there are differences in unit cell parameters and Ln-O bond length.

Both structures show the disorder of two sites, and each structure shows two sets of ligands and metal sites that are equally occupied. For clarity, Figure 1 shows only half of the average, disordered NdHHTP structure. Similar to the reported transition metal analogues, Ln3+ ions combine triangular HHTP ligands into a honeycomb two-dimensional network structure with soluble pores. The crystal diameter of NdHHTP is about 1.94 nm. By fitting the N2 adsorption isotherm at 77 K, the experimental pore diameters of the four materials are all ~1.6 nm, which is consistent with the crystallographic value.

The transition metal and the organic ligand are on the same plane, forming a strict two-dimensional (2D) plane, and here, the Ln3+ ions are located between the planes of the organic ligand, thereby connecting the latter into a three-dimensional network. The lanthanide itself is connected to 6 oxygen atoms and 1 water or hydroxide radical in the adjacent ligand. They are bridged into infinite chains, which can be expressed as cap-shaped triangular prisms sharing edges, similar to the coordination environment and extended structure of rare earth oxides including Nd2O3. Whether it is NdHHTP or YbHHTP, the positions of lanthanides are not completely occupied, and almost one-third of the positions are empty, which is the result of calculation according to the formula.

Key point 2: Calculation of electronic band structure

The close π-stacking promotes the orbital overlap of effective ligands in the direction of crystal c, thereby promoting the normal two-dimensional flakes of charge transport. The researchers used density functional theory (DFT) to investigate their electronic structure materials. In order to avoid the prohibitively computationally intensive spin-polarization calculations, the researchers calculated the closed-shell LaHHTP structure and the hypothetical LuHHTP structure, the latter being used as a model for the smaller lanthanides Yb3+ and Ho3+.

These calculations indicate that the material should exhibit metallic behavior along the c direction (the Brillouin zone vector A to Γ in Figure 2), which can be seen from the energy band crossing the Fermi level. In contrast, the Fermi level in the energy band lies in the a-b plane (Γ-K-M), and exhibits semiconductor behavior in the plane parallel to the organic layer (Γ-K-M). As expected, the valence layer orbitals of the lanthanides do not contribute much to the energy band around the Fermi level, which indicates that there is little in-plane electronic communication between the ligands.

In fact, the energy band is basically flat along the Γ-K-M vector, and the formed "in-plane" band gap (range 1.2 ~ 0.7 eV) strongly depends on the stacking distance. In sharp contrast, the band structure is along the A to Γ vector (forming a "cross plane"), where the bands accumulate significantly, resulting in a low density of states (DOS). In addition, the modeling of the structure by different metal atoms results in an insignificant change from La to Lu.

Figure 2. Electronic energy band structure and DOS diagram of LaHHTP.

Point 3: Spectral research

The diffuse reflectance spectra of the four materials (Figure 3a) showed clear absorption edges in the range of 0.7-1eV, and the researchers used it as the in-plane band gap predicted by the DFT calculation. The four MOFs also show absorption characteristics below this edge, with La and Nd about 0.3 eV, and Ho and Yb about 0.7 eV. However, these absorption features do not conform to the linear region when drawn on the Tauc line type of the direct or indirect band gap, and therefore can be attributed to the absorption of defects.

The absorption characteristics at 2000-3000 nm in the [Ru3HHTP]2+ complex are attributed to the charge transfer within the ligand. Directly allowed transitions plotted in Tauc coordinates (Figure 3b), the spectrum shows a surprising trend from 0.85 eV for the larger lanthanide element LaHHTP to 0.73 eV for the smaller lanthanide element YbHHTP. Reduce the optical band gap.

This trend is contrary to the phenomenon observed in most classic semiconductors (including GaAs, InP, and Ge): In these semiconductors, stressing and lowering the cell parameters will cause the band gap to expand. However, the trend of LnHHTP materials is consistent with the situation observed in semiconductors such as PbE (E = S, Se, Te) with band inversion and some indirect gap semiconductors including Si. The precise positioning of this trend requires further system experiments.

However, the researchers noticed that the DFT calculation here positions the Fermi energy on the ligand oxidation of HHTP3-, and the experimental formula shows that this oxidation state slightly changes to more reduced ligands, which will make the Fermi energy Can become higher.

Figure 3. Diffuse reflectance spectrum of LnHHTP (Ln = La, Nd, Ho, Yb).

Point 4: Conductivity

The electrical conductivity of polycrystalline microspheres made of LnHHTP material is comparable to that of porous MOFs with the best conductivity so far, ranging from 0.9×10-4 S cm-1 for LaHHTP to 0.05 S cm-1 for HoHHTP (Figure 4a). Although the batch-to-batch difference of each material is relatively large, for smaller lanthanide elements (Ho, Yb), higher average and higher device conductivity can be obtained, which is due to these elements Produced a denser stacked structure.

It is worth noting that the Ln-O bond is covalently lower in the LnHHTP material. Importantly, the value of all LnHHTP MOFs may be seriously underestimated due to the potential anisotropy of charge transport and the additional contact and grain boundary resistance. Although these contributions are difficult to quantify, since all MOFs exhibit similar morphology and grain size, they are assumed to have a certain similarity to different materials. Compared with the double probe microspheres, the conductivity of single crystal MOFs is improved by more than two orders of magnitude.

The variable temperature conductivity measurement results reveal the thermally activated transport of the four materials (Figure 4b). The Arrhenius equation fits the equation between conductance and temperature as G = G0exp (EA/kT), where G is conductance, G0 pre-factor, EA is activation energy, k Boltzmann constant and T temperature, at 225 In the -300k temperature range, for all four materials, a similar EA value is displayed, about 0.25 eV. These values are consistent with the reported values of other highly conductive MOFs, including the chemically related 2,5-dihydroxybenzoquinone and lanthanide-based materials.

Figure 4. The conductivity of LnHHTP (Ln = Yb, Ho, Nd, La).

summary

The current interest in 2D materials is mainly focused on the in-plane electronic properties, while the out-of-plane transport properties are relatively less concerned. Compared with traditional inorganic 2D materials, MOFs can produce functional materials through simple connections, and the 2D layers of these functional materials are connected by strong bonds.

The researchers demonstrated that the reaction of lanthanide ions with ligands traditionally used to synthesize two-dimensional MOFs can produce layered materials, in which organic ligands form flakes, and in-plane electronic communication is not important. Electrical transport mainly occurs in the direction perpendicular to the plane and is affected by the highly adjustable interlayer stacking distance, which changes in proportion to the radius of the lanthanide cations. These results provide another point of view for the canonical interpretation of transmission in the two-dimensional MOFs of conductibility, which is almost entirely focused on plane parameters. The findings of this study expand the range of interactions that may produce effective transport in these porous materials, thereby providing additional design strategies for MOFs with recorded conductivity and charge delocalization in all three dimensions.

references:

Skorupskii, G., Trump, B.A., Kasel, T.W. et al. Efficientand tunable one-dimensional charge transport in layered lanthanidemetal--organic frameworks. Nat. Chem. 12, 131–136 (2020).

DOI: 10.1038/s41557-019-0372-0

https://doi.org/10.1038/s41557-019-0372-0
This information is from the Internet for academic exchanges. If there is any infringement, please contact us and delete it immediately