Thermal conductivity of a metal-organic framework (MOF-5): Part II. Measurement

https://doi.org/10.1016/j.ijheatmasstransfer.2006.10.001Get rights and content

Abstract

The thermal conductivity of MOF-5 single crystals is measured over a wide temperature range between 6 K and 300 K, using the longitudinal, steady-state heat flow method. Between 6 K and 20 K, the thermal conductivity increases with the increase in temperature and exhibits a peak near 20 K. This peak results from the crossover between the decreasing mean free path and the increasing phonon specific heat with the increasing temperature. From 20 K to 100 K, the thermal conductivity decreases rapidly with increasing temperature. Above 100 K, the thermal conductivity is nearly temperature independent, and its value at 300 K is 0.32 W/m K, a rather low value for crystals. The mean free path analysis shows that at high temperature, the phonon mean free path is minimized to the cage size due to the porous, flexible structure of MOF-5.

Introduction

The crystalline metal-organic frameworks (MOFs) are a new sub-family of the nanoporous crystals. They are currently receiving attentions because of their high adsorption surface area and large free cage volume [1], [2]. MOFs are most attractive for their high capacity for hydrogen absorption and storage [3]. Many members of the MOF family are synthesized in recent years. Unlike other nanoporous crystals with inorganic host frameworks, MOFs have three-dimensional hybrid frameworks, which are comprised of metal-oxygen cages connected by a variety of organic bridges and lead to designable pore size, shape and functionality [3]. MOF-5, is the first stable cube-like structure in the family. MOF-5 has a regular, three-dimensional cubic lattice with 1,4-benzenedicarboxylate (BDC) as edges and Zn4O cluster as vertexes (see Fig. 1). The thermal properties of MOF-5 are important for gas storage and other potential applications.

Microporous crystals, such as zeolites, characterized by large unit cells and angstrom sized pores and linkers, normally have very low thermal conductivities [4], [5], [6]. Generally these crystals cannot be grown to a big-size single crystal and their thermal conductivities are often extracted indirectly from the measurement of loose or compacted powder [4]. Since there are many factors affecting the effective thermal conductivity and currently no model can account for all of them, the uncertainty in the effective crystal thermal conductivity is large [4]. MOF-5, however, has been grown up to a linear dimension of 1–2 mm, making the direct measurement possible. MOF-5 has a longer linker and a larger pore size, compared to most nanoporous crystals (including zeolites). For example, the number density of MOF-5 is 2.46 × 1028 atoms/m3, much less than those of zeolites [sodalite (5.13 × 1028 atoms/m3) and zeolite-A (4.10 × 1028 atoms/m3)] [7]. Therefore, MOF-5 is expected to have a even lower thermal conductivity than most nanoporous crystals. Also, the special cage–bridge structure of MOF-5 makes it an ideal object to study the effects of the cage and bridge structure on the microscale energy transport. Considering the freedom in the construction of MOFs [3], such a study may lead to the first step in the systematic design of structures with the desired thermal properties.

Here we report the measurement of the thermal conductivity of MOF-5 over a wide temperature range, from 6 K to 300 K, using the longitudinal, steady-state heat flow method. Then the low thermal conductivity of MOF-5, and its weak temperature dependence at high temperature, are discussed, which are shown to be due to the minimization of the phonon mean free path.

Section snippets

Crystal preparation

We synthesized large single cubic crystals by mixing 8.38 g Zn(NO3)2 · 4H2O (32.0 mmol) and 1.77 g terephthalic acid (10.7 mmol) dissolved in 100 mL DEF in a glass beaker and sonicating the mixture for 15 min. The solution was dispensed evenly into 20 scintillation vials (20-mL size) by using a plastic syringe equipped with a PTFE filter (Whatman, 0.45 μm pore size). The vials were then tightly capped and placed in an isothermal oven. The reactions were stopped after being heated at 368 K for 72 h. The

Results and discussion

Fig. 4 shows the variation of the MOF-5 thermal conductivity with respect to temperature, from 6 K to 300 K. The experimental uncertainty of the absolute thermal conductivity is within ±15% (estimated by the standard error relation [13]). The uncertainty mainly results from the difficulty in the accurate determination of the effective cross-section area A (due to the small size and irregular shape of the sample) and the effective length of the heat flow path d (due to the junctions).

Since MOF-5

Conclusions

The thermal conductivity of MOF-5 is measured for the first time over a wide temperature range, from 6 K to 300 K. The peak appears at about 20 K, and above 100 K the thermal conductivity is nearly temperature independent. The analysis of the mean free path suggests three regimes, namely, below 35 K, the lattice-defect scattering is an important scattering mechanisms. From 35 K to 100 K, the interphonon scattering dominates. Above 100 K, the phonon mean free path reaches its minimum, which is also

Acknowledgements

This work has been supported by the US Department of Energy, Office of Basic Energy Sciences under grant DE-FG02-00ER45851.

References (17)

There are more references available in the full text version of this article.

Cited by (0)

View full text