Screening Aluminum-Based Compounds as Low-κ Dielectrics for High-Frequency Applications

Abstract

Advances in telecommunications require electronics that operate at ever-increasing frequencies, exemplified by 5G or fifthgeneration technologies that operate in the GHz regime. At high frequencies, electrical circuits are plagued by so-called RC delays, arising from the time constant τ = RC for electrical signals which is the product of the resistance R and the capacitance C, respectively, of conductors and their insulating substrates. Besides using high quality, low-R electrical conductors such as high-purity Cu with low surface roughness, small RC delays are achieved by lowering the dielectric constant κ of the materials used in printed circuit board substrates. These largely comprise particles of an inorganic material, notably functionalized SiO 2 , embedded in a polymer-based matrix. The value of κ of the composite is primarily dictated by κ of the inorganic material. The properties of the inorganic component also impact other relevant parameters such as the quality factor, mechanical strength, and thermal expansion of the substrate. Here, we ask whether there are inorganic compounds with dielectric constants (measured at 10 GHz) that are lower than that of SiO 2 and potentially replace it in electronics. We describe the key characteristics for low-κ materials and develop a framework for screening such compounds by employing some guiding principles, followed by using a combination of empirical estimates and density functional perturbation theory-based calculations. We then report experimental results on two promising aluminum-based low-κ compounds for high-frequency applications. The first is the cristobalite form of AlPO 4 . The second is the simplest 3D metal−organic framework, aluminum formate Al(HCOO) 3 . The measured values of κ at 10 GHz, which are 4.0 for AlPO 4 and 3.8 for Al(HCOO) 3 , compare well with what is measured on SiO 2 particles.