Material survey for millimeter-wave absorber using 3-D printed mold

TL;DR

This paper surveys materials for millimeter-wave absorbers made with 3D-printed molds, identifying optimal materials and designs to achieve low reflectance across 20-300 GHz for cryogenic astronomical receivers.

Contribution

It introduces a new 3D-printed mold technique for shaping RAM surfaces and evaluates multiple materials, identifying a mixture of epoxy and carbon fiber as optimal for low reflectance.

Findings

Epoxy and carbon fiber mixture achieves best performance.

The designed RAM meets reflectance requirements above 20 GHz.

Simulation confirms effectiveness at cryogenic temperatures.

Abstract

Radio absorptive materials (RAMs) are key elements for receivers in the millimeter-wave range. For astronomical applications, cryogenic receivers are widely used to achieve a high-sensitivity. These cryogenic receivers, in particular the receivers for the cosmic microwave background, require that the RAM has low surface reflectance ($\lesssim 1\%$) in a wide frequency range (20--300 GHz) to minimize the undesired stray light to detectors. We develop a RAM that satisfies this requirement based on a production technology using a 3D-printed mold (named as RAM-3pm). This method allows us to shape periodic surface structures to achieve a low reflectance. A wide range of choices for the absorptive materials is an advantage. We survey the best material for the RAM-3pm. We measure the index of refraction ($n$) and the extinction coefficient ($\kappa$) at liquid nitrogen temperature as well as at room temperature of 17 materials. We also measure the reflectance at the room temperature for the selected materials. The mixture of an epoxy adhesive (STYCAST-2850FT) and a carbon fiber (K223HE) achieves the best performance. We estimate the optical performance at the liquid nitrogen temperature by a simulation based on the measured $n$ and $\kappa$. The RAM-3pm made with this material satisfies the requirement except at the lower edge of the frequency range ($\sim$20 GHz). We also estimate the reflectance of a larger pyramidal structure on the surface. We find a design to satisfy our requirement.