arXiv2025
This study presents a prototype D-band waveguide orthomode transducer (OMT) fabricated using chemically etched brass platelets. This method offers a fast, cost-effective, and scalable approach for producing waveguide OMTs above 100 GHz, making it well-suited for current and future Cosmic Microwave Background polarization experiments, where large focal planes with thousands of receivers are required to detect the faint primordial \textit{B}-modes. Chemical etching has already demonstrated its effectiveness in manufacturing corrugated feedhorn arrays with state-of-the-art performance up to 150 GHz. Here, we evaluate its applicability to more complex structures, such as OMTs. We designed a single OMT prototype operating in the 130-170 GHz range, fabricated by chemically etching 0.15 mm-thick brass plates, which were then stacked, aligned, and mechanically clamped. Simulations based on metrological measurements of the OMT profile predict return losses below $-$10 dB, isolation better than $-$30 dB, and transmission around $-$0.5 dB. The measured transmission and isolation, however, is around $-$1.5/$-$2 dB and $<-$20 dB, respectively. Further simulations show that the degradation in the transmission is related to defects and roughness along the etched profile ($\mathrm{RMS}\simeq$3 $\mu$m), which is a typical and unavoidable effect of chemical etching. The discrepancy in isolation, instead, could arise from a slight rotation ($\sim$3$^{\circ}$) of the polarization angle within the measurement chain. Our results show that chemical etching is a fast, low-cost, and scalable technique for producing waveguide OMTs with state-of-the-art performance in terms of return loss and isolation. However, at frequencies above 100 GHz the transmission coefficient may degrade due to the mechanical precision limitations of chemical etching.