Lightweight broadband electromagnetic wave absorber based on a three-dimensional pixel-optimized honeycomb structure

Highlights

• Designed and optimized 3D pixelated honeycomb absorber via genetic algorithm.

Introduction

As stealth technology has become a critical factor in determining the survivability of modern military and aerospace platforms, development of structures with limited electromagnetic detectability across various frequency ranges has become essential. Stealth fighter aircrafts, guided munitions, and high-speed unmanned aerial vehicles, which operate in multifrequency environments such as the S-, C-, X-, and Ku-bands, require low-visibility structures capable of responding effectively to threats. However, in conventional absorber designs, a fundamental trade-off exists among high absorption efficiency, broad bandwidth, and reduced thickness, resulting in structural limitations. ¹

Conventional electromagnetic wave absorbers are typically based on composite structures incorporating magnetic loss materials^2–4 or possessing multilayered Jaumann structures.^5–8 These approaches aim to achieve high absorption through the use of magnetic materials or to realize broadband absorption via multilayered designs; however, they often suffer from structural complexity and increased weight. Consequently, the demand for developing new material combinations and structural design strategies to achieve broadband impedance matching, without relying on magnetic materials, is surging. Resonator-based metamaterials,^9–15 pixelated metamaterial surfaces,^16–21 and multilayered dielectric-coupled structures have been explored to realize electromagnetic wave absorption without the use of magnetic components, leveraging design flexibility and frequency selectivity to increase absorption. Among these, honeycomb-based structures have attracted particular attention as next-generation absorbers, as they simultaneously satisfy lightweight and structural stiffness requirements while increasing electromagnetic absorption bandwidth through multiple reflections and path delay effects.^22–24 In addition to these structures, various sandwich-type radar absorbing structures have been investigated to combine electromagnetic absorption capability with structural load-bearing performance. For example, polyurethane-foam-based sandwich absorbers, incorporating conductive fillers, reportedly achieved broadband attenuation across microwave frequency ranges. ²⁵ Similarly, nanocomposite sandwich structures, combining CNT-filled composite face sheets with honeycomb cores, exhibited broadband microwave absorption while maintaining structural stiffness. ²⁶ More recently, multifunctional sandwich absorbers based on lightweight polymer foam cores have been proposed, exhibiting electromagnetic absorption capability along with mechanical load-bearing performance. ²⁷ Furthermore, structural design approaches considering electromagnetic compatibility, such as composite inserts for sandwich radome structures and additive-manufacturing-based multilayer radar absorbing structures, have been explored to improve structural integration and fabrication flexibility.^28,29 However, despite these advantages, they still lack sufficient design freedom to achieve broadband performance.

Based on these findings, in this study, we propose a lightweight honeycomb-based structure capable of achieving broadband electromagnetic absorption without magnetic fillers or multilayer stacking. The approach introduces a binary pixel grid onto the internal walls of the aramid honeycomb structure, where each pixel is either filled or left unfilled with a conductive ink and optimized using a genetic algorithm (GA). Unlike conventional planar pixel absorbers, the proposed method implements pixel optimization directly within the three-dimensional (3D) honeycomb architecture, enabling broadband absorption while maintaining a structurally lightweight composite design. This integration of pixel-based optimization with a honeycomb framework enhances impedance matching and frequency selectivity while remaining compatible with scalable roll-to-roll fabrication. Consequently, the proposed structure demonstrates high potential as a lightweight and structurally integrable electromagnetic absorber for aerospace composite applications.

Design and optimization

Fabrication and measurement

Several fabrication steps were required to produce an optimized pixelated honeycomb structure. The fabrication process included (1) printing conductive ink patterns, (2) sheet cutting and honeycomb stacking, (3) honeycomb expansion and epoxy resin dipping followed by curing, and (4) specimen machining. The epoxy resin was primarily used to maintain the extended honeycomb geometry and provide structural stability during fabrication. The conductive ink patterns were printed on a 0.05-mm aramid paper using a thermally curable carbon-based ink via a continuous roll-to-roll method. The printing equipment incorporated an inline drying unit, enabling immediate drying and precise control of the ink thickness. A critical design constraint was the minimum printable resolution of the conductive ink, which was less than approximately 0.4 mm in width; therefore, the pixel size was set to 0.5 mm × 0.5 mm. The specimens were machined to a 300 mm × 300 mm configuration suitable for the Naval Research Laboratory (NRL) arch method (Figure 5), ensuring sufficient area to avoid measurement interference and to maintain planar wave conditions during testing.OPEN IN VIEWER

The electromagnetic wave absorption characteristics were measured over a frequency range of 1–18 GHz using the NRL arch method. The S-parameters were obtained using horn antennas and a vector network analyzer. Measurements were performed in two frequency segments, 1–3 and 3–18 GHz, using horn antennas optimized for each band. The specimen was placed at the center of the arch path to minimize interference between the reflected and direct waves. To evaluate polarization characteristics, HH and VV conditions were measured separately by adjusting the specimen. Figure 6 shows a schematic of the NRL arch measurement setup.OPEN IN VIEWER

Results and discussion

Figure 7 compares the reflection loss results of the 20-mm and 40-mm specimens obtained using the NRL arch method. For the 20-mm specimen, simulations predicted broadband absorption with a reflection loss below −10 dB across 2.5–20 GHz. Experiments indicated absorption below −5 dB from approximately 1.5 GHz, with a −10 dB reflection loss bandwidth of approximately 14.3 GHz in the 1–18 GHz range. Although the measured resonance frequencies were consistent with the simulated ones, the overall reflection loss was reduced and effective bandwidth was slightly narrower.OPEN IN VIEWER

For the 40-mm specimen, fabricated by vertically repeating the same pixelated pattern, simulations predicted absorption below −10 dB over 3.5–20 GHz. Experiments confirmed absorption below −10 dB beginning near 4 GHz, and this absorption was consistently maintained up to the Ku-band. Although the measured performances of both the specimens were lower than those obtained through simulations, the experimental results verified that broadband electromagnetic wave absorption could be achieved with a lightweight, nonmagnetic, and single-layer honeycomb structure. The performance degradation is likely attributable to fabrication-related factors such as pixel misalignment and variations in the conductive ink thickness. To further investigate this degradation, the following section presents simulation-based analyses that estimate the effect of pattern misalignment and provide qualitative evidence for its influence on the absorption characteristics.

This performance degradation is primarily attributed to pattern-printing deviations and alignment errors that occur during fabrication. In particular, the positional misalignment of patterns on the cell walls, non-uniform thickness of the conductive ink coating, and fine gaps introduced during the printing process can lead to uneven surface resistance, which may reduce the absorption performance in specific frequency bands.

Figure 8 presents a simulation-based analysis of fabrication-induced pattern misalignment, verifying this assumption. The coordinates of the top and bottom pixel positions are defined as (xi, yi) and (xj, yj), respectively, with relative offsets along each axis introduced as variables.OPEN IN VIEWER

A total of 627 offset cases were generated by independently considering misalignments of the top and bottom pixels, and the corresponding reflection loss curves were plotted. This analysis enabled evaluation of the influence of the pixel offset on the reflection loss and estimation of the error range, thereby providing a reference for explaining the observed performance degradation. Furthermore, the simulation results revealed some variations across the frequency bands, depending on the type of solver used (T-solver vs F-solver) (Figure 9). For the F-solver, the results aligned more closely across the entire frequency range, indicating that frequency-domain simulations more accurately captured the electromagnetic response of the proposed structure.OPEN IN VIEWER

These results suggest that the proposed structure has high potential for practical applications in aerospace and defense systems, in which lightweight characteristics and broadband low observability are critical. In particular, the ability to directly integrate absorption functionality into complex structures—such as aircraft skins, engine inlets, and unmanned-aerial-vehicle platforms—highlights its potential as a structurally functional composite. However, the printing-based process is sensitive to pixel alignment and coating thickness variations, necessitating precise control techniques to ensure stable performance in large-area fabrication. Moreover, as this study primarily focuses on the electromagnetic performance, the mechanical strength and durability of the honeycomb composite remain to be investigated. Extending the developed single-layer design to multilayer structures or more complex geometries, coupled with process optimization, will further enhance its feasibility for integration into operational aerospace platforms.

Conclusion

In this study, we experimentally demonstrated broadband electromagnetic wave absorption using a lightweight, single-layer structure that was 3D pixelated with a conductive ink. A binary pixel grid was introduced onto the internal walls of the aramid honeycomb structure, where each pixel was either filled or left unfilled with the conductive ink and optimized using a GA. The results showed that increasing the thickness from 20 to 40 mm while retaining the same pixel-based design only increased the volume, with limited improvement in the reflection loss performance. This result indicates that simply increasing thickness does not significantly enhance broadband electromagnetic wave absorption. By contrast, the optimized 20-mm-thick structure achieved broadband absorption while maintaining a lightweight configuration, demonstrating that the proposed 3D pixel optimization strategy, rather than simply increasing thickness, is critical for achieving the desired broadband performance. Therefore, the proposed design provides a viable strategy for mitigating the inherent trade-off between thickness and bandwidth. Moreover, the absorber maintained its performance even when the optimized pixel pattern was applied. The density of pure aramid honeycomb was 47.8 kg/m³, while that of the fabricated 20-mm-thick specimen was 50.6 kg/m³, demonstrating that improved functionality can be achieved while maintaining an ultralightweight nature. This result indicates that broadband electromagnetic absorption can be realized with only a minimal increase in the weight relative to the pristine honeycomb structure.

These results validate the practical applicability of the proposed design to complex structural systems, including next-generation stealth aircraft inlets, weapon platforms, and unmanned aerial vehicles. In future studies, more precise control of absorption characteristics and extended performances across wider frequency ranges can be achieved by adjusting the pattern repetition period, refining the pixel resolution, and improving the fabrication stability. Furthermore, the proposed GA-based optimization framework can be extended to incorporate multiple incidence angles and polarization conditions, yielding absorber designs that more closely reflect real operational environments.