Conference Agenda

Accurate insect classification is essential for applications in agriculture, biodiversity monitoring, and pest control [1,2]. This work investigates the use of terahertz (THz) circular synthetic aperture radar (CSAR) for non-invasive insect identification by analyzing radar cross-section (RCS) data. THz radar systems offer significant advantages over traditional optical and microwave techniques, particularly in resolving fine structural features of small insects under varying environmental conditions. We demonstrate the feasibility of using a THz time-domain spectroscopy (THz-TDS) system for capturing detailed RCS profiles of insects, exemplified by high-resolution measurements of a 3D-printed honeybee model. A spherical inverse SAR configuration allows for diverse angular perspectives, and back-projection algorithms reconstruct the spatial features of the insect. However, the time-intensive nature of these measurements (~30 hours per specimen) presents a bottleneck. To address this, we introduce a radar digital twin simulation framework that uses full-wave electromagnetic modeling to generate synthetic RCS datasets. These simulations emulate a stepped-frequency THz radar (100– 1100 GHz) scanning an insect target in 360°, enabling fast, cost-effective training data generation for machine learning algorithms. Figure 1 illustrates the impact of bandwidth on image resolution for two insects: (a) Aedes aegypti mosquito and (b) honeybee. While a bandwidth of 200 GHz is sufficient to resolve most structural details of the larger honeybee, capturing the finer features of the smaller mosquito requires the full 1000 GHz bandwidth. Our findings highlight the benefits of THz radar's broad bandwidth in resolving insect anatomy and showcase the potential of digital twins for scalable, efficient classification systems. This approach paves the way for future integration with artificial neural networks and realtime insect monitoring technologies.

Cancer is a leading life-threatening disease worldwide with an increasing mortality rate. Especially malignant melanoma, the most aggressive skin cancer, poses a high risk due to its high mutation rate. Despite recent therapeutic advances, the survival rate of cancer patients remains suboptimal due to delays in diagnosis and the poor prognosis associated with the disease. Cancer is a multistage disease, and the onset and progression of the disease are associated with a complex array of genetic or epigenetic alterations resulting in tumorigenic transformation and malignancy. Hence, an early diagnosis and a personalized therapy are essential to improve the patient’s survival rate. Conventional methods of cancer detection, including ultrasound, magnetic resonance imaging, serum analysis and biopsy, are rather inefficient for early-stage cancer detection or early detection in tumor follow up.

Therefore, terahertz (THz) biosensorics becomes more and more attractive for the detection of malignant melanoma. With its benefits of delivering non-invasive and fast results with a small sample volume, it could be a good approach for early cancer detection. A big challenge for its clinical relevance is the specificity and sensitivity of the proposed sensor principle, which is generally based on so-called frequency-selective surfaces designed for the THz range.

In previous studies, we focussed on the detection of melanoma mRNAs (MIA) and revealed a high sensitivity based on the label-free detection of DNA samples with concentrations as low as 1.55 x 10^-12 mol/l [1]. Newly, we wanted to develop a THz-based sensing-principle to also determine melanoma-associated proteins. We chose the early growth response protein 2 (EGR2), a transcription factor with an increased activity in melanoma cells, as the biomarker of choice and established a specific measurement method from a complex sample of cellular proteins [2]. Highly specific detection of EGR2 was achieved by coupling the double-stranded DNA binding site to the THz-biosensor and measuring EGR2 binding. Guided by the measurement of positive and negative controls, we revealed sensitive and specific measurement of a protein, which is a new and important step in the usage of THz diagnostics. Encouraged by this step forward, we actually try to establish a measurement method to determine tumor-derived vesicles, because in melanoma, melanoma-derived extracellular vesicles play an important role in tumour progression. In a work in progress study, we were able to detect differences in the concentration of vesicles. To ensure specificity we are now developing defined binding sites on the THz biosensor to clearly distinguish between different vesicular structures. This will be an important step in the development of THz-biosensor measurements in clinically relevant samples, not only for malignant melanoma but also for other diseases.

Reference:

[1] C. Weisenstein et al. Biomed Opt Express, 11(1):448-460 (2019)

[2] M. Richter et al. Scientific reports, 13:20708 (2023).

Terahertz (THz) metamaterial sensors have gained increasing significance for the detection and analysis of biomolecules in the last decades [1]. Following the successful demonstration that sensitive and selective biosensors can facilitate meaningful biological experiments [2], current research efforts are increasingly directed toward optimizing and enhancing the sensor’s sensitivity. A promising approach was proposed in [3] where a strategy for optimizing the design of a metamaterial sensor used for the detection of small amounts of dielectric materials is presented. Important design parameters such as the overlap volume between the analyte and the high-electric-field region of the sensor have been established. This overlap closely correlates with the actual sensitivity of a sensor and is therefore crucial for the detection of biological samples. Based on this design optimization, a new metamaterial sensor design, a so-called interdigitated electrical split ring resonator (ID‑eSRR), which exhibits a meandering gap structure with 600 nm small structures, was developed and manufactured [3]. Following a successful proof-of-principle experiment that demonstrated increased sensitivity through the detection of a SiO₂ thin film [3], we performed an initial experiment with biological samples. This first experiment was conducted with the aim of detecting a protein with an increased activity in melanoma cells. Therefore, we have chosen the transcription factor early growth response protein 2 (EGR2). The required bio-functionalization process involved attaching a specific double stranded DNA sequence to the gold surface of the sensor via a thiol anchor, prior to the application of a protein mixture, to ensure the selective binding of the target protein to the corresponding DNA recognition sequence. A more detailed description of this process can be found in [2]. Our measurements demonstrated that the functionalization stage with only DNA bind to the biosensors surface (prior to the application of the protein mixture) led to the proposed enhancement in sensitivity, increasing it by a factor of three compared to the metamaterial structures used in [2]. However, the measurements using DNA also showed that the sensor is not only sensitive to the selected biomolecules, but also reacts strongly to environmental and minor functionalization variations. In order to assess the sensor’s behavior towards the target biomarker EGR2, the ID-eSRR was exposed to a mixture of melanoma nucleus proteins (including EGR2). At this stage, no conclusive measurements with the proteins bound to the DNA could be shown and it was observed that more non-selective binding of the proteins occurred over the entire sensor structure rather than binding to the desired DNA recognition sequence. Since the theory, the SiO₂ thin film experiment, and the experiment in which only DNA is bound to the biosensor all yield the expected results, we conclude that a significantly more stable functionalization is necessary for more complex biological experiments, such as the detection of EGR2. In the present case, the nanometer-scale structures that significantly enhance sensitivity become problematic if our established functionalization process cannot reliably access these highly-sensitive nanoscale regions. This indicated that the ability to functionalize the biosensor should be considered an important design parameter as well, to develop sensors that not only exhibit enhanced sensitivity but are also suitable for meaningful biological experiments in the future.

Acknowledgements:

The authors thank Maira Pérez Sosa, Dr. Alaa Jabbar Jumaah and Prof. Dr. Shihab Al-Daffaie from the Department of Electrical Engineering at Eindhoven University of Technology for the fabrication of the utilized interdigitated metamaterial sensor structures.

Reference:

[1] Zhang et al. Nanotechnol. Rev. 13, 20230182 (2024).

[2] Richter et al. Sci. Rep. 13, 20708 (2023).

[3] Cao et al. Photonics Res. 12, 1115–1128 (2024).

Plasmonic exceptional points for nanoscale sensing

Elham Talvari¹, Lei Cao², Fanqi Meng¹, and Hartmut G. Roskos¹

¹Physikalisches Institute, Goethe-Universität Frankfurt, 60438 Frankfurt am Main, Germany

²State Key Laboratory of Advanced Electromagnetic Technology, Huazhong University of Science and Technology, Wuhan 430074, China

Corresponding author: talvari@physik.uni-frankfurt.de

Open systems inherently imply interaction with the surrounding space via the exchange of energy. They can be found in both classical and quantum systems, involving acoustical or optical waves . Open systems can be described with non-Hermitian Hamiltonians. A remarkable consequence of non-Hermiticity is the emergence of “exceptional points” (EPs). EPs are spectral singularities at which at least two eigenvectors and their corresponding eigenvalues coalesce as parameters such as coupling strength, loss, or gain are varied which is shown in Figure 1a. In close proximity to an EP, the eigenvalues exhibit a strongly enhanced sensitivity to perturbations, making them prospective for sensing applications[1].

The simplest example of non-Hermitian optics is two coupled resonators with gain and/or loss[2]. In this regard, the coupling of metamaterial (MM) and photonic cavities in the terahertz frequency range provides an appropriate platform for us to investigate the EP [3]. We designed a practical coupled-MM-cavity system (Figure 1b). The cavity is a Fabry-Perot-type resonator composed of a metallic Babinet-comple­men­tary MM (CMM) reflector and a metal grating as the second reflector, both separated by a high-resistivity Si spacer. The coupling strength between the CMM and the cavity can be changed by rotating one of the metallic structures around the beam axis. At a specific rotation angle, the existence of the EP was verified through the measurements of the transmission spectra both along the polarization of the cavity mode and perpendicular to it and it is clear at Figure 1c.

Figure 1: (a) Real parts (solid lines) and imaginary parts (dash-dotted lines) of the eigenvalues of a two-coupled-mode system as a function of the coupling strength. Red and blue colors represent two respective eigenmodes. (b) C-shaped CMM coupled with a Fabry-Perot-type cavity formed by the CMM and a grating, separated by a Si spacer. (c) Measured transmission spectra of the coupled system as a function of the rotation angle of the C-shaped CMM.

Reference:

[1] J. Park, et al. Nat. Phys. vol. 16, 462-468, 2020.

[2] El-Ganainy, et al. Nat. Phys. vol. 14, 11-19, 2018.

[3] F. Meng, et al. Nanophotonics. vol. 13, 2443-2451, 2024.

Infectious diseases caused by bacteria and viruses continue to present a significant public health challenge in the 21st century. Rapid, sensitive, point‑of‑care (POC) diagnostics are essential for pathogen detection and outbreak prevention. Biosensors based on terahertz (THz) technology, which offer the advantages of label-free detection, short wavelengths, and low photon energy, make them a compelling solution to this challenge. However, the majority of research has focused on THz metasurface biosensors, whose low quality factors (Q‑factors) and weak field confinement limiting their sensing performance [1]. In contrast, the present study demonstrates that THz photonic crystal (PhC) resonators exhibit a high Q-factor and high field concentration, resulting in a higher figure-of-merit (FOM), that is a metric evaluating the overall sensing capacity, despite operating at a lower frequency [2].

The PhCs are composed of alternating dielectric materials and can be fabricated on silicon wafers using silicon processes by 3D printed alumina. The high degree of design flexibility exhibited by these components allows for the optimization of Q-factors, the enhancement of field concentration, the realization of multiple channels, and wireless reading capacity. In our previous paper, the PhC slot resonator demonstrated a FOM of 5.1 /RIU/µm, which exceeds the FOM of conventional THz metasurface biosensors [1]. Furthermore, as shown in Fig. 1(a), a dual-channel structure was designed and fabricated to compensate for environmental effects with a single measurement, thereby enhancing the detection robustness [2]. Moreover, the PhC resonator was combined with an antenna for remote biosensing applications, as shown in Fig. 1(b). This configuration has the potential to detect airborne pathogens in the ambient air before people get infected. The remote biosensors have a reading range of up to 90 cm and an acceptance angle of 90° [3].

In summary, due to the advantages of high FOM, compactness, and compatibility with silicon processes, THz PhCs show considerable potential for biosensing applications. In future works, the sensing capacity of the PhC resonators with higher resonant frequency will be further studied.