Airbone Greenhouse Gas Detection Metalens (AGDM) Camera

The aim of the project was to explore the use of optical metasurfaces for the detection of greenhouse gases. The project was initiated during the postdoctoral appointment of Prof. Meretska in the group of Federico Capasso, where she conceived the core idea. At this stage, the metasurface was designed, nanofabricated, and optically characterized. Within the framework of the YIN grant, the project was extended to include a detailed characterization of the optical performance in the laboratory of Prof. Butz, who specializes in environmental physics. During the course of the project, Prof. Meretska also established an additional collaboration with Dr. V. Sokurenko, who contributed expertise in optical design and ray-tracing modeling.

The results of this investigation are currently compiled in a manuscript planned for publication. The YIN grant not only enabled the establishment of new collaborations but also helped to maintain and strengthen existing ones. The outcomes of this project further served as the basis for an application to the Inno Space Master Challenge, which awarded Prof. Meretska €500k in funding to further explore metasurface-based solutions for space applications. This grant will support the employment of a postdoctoral researcher.

The project’s public-relations activities attracted a highly qualified recent master’s graduate specializing in space optics, who subsequently applied for a Max Planck School of Photonics PhD scholarship to join the Meretska group. The results of this application are expected to be announced in the summer. In addition, a further DFG proposal is planned for submission in 2026 to secure additional funding for a PhD position.

In this report, I present a summary of my work on the project titled “Next Generation Spectroscopic tools of Greenhouse gas Detection,” a collaborative research project between the Karlsruhe Institute of Technology (KIT) and Heidelberg University. Greenhouse gas sensing plays a critical role in understanding and addressing climate change. Accurate monitoring of these gases is essential for tracking atmospheric variations and supporting mitigation strategies. For this purpose, imaging spectrometers are commonly deployed on satellites to observe greenhouse gases across multiple wavelengths. These remote sensing instruments can detect gases such as CO₂, CH₄, and CFCs, each of which has it’s own characteristic absorption lines in the near-infrared (NIR) to mid-wave infrared (MWIR) spectral regions [1]. For example- CO₂, absorption bands at 0.76 μm, 1.61 μm, and 2.01 μm are frequently utilized.

Conventional imaging spectrometers typically consist of a light-collection system, dichroic mirrors for spectral separation, and multiple detection optics. These systems are designed based on different traditional optical components [2,3,4,5,6,7,8,9], which makes them relatively bulky and heavy. To address this limitation, an alternative solution is proposed: a flat-lens-based optical system capable of performing multiple functions within a single compact element. This approach aims to replace traditional spectrometers with a more portable and lightweight solution for gas concentration measurements.

In this project, a metalens was designed using an inverse design algorithm and fabricated in a nanofabrication laboratory. The device was then characterized under various realistic environmental conditions to evaluate its capability for CO₂ detection. The metalens was engineered with three distinct focal lengths corresponding to the three CO₂ absorption wavelengths in the NIR and MWIR regions (0.76 μm, 1.61 μm, and 2.01 μm). Multiple characterization experiments were conducted to evaluate the efficiency and performance of the metalens. Among these, I was directly involved in an experiment using a CO₂ gas tube, where the primary objective was to measure the concentration of carbon dioxide. 

My experimental setup [Fig 1A] consists of a halogen lamp used as the light source, positioned at the entrance of the gas tube. A narrowband optical filter with a bandwidth of 12 nm is placed at the exit of the tube. The transmitted light is then focused by a metalens with a focal length of 5 cm, arranged after the filter. Finally, an infrared camera is set up behind the metalens to record the resulting image.

Figure 1: Experimental arrangement of spectroscopic CO2 detection

In the experiment, measurements were carried out in two stages by calculating the pixel difference between captured images. In the first stage, images were recorded with the gas tube filled separately with air and CO₂, without using any optical filter, and the corresponding pixel differences were computed. In the second stage, the same procedure was repeated using a 12 nm bandpass filter centered at a wavelength of 1.6 μm [Fig 1C], where CO₂ reflects its strongest absorption resonance. The results indicate that when measurements are performed around the resonance peak using the optical filter, the pixel difference starts from 3.5% whereas without filter it starts from around 1% [Fig 1B]. This demonstrates that by changing the spectral window around the absorption resonance enhances the sensitivity of the metalens to changes in gas concentration.

To validate the experimentally observed 3.5% change, a theoretical model was needy for a base model. Hence, A theoretical model was constructed using the HITRAN database and its API, replicating the experimental conditions with a tube length of 1 m. Bandpass filters of different bandwidths were incorporated into the model to investigate how the detectable concentration difference varies with spectral window size. The simulation results indicate that, when a 12 nm filter is applied, a concentration change of approximately 1.7% is observed.

[1] R. E. Dickinson and R. J. Cicerone, “Future global warming from atmospheric trace gases,” Nature 1986 319:6049, vol. 319, no. 6049, pp. 109–115, 1986, doi: 10.1038/319109a0.
[2] S. M. Miller, A. M. Michalak, V. Yadav, and J. M. Tadié, “Characterizing biospheric carbon balance using CO2 observations from the OCO-2 satellite,” Atmos Chem Phys, vol. 18, no. 9, pp. 6785–6799, May 2018, doi: 10.5194/ACP-18-6785-2018.
[3] T. E. Taylor et al., “Evaluating the consistency between OCO-2 and OCO-3 XCO2 estimates derived from the NASA ACOS version 10 retrieval algorithm,” Atmos Meas Tech, vol. 16, no. 12, pp. 3173–3209, Jun. 2023, doi: 10.5194/AMT-16-3173-2023.
[4] T. Chiba et al., “Measuring regional atmospheric CO2 concentrations in the lower troposphere with a non-dispersive infrared analyzer mounted on a UAV, Ogata Village, Akita, Japan,” Atmosphere (Basel), vol. 10, no. 9, Sep. 2019, doi: 10.3390/ATMOS10090487.
[5] M. Ouchi et al., “Development of a balloon-borne instrument for CO2 vertical profile observations in the troposphere,” Atmos Meas Tech, vol. 12, no. 10, pp. 5639–5653, Oct. 2019, doi: 10.5194/AMT-12-5639-2019.
[6] R. M. Duren et al., “California’s methane super-emitters,” Nature 2019 575:7781, vol. 575, no. 7781, pp. 180–184, Nov. 2019, doi: 10.1038/s41586-019-1720-3.
[7] D. H. Cusworth et al., “Intermittency of Large Methane Emitters in the Permian Basin,” Environ Sci Technol Lett, vol. 8, no. 7, pp. 567–573, Jul. 2021, doi: 10.1021/ACS.ESTLETT.1C00173.
[8] F. Chevallier, L. Feng, H. Bösch, P. I. Palmer, and P. J. Rayner, “On the impact of transport model errors for the estimation of CO2 surface fluxes from GOSAT observations,” Geophys Res Lett, vol. 37, no. 21, Nov. 2010, doi: 10.1029/2010GL044652.
[9] J. Yang et al., “Erratum: The role of satellite remote sensing in climate change studies (Nature Climate Change (2013) 3 (875-883)),” Nat Clim Chang, vol. 3, no. 11, p. 1001, Nov. 2013, doi: 10.1038/NCLIMATE2033.