Expertise

Terahertz research sits between photonics and high-frequency electronics. That position is scientifically productive, but it also makes system design unusually interdependent: the useful information does not come from a source, detector or algorithm in isolation. It emerges from a measurement chain whose bandwidth, geometry, calibration and reconstruction method have been chosen for a specific interaction with the sample.

The expertise developed at IMS follows this complete chain. Experimental work on the ATLAS platform is combined with electromagnetic modelling, component development and computational methods including tomography, holography and synthetic-aperture reconstruction. Depending on the question, the work may begin with a fundamental propagation or coupling problem, or with an application constraint such as contactless thickness measurement, subsurface inspection, real-time imaging or access to a confined region.

This breadth should not be read as a claim that one terahertz instrument fits every problem. Each study begins by identifying the expected contrast, the relevant length scale and the reference against which the result can be tested. The objective is to establish what can be measured reliably, under which conditions, and with which remaining uncertainties.

Jean-Paul Guillet is an associate professor at the University of Bordeaux and a researcher at the IMS Laboratory. His work connects terahertz instrumentation, electromagnetic components, computational imaging, material characterization, and application-oriented studies.

The research approach is deliberately system-oriented. It connects electromagnetic design, sources and detectors, experimental acquisition, calibration, signal processing, reconstruction, and validation on representative samples.

A terahertz study usually begins before the first scan. The expected contrast must be related to a physical property of the sample, and the spatial scale of that contrast must be compared with wavelength, penetration, bandwidth, numerical aperture, and available dynamic range. This preliminary analysis determines whether the relevant experiment should use pulsed spectroscopy, coherent radar, full-field imaging, near-field probing, guided propagation, or a combination of modalities.

Capabilities

Experimental and computational continuity

The source dossier describes work using the IMS ATLAS platform, electromagnetic simulation, and processing methods including tomography, holography, synthetic-aperture approaches, and phase retrieval.

Instrumentation and computation are treated as one measurement problem. Calibration, positioning, environmental stability, reference samples, reconstruction assumptions, and uncertainty all influence what can be concluded from a dataset. This is especially important when an image is intended to support a material, industrial, heritage, or biomedical interpretation rather than serve only as a visual demonstration.

The portfolio spans fundamental propagation and component studies, laboratory prototypes, advanced imaging methods, and application-oriented experiments. The source dossier records work across technology-readiness levels from early concepts to representative demonstrations, but the maturity of each result must be assessed individually. A successful experiment on a controlled sample is not automatically a validated operational system.

Discuss a research question

Collaboration can begin with a sample, a measurement challenge, a component need, or a reconstruction problem.

How an expertise study is framed

A first technical assessment defines the target information, the expected physical contrast, the representative samples, and the reference method. It then identifies the frequency range, geometry, dynamic range, spatial sampling, and reconstruction strategy that could answer the question. When the available evidence suggests that terahertz methods are unsuitable, that conclusion is part of a responsible feasibility process.