Feasibility of Using a 300 GHz Radar to Detect Fractures and Lithological Changes in Rocks

Research publication · Sub-terahertz radar for rock inspection

Feasibility of Using a 300 GHz Radar to Detect Fractures and Lithological Changes in Rocks

Federico Sanjuan, Frederic Fauquet, Bertrand Fasentieux, Patrick Mounaix and Jean-Paul Guillet · Remote Sensing · 2023 · Volume 15, issue 10, article 2605 · DOI: 10.3390/rs15102605

Rock fractures influence structural integrity, permeability and weathering, but no single inspection method offers unlimited depth, resolution and survey area. Conventional ground-penetrating radar reaches useful depths at comparatively low frequencies, while X-ray microtomography provides fine detail within specimens that fit its geometry. This paper examines a different operating point: a frequency-modulated continuous-wave radar near 300 GHz, where the wavelength is short enough to resolve millimetre-scale interfaces but attenuation restricts penetration. Through controlled experiments on limestone, dolomite and granite blocks, the authors evaluate whether such a radar can locate artificial fractures, distinguish lithological contacts and reconstruct a simple three-dimensional gap.

The paper is explicitly a feasibility study on prepared samples. The fractures are air gaps formed by placing machined rock blocks together, and their surfaces are smoother than many natural fractures. Within those conditions, the radar detects interfaces through approximately one centimetre of rock and responds to gaps down to about 500 micrometres. Those findings establish measurable high-frequency radar contrast in several rocks. They do not yet show equivalent performance in a heterogeneous outcrop, a reinforced structure or a fluid-filled natural fracture, where roughness, water, mineral infill and access geometry may change both propagation and reflection.

Featured visual: Image 1 from the Airtable record associated with this publication. Consult the original paper for the authoritative figure caption and interpretation. Source publication.

Visuals are drawn from the Airtable research archive. Figure numbering, rights and interpretation should be checked against the original publication before republication outside this site.

Raster imaging with a wideband FMCW radar

The Synview radar operates with a centre frequency around 275 GHz and a 90 GHz sweep bandwidth. In an FMCW system, the delay of a reflected chirp appears as a beat frequency, allowing echoes from different depths to be separated. With the wave travelling in air, the stated bandwidth corresponds to an axial resolution of 1.7 mm. A 100 mm focal-length Teflon lens gives a measured lateral resolution of about 2.5 mm and a Rayleigh length of 4.9 mm, a compromise between focusing and tolerance to depth variation. The transceiver is moved on an XY stage whose positioning accuracy is 0.5 mm.

Before imaging, reflections from a metal plate and from an open path define the high and low amplitude references. The samples are then raster-scanned over an 11 by 11 cm area in 1 mm steps. Acquiring that 121 cm2 surface, with a 3 cm depth window, takes about 15 minutes. Each rock type is represented by two blocks of controlled thickness: limestone blocks of approximately 1 and 2 cm, granite blocks of 0.9 and 1.2 cm, and dolomite blocks of 0.5 and 0.8 cm. Positioning two blocks with an adjustable gap produces a known fracture geometry against which the radar image can be compared.

The front and rear faces of the first block, the air gap and the front face of the second block produce separable echoes when contrast and spacing are sufficient. Propagation through rock shortens the wavelength according to its refractive index. Values derived from measured optical thickness are approximately 2.8 for limestone, 2.66 for dolomite and 2.45 for granite, in agreement with the order expected from published data. This shorter wavelength can improve effective spatial resolution inside the material, but it comes with greater attenuation. The experiment therefore illustrates the basic trade-off of moving radar toward sub-terahertz frequencies: finer localization over a smaller accessible depth.

Fractures, rock interfaces and a wedge-shaped gap

The radar identifies millimetre-scale artificial fractures behind roughly 1 cm of rock and remains sensitive to a 500 micrometre separation under the reported conditions. Interfaces between blocks also change the reflected amplitude. Contacts involving limestone and dolomite are visible, including some joints made from two pieces of the same lithology. Granite interfaces are less readily distinguished in the tested arrangement, showing that detectability is not governed by geometric depth alone. Dielectric contrast, absorption, porosity, moisture and internal texture all influence whether an interface yields a resolvable echo.

For a three-dimensional test, the researchers create a wedge-shaped air gap between limestone blocks and stack successive radar slices in FIJI. The reconstructed volume follows the inclination of the second surface. A thickness change of 1.43 mm is obtained from the image, compared with 1.8 mm in the physical setup. The reconstructed slope is 6.5 degrees, while the independent experimental estimate is 9 plus or minus 2 degrees. The slope becomes clear when the fracture approaches 1.4 mm, close to the system’s 1.7 mm axial resolution, although evidence of a gap is visible at around 500 micrometres. These comparisons are useful precisely because they expose the difference between detecting a contrast and measuring its geometry accurately.

Natural rock presents a harder problem. A fracture may contain water, clay, cement or mineral growth instead of air, and each filling changes the dielectric boundary. Surface roughness can scatter the focused beam; layering and anisotropy can create signals that resemble discontinuities; and moisture can sharply reduce penetration at these frequencies. The limited set of lithologies in this study cannot represent that full variability. Heating or drying may increase penetration in a laboratory sample, but such preparation is not necessarily available in field or heritage inspection.

Where high-frequency radar could add value

The results position 300 GHz FMCW radarFMCW radar transmits a continuously swept frequency and measures the beat signal produced by delayed reflections, enabling distance, thickness, and depth-resolved imaging with compact coherent hardware. More between deep, lower-resolution geophysical sensing and fine laboratory tomography. It offers contactless access to a sizeable surface, millimetre-scale depth discrimination and a straightforward path to volumetric display. Possible uses include examining thin stone elements, prepared cores, construction specimens or cultural-heritage objects whose materials and thicknesses fall inside the radar’s penetration range. The evidence supports further application-specific trials; it does not establish a general method for mapping deep fracture networks or certifying structural safety.

Productive collaboration would combine radar engineers with geologists, civil engineers or conservators who can define realistic defect populations and reference measurements. A larger validation programme should vary lithology, water content, roughness, fracture filling and orientation, then quantify false positives and repeatability rather than relying only on selected images. Faster mechanics or array-based acquisition would also matter when moving beyond laboratory raster scans. By reporting both successful detections and dimensional discrepancies, this study provides a grounded starting point for that work and clarifies the physical window in which sub-terahertz radar may complement established rock-characterization tools.

This publication also connects with our broader work on geosciences and civil engineering and FMCW radar expertise.

Bibliographic reference