Corrosion of steel in concrete is a key durability issue for reinforced concrete structures, leading to significant economic costs and safety risks. Research suggests that moisture within voids at the steel-concrete interface (SCI) plays a critical role in corrosion. However, experimental evidence clarifying this role remains limited. Here, we systematically investigated the influence of macro-voids at the SCI by exposing reinforced concrete specimens to different wet-dry cycles and to continuous submersion in chloride solution, thereby generating distinct moisture conditions within the interfacial voids. X-ray computed tomography (XCT) was used to monitor the evolution of the gas-liquid configuration inside these voids along with steel corrosion over 9 months. Under wet-dry cycles, corrosion initiated at lower chloride levels than under continuous immersion. XCT revealed that interfacial voids became fully saturated after prolonged immersion but remained partially saturated during wet/dry exposure. These findings are interpreted through the lens of corrosion science and gas-liquid interactions in porous materials. We suggest a mechanism in which the dissolution of pressurized gas bubbles trapped within interfacial voids leads to an increase in dissolved oxygen concentration near the void, thereby locally elevating the steel corrosion potential (E_corr). Together with chloride accumulation, which gradually lowers the pitting potential, this shift in E_corr enhances the likelihood for local corrosion initiation under wet-dry cycles. Overall, this study contributes to understanding how different water exposure conditions affect the local environment at the SCI, providing insight into the mechanistic links between interfacial void gas-liquid content and corrosion in reinforced concrete.

Corrosion of steel in concrete is a key durability issue for reinforced concrete structures, leading to significant economic costs and safety risks. Research suggests that moisture within voids at the steel-concrete interface (SCI) plays a critical role in corrosion. However, experimental evidence clarifying this role remains limited. Here, we systematically investigated the influence of macro-voids at the SCI by exposing reinforced concrete specimens to different wet-dry cycles and to continuous submersion in chloride solution, thereby generating distinct moisture conditions within the interfacial voids. X-ray computed tomography (XCT) was used to monitor the evolution of the gas-liquid configuration inside these voids along with steel corrosion over 9 months. Under wet/dry cycles, corrosion initiated at lower chloride levels than under continuous immersion. XCT revealed that interfacial voids became fully saturated after prolonged immersion but remained partially saturated during wet/dry exposure. These findings are interpreted through the lens of corrosion science and gas-liquid interactions in porous materials. We suggest a mechanism in which the dissolution of pressurized gas bubbles trapped within interfacial voids leads to an increase in dissolved oxygen concentration near the void, thereby locally elevating the steel corrosion potential (E_corr). Together with chloride accumulation, which gradually lowers the pitting potential, this shift in E_corr enhances the likelihood for local corrosion initiation under wet-dry cycles. Overall, this study contributes to understanding how different water exposure conditions affect the local environment at the SCI, providing insight into the mechanistic links between interfacial void gas-liquid content and corrosion in reinforced concrete.

Corrosion of steel in concrete is a key durability issue for reinforced concrete structures, leading to significant economic costs and safety risks. Research suggests that moisture within voids at the steel-concrete interface (SCI) plays a critical role in corrosion. However, experimental evidence clarifying this role remains limited. Here, we systematically investigated the influence of macro-voids at the SCI by exposing reinforced concrete specimens to different wet-dry cycles and to continuous submersion in chloride solution, thereby generating distinct moisture conditions within the interfacial voids. X-ray computed tomography (XCT) was used to monitor the evolution of the gas-liquid configuration inside these voids along with steel corrosion over 9 months. Under wet-dry cycles, corrosion initiated at lower chloride levels than under continuous immersion. XCT revealed that interfacial voids became fully saturated after prolonged immersion but remained partially saturated during wet/dry exposure. These findings are interpreted through the lens of corrosion science and gas-liquid interactions in porous materials. We suggest a mechanism in which the dissolution of pressurized gas bubbles trapped within interfacial voids leads to an increase in dissolved oxygen concentration near the void, thereby locally elevating the steel corrosion potential (E_corr). Together with chloride accumulation, which gradually lowers the pitting potential, this shift in E_corr enhances the likelihood for local corrosion initiation under wet-dry cycles. Overall, this study contributes to understanding how different water exposure conditions affect the local environment at the SCI, providing insight into the mechanistic links between interfacial void gas-liquid content and corrosion in reinforced concrete.

Corrosion of steel in concrete is a key durability issue for reinforced concrete structures, leading to significant economic costs and safety risks. Research suggests that moisture within voids at the steel-concrete interface (SCI) plays a critical role in corrosion. However, experimental evidence clarifying this role remains limited. Here, we systematically investigated the influence of macro-voids at the SCI by exposing reinforced concrete specimens to different wet-dry cycles and to continuous submersion in chloride solution, thereby generating distinct moisture conditions within the interfacial voids. X-ray computed tomography (XCT) was used to monitor the evolution of the gas-liquid configuration inside these voids along with steel corrosion over 9 months. Under wet/dry cycles, corrosion initiated at lower chloride levels than under continuous immersion. XCT revealed that interfacial voids became fully saturated after prolonged immersion but remained partially saturated during wet/dry exposure. These findings are interpreted through the lens of corrosion science and gas-liquid interactions in porous materials. We suggest a mechanism in which the dissolution of pressurized gas bubbles trapped within interfacial voids leads to an increase in dissolved oxygen concentration near the void, thereby locally elevating the steel corrosion potential (E_corr). Together with chloride accumulation, which gradually lowers the pitting potential, this shift in E_corr enhances the likelihood for local corrosion initiation under wet-dry cycles. Overall, this study contributes to understanding how different water exposure conditions affect the local environment at the SCI, providing insight into the mechanistic links between interfacial void gas-liquid content and corrosion in reinforced concrete.

This dataset contains reconstructed neutron computed tomography (CT) volumes of a reinforced concrete sample, acquired at 7 systematically varied exposure times: 0.5s, 1s, 2s, 5s, 10s, 20s, and 50s per projection. The reconstructions were performed using the filtered backprojection algorithm implemented in MuhRec, an open-source software for neutron CT processing.Each reconstruction is provided as a ZIP archive (e.g., 01_ct05s.zip, 07_ct50s.zip) corresponding to a specific exposure time. When unzipped, each folder contains:A series of reconstructed 2D axial slices in 16-bit TIFF format (e.g., ct50s0404.tif, where the last 4-digit number indicates the vertical slice index),A configuration file used for the reconstruction in MuhRec (for reproducibility),A citation file acknowledging MuhRec.Physical meaning: The reconstructed voxel values are scaled according to the known pixel size and represent the linear attenuation coefficient in cm⁻¹, reflecting neutron attenuation (primarily due to the total macroscopic neutron cross section). This enables quantitative analysis of internal features, such as water content and corrosion products.Note: Only slices containing the sample volume are included to reduce data size. Slice indices do not start at 0001.This dataset supports research in neutron CT image quality, noise reduction, reconstruction algorithms, and low-dose imaging optimization. It complements a separate dataset of raw projections (see related dataset DOI).

This dataset contains reconstructed neutron computed tomography (CT) volumes of a reinforced concrete sample, acquired at 7 systematically varied exposure times: 0.5s, 1s, 2s, 5s, 10s, 20s, and 50s per projection. The reconstructions were performed using the filtered backprojection algorithm implemented in MuhRec, an open-source software for neutron CT processing.Each reconstruction is provided as a ZIP archive (e.g., 01_ct05s.zip, 07_ct50s.zip) corresponding to a specific exposure time. When unzipped, each folder contains:A series of reconstructed 2D axial slices in 16-bit TIFF format (e.g., ct50s0404.tif, where the last 4-digit number indicates the vertical slice index),A configuration file used for the reconstruction in MuhRec (for reproducibility),A citation file acknowledging MuhRec.Physical meaning: The reconstructed voxel values are scaled according to the known pixel size and represent the linear attenuation coefficient in cm⁻¹, reflecting neutron attenuation (primarily due to the total macroscopic neutron cross section). This enables quantitative analysis of internal features, such as water content and corrosion products.Note: Only slices containing the sample volume are included to reduce data size. Slice indices do not start at 0001.This dataset supports research in neutron CT image quality, noise reduction, reconstruction algorithms, and low-dose imaging optimization. It complements a separate dataset of raw projections (see related dataset DOI).

Raw projection data in fits format ,containing 7 acquisitions, exposure times at 0.5s, 1s, 5s, 10s, 20s, and 50s.For each acquisition, there are 5 open beam images named as ob_0000x.fits, 5 dark current images named as dc_0000x.fits, 6 open beam images with blackbody grid (bbob_0000x.fits), and 26 sample images with blackbody grid ('bbs_0000x.fits) for bias and scattering correction, as well as 1125 raw projections (ct_0000x.fits`) over a full rotation of 360 degrees evenly. recording setup: Andor iKON-CCD camera coupled with a Zeiss 100 mm f/2.0 macro lens and a 30 μm thick gadolinium oxysulfide (Gadox) scintillator

Raw projection data in fits format ,containing 7 acquisitions, exposure times at 0.5s, 1s, 5s, 10s, 20s, and 50s.For each acquisition, there are 5 open beam images named as ob_0000x.fits, 5 dark current images named as dc_0000x.fits, 6 open beam images with blackbody grid (bbob_0000x.fits), and 26 sample images with blackbody grid ('bbs_0000x.fits) for bias and scattering correction, as well as 1125 raw projections (ct_0000x.fits`) over a full rotation of 360 degrees evenly. recording setup: Andor iKON-CCD camera coupled with a Zeiss 100 mm f/2.0 macro lens and a 30 μm thick gadolinium oxysulfide (Gadox) scintillator