Decontamination methods that effectively remove the radioactive surface contamination is needed for concrete facilities at nuclear decommissioning sites to reduce the amount of concrete waste requiring regulated disposal. Ultimately, a remote decontamination tool that can safely and effectively peel off the highly contaminated surface layer from bulk concrete is required to guarantee work safety and reduce significant disposal costs . Scabbling technology generally refers to the removal of a layer from the concrete surface . Among surface-removal technologies, laser scabbling technique coupled with a high-power fiber laser may be a convenient approach for nuclear site decommissioning in that it permits a remote controllability and emits only a small amount of secondary waste . In comparison with other mechanical tools, this technique allows easier remote operation because there is no reaction force [1, 4].
The application of laser scabbling technology is possible because concrete is a porous material. In addition, concrete inherently contains a considerable amount of moisture . Concrete absorbing the intense laser energy undergoes a rapid increase in surface temperature and a sharp increase of vapor pressure in the pores . This vapor pressure beyond the threshold gives rise to an explosive release, resulting in spalling. The possibility of explosive spalling is dependent on the volume of capillary pores. A small volume of capillary pores enhances the impermeability, contributing to the formation of vapor pressure beyond the threshold level that pore structure can withstand . Concrete with a low porosity can be achieved by a small amount of water mixed with cement in addition to higher degree of hydration reaction [5, 8].
In this study, concrete blocks were manufactured with a designed normal compressive strength of 24 MPa . The scabbling experiment on the concrete blocks was performed using a high-power fiber laser under both the stationary and moving conditions. Simultaneously, the temperature distribution on the spalled or glazed surface of concrete was measured using a thermal imaging camera.
2.1 Laser Scabbling System
A schematic illustration of the laser scabbling apparatus for a concrete block is shown in Fig. 1(a). A 10-kW fiber laser (IPG Photonics, YLS-10000, λ=1,070 nm) was incorporated into the laser scabbling system. However, the actual output power was limited to 8.0-kW because of aging. A high-power laser beam was fed via process fiber (core diameter: 150 μm, length: 20 m) into the scabbling head. The optical head contained a collimator (ƒ = 160 mm), an aspherical lens (ƒ = 160 mm), and a protection window. Its beam parameter product (spot size) was measured to be 7.05 mm∙mrad (200 μm). The focused beam passed through the nozzle tip, which had a diameter of 3.0 mm. Compressed air was simultaneously discharged from the nozzle, while the beam was irradiated on the concrete block, to prevent debris from scattering toward the optical head. The mass flow rate was approximately 500 L·min−1. The stand-off distance was set to 510 mm for all trials. The stand-off distance was defined as the distance between the nozzle tip and the front surface of the concrete block. A laser beam size with a diameter of 74 mm was incident on the concrete surface for all the trials. A thermal imaging camera (FLIR SC 640) was placed 4.5 m away from the concrete block to protect it from flying debris in the scabbling process.
2.2 Specimen Preparation
We designed concrete with a 28-day compressive strength of 24 MPa. The main composition was ASTM Type I ordinary Portland cement (OPC), water, fine aggregate, and coarse aggregate. Ordinary sand and crushed stone with the nominal maximum size of 25 mm were used as fine and coarse aggregates, respectively. All aggregates used in the mixing of concrete were in the saturated-surface-dry condition. The mix proportions of the concrete are listed in Table 1. The water-to-binder ratio of the concrete was set at 0.5. The replacement ratio of blast furnace slag and fly ash was 15% by the weight of Portland cement and a high range water reducing agent (SP450, FOSROC Kora Inc.) was used to achieve a slump of 150 mm.
A common pan mixer (Hanshin Kumpung Co. Ltd., Korea, 50 L) was used to prepare the concrete. The cementitious binder and aggregate were poured into the mixing bowl and dry-mixed for 60 s at a modest speed. Water was added to the dry blend and mixed at a modest speed for 90 s. Finally, a high range water-reducing agent was added during mixing and mixed for an additional 60 s.
As soon as the concrete was mixed, the concrete was poured into two types of molds. One was cuboid molds (300 mm × 300 mm × 80 mm) for the scabbling test, and the other was cylindrical molds (ø100 mm × H 200 mm) for measuring the compressive strength. After casting, the top of each mold was covered using plastic wrap to prevent moisture evaporation. The concrete was removed from the molds three days after mixing at ambient temperature, and placed in lime-saturated solution for further curing. The total curing time was 28 d. To minimize experimental error, all materials were preconditioned in the testing laboratory (23 ± 0.5℃, 46% relative humidity) for at least one day until the day of testing. The 28-day compressive strength measurements were performed using a universal testing machine (S1 Industry Co., Korea, S1-1471D). The loading rate of the specimens was fixed at 2.5 mm∙min−1. The actual compressive strength after curing was confirmed to be 30 MPa, based on the uniaxial compression test.
3. Results & Discussion
3.1 Laser Scabbling Using 4.8-kW Laser Beam
Laser scabbling on concrete was performed under a stationary 4.8-kW high-power laser beam to determine the optimum scan speed. Concrete age was over a 60 d. A laser power density of 112 W·cm−2 was incident on the concrete surface at a 510 mm stand-off distance. Fig. 2(a) and (b) show the scabbling scenes arranged by the time elapsed after the laser was fired, which were captured under the conditions of laser beam exposure for 25 s and 40 s on the concrete block, respectively. Frequent ejection of smaller fragments was observed as soon as the laser was fired. This indicates that the concrete had low porosity and a smaller pore size. These features contributed to the formation of sufficient vapor pressure at the pores to induce explosions. An intense blaze usually appeared after a 30 s laser interaction time, resulting in glazing after cooling, as indicated by the black arrow in Fig. 2(b) . Even after laser firing stopped, burning trace remained because of the high temperature of the surface, as indicated by the white arrow in Fig. 2(b).
A thermal camera was used to investigate the temporal evolution and spatial distribution of surface temperatures during the scabbling process . Fig. 3 shows the thermal images arranged by the time elapsed, similar to the scabbling experiment photos in Fig. 2. However, it is noted that the elapsed times depicted in Figs. 2 and 3 are not exact matches because there was no synchronization between the thermal camera and laser equipment.
At the onset of laser action on the concrete, the surface temperature continuously increased and reached 800℃ within 2 s. It subsequently underwent frequent spalling with a rattling sound, and a sudden decrease in temperature appeared in the peeled region. The temperature in the peeled region decreased between 300℃ and 400℃, as shown in the thermal image captured at 5 s in Fig. 3(a). This process was repeated several times. The frequent ejection of smaller fragments restricted a further increase of temperature at the impact region. After peeling off the mortar on the concrete surface, the laser beam raised the temperature of the exposed coarse aggregate beyond 1,000℃. It is noted that part of the heated coarse aggregate was also removed along with the mortar during the explosion process. After a significant portion of the coarse aggregate was exposed to the outside as a result of long interaction with the laser beam, the coarse aggregate became the dominant part in direct contact with the central region of the laser beam with high intensity. It led to the further rise in temperature and subsequently melting on the concrete surface, as shown in the thermal image captured at 33 s in Fig. 3(b). In turn, it indicates that it is difficult to induce spalling on coarse aggregate because of different void structures.
Fig. 4(a) shows laser-produced craters arranged by the interaction time between the 4.8-kW laser beam and the concrete. The exposure time varied from 3 s to 40 s. An obvious vitrification resulting from melting at an exposure time of over 35 s was clearly observed. The white arrow in Fig. 4(b) shows the coarse aggregate that has been vitrified. It is easily noticeable that the color of the vitrified region was changed to black after melting. When exposed to a high-power laser beam, the coarse aggregate in concrete was prone to phase transition to melting without the explosion process. Little vitrification and a shallower scabbled depth appeared with laser action of 10 s or less. It indicates that the exposure time is the dominant variable, which determines the degree of the vitrification on the concrete surface under the experimental conditions of the same laser power. Notably, weak vitrification and relatively deeper depth of the crater were produced at a 30 s laser exposure time. The diameter and central depth of the crater produced at a 30 s laser exposure time were measured to be approximately 95 mm and 17.8 mm, respectively. The depth was evaluated using an MCD233 precision Niigataseiki micrometer with a measuring range of 0–25 mm. It was determined that a laser beam of 74 mm exposed to the concrete surface for 30 s was equivalent to an approximately 150 mm∙min−1 scan speed of a moving optical head.
Fig. 5(a) and (b) show the concrete block after scabbling and the scabbling route used. A scan speed was a 300 mm∙min−1. The scabbling head moved on the concrete surface under the fixed condition of a 510 mm stand-off distance. A 4.8-kW laser beam was initially moved from left to right for a length of 200 mm, after which it was moved downward for a length of 40 mm. The laser beam was then moved back to the left by 200 mm. The head moved back and forth for five times in this manner. The manipulation device that moved the scabbling head was set to stop for 2 s each time when the horizontal or vertical direction was changed. The track partially overlapped approximately half the diameter of the laser beam radiated on the concrete surface . The purpose of the overlap was to flatten the conical depth profile of the laser-produced craters. The laser beam was effectively swept twice over each horizontal track, except for the first and last paths. Fig. 5(c)–(e) show the sequential images of the laser scabbling process. An infrared shielding filter was attached to the front of the camera to protect it from the intense laser light and scattered concrete debris. The scan speed was set to 300 mm∙min−1, which was deduced from laser exposure for 30 s under the stationary condition and takes double pass into account. The spalling rates were calculated on the basis of the mass difference before and after scabbling, density values of the employed concrete block (2.25 g·cm-3), and the overall laser firing time. The concrete mass removed by the scabbling process was measured to be 815 g using a Radwag WLC120 precision balance with an error of ± 2 g. The spalling rate was calculated to be 87 cm3∙min−1. The spalled length and width obtained from the measurements of the scabbled concrete surface were estimated to be 250 mm and 210 mm, respectively. The total scabbled area was approximated by multiplying its length and width. The removed volume in the scabbling process was estimated from the mass loss divided by the density. The average depth was estimated to 6.9 mm.
3.2 Laser Scabbling Using a 6.8-kW Laser Beam
The laser power radiated on the concrete surface was increased to 6.8-kW to enhance the spalling rate. The spalling behavior was initially investigated under the stationary laser beam. Fig. 6 shows a concrete block scabbled from a stand-off distance of 510 mm. The laser power density was 158 W·cm−2 on the concrete surface, glazing on the scabbled surface appeared after a 20 s exposure time. Comparing the exposure times of 25 s and 40 s in Figs. 6(b) and (c), the degree of vitrification was more pronounced at the exposure time of 25 s, despite the shorter interaction time. This might be attributed to the sparse distribution of coarse aggregates near the central point of the laser beam at the highest intensity, as shown in Fig. 6(c). This indicates that the spalling behavior was greatly affected by the coarse aggregates. The central depths of the laser-produced craters shown in Figs. 6(b) and (c) were measured to be approximately 8.9 mm and 20.7 mm, respectively.
It was noted that there was little vitrification at a laser interaction time of 15 s. The scan speed was calculated as 300 mm∙min−1 on the basis of a 74 mm beam diameter incident on the surface and a 15 s laser action time. Considering the overlap of approximately half the diameter of the laser beam on the concrete surface, scan speed was set to 600 mm∙min−1. As indicated in Fig. 7, a 6.8-kW laser beam traveled along the red path marked on the surface. The horizontal and vertical path lengths were 200 mm and 40 mm, respectively, as before. The stand-off distance was fixed at 510 mm. Fig. 7(a) shows the concrete block scabbled at scan speed of 600 mm∙min−1. Little vitrification was observed. The mass removed from the scabbling process was 1,165 g. The overall laser action time on the concrete block was 2.28 min. The spalling rate was calculated to be 227 cm3∙min−1. Furthermore, the average depth was estimated to be 9.8 mm.
The scan speed was reduced to investigate the effect of vitrification on the spalling rates. Fig. 7(b) shows the concrete block scabbled at scan speed of 300 mm∙min−1. Vitrification on the surface was observed to some extent. This is attributed to the extended interaction time between the laser beam and the concrete. The mass loss after scabbling was 1,210 g. The overall laser action time on the concrete block was 4.15 min. The spalling rate was reduced to be 129 cm3∙min−1. Overexposure to the laser beam was less effective for scabbling because some laser energy is consumed in unwanted melting of the concrete, thereby not causing the surface of the concrete to explode.
Experimental conditions focusing on the removal rate of the concrete surface by a laser-induced explosion without vitrification were investigated. It was confirmed that improvement in the removal rate depended on the increase in laser power acting on the concrete surface and was associated with the proper selection of laser interaction time. We obtained a removal rate of 87 cm3∙min−1 when a 4.8-kW laser beam was irradiated to a concrete surface at a scan speed of 300 mm∙min−1. The removal rate was improved to 227 cm3∙min−1 when the 6.8-kW laser beam was irradiated to the concrete surface at a scan speed of 600 mm∙min−1. The laser power densities irradiating on the concrete surface were 112 W·cm−2 and 158 W·cm−2, respectively, which indicates that vitrification-free scabbling is possible by controlling the scan speed at a given laser power. In contrast, the removal rate of concrete decreased to 129 cm3∙min−1 under the conditions of the 6.8-kW laser beam power and 300 mm∙min−1 scan speed. Some vitrification on the surface was observed. The slower scan speed implies an increase in the exposure time of the laser beam on the concrete surface. A longer exposure of the laser beam led to an excessive deposition of laser energy on the concrete and resulted in vitrification on the surface. In turn, this indicates that vitrification occurs according to the length of exposure time even at the same laser power.