A Comparative Study on Laser Cutting Performance with Varying Speeds at 10 M Underwater

Abstract

Despite the dismantling structures that are submerged to significant depths of water during the decommissioning of nuclear power plants, there is limited research on deep-water laser cutting processes. A self-designed pressurized chamber was used in this study and successfully conducted the world’s first laser cutting experiment in a simulated 10 m water depth environment. laser cutting was performed in a 10 m underwater environment, and the cutting efficiency was compared to that observed in a 1 m underwater environment. Therefore, A 100 mm thickness of 304 stainless steel was successfully cut underwater, and the highest cutting speed of 100 mm/min was achieved. The result indicates that, as the cutting speed increased during underwater laser cutting, both the heat input and the mass flow rate of the assist gas decreased, resulting in a narrower rear kerf width and an ineffective evacuation of the molten metal.

1. Introduction

Laser technology has been continuously researched for its application in various fields such as heat treatment, welding, and cutting due to its numerous advantages [1,2,3,4,5,6,7,8,9]. Among these, laser cutting has been widely adopted across diverse industries such as marine and nuclear due to several benefits, including remote operation, fast cutting speeds, and low heat input [10,11,12,13,14,15]. Particularly, as the operational lifespans of nuclear power plants in Korea are reaching their end, laser cutting technology is being considered a next-generation solution for decommissioning [16,17,18]. Laser cutting in nuclear power plant has several advantages, including the potential for high-speed cutting, the ease of remote control, and the generation of minimal secondary waste.

In recent years, with the increasing focus on work safety, nuclear power plant decommissioning has required underwater cutting to prevent human workers from being exposed to radioactive environments. As a result, many researchers have developed their own underwater laser cutting technologies [19,20,21]. However, the underwater cutting process is technically more challenging than cutting in air, as there are many more factors to consider [22]. In particular, underwater laser cutting requires the successful creation of a local dry zone, and the waterproofing of the laser head must be excellent [23]. As the water depth increases, it becomes more difficult to form a stable cavity due to the buoyancy and water pressure. Therefore, securing the technology to ensure stable laser emission and effective gas flow to expel molten material through a stable air tunnel is of great importance. As such, numerous researchers have been conducting studies to achieve laser cutting in underwater environments.

Jae sung Shin et al. conducted a study on improving the initial cutting speed when cutting steel plates thicker than 40 mm using an underwater laser [22]. They proposed the application of an angled cutting technique, and their findings demonstrated that this method significantly improved the efficiency of underwater laser cutting, particularly for processes that require cutting thick steel plates. To overcome the limitations of single-nozzle cutting, another study applied a dual-nozzle system to cut steel plates thicker than 80 mm [20]. Jae sung Shin et al. stabilized the gas flow through an auxiliary nozzle, improving the initial cutting process, and successfully cut plates up to a 100 mm thickness using a 9 kW laser. Seong Y. Oh et al. used a 6 kW fiber laser to cut stainless steel plates of 50, 60, 70, and 80 mm in thickness, also applying a dual-nozzle system to enhance cutting efficiency [24]. The results revealed that the dual-nozzle system was effective in improving cutting accuracy, and the narrow kerf achieved through laser cutting significantly reduced the amount of secondary waste compared to other methods, such as plasma cutting or water jet cutting. Kwan Kim et al. analyzed the effects of laser cutting parameters on kerf width, drag line formation, and surface roughness when cutting 50 mm thick stainless steel plates underwater [25]. They identified the optimal conditions for achieving high-quality cuts and concluded that the optimal laser cutting parameters for cutting 50 mm thick stainless steel underwater were a focal position of −30 mm, a laser power of 9 kW, and a cutting speed of 30 mm/min. In another study, Jan Leschke et al. investigated the underwater cutting of 3 mm thick stainless steel using an Yb laser [26] source and evaluating the cutting efficiency. Their goal was to minimize material loss and secondary waste, which plays a crucial role in reducing air and water filtration costs during decommissioning operations. They concluded that underwater laser cutting could effectively reduce material loss and secondary waste, particularly when a ball-type dross formation occurred at low gas pressure and high laser power. This method was shown to offer significant improvements over conventional cutting techniques by reducing both secondary waste and overall costs.

Despite the fact that, in actual nuclear power plant decommissioning, the reactor vessel internals (RVIs) are submerged at significant depths, current studies have not adequately considered the laser cutting process at greater depths. As mentioned earlier, overcoming water resistance and ensuring the waterproofing of the laser head are critical challenges in underwater laser cutting. While it becomes increasingly difficult to maintain a stable cavity due to water pressure at greater depths, there is a notable lack of research exploring laser cutting in deep-water environments. Therefore, in this study, we conducted laser cutting in a simulated 10 m underwater environment and compared the cutting efficiency to that observed in a 1 m underwater environment. To achieve this goal, our research team designed and constructed a custom pressurized tank capable of simulating a 10 m water depth and, for the first time globally, successfully conducted laser cutting experiments in a 10 m water depth environment.

2. Materials and Experimental Procedures

In this study, AISI 304 stainless steel (heat no.17SD63345, DK Corporation), which is widely used in nuclear power plants [27,28,29], was selected as the test material, and its chemical composition is shown in

Table 1. Chemical compositions of cutting material (wt %).

Materials	Fe	C	Si	Mn	P	S	Cr	Ni	Mo	N
AISI304	Base	0.021	0.39	1.63	0.03	0.006	18.19	8.10	0.16	0.077

. The specimens were prepared with a width of 80 mm, height of 80 mm, and thickness of 100 mm with a laser cutting thickness of 100 mm. Detailed sample schematic diagram is attached in the Appendix A. A CW fiber laser with a maximum output power of 20 kW was used as the cutting laser source (IPG, YLS-20000-S2T, λ = 1070 nm), and the end of the process fiber was connected to a custom-designed laser cutting head. The beam parameter product (BPP) was measured to be 8 mm·mrad, with the laser spot diameter measured to be approximately 750 μm at the focal point and 2.1 mm at the specimen surface. The cutting head used in this experiment was specifically designed to withstand the pressure of a simulated 10 m underwater environment. Additionally, it was sealed to ensure excellent waterproofing, as the laser cutting head system needed to be fully submerged.

Figure 1 shows the experimental setup of the custom-designed underwater laser cutting system used in this study. The underwater laser cutting equipment consists mainly of a pressure chamber, monitoring sensors, a door, a control box, and a monitor. The pressure chamber was specifically designed to simulate a 10 m underwater environment, using compressed air to apply and withstand a pressure of 1 bar. The door is designed to be opened and closed, providing a passage for attaching and removing specimens. Additionally, a window made of transparent acrylic material was installed to enable a real-time observation of the cutting process and allow for immediate termination in the event of a problem. Both the laser head and specimen were fixed to a Z-axis moving stage, which moved along the cutting direction (Z-axis) within the pressurized tank. This stage was computer-controlled, allowing for the remote operation of all cutting processes, including turning the laser power on and off, setting the laser power, moving the cutting head, moving the specimen, and controlling the assist gas. A nozzle was also designed to generate a stable gas flow to form a cavity and was integrated into the cutting head. The collimating focal length was 70 mm, and the focusing focal length was 300 mm. In all tests, the laser output was fixed at 15 kW, the cutting nozzle had an exit diameter of 3 mm, and nitrogen was used as the protective gas. The shielding gas for the laser cutting was fixed at 15 bar, with the cutting speed and chamber pressure used as test parameters. The focal position was set at 54 mm inside from the front surface of the specimen, based on previous research that demonstrated wider upper kerf formation when the focal point was positioned deeper within the specimen [30]. The detailed cutting conditions are provided in

Table 2. Underwater laser cutting experiment condition in this study.

Test No.	Shielding Gas Pressure [bar]	Focus Position [mm]	Output Laser Power [kW]	Chamber Pressurization [bar]	Cutting Speed [mm/min]
1	15	−54	15	0	20
2				0	50
3				0	70
4				0	100
5				1	20
6				1	50
7				1	70
8				1	100

.

The underwater cutting experiment was carried out as follows: water was poured into the custom-designed pressurized tank, and the specimen was fixed in place using a jig. After inputting the laser cutting parameters, the cutting process commenced, during which compressed air, serving as a protective gas for the laser head, was emitted from the nozzle. As the cutting head and specimen descended below the water surface, nitrogen gas was emitted to form an air tunnel for stable cavity formation. Upon reaching the starting point, the laser was directed onto the material through the air tunnel, initiating the cutting process at a slow speed with piercing. Subsequently, the specimen moved along the predetermined cutting path at the specified cutting speed. Once the set distance was achieved, the laser was turned off, and both the cutting head and specimen were lifted back to the water surface.

After the cutting experiment, the front and back of the laser-cut specimen, the cut surface, and the cut kerf cross-section were observed according to the changes in the speed and pressure conditions in the underwater environment to evaluate the cutting quality. The cut surface forms a repetitive stripe-shaped drag line along the cutting direction of the specimen. It can be largely classified into three regions: the upper region where cutting by the laser beam is dominant, the middle region where melting is dominant near the focus position, and the lower region where cutting by the heat transfer of the molten metal is dominant [25].

In order to observe the cross-section of the cut kerf and the microstructure of the cut surface, the specimen was ground using 2000 grit abrasive paper and then etched using an etching solution consisting of a mixture of 300 mL of HCl and 100 mL of HNO3. The cross-section of the kerf width and the microstructure of the cut surface were then observed using a digital microscope (KEYENCE, VHX-7000, Osaka, Japan). Since a more in-depth study of the microstructure of the cut specimens will be conducted in future studies, this study focuses on the new cutting technique rather than the microstructural aspects.

3. Results and Discussion

3.1. Evaluation of Cutting Performance at a Depth of 1 M Underwater

First, to assess the cutting performance at a depth of 1 m underwater, a laser cutting test was performed without applying pressure to the chamber.

Figure 2 shows a schematic diagram of the underwater laser cutting process and detailed dimensions of the cut specimen. In the cutting process, the first 15 mm section was cut at a slow speed (6 mm/min), followed by a rapid cutting of the subsequent 50 mm section. This method was adopted because, as demonstrated in previous studies, once the specimen is fully penetrated, the molten pool is able to escape from the back of the specimen, allowing high-speed cutting to proceed smoothly [22].

Table 3. Laser cutting conditions, cutting status, and kerf width of 100 mm thick specimens tested in 1 m underwater simulation environment.

Cutting Speed [mm/min]	Chamber Pressure [bar]	Shielding Gas Pressure [bar]	Cutting Status	Average Kerf Width [mm]
				Upper	Lower
20	0	15	Ο	2.09	3.91
50	0	15	Ο	2.83	2.08
70	0	15	Ο	2.93	1.86
100	0	15	Ο	3.17	1.92

presents the experimental conditions, results of the underwater laser cutting test, and the kerf widths of the cutting specimens conducted at a depth of 1 m. When the 100 mm thick specimen was laser cut at a depth of 1 m underwater, the maximum cutting speed achieved was 100 mm/min. However, at 100 mm/min, the cutting was just barely completed under the given boundary conditions. Figure 3 shows the front and rear surfaces of the specimen with laser cutting at various speeds at a depth of 1 m underwater.

Figure 3 depict the front and rear kerf of cutting specimens with various speeds of laser cutting. When laser cutting is performed underwater, the cooling rate is higher compared to cutting in air due to the cooling effect of the water. Therefore, if the molten metal is not sufficiently blown away with gas as it escapes through the rear, it tends to accumulate on the rear kerf due to the rapid cooling rate. As seen in Figure 3, as the cutting speed increases, the molten metal is not fully expelled, resulting in accumulation on the rear kerf. Under the conditions of this cutting experiment, it was found that a cutting speed of 50 mm/min produced the cleanest kerf shape on both the front and rear surfaces. Additionally, when the cutting speed is slower, the heat accumulation leads to a larger amount of re-melted metal, which is not fully removed by the shielding gas, causing the rear kerf width to narrow. Further details on this phenomenon are provided in Section 3.3.

3.2. Evaluation of Cutting Performance at a Depth of 10 M Water

To investigate the effect of varying cutting speeds on the laser cutting performance in a 10 m underwater environment, a custom-made chamber was used, where compressed air was injected to create the equivalent pressure of a 10 m water depth. After setting the focal length to −54 mm from the specimen’s surface and fixing the laser power at 15 kW, the cutting speed was gradually increased to achieve cutting performance similar to that of 1 m underwater depth conditions at a 10 m underwater depth.

Figure 4 shows the front and rear kerf of the specimen after laser cutting. The front kerf shape shows little difference with increasing cutting speed. However, when comparing the rear kerf width, it was observed that the kerf width narrowed as the cutting speed increased. Additionally, there was a tendency for more dross to accumulate on the rear surface as the molten metal was not fully expelled. This phenomenon is similar to that observed in 1 m depth underwater cutting, where the molten metal exits the bottom but cannot be fully removed due to the insufficient gas flow at higher cutting speeds, leading to accumulation in the rear kerf. In the simulated 10 m underwater environment, the added water pressure makes it even more difficult to expel the molten metal compared to 1 m underwater cutting, resulting in a generally narrower kerf width. Specifically, under the cutting speed condition of 100 mm/min, a significant amount of dross was found to accumulate on the rear surface, indicating that the molten metal was not effectively expelled, suggesting that this condition approaches the cutting boundary.

Table 4. Laser cutting conditions, cutting status and kerf width of 100 mm thick specimens tested in a 10 m underwater simulation environment.

Cutting Speed [mm/min]	Chamber Pressure [bar]	Shielding Gas Pressure [bar]	Cutting Status	Average Kerf Width [mm]
				Upper	Lower
20	1	15	Ο	2.29	2.72
50	1	15	Ο	2.66	2.27
70	1	15	Ο	2.83	1.54
100	1	15	Ο	2.94	1.42

presents the experimental conditions, the results of the underwater laser cutting test, and the kerf widths of the cutting specimens conducted at a depth of 10 m.

Figure 5 shows sequentially captured images from a video of the specimen being laser-cut in a simulated 10 m underwater pressurized chamber. The experimental parameters used were identical to those in the 1 m underwater cutting, with a cutting speed set to 20 mm/min. After 80 s, the laser was observed to penetrate the rear surface of the specimen for the first time. Subsequently, the specimen was lowered, while the laser cutting continued. As seen in the images, the shielding gas successfully created a cavity through which the laser could be emitted, allowing the molten metal to be expelled from the rear and the cutting process to proceed. Through multiple experiments, it was concluded that the key factor for successful underwater cutting is the laser’s ability to penetrate the specimen during the initial cutting phase.

Figure 6 illustrates the failed experimental conditions in which laser cutting in a simulated 10 m underwater environment under an assist gas pressure of 9 bar. When the assist gas pressure was set to 9 bar, insufficient gas flow was provided to expel the molten metal generated during cutting from the rear of the specimen, leading to accumulation on the rear surface. This accumulation blocked the rear, preventing the laser from fully penetrating the material. Excessively high assist gas pressure can result in increased costs and a rise in secondary waste. Therefore, establishing an optimal assist gas pressure is crucial in underwater laser cutting processes for nuclear power plant decommissioning.

In this study, the optimal conditions were determined to balance secondary waste generation and cutting quality. Based on previous studies, an assist gas pressure of 15 bar was selected, and the detailed results of this investigation will be presented in a subsequent paper.

3.3. Comparative Analysis of Cutting Performance According to Water Depth

Figure 7 presents the cut surfaces of laser-cut specimens in both 1 m and 10 m underwater environments at various cutting speeds. A cutting speed of 100 mm/min was considered a boundary condition and thus excluded from observation in this study. As noted previously, the cut surface forms repetitive drag lines along the cutting direction of the specimen, which can be categorized into three primary regions based on drag line morphology: the upper, middle, and lower regions [25]. The upper region is where cutting occurs directly from the laser beam; the middle region is near the focal position where the metal melts and cuts due to the laser’s energy; and the lower region is where residual molten material, not fully expelled by the assist gas, remains on the cut surface.

Striation patterns on the cut surfaces varied between the two environments. The striations on specimens cut at a depth of 10 m appear relatively straighter than those cut in the 1 m underwater condition, likely due to increased gas density influenced by water pressure. This result suggests that sufficient assist gas was delivered across the cutting areas, resulting in effective gas flow. In this experiment, the laser focus was set at −54 mm, leading to the significant solidification of molten metal near the focal region. Furthermore, lower cutting speeds increased heat input, generating more molten material. When this molten metal is not fully removed by the shielding gas, it leaves residues on the cut surface. This outcome is observable in the cross-section of the specimen cut at 20 mm/min (Figure 7), which shows a larger melting and re-solidification area compared to specimens cut at higher speeds. To examine the drag lines in greater detail at higher cutting speeds, the topographic color distribution and arithmetic mean height values of the cut surface at a laser cutting speed of 70 mm/min were measured according to underwater depth in Figure 8. The areas of the topographic color distribution were highlighted with yellow boxes in the cut surface OM images. To measure the arithmetic mean height values according to the three previously mentioned drag line patterns, the cut surface was divided into three regions (1, 2, and 3), with the average height value calculated for each region. The upper region was defined as the area where striation lines appear clearly as straight lines, the middle region extended to the focal depth of −54 mm, and the lower region included the area from −54 mm to the end of the specimen.

In Figure 8, it is evident that the straight-line striations in the upper region (1) are longer in the specimen cut in the 10 m underwater environment than in the 1 m underwater condition. Previous studies [30] have demonstrated that the length of the upper region increases as the assist gas more effectively reaches the cut surface. In the 10 m underwater condition, sufficient assist gas was delivered throughout the cutting area due to the influence of water pressure, resulting in effective gas flow. This effect is further supported by the molten metal distribution observed in the middle region (2). In the 1 m underwater laser cutting process, the middle region of the cut surface shows an accumulation of molten metal flowing downward. The Sa values confirm this observation, with the middle region in the 1 m condition exhibiting a Sa value of 172.1 μm, higher than the Sa value of 149.93 μm observed in the 10 m underwater specimen. Additionally, Sa values in the lower region (3) are 207.13 μm and 189.9 μm for the 1 m and 10 m underwater environments, respectively, indicating that the 10 m underwater cutting environment has a lower Sa value in the lower region. This finding suggests that the molten metal generated in the middle region was not fully expelled by the assist gas. Consequently, it can be inferred that the assist gas penetrated more effectively into the specimen in the 10 m underwater cutting environment compared to the 1 m underwater condition.

Figure 9 illustrates the kerf cross-section. While the performance of the underwater-cut specimens showed minimal differences between the 1 m and 10 m underwater environments, it was observed that, in the 1 m underwater condition, the kerf near the focal position was noticeably wider at a cutting speed of 20 mm/min. Additionally, in all conditions, the kerf width increased toward the bottom of the specimen, starting from the vicinity of the laser focus. Figure 9 demonstrates that, as the cutting speed increased, the kerf width decreased, with an overall trend of narrower kerf widths in the 10 m underwater environment.

The variation in kerf width was further examined through changes in material weight loss. As shown in Figure 10, lower cutting speeds resulted in greater material weight loss, indicating a wider kerf. For the specimens cut in the 1 m underwater environment, material loss decreased significantly from 168.6 g at a cutting speed of 20 mm/min to 76.3 g at 100 mm/min. This finding is due to increased heat input at lower cutting speeds, resulting in more molten material generation and allowing more assist gas to enter the cutting area, effectively blowing away the molten metal. This trend was more pronounced in the 1 m underwater cutting condition than in the 10 m condition.

In the 10 m underwater environment, increased water pressure compresses the assist gas volume, increasing its density and reducing the cavity size through which the laser passes, thereby producing a relatively narrower kerf width. For specimens cut in the 10 m underwater environment, material loss similarly decreased from 89.7 g at a cutting speed of 20 mm/min to 66.8 g at 100 mm/min; though overall, specimens cut in the 10 m underwater condition exhibited lower material loss.

A magnified image of the upper kerf width is presented in Figure 11, demonstrating a sharper kerf shape in the pressurized environment. According to Boyle’s law, at a constant temperature, increased pressure reduces gas volume. In this case, the water pressure at a depth of 10 m resulted in decreased gas volume and increased gas density, enhancing its concentration. This effect produced a relatively stable and sharp kerf in the upper region, leading to an overall narrower kerf width. Figure 10 presents the average kerf width of the underwater-cut specimens, including upper kerf width, lower kerf width, and the ratio between the two. The upper kerf width remained relatively consistent across different cutting speeds; however, at lower speeds, the increased heat input caused more molten material to accumulate in the kerf, resulting in a narrower width. This phenomenon is illustrated in Figure 12, where the molten layer thickness decreased from 1.3 mm at a cutting speed of 20 mm/min to 0.6 mm at 50 mm/min in the 1 m underwater condition, and from 0.5 mm at 20 mm/min to 0.3 mm at 50 mm/min in the pressurized condition.

For the lower kerf width, the pressurized specimens displayed a relatively narrower kerf compared to those cut in the 1 m underwater condition. As cutting speed increased, the kerf width further decreased. This trend is attributed to the lower heat input at higher speeds, which limited the assist gas’s capacity to fully reach the rear of the specimen and effectively expel the molten material. Additionally, the kerf width in the 10 m underwater condition was relatively narrower than that observed in the 1 m underwater condition due to the reduced gas volume under pressure.

4. Conclusions

In this study, a pressurized chamber was designed to simulate the 10 m water depth environment of nuclear power plant decommissioning facilities, and underwater laser cutting was conducted. Additionally, the laser cutting performance was compared with that conducted in a 1 m water depth environment to analyze the differences in cutting performance. The kerf width, cut surface analysis, and mass loss were investigated, and the following conclusions were drawn:

• In underwater laser cutting, as the cutting speed decreases, the heat input to the material increases, generating a greater amount of molten metal and further reducing the upper kerf width. This phenomenon is more pronounced in the 1 m underwater cutting environment.
• During underwater laser cutting, as the cutting speed increases, both the heat input and the assist gas flow rate decrease, resulting in a narrower rear kerf width and a less effective expulsion of molten metal.
• Observations of the topographic color distribution on the cut surface indicate that the linear striation lines in the upper cutting region are noticeably longer in the 10 m underwater cutting environment. This finding suggests an effective delivery of the assist gas to the cut surface.
• In the 10 m underwater cutting environment, increased pressure reduces the assist gas volume and raises gas density, thereby enhancing concentration. This finding results in a relatively stable and sharp kerf shape in the upper region, leading to an overall narrower kerf width.

In the future, more in-depth studies will be conducted on the microstructure of underwater laser-cut specimens, focusing on the effects of water depth.