Influence of the Cutting Feed Rate on the Hardness and Microstructure of Copper Using Plasma Arc Machining (PAM)

: This study investigated the inﬂuence of the cutting feed rate on the hardness and microstructure of copper machined using a plasma arc (PA) to examine the resulting changes and their impact on the quality of the cut surface. Various constant cutting feed rates and amperage values were used as parameters to measure the cutting performance. Pre-and post-cut hardness measurements and scanning electron microscope (SEM) images were taken. The hardness of the copper surface was the same before and after plasma arc cutting (PAC). PAC did not affect the copper’s hardness or the microstructure of the thermally affected cutting zone. The copper from the cut surface was melted by the PA operation near the edge of the cutting surface with no change in the microstructure. SEM imaging of the cut conﬁrmed this. Thus, the quality of the cutting surface was not affected. In addition, the microstructure of the copper’s thermally affected cutting zone did not alter the cutting surface’s quality. Hardness measurements post-cutting yielded 69.28, 71.65, 70.15, and 60.09 HB for four tests at 500 mm/min and 30 A. The lowest cutting width was 1.504 mm at 12,000 mm/min, and the surface roughness was 2.5 µ m at 500 mm/min.


Introduction
Plasma arc machining (PAM) is an effective manufacturing method widely used today. Plasma cutting (PC) is an efficient method for cutting hard metals. PAC in sheet-metalcutting manufacturing units is widespread, and the selection of operating parameters is essential for efficient and fast cutting [1].
PAC is a thermoelectric cutting process that melts and separates metal using a focused jet of high-temperature plasma gas. Many studies have assessed the surface quality of such manufactured metal parts. The heat-affected zone is another critical characteristic that depends on the quality and performance of PA-machined surfaces. The heataffected zone is a distinctive characteristic of PA-machined surfaces that affects quality and performance [2,3]. Another technique that examines the surface microstructure is SEM [4]. A schematic of the PAC process is shown in Figure 1 that comprises the electrode, the torch body, the outer casing (shell), the locations of both the primary and secondary gas, the work part, the molten metal removed, the PA, and the cutting (kerf) width.
Various studies have reported on the effects of PAC on different materials. Senthilkumar et al. studied the influence of arc cutting operating variables such as the voltage, current, gas pressure, feed rate, standoff distance, and gas flow rate on the cutting quality of stainless steel (SS304). They found that PAC affected the feed rate, amperage, and pressure of the SS304 stainless steel [5]. Ramakrishnan et al. investigated the cut quality characteristics on stainless steel (SS321) using PAC [6]. Patel et al. studied the influences of PA process factors on the cutting of quard-400 martensitic steel with a hardness of 400 HB [3]. PAC operating elements for EN 31 steel were also evaluated for their influences and parametric improvement. The following variables were taken into consideration: gas pressure, arc current, and high flame. In a study by Das et al., the surface morphology was examined via scanning electron microscopy [7]. The influence of various process parameters on the heat-affected zone (HAZ) was determined. A HAZ analysis was done by measuring hardness changes on the specimen cross-sections and via microscopic observation of the material structure [8]. Kountouras et al. investigated the influence of microstructural aspects on CNC factor design for carbon steel sheets in PAC. At various distances from the cutting edge, a fine stiffness was measured. The results showed that the values, which included the optimum factors derived from metallographic aspects, were determined based on the precise hardness and microstructure of the material [9]. Akkurt et al. studied the hardness and microstructures of pure aluminum surfaces of Al 6061 as they were cut using techniques such as sawing, milling, submerged plasma, PA, laser, wire electric discharge machining, oxy-fuel, and abrasive water jet cutting. Because various cutting techniques exist, the hardness changes and microstructures of the cut surfaces were investigated. The results showed that the type of cut affected the hardness and surface quality of the cut surfaces. The microstructure was affected except in the abrasive water jet cutting, and for each cutting process, microscopic changes were discovered. As long as there is not a significant reduction in the hardness or microstructure, abrasive water jets can be effective in industrial applications [10].
For the heat-affected zone's characteristics, various studies have been reported. Celik et al. studied the heat-affected area, cutting width, and roughness of plasma-cut stainless steel 309. These three parameters were optimized through experimental results. In addition, the three factors that affected the cutting quality and efficiency were identified in this study: the cutting width, surface roughness (Ra), and heat-affected zone. As reported in [11], controlling the influence of PAC factors on the heat-affected zone is possible. Further, the influence of cutting factors on materials cut via digital plasma control was investigated for steel alloys. Gautam and Gupta analyzed various operating factors such as the heat-affected zone of Monel 400 [12]. Masoudi et al. reported the optimized conditions of the plasma's heat-affected area, cutting width, and roughness when cutting stainless steel 309. The cutting width, Ra, and heat-affected zone were three determining factors that affected the cutting quality and efficiency on 309 stainless steel alloys [13]. As reported in [14], the influence of PAC elements on the heat-affected zone for low-carbon steel could be controlled. A microstructure analysis was conducted to determine the formation of the cutting surface under different cutting conditions for Monel 400 (a nickel-copper alloy) [2]. Abouzaid et al. investigated the effect of the machining parameters of thin brass and found that the hardness parameter's influence was minor and negligible because the decreased hardness depth was as low as 150 µm. Moreover, they showed grain growth near the cutting edge at 30, 45, and 60 A and 2000 mm/min (the cutting feed rate), while the re-melted edge displayed a very thin layer with a dendritic structure [15]. Cegan et al. examined the influences of PAM parameters on the microstructure of Nb-Al and W-Al alloys [16]. In addition, multiple melting methods were used to subject NiTi to PAM within vacuum induction melting, as was reported in [17]. The microstructure, phase composition, and hardness of Ti-48Al-2Cr-2Nb were investigated. The results showed that the alloy could be purified after EBM by separating and removing the impurities [18]. For S235JR steel, Irsel and Guzey investigated the microstructure and hardness (HV 0.1) before the cutting process for tensile specimens. The hardness of the cut surfaces in PAC increased from 150 HV to 230 HV. PAC is more cost-effective [19]. Deposited specimens of 7075 aluminum were found to have a microhardness greater than that of Mg 2 Si and (Mg (Zn, Cu, Al 2 ) as determined by Hu et al. [20]. Kumar determined how the cutting system and feed rate affected the material structure of Hardox plate and how the hardness decreased from the outer surface to the core [21]. Most reported studies on the effect of the process parameters of PAC were done on steel and Al alloys, and few studies reported Cu alloys.
This study examined the effect of differences in the cutting feed rate on the hardness, kerf width, surface roughness, and microstructure in the operation of a PA with copper. Taking advantage of changes in these characteristics could occur in the operation. These changes could adversely affect the quality of cutting surfaces of copper machined using PA. This study aimed to determine the impact of cutting feed rate values on the hardness and microstructure of copper machined using PAC. Table 1 shows the chemical composition of the specimens of the copper material used in carrying out the cutting experiments. It was determined using a Foundry-Master Pro device for the chemical analysis. This study used an automated gas control PAC machine (Hypertherm, HyPerformance ® Plasma Loyal Mak Max 130XD) to perform 13 experimental runs to cut 13 specimens of copper with a 1 mm thickness. Table 2 shows the values of the operating factors for a current capacity of 30 A, cutting feed rates of 500 to 12,000 mm/min, a constant arc voltage of 130 V, and air pressure used for the cutting gas at 60 Pa. Table 2. Values used in the cutting experimental study for specimens 1 to 13 of copper for the PAC operating factors of cutting feed rates in mm/min, cutting current in amperes (A), voltage in volts (V), and gas pressure (Pa). All experimental specimens were organized into one group comprising the 13 specimens. The study used 13 cutting feed rates graded from the lowest to the highest (500 mm/min to 12,000 mm/min). A constant cutting current capacity of 30 A was used for all experiments.

Experiment/Specimen
All tests were conducted in the laboratory of the Central Metallurgical Research and Development Institute (CMRDI) in Egypt. The HB hardness showed copper number 1 (a) and 1 (b) specimens. Since Brinell has a recommended hardness, the Instron Wolpert GmbH (UK) was used to examine and measure the hardness of separate locations on the specimens' surfaces. The kerf width of the specimens was measured using the projector zoom apparatus at all points from 1 to 18 shown in Figure 2. The surface roughness of the cutting surface edge was examined and measured. A surface roughness measuring instrument (Surftest SJ-201, Mitutoyo) was used to check the cutting-edge surface roughness (surface roughness inspection) for specimens 1 and 13. A FEG-SEM microscope (Quanta FEG 250) was used.

Hardness
In Figure 3 and Figures 6-12 show the experimental specimens' after PC and arranged according to the cutting feed rates used in the process (from minimum to maximum). Table 3 shows the measurement readings before PAC. The readings for the copper-1 specimen measurements after PAC and in locations close to the cutting edge using the same diameter and load are shown in Table 4. Table 3. Standard deviation (SD), statistics, and pre-cut hardness readings.
Since the PA would have a thermal influence on the area surrounding the cutting path, we had to measure the hardness before and after cutting. According to the readings for the copper-1 specimen measurements, we found that the differences in the readings in both cases were small. The PAC was conducted with a low feed rate and a lower amperage. The hardness of the copper only showed a very slight change that could be ignored. Figure 3 shows the PAC views of Specimen 1 of copper-1 (a) from the front and 1 (b) from the back at 500 mm/min and a current of 30 A.
The hardness test was conducted at the lowest cutting feed rate of 500 mm/min due to the higher cutting heat of the plasma. The differences in the hardness readings were slight before and after cutting. A Specimen 1 (a), (b) was divided into four parts and a total of nine tests were conducted on each part. Hardness measurements were taken at different points on a track (labeled as numbers 1 to 18 in Figure 2) divided into the four parts of the specimen. The hardness test was carried out on one of the four parts after their preparation. The test was carried out at separate points from and on more than one point. These points were on all types of lines on the cutting path in the HAZ near the cutting edge. The results of the measurements were approximately the surface of the specimen; it was found that the readings were very close in terms of values. The results of the characteristic specimens were determined by averaging; each part represented a hardness test on the entire cutting surface.
Based on the readings given below for Specimen 1 (Figure 4a-h), we found that the hardness did not change at the minimum cutting feed rate and at the current capacity that was used.  The specimen was chosen based on the cutting feed rate of at least 500 mm/min and the cutting capacity of 30 A. The presence of internal microshrinkage (pores) on the surface of the specimen was due to the emission of gases that were formed during the casting. The preparation of the copper raw material was taken into consideration.
The measurement before PC was made, and the measurements in locations before the PAC in the equation were recorded. Because the ball diameter was in mm and the load was in kg, the following diagonal was 20 kg/2 mm.
The equation used for calculating the Brinell hardness was as shown in Equation (1): where BHN stands for the Brinell hardness number, p is the load in kg, D is the diameter of the ball used as a penetrator, and d is the impact diameter on the surface of the specimen. The Brinell hardness meter automatically displayed the result after the test. Figure 4a-h show the hardness testing data results and images of the indentation of the ball on the copper for Specimen 1 at a feed rate of 500 mm/min and a current capacity of 30 A. The hardness of the cut surface of Specimen 1 was examined and measured before and after cutting. All readings were in a narrow range between 60 and 72 HB.
Graphical representations of the relationship between the Brinell HB hardness of specimen copper-1 and the cutting feed rate and amperes used for four pre-and post-PAC tests at points on the cutting-edge surface are shown in Figure 5. In Figure 5a,b, the Brinell hardness measurements at different test points on the cutting surface at a cutting feed rate of 500 mm/min and 30 A for Specimen 1 before and after cutting are shown. Since the hardness measurements were taken on different point scales, changes between the measurements before and after cutting could be ignored. The standard deviations were 2.44 and 5.23; both were very small compared to the reading. In addition, based on the graphical representation, we verified that the differences in the readings in the two cases were completely non-existent, while reading differences in the tests were very slight. We could neglect the differences in the readings because there were no significant changes in the hardness before and after cutting. Abouzaid et al. proved that hardness decreased near the cutting edge due to grain growth. This decrease in hardness was very small and its effect could be neglected because the decreased hardness depth was as low as 150 µm [15]. The HAZ, kerf width, and surface roughness are important properties that influence the quality and performance of plasma cut surfaces [2,6]. The hardness near the outer surface, which was affected by the high heat that occurred due to the plasma cutting, was increased [21].

Kerf Width
The higher the cutting feed rate, the lower the kerf width and the higher the cutting quality at the maximum feed rate of 12,000 mm/min. The lower the cutting feed rate, the greater the kerf width and the lower the cutting quality at the minimum feed rate of 500 mm/min. Figure 2 and Tables 5 and 6 show the values of the readings of the cutting paths' width measurements generated when cutting the specimens. The cut-off was performed for all specimens at 30 A; the points at which the measurements were made are indicated with arrows. Below are the readings that showed an increase in the cutting feed rate, a decrease in the cutting width, and an increase in the cutting quality. As is known, the quality of the process and machining depends on the surface of the cut and kerf width [4,22]. Figures 3  and 6 shows the width of the cutting path for 500 mm/min from both sides (the front and the back). The beginning of the cutting path is represented by Point 1 ("Key Hall"); at this point the largest reading of the cutting width was 2212 mm (which was less than the average). Compared with the last result at 2000 mm/min, the cutting width at the same point recorded a reading of 1.453 mm. We found that decreasing the cutting feed rate increased the cutting width at the same point and thus reduced the quality of the cut. We also noted that the cut-off at the angle that resulted from the intersection of the vertical line and the curve of the circle produced a larger cut-off width. Figure 7 shows a comparison at the angle resulting from the convergence of the vertical line with the curve of the circle on the path represented by Point 2. The minimum width of the cut-off path was recorded at 2819 mm at 4000 mm/min. The same point recorded the highest cutting width with a value of 5.695 mm (a larger width value) at 500 mm/min. There was an increase in the cutting width as a result of reducing the feed rate and reducing the quality of the cut. The irregular cutting of the circumference of the circle was caused by the lack of true hole technology in the machine and was not due to the plasma technology. Points 3, 4, 5, and 6 in Figure 2, which were located on the circumference of the circle on the cutting path, recorded cutting widths with readings of 1.286, 1.666, 1.404, and 1.521 mm at 7000 mm/min (Figure 8). The same points recorded higher cut-off readings with respective values of 1.605, 2.293, 2.327, and 2.331 mm (lower and below-average width values) at the minimum 500 mm/min. At 12,000 mm/min, at Point 7 ( Figure 2) the cutting width with the lowest reading of the obtuse angle had a value of 1316 mm (the cutting width was at the minimum, as were all the readings for the rest of the points). This was already explained in [14]. We found that the feed rate at the maximum produced a minimum cutting width and a higher cutting quality. Compared to the minimum feed at 500 mm/min, it had the highest cutting width, which was 2117 mm (Figure 9). At 12,000 mm/min, Point 8 (the obtuse angle on the cutting path) recorded a lower cutting width of 1.754 mm (Figure 9). At 4000 mm/min, the same point recorded (Figure 7) an average reading of the cutting width of 2.591 mm (most of the readings for the rest of the specimens were below average). This also confirmed that the cutting width at the obtuse angle was increased by decreasing the cutting feed rate and decreasing the cutting quality. The cutting width of Point 9, which was located on the straight line on the cutting path, was recorded. In Figure 9, the lower cutting width was 1.504 mm at 12,000 mm/min. The same point in Figure 8 recorded a lower reading for the cutting width with a value of 1.665 mm at 7000 mm/min. This indicated that the width of the straight line cut increased slightly by decreasing the cutting feed rate and by decreasing the quality of the cut. The lower feed rates recorded below-average readings, and the higher feed rates recorded lower readings and produced incomplete cuts for some parts of the cutting path. The small thickness of the copper resulted in a high melting point. The large heat input was obtained from the PA. The molten copper was removed from the kerf [3]. The large heat contributed to the occurrence of melting of the metal, and with high feed rates, it produced incomplete cuts at some parts on the cutting path. This was proven in brass [15]. The current paper agreed with the results of the study in [10] in that plasma was not the optimal technique for cutting copper.
With the decrease in the cutting feed rate, the acute angle cutting width increased significantly, and the cutting quality decreased along with it and produced the largest cutting width. The cutting width was of poor quality at all acute angles on the cutting path. At 5000 mm/min, Point 10 (the acute angle on the cutting path at 500 mm/min) ( Figure 6) recorded a maximum cutting width of 7.141 mm (most of the feed rates recorded high and above-average readings). In comparison, the same point at 5000 mm/min ( Figure 10) showed the lowest reading for the cutting width with a value of 3.762 mm. At 500 mm/min, Point 11 (the obtuse angle on the cutting path) recorded a lower cutting width of 1.494 mm ( Figure 6). In comparison, at 4000 mm/min, the same point was recorded (Figure 7) as a below-average reading when the cutting width was 2.023 mm. This result confirmed that the obtuse angle's cutting width increased slightly by decreasing the cutting feed rate and that the cutting quality decreased with it. However, the rest of the points recorded lower readings for the cutting width. Figure 2 also showed the right angle to the cutting path shown in Figure 7 with a maximum cutting width of 2.718 mm at 4000 mm/min. The same point at 12,000 mm/min ( Figure 9) recorded a lower reading of the cutting width with a value of 1.937 mm. This showed that the width of the right-angle cutting increased normally while decreasing the cutting feed rate. The cutting quality decreased with the cutting feed rate and produced an average cutting width at the right angle on the cutting path. Point 13 in Figure 2 represents the angle resulting from the convergence of the horizontal line with the curve of the circle on the cutting path. It showed the lowest reading for the cutting width with a value of 2.161 mm at 7000 mm/min (Figure 8). However, it recorded the highest reading for the cutting width of 2.824 mm at 8000 mm/min ( Figure 11). The readings for the rest of the points were medium and close, and the type of feed rate used did not cause a large variation. Figure 11. The cutting path width of (i1,i2) at 8000 mm/min at 30 A for Point 13.

Point 12 in
Points 14, 15, and 16 in Figure 2, which were located on the curved line on the cutting path, recorded the lowest readings of 1.403, 1.14, and 1.49 mm at 12,000 mm/min ( Figure 9). However, as shown in Figure 6, the same points recorded higher readings at a lower 500 mm/min with values of 2.247, 1.659, and 2.43 mm, respectively. Most of the readings at these points for all points indicated that the lowest cut width accompanied the highest cut feed rates. The readings of the cutting width were lower than average with lower cut feed rates, which provided a good cut quality. Point 17 in Figure 2 represents the angle that resulted from the convergence of the circle curve with the horizontal line on the cutting path, which recorded the lowest reading for the cutting width with a value of 1.951 mm for 2000 mm/min ( Figure 6). However, as shown in Figure 6, the reading of the cutting width at the same point increased by 2.67 mm when the feed rate was reduced to 500 mm/min, and this decreased the quality of the cut. All the readings for the rest of the points were less than average. Point 18, which represented the end of the cutting path, recorded the lowest reading of the cutting width with a value of 2.068 mm at 10,000 mm/min. The same point recorded the highest reading for the cutting width with a value of 2.675 mm at 6000 mm/min ( Figure 12). All readings were average for the rest of the points.
Based on the above data, the study showed that the higher feed rates produced cutting widths at the minimum. This had been reported previously [23]. They were good at the beginning of the path, on the straight and curved lines, at obtuse and right angles, and where the horizontal line met the curve of the circle. The higher feed rates provided a cutting width in the upper limit in the angle resulting from the connection of the vertical line with the curve of the circle. This was also true of the acute angle as well. A lower cutting feed rate also produced a slightly larger cutting width except for the angle caused by the vertical line connecting the curve of the circle and at the acute angle, where the width was at its maximum. The incomplete cutting resulting from the higher and medium feed rates was due to the plasticity of copper caused by its high melting point.   It is clear in the two series columns (columns 2 and 10 in the graphs in Figure 13a,d) that the largest values for the cutting width were at Points 2 and 10 on the cutting path, respectively, at the minimum feed rate of 500 mm/min. The lowest values for the cutting widths were at the 3, 7, and 15 series curves at Points 3, 7, and 15 in the graph in Figure 13a,c,e, which were recorded at the maximum 12,000 mm/min.

Surface Roughness
The types of cutting-surface edge roughness were minimal at the lower cutting feed rate of 500 mm/min, and they produced a higher cutting quality. All of the types of roughness were at their maximum at the higher feed rate of 12,000 mm/min, and they produced a lower cutting quality at all points when using a 30 A current. A significant effect of the feed rate and surface roughness on the quality characteristics was observed previously [24]. Table 7 illustrates the results of the readings of the roughness measurements at the edge of the cutting surface for Points 1 and 2 on the straight lines for 500 and 12,000 mm/min for all types of roughness (Ra, Ry, Rz, and Rq). The mean readings for the roughness (Ra), the maximum height of the roughness (Ry, Rz (Rt)), and the mean square root of the roughness deviations (Rq) were taken for the surface of the material before cutting for all types of roughness for 500 and 12,000 mm/min. Previous research showed that the surface roughness is one of the preferred features in quality studies of surfaces [4].
The readings were as follows: Ra = 0.33 µm, Ry, Rz = 1.44, and Rq = 0.39. After using 500 and 12,000 mm/min for Points 1 and 2 and based on our examination thereof, the following was found.
The maximum reading of the average roughness of the cutting surface edge for 500 mm/min was Ra = 2.82 µm at Point 1. The lowest reading of Ra = 2.5 µm at Point 2 was also located on the edge of the cutting surface. It returned the lowest reading compared to all readings at 12,000 mm/min, which was the maximum reading of the average roughness of the cutting surface at 12,000 mm/min with Ra = 8.9 µm at Point 2, while the lowest reading of the average cutting edge roughness for the same feed rate was Ra = 4.08 µm at Point 1.
Accordingly, we concluded that the average roughness or the so-called Ra was low and that the quality of the cut reached the highest level at a cutting feed rate of 500 mm/min, while the average roughness was high and the quality of the cut was the lowest at 12,000 mm/min. At Point 1, which was located on the edge of the cutting surface at 500 mm/min, the highest reading of the maximum roughness height was Ry, and Rz = 14.12 µm. At Point 2 for the same feed rate, the lowest reading of the roughness height was Ry, Rz = 13.93 µm at a cutting feed rate of 500 mm/min, whereas the maximum reading of the roughness height at Point 2 for 12,000 mm/min was Ry, Rz = 46.37 µm. The lowest reading of the roughness height for the same cutting feed rate at Point 1 was Ry, Rz = 21.22 µm at 12,000 mm/min. Therefore, we found that the surface roughness or the so-called Ry, Rz was higher at the maximum 12,000 mm/min. Similar results were recently found [23]. The quality of the cutting surface was at the lowest level at the lowest cutting feed rate. The surface roughness (Ry, Rz) was lower at the minimum 500 mm/min. Therefore, the quality of the cutting surface was higher. This result was also proved at 400 mm/min [12].
The highest reading of the mean square root of the surface roughness deviations at Point 2 was Rq = 3.3 µm at 500 mm/min. The lowest reading was for Rq = 3.42 µm at Point 1. By comparing the result readings of 500 mm/min with the readings of 12,000 mm/min, it was found that the highest reading of the mean square root of the surface roughness deviations at Point 2 was Rq = 11.04, and the lowest reading at Point 1 was Rq = 5.11. It followed that the lower feed rate produced fewer surface roughness deviations than the higher feed rate (500 and 12,000 mm/min, respectively).
In addition, when considering all kinds of roughness at all of the measuring points, 500 mm/min was the best rate to achieve a higher cut quality. Figure 14 shows the surface roughness locations for 500 and 12,000 mm/min. All types of roughness at all measurement points at 12,000 mm/min were the worst for cutting quality and the cut view of Specimen 13 from the front and back. Figure 15 shows a graph of the results of the readings of the cutting surface roughness measurements for the types of roughness (Ra, Ry, Rz, and Rq) for feed rates of 500 and 12,000 mm/min; two points on a straight line were measured as a representative of Specimens 1 and 13. When interpreting the graphs, it was found that all the readings for Points 1 and 2 with the lowest feed rate had small values at 500 mm/min and showed a lower surface roughness and a higher cutting surface quality. The readings for Points 1 and 2 at the maximum at 12,000 mm/min were large and showed a greater surface roughness.
The lowest reading of all types of roughness for specimen 1 at Point 2 was 2.5 µm, which showed that these parameters yielded the highest cutting quality at the lowest cutting feed rate.
The graph above also shows that all the readings of the roughness measurements varied to some extent from low to high and from the lowest cutting feed rate to the highest feed rate.

Microstructure
As shown in Figure 16a-h, the imaging of the surface of the affected area was graded to display the cross section of the cutting edge of the surface before and after cutting at 500 mm/min with varying areas of 200, 100, 50, and 40 µm. The PAC resulted in a melting area with a depth of the metal thickness of 1000 µm below the cutting edge. When examining the morphology of the metal before and after cutting at the cross section, we found that there was no change in the microstructure of the metal. The cross section of the cutting areas under the SEM were displayed in slides that are shown in Figure 16 below. We noted that there was no change in the shape of the copper microstructure after cutting. We observed that the surface quality had not changed. We also found that the heat input did not affect the microstructure of the copper after cutting with the PA. However, areas of the microstructure were melted. In a comparison between the shapes shown in Figure 16a-h using SEM observation, there were no differences found between them.
The images in Figure 16a-d show the microstructure of the base metal. After using cutting feed rates, the cross sections of the cutting were imaged at a rate of 500 mm/min (Figure 16e-h). The feed rate did not have a significant effect on the shape of the structure, and the cutting surface was not significantly affected by the feed rate. The part close to the cut showed almost the same shape as the base metal, as well as molten metal and less surface roughness. The cutting quality was good at the same rate as presented previously. Images e, f, g, and h in Figure 16 were the best images taken at the cross section of the cutting.
We found that the microstructure of the area surrounding the surface of the copper-1 cut, which was thermally affected, did not change significantly. The melting zone result was shown when the edge of the cutting surface was displayed along the cutting path. Therefore, the PA operation of copper resulted in melting at the edge of the cutting surface.