Cleaning of Graphite Particles Embedded in the Surface of Ductile Iron by Using a Novel Method

Ductile iron has unique mechanical property and has been widely used in many industrial applications, e.g., engine cylinder covers, crank axles, and machine tool beds and cams. +e welding performance of ductile iron is normally influenced by graphite particles in the surface of ductile iron, which needs to be removed before welding. In this article, laser cleaning technique was developed to remove graphite particles implanted in the surface of ductile iron. Laser cleaning parameters and the damage threshold value of the substrate were investigated by using a pulsed Nd: YAG laser. +e optimized laser cleaning parameters were obtained to achieve high-quality cleaning effect and avoid the formation of the oxide layer. Surface morphologies and elemental compositions of specimens before and after laser cleaning were characterized by scanning electron microscopy and energydispersive X-ray spectroscopy. Results indicated that graphite particles implanted in the surface of ductile iron were removed completely and efficiently by using the pulsed laser without the protection of inert gas atmosphere, and surface oxidation was not observed during the laser cleaning process.


Introduction
Ductile cast iron is used extensively in different industrial fields for the manufacturing of engine cylinder covers, crank axles, machine tool beds and cams, and so on. It is a potential replacement of steel in some applications due to its excellent casting performance, extraordinary abrasive resistance, outstanding corrosion resistance, and the well combination of toughness and strength [1][2][3][4]. Phase compositions of ductile cast iron are pearlite, ferrite, and graphite phases, with ferrite as the matrix. However, graphite particles implanted in the surface of ductile iron could affect the quality of coating, reduce the welding performance, and short the service life of the workpiece. ose particles need to be removed completely before welding [5]. Techniques have been developed by researchers to remove graphite particles, in which the sand-blasting method is the most commonly used. Sand blasting [6] is an efficient and effective technique for the cleaning of the large surface area. However, it is time and energy consuming for the cleaning of small workparts of ductile iron. Chemical etching [7] is not suitable for the removal of graphite particles due to their chemical stability. Recently, laser cleaning technique has been developed as a novel method to remove surface contaminants [8][9][10][11]. Comparing with traditional surface treatment techniques, laser cleaning is environmentally friendly for direct removal of contaminants without altering or affecting properties of matrix.
Different from continuous lasers for cutting, welding, and cladding, pulsed laser source was employed for laser cleaning. Laser cleaning is a multifactor process with mechanism of evaporation, ablation, acoustic effect, and impact effect (Figure 1) [12]. Intensity and spatiotemporal characteristics of radiation play the main role during laser cleaning process. Cleaning effect is also related to the absorbing ability of the material, which strongly depends on temperature, thermal conductivity, and other factors.
When laser irradiates on metal surface, the metal partially absorbs the laser radiation, and it will be heated, melted, partially evaporated, and ionized. A rigorous mathematical description of the interaction taking into account all participating physical processes is complex.
Depending on the power density of the radiation impacting the surface, laser cleaning can be implemented by evaporation process, impact process, or acoustic and thermoplastic forces [12,13]. Evaluations indicate that effective cleaning due to evaporation for most materials (Figure 1(a)) is achieved at the power density of 10 7 -10 8 W/cm 2 . A low threshold thermal-chemical impact ablation mechanism of the film can be implemented under the effect of pulsed laser radiation with power density lower than 10 7 W/cm 2 [12,14] ( Figure 1(b)). Laser cleaning technique has been successfully applied to removal of paint and oxide layer on metal surfaces, cleaning of optical and electronic elements, and conservation of cultural heritage [15][16][17][18][19][20][21][22]. However, few research studies have been carried out on laser cleaning of ductile iron.
Inspired by the effectiveness of the laser cleaning technique for the removal of surface contaminants, in this work, the infrared high-repetition nanosecond pulsed laser was employed to remove graphite particles implanted in the surface of ductile iron. e technical details of laser cleaning and removal possibilities of graphite particles with different implantation forms were discussed. Cleaning effects of different laser parameters were investigated. Surfaces before and after laser treatments were characterized by scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). e optimized cleaning parameters and the damage threshold value of the substrate were obtained. Results indicated that laser cleaning is an effective and environmentally friendly technique for the removal of graphite particles, which will not lead to degradation of essential performance of the ferrite matrix.

Sample Preparation.
Different from the contaminants adhered to the surface of substrate, for ductile iron, graphite particles are embedded in the ferrite matrix, and it is technically more difficult to be removed. e cleaning process and laser parameters are different from those for cleaning of the dusts, rubber particles, and paints from the substrate. If the shape of graphite particles is assumed as a regular sphere, the implanted particles have 4 forms in the iron substrate as shown in Figure 2.
It is obvious that shallowly embedded particles can be removed easily. Particles A and B in Figure 2 can be removed by laser ablation. However, for particle C, evaporation should be the main mechanism. A few particles like D, because of their small exposed area, are difficult to be removed completely by laser cleaning or sand blasting. Furthermore, it is difficult to calculate the adhere force between graphite particles and matrix, which is necessary to decide laser parameters when ablation is the main mechanism in a cleaning process. In this case, evaporation mechanism is more suitable for ductile iron cleaning. e boiling point of graphite is higher than that of iron, and it is much higher than melting point of iron; therefore, iron substrate would melt and gasify during laser cleaning process. Melting and gasification of iron substrate are acceptable phenomenon. If a power intensity higher than 10 8 W/cm 2 is used during laser cleaning process, oxidation may be induced, which is not acceptable for further treatment of iron substrate. Iron oxide will be formed quickly in atmosphere when the temperature is higher than 5000°C.
In this work, a ductile iron club was chosen as the research target, and it was cut into small specimens with a dimension of ϕ20 mm × 8 mm. Specimens were cleaned by using alcohol to remove particulate and organic contaminants from the surface.

Experimental Setup.
e schematic diagram of the laser cleaning system is shown in Figure 3. Specimens were fixed on a platform moving along Y-axis controlled by a computer. Laser beam movement on surfaces of specimens was controlled by a directional control system in X-axis, and its movement in Y-axis was achieved by the movable platform. High power densities were guaranteed by an F-θ lens, which focused the laser beam on the surface of specimen with a spot diameter of 500 μm at focal plane. Values of fluence were adjusted by varying the output power while keeping the repetition rate and pulse duration the same. To obtain the threshold fluence, several fluence values were chosen to determine the optimal condition for the graphite removal. Surfaces were cleaned several times to remove the implanted graphite particles completely. Inert gas was not applied to the irradiated area during the laser cleaning process.
A Q-switched Nd:YAG laser system with a maximum power of 400 W, repetition rate of 5-100 kHz, and pulse duration of 60-100 ns was employed in laser cleaning experiments.
e diameter of laser spot at focal plane was 500 μm. During the experiment, the repetition rate of the laser was much higher than the scanning rate, ensuring the complete surface coverage during cleaning process. Meanwhile, the scanning rate was matched to the repetition rate, ensuring a small overlap rate (20%) to reduce the influence of energy accumulation.

eoretical Calculation.
Considering that the radiation is absorbed in the near-surface metal layer, while parameters such as diffusivity α t and thermal conductivity K are constant, the heat conduction equation can be written as follows [12]: where T(z, t) is the temperature of the material at depth z after time t, P is the power of a single pulse, S is the area of spot, T 0 is room temperature, R is reflection of ductile iron, and ierfc () is the error function. e value of ierfc () varies with values of depth z and time t. Critical density of the radiation flux to attain the boiling point on the surface of the material can be determined by the temperature field. Critical temperature of Fe substrate can be obtained by using equation (1), and calculated results are shown in Figure 4. ermal parameters of substrate are given in Table 1. Because inert gas is not applied, cleaning parameters especially oxidation threshold of laser power intensity should be obtained and optimized. According to the calculated results in Figure 4, oxidation threshold of laser power intensity is about 10 J/cm 2 .

Experimental Demonstration.
According to equation (1), a high temperature is needed to completely remove graphite particles. us, a high laser power intensity is required. ere are two routes to increase the power intensity, one is to increase the output laser power and the other is to decrease the spot area. Increasing output power is much easier in comparison with decreasing the spot area. In this research, the average power was increased while maintaining other experimental parameters the same as shown in Table 2.
e repetition rate was 15 kHz, the pulse duration was 100 ns, and the scanning velocity was 3200 mm/s. e surface of the specimen was placed at the focal plane of the laser beam. e cleaning passes were carefully chosen. Cleaned specimens were judged by their appearance first and then evaluated by the combination of SEM and EDS analysis.
Based on previous analysis and the conclusion of reference [14], initial testing experiment on S0 is performed with a laser radiation power density of 6.79 ✕ 10 7 W/cm 2 (6.79 J/cm 2 ). As shown in Figure 5, region A is the surface before cleaning and region B is the surface after cleaning. e color of cleaned surface is silver white, much brighter than that of original surface, and is approaching to the color of iron. It seems that the expected appearance has been achieved by only one-step laser cleaning. Figure 6 shows detailed views of surface morphologies of initial ( Figure 6(a)) and laser cleaned ( Figure 6(b)) specimens. As previously mentioned, different from particulate contaminants adhere to the surface, Figure 6(a) shows that graphite particles (black area, e.g., C) with diameters ranging from 10 μm to 50 μm embedded in the matrix of ductile iron (gray area, e.g., D). Figure 6(b) indicates that graphite particles with diameters less than 10 μm were completely removed (e.g., site 1), leaving implanted holes. Deeply embedded particles with diameters between 10∼50 μm were not completely removed (e.g., site 3). Shallowly implanted large particles were also completely removed (e.g., site 2). Comparing with Figure 6(a), it can be seen that the ferrite matrix melted due to the high laser radiation power density, which is in agreement with our previous analysis. In this case, evaporation is the main mechanism for ductile iron cleaning. e boiling point of graphite (4830°C) is higher than that of iron (2750°C), and it is much higher than melting point of iron (1535°C); therefore, iron substrate melt and be gasified during laser cleaning process. Melting and gasification of iron substrate are acceptable for further welding treatment of ductile iron.
EDS analysis was employed for further investigation of the oxidation process during laser cleaning. Figure 7 shows the EDS results of regions C and D of initial specimen (Figure 6(a)) and region E of the cleaned specimen ( Figure 6(b)). It can be seen from Figure 7(a) that the main elements of region C (graphite particle) and D (ductile iron substrate) are carbon and iron, respectively, which are consistent with the composition of ductile iron. Comparing with Figure 7(a), Figure 7(b) shows that the composition of iron substrate after laser cleaning is similar to that of the initial status, which demonstrates that oxidation does not occur on iron substrate during laser cleaning process. e result indicates that laser cleaning is a promising method to remove graphite particles implanted in ductile iron while preserving the mechanical property of the substrate.     However, deeply embedded graphite particles still exist after laser cleaning. In following experiments, laser cleaning parameters will be optimized to improve the remove efficiency. Figure 8(a) is the specimen cleaned by using the laser with the power of 340 W (region F) and 320 W (region H). Damage is observed on the surface of the specimen cleaned by using the 340 W laser. SEM/EDS analysis of the circled part of region F was performed, and its results are shown in Figure 8(b). It can be seen that the atomic ratio of O is 58.08%, of O is 58.08 at.%, demonstrating that oxidation occurred during 340 W laser cleaning process. e SEM image in Figure 8(b) reveals that graphite particles were removed but the appearance of the substrate also changed. Some loose materials formed on the surface such as region I could be iron oxide. e appearance of region H is similar to that of region B in Figure 5.

Mathematical Problems in Engineering
e SEM image of region H is shown in Figure 8(c). It can be observed that there are still many residual graphite particles. e number of removed particles is more than that of the specimen in Figure 6(b), demonstrating that increasing power intensity to a certain value below damage threshold is a proper way to obtain a better cleaning performance. According to above results, it can be concluded that the safe power intensity for the cleaning of ductile iron substrate is about 320 W, corresponding to 10.86 J/cm 2 (1.086 × 10 8 W/cm 2 ), which is close to the calculated result. According to literatures [10,[23][24][25], the thick surface layer can be completely removed by increasing cleaning times. us, in following experiments, average laser power is adjusted to 320 W and other parameters are kept the same with prior experiments. e number of laser passes will be increased and tested to improve the cleaning efficiency. Figure 9(a) is the SEM image of a specimen cleaned 3 times by using the 320 W laser with 60 s time gap.

Mathematical Problems in Engineering
Comparing with Figure 8(c), it can be seen that most of graphite particles were removed except those with large diameters or deeply embedded ones. e increasing of cleaning times is a feasible way to obtain an excellent cleaning result. Figure 9(b) shows EDS analysis of the surface after the third pass. e ingredients of substrate (region J) are similar to those of the initial status (Figure 7(a)). However, the ingredients of the graphite area (region K) are different from those of the initial surface (Figure 7(a)), indicating that surface graphite particles were efficiently removed. Oxidation was not observed during laser cleaning process. erefore, the power intensity of 1.087 × 10 8 W/cm 2 is a safe parameter for the cleaning of ductile iron substrate. Time duration more than 60 s between cleaning passes is necessary. Figure 9(c) shows the SEM image of a specimen cleaned 3 times by using the 320 W laser but without time gap, which is similar to the image inset in Figure 8(b). Loose materials (e.g., region L) are observed on the surface of ductile iron. It can be concluded that oxidation occurred during laser cleaning process without time gap.
Decreasing the output power is extremely important for the real application of laser cleaning technique, which may save power, shorten the cleaning time, and decrease the oxidation rate during continuous laser cleaning process without time gap. us, in following experiment, the output power was decreased to 290 W (9.85 × 10 7 W/cm 2 ) and cleaning processes were increased to 5 times without time gap between passes. Other experimental parameters were kept the same. e SEM image of the cleaned specimen is shown in Figure 10. Results showed that most of graphite particles were removed, and no region like region L in Figure 9(c) was observed. It can be concluded that more passes should be performed at a lower output power to optimize laser cleaning effect.     Figure 9: (a) e SEM image of a specimen cleaned 3 times by using the 320 W laser with 60 s time gap; (b) EDS analysis of a specimen cleaned 3 times by using the 320 W laser with 60 s time gap; (c) the SEM image of a specimen cleaned 3 times by using the 320 W laser without time gap.

Conclusions
In this article, a novel and environmental friendly method is developed to remove graphite particles embedded in the surface of ductile iron to improve its welding performance. Oxidation threshold of laser power intensity of 10 J/cm 2 was obtained by calculation. Graphite particles implanted in the surface of ductile iron were removed effectively by using proposed laser technique. Experimental results showed that oxidation of substrate occurred during laser cleaning process when power intensity of single pulse is higher than 1.087 × 10 8 W/cm 2 . Most graphite particles were removed after three laser scanning passes with a power intensity of single pulse of 1.087 × 10 8 W/cm 2 . e time gap between passes is necessary in this case due to the occurrence of oxidation. e optimize power intensity of single pulse is 9.85 × 10 7 W/cm 2 , and cleaning process is 5 times with continuous laser scanning without time gap. is investigation contributes to the improvement of welding performance of ductile iron.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest regarding the publication of this paper.