A New Method to Recycle Stainless–Steel Duplex UNS S31803 Chips

: Due to the increased consumption of raw materials, energy, and the waste it generates, recycling has become very important and fundamental for the environment and the industrial sector. The production of duplex stainless–steel powders with the addition of vanadium carbide in the high energy mechanical milling process is a new method for recycling materials for the manufacture of components in the industrial sector. This study aims to reuse the chips from the duplex stainless–steel UNS S31803 by powder metallurgy with the addition of Vanadium carbide (VC). The mechanical milling was performed using a planetary ball mill for 50 h at a milling speed of 350 rpm and a ball-to-powder weight ratio of 20:1, and the addition of 3 wt % of VC. The material submitted to milling with an addition of carbide has a particle size of less than 140 µ m. After milling, the sample went through a stress relief treatment performed at 1050 ◦ C for 1 h and the isostatic compaction process loaded with 300 MPa. The sintered powders and material was characterized by scanning electron microscopy, X-ray diffraction, and micro-hardness tests. The milling process with an addition of 3% VC produced a particle size smaller than the initial chip size. The measurement of micrometric sizes obtained was between 26 and 132 µ m. The sintered material had a measurement of porosity evaluated at 15%. The obtained density of the material was 84% compared to the initial density of the material as stainless–steel duplex UNS S31803. The value of the microhardness measurement was 232 HV. The material submitted for grinding presented the formation of a martensitic structure and after the thermal treatment, the presence of ferrite and austenite phases was observed. Thus, in conclusion, this study demonstrates the efﬁcacy in the production of a metal-ceramic composite using a new method to recycle stainless–steel duplex UNS S31803 chips.


Introduction
Nowadays, a great motivation for metal research is developing ways to optimize steel for the property of application through cost-effective processing routes [1]. In the stainless-steel family, the stainless-steel duplex UNS S31803 has been widely used in many industrial sectors because of its surface physicochemical properties and resistance to corrosion, abrasion and antioxidation [2,3]. Duplex stainless-steels are steels in which the microstructure consists of two main phases: austenite (y-FCC) and ferrite (α-BCC) in equal proportions, which can combine the good properties of the ferritic and austenitic phases [3][4][5][6].
conductivity [34,35]. The addition of these carbides has also been shown to improve the efficiency of the milling process. Kufffer et al. (2015) studied the addition of niobium carbide in the high-energy milling of a 52,100 steel causes an increase in the efficiency of the milling process with a greater reduction of particle size compared to carbide-free milling [36]. Dias et al. (2018) added carbides which resulted in increased efficiency of the high energy milling process used in the recycling of aluminum bronze alloy chips [37].
The aim of the present work was to evaluate a new method to recycle stainless-steel duplex UNS S31803 chips. Thus, the result is to obtain reducing the cost of manufacturing in the industrial sector and recycling the materials discarded in the machining process. In this study, was evaluated the feasibility of reusing the UNS S31803 duplex stainless-steel chip by high-energy milling with the addition of vanadium carbide. Also, the isostatic press was utilized to produce a metal-ceramic composite. The obtained composite was investigated using scanning electron microscopy, X-ray diffraction, and micro-hardness tests.

Materials and Methods
The raw material in the form of chips used in the high-energy milling process was obtained by machining sheet metal of a duplex stainless-steel UNS S31803 with low rotation and no use of lubricants, thus avoiding contamination with soluble oil. The average size of the chips was characterized by binocular stereoscopic microscopy (Model Olympus SZ61, Tokyo, Japan). For the milling, 3 wt. % VC with an initial size between 5 µm and 10 µm was added in a high energy mill of the Noah-Nuoya model NQM 0.2 L (Yangzhou Nuoya Machinery CO., LTD, Yangzhou, Hanjiang District, China), the speed was 350 rpm, mass/ball ratio of 1:20 and milling time of 50 h with argon atmosphere to avoid the oxidation of the powders. After milling, a pre-grading was carried out through a 300 µm sieve. A particle size analyzer (Microtrac model S3500, Microtrac Global Location, Montgomeryville, PA, USA) was used to collect the particle sizes and their distribution. The characterization of the obtained powders was carried out using a scanning electron microscope-Carl Zeiss EVO MA 15 (ZEISS Microscopy, Cambridge, UK)-in the secondary electron mode (SE), backscatter and energy dispersion (EDS). The sample was polished with colloidal silica before microstructural analysis.
Thereafter, a stress relief treatment was performed at 1050 • C for 30 min on the furnace. The samples were quenched in water. For the compaction process, the powder was submitted initially to the uniaxial press (SCHULZ-Mod. 30T, Joinville, Brazil) at a load of 300 MPa to mold the initial body. After that, the samples were subjected to isostatic at compaction (model WT Industria, São Carlos, Brazil). Then, the samples were wrapped with latex in a vacuum environment (10 −1 mbar). The pressing process was done at room temperature in three steps. In the first step, the samples were allowed to stand for 5 min at 150 MPa pressure, then 250 MPa for 5 min and finally at 350 MPa for 20 min. Samples were placed in one oil-encased chamber. So, a uniformly distributed load was applied. The samples were sintered in a vacuum muffle (Fortelab, São Carlos, Brazil) resistive furnace for 1 h at 1250 • C with a rated heating of 5 • C/min and they are left to cool in the furnace.
The identification of the phases in the microstructure was made using the X-ray diffraction technique (PANalytical X'Pert Pro model B.V., Almelo, The Netherlands).The parameters used were: scan angle from 45 to 135 • , a pitch of 0.02 • /s, and cobalt tube (λ = 1.7890 A • ). For the conditions as received, and sintered, 10 measurements of Vickers microhardness with a load of 50 gf were carried out in a Micro-hardness, Tester HV 1000 of the Digimess (São Paulo, Brazil). The dwelling time of the micro-hardness test is 15 s. In order to do the analysis of the grain boundaries, the samples were sanded, polished and subjected to an electrolytic attack at 5 V for 10 s with the oxalic reagent.
The green density of the samples was obtained by mass/volume ratio. The density after sintering was taken by mass/volume ratio-and also by the Archimedes principle-according to the ASTM STANDARD C 20 [38]. For the determination of the percent porosity in the sintered samples, five images of each sample were taken and an average was calculated. The images obtained by Optical microscopy were analyzed with the (Stream Basics 1.9.1. software, Olympus, Waltham, MA, USA). Figure 1 shows the initial characterization of UNS S31803 steel chips obtained by machining. Basically, chip formation is a shear process involving plastic deformation within the shear zone [14,39]. The encircled region in Figure 1 shows the shear area of a particle caused by plastic deformation. This process produced helical-shaped chips. It is observed that the obtained chips have an average size of 8 mm. Figure 2 shows an enlargement of the regions of plastic deformation resulting from the shear caused by the machining.  Figure 1 shows the initial characterization of UNS S31803 steel chips obtained by machining. Basically, chip formation is a shear process involving plastic deformation within the shear zone [14,39]. The encircled region in Figure 1 shows the shear area of a particle caused by plastic deformation. This process produced helical-shaped chips. It is observed that the obtained chips have an average size of 8 mm. Figure 2 shows an enlargement of the regions of plastic deformation resulting from the shear caused by the machining.   Figure 3 shows the particle morphology obtained by SEM after high-energy milling with the addition of 3% vanadium carbide, the chip-shaped material was transformed into particles with irregular morphology and heterogeneous size, ranging from 10 to 150 μm. The material submitted to milling with the addition of carbide has a particle size of fewer than 140 μm and an average size of 67.42 μm.  Figure 1 shows the initial characterization of UNS S31803 steel chips obtained by machining. Basically, chip formation is a shear process involving plastic deformation within the shear zone [14,39]. The encircled region in Figure 1 shows the shear area of a particle caused by plastic deformation. This process produced helical-shaped chips. It is observed that the obtained chips have an average size of 8 mm. Figure 2 shows an enlargement of the regions of plastic deformation resulting from the shear caused by the machining.   Figure 3 shows the particle morphology obtained by SEM after high-energy milling with the addition of 3% vanadium carbide, the chip-shaped material was transformed into particles with irregular morphology and heterogeneous size, ranging from 10 to 150 μm. The material submitted to milling with the addition of carbide has a particle size of fewer than 140 μm and an average size of 67.42 μm.  Figure 3 shows the particle morphology obtained by SEM after high-energy milling with the addition of 3% vanadium carbide, the chip-shaped material was transformed into particles with irregular morphology and heterogeneous size, ranging from 10 to 150 µm. The material submitted to milling with the addition of carbide has a particle size of fewer than 140 µm and an average size of 67.42 µm.

Results and Discussion
Thus, the particle size obtained after 50 h of milling with 3% vanadium carbide was the measurement by particle size analyzer and was found to be smaller than the initial chip size. Mendonça et al. (2017) show that the addition of carbides increases the efficiency of the milling process over carbide-free milling-this is due to the carbide particles, a material with a high hardness, which collides with the chip. This collision causes a region of generalized stresses in the ductile particle, leading to rupture through the combination of the ductile-brittle process [40]. Thus, the particle size obtained after 50 h of milling with 3% vanadium carbide was the measurement by particle size analyzer and was found to be smaller than the initial chip size. Mendonça et al. (2017) show that the addition of carbides increases the efficiency of the milling process over carbide-free milling-this is due to the carbide particles, a material with a high hardness, which collides with the chip. This collision causes a region of generalized stresses in the ductile particle, leading to rupture through the combination of the ductile-brittle process [40].  Figure 4 shows the particle size analysis, which reveals a distribution region with sizes between 20 and 200 μm. Analyzing the graph, the particle size for the D50 factor with VC was 53.07 μm with 50 h milling, which indicates that 50% of the particles have up to this size. Likewise, the reported values for D10 and D90 are 26.45 μm and 132.90 μm, indicating that 10% of particles have dimensions less than 26.45 μm and 90% of particles are below the size of 132.9 μm, respectively.  Table 1 shows the average green densities which were 5.08 g/cm³. Also, the value measurement by the Archimedes method after the sintering. After sintering, it was verified with the Archimedes method that the average density of the samples produced with adding 3% vanadium carbide was 6.57 g/cm³. It is 16% lower than that of the steel produced by the melting process with the average density was 7.8 g/cm³.  verified an increase in density values for the duplex stainless-steels starting from Fe-Cr-Ni powders produced by powder metallurgy at temperatures of  Figure 4 shows the particle size analysis, which reveals a distribution region with sizes between 20 and 200 µm. Analyzing the graph, the particle size for the D50 factor with VC was 53.07 µm with 50 h milling, which indicates that 50% of the particles have up to this size. Likewise, the reported values for D10 and D90 are 26.45 µm and 132.90 µm, indicating that 10% of particles have dimensions less than 26.45 µm and 90% of particles are below the size of 132.9 µm, respectively. Thus, the particle size obtained after 50 h of milling with 3% vanadium carbide was the measurement by particle size analyzer and was found to be smaller than the initial chip size. Mendonça et al. (2017) show that the addition of carbides increases the efficiency of the milling process over carbide-free milling-this is due to the carbide particles, a material with a high hardness, which collides with the chip. This collision causes a region of generalized stresses in the ductile particle, leading to rupture through the combination of the ductile-brittle process [40].  Figure 4 shows the particle size analysis, which reveals a distribution region with sizes between 20 and 200 μm. Analyzing the graph, the particle size for the D50 factor with VC was 53.07 μm with 50 h milling, which indicates that 50% of the particles have up to this size. Likewise, the reported values for D10 and D90 are 26.45 μm and 132.90 μm, indicating that 10% of particles have dimensions less than 26.45 μm and 90% of particles are below the size of 132.9 μm, respectively.  Table 1 shows the average green densities which were 5.08 g/cm³. Also, the value measurement by the Archimedes method after the sintering. After sintering, it was verified with the Archimedes method that the average density of the samples produced with adding 3% vanadium carbide was 6.57 g/cm³. It is 16% lower than that of the steel produced by the melting process with the average density was 7.8 g/cm³.  verified an increase in density values for the duplex stainless-steels starting from Fe-Cr-Ni powders produced by powder metallurgy at temperatures of  Table 1 shows the average green densities which were 5.08 g/cm 3 . Also, the value measurement by the Archimedes method after the sintering. After sintering, it was verified with the Archimedes method that the average density of the samples produced with adding 3% vanadium carbide was 6.57 g/cm 3 . It is 16% lower than that of the steel produced by the melting process with the average density was 7.8 g/cm 3 .  verified an increase in density values for the duplex stainless-steels starting from Fe-Cr-Ni powders produced by powder metallurgy at temperatures of 1000 • C, 1200 • C and 1400 • C. These authors found the highest values of sintered density to ferritic stainless-steel at 89% to 1400 • C following for 1200 • C with 83% of sintered density [11]. Table 1. Values of green density, sintered material density and Archimedes density of stainless-steel with the addition of 3% VC.

Sample
Density in Green (g/cm 3 )

Density of Arquimedes (g/cm 3 )
Duplex stainless-steel/3 wt % VC 5.08 ± 0.06 6.57 ± 0.08 Figure 5 shows the micrograph of the sintered sample at 1250 • C for 1 h. Furthermore, it verified the presence of uniform pore shapes distributed in the microstructure. The pores have dimensions ranging from 3 to 20 µm. The porosity is related to the particle size distribution. This facilitates atomic diffusion and densification at the moment of sintering and resulting in a microstructure with a lower porosity. However, in this study, the obtained porosity increased, although the process presented a good distribution. Abdoli et al. (2009) [41] performed the milling of Al-X wt % AlN (X = 0, 2.5, 5 and 10) composite powders in a planetary ball mill for 25 h. They found the sinterability was degraded when increasing the reinforcement content material. Due to a decrease in the compressibility and green density of the composite when increasing the amount of reinforcing material because this causes less metal-to-metal contact.
Varol and Canakci (2013) [42] investigated the effects of weight percentage and particle size of B 4 C reinforcements on the physical and mechanical properties of powder metallurgy Al 2 0 24 -B 4 C composites. Based on their study, the relative density decreased with increasing amounts of B 4 C particles in the consolidated samples. They have more corrosion in the regions with pores. 1000 °C, 1200 °C and 1400 °C. These authors found the highest values of sintered density to ferritic stainless-steel at 89% to 1400 °C following for 1200 °C with 83% of sintered density [11]. Table 1. Values of green density, sintered material density and Archimedes density of stainless-steel with the addition of 3% VC.

Sample Density in Green (g/cm 3 )
Density of Arquimedes (g/cm 3 ) Duplex stainless-steel/3 wt % VC 5.08 ± 0.06 6.57 ± 0.08 Figure 5 shows the micrograph of the sintered sample at 1250 °C for 1 h. Furthermore, it verified the presence of uniform pore shapes distributed in the microstructure. The pores have dimensions ranging from 3 to 20 μm. The porosity is related to the particle size distribution. This facilitates atomic diffusion and densification at the moment of sintering and resulting in a microstructure with a lower porosity. However, in this study, the obtained porosity increased, although the process presented a good distribution. Abdoli et al. (2009) [41] performed the milling of Al-X wt % AlN (X = 0, 2.5, 5 and 10) composite powders in a planetary ball mill for 25 h. They found the sinterability was degraded when increasing the reinforcement content material. Due to a decrease in the compressibility and green density of the composite when increasing the amount of reinforcing material because this causes less metal-to-metal contact.
Varol and Canakci (2013) [42] investigated the effects of weight percentage and particle size of B4C reinforcements on the physical and mechanical properties of powder metallurgy Al2024-B4C composites. Based on their study, the relative density decreased with increasing amounts of B4C particles in the consolidated samples. They have more corrosion in the regions with pores.  Figure 6a shows the distribution of vanadium particles on the stainless-steel particles. It was observed that the vanadium particles are located in both the matrix and the pores. Figure 6b identified the particles using EDS. It shows the presence of a vanadium chemical element, as well as the presence of iron, chromium and nickel elements, which are constituent elements of duplex stainlesssteel. In this study, the temperature used was 1250 °C, which is below the melting temperature of the vanadium carbide of 2800 °C so that the dissociation of carbon and vanadium elements does not occur. Figure 6a shows the presence of the vanadium carbide on green color in the pores. The presence of carbide on the surface of the particles makes the contact between the metal-metal particles  Figure 6a shows the distribution of vanadium particles on the stainless-steel particles. It was observed that the vanadium particles are located in both the matrix and the pores. Figure 6b identified the particles using EDS. It shows the presence of a vanadium chemical element, as well as the presence of iron, chromium and nickel elements, which are constituent elements of duplex stainless-steel. In this study, the temperature used was 1250 • C, which is below the melting temperature of the vanadium carbide of 2800 • C so that the dissociation of carbon and vanadium elements does not occur. Figure 6a shows the presence of the vanadium carbide on green color in the pores. The presence of carbide on the surface of the particles makes the contact between the metal-metal particles impossible at the moment of compaction. This situation then hinders the atomic diffusion process. Due to the impacts generated in the milling process, a decrease in the size of the carbide can be observed, with particles having sizes smaller than 1 µm. impossible at the moment of compaction. This situation then hinders the atomic diffusion process. Due to the impacts generated in the milling process, a decrease in the size of the carbide can be observed, with particles having sizes smaller than 1 μm.  Figure 7 shows the micrograph of the sample of stainless-steel compacted in the isostatic press and then sintered at 1250 °C. In order to obtain an average of the samples, five images obtained by optical microscopy were analyzed. In these images, 15% of porosity on sintered samples could be observed.
In Figure 7, the red spots represent the pores, while the white spots represent the sintered duplex stainless-steel. As reported above, at the moment of the compaction and sintering process, if this ceramic is between two particles, it hinders the diffusion process of the samples, leading to an increase in porosity.   Figure 7 shows the micrograph of the sample of stainless-steel compacted in the isostatic press and then sintered at 1250 • C. In order to obtain an average of the samples, five images obtained by optical microscopy were analyzed. In these images, 15% of porosity on sintered samples could be observed.
In Figure 7, the red spots represent the pores, while the white spots represent the sintered duplex stainless-steel. As reported above, at the moment of the compaction and sintering process, if this ceramic is between two particles, it hinders the diffusion process of the samples, leading to an increase in porosity. impossible at the moment of compaction. This situation then hinders the atomic diffusion process. Due to the impacts generated in the milling process, a decrease in the size of the carbide can be observed, with particles having sizes smaller than 1 μm.  Figure 7 shows the micrograph of the sample of stainless-steel compacted in the isostatic press and then sintered at 1250 °C. In order to obtain an average of the samples, five images obtained by optical microscopy were analyzed. In these images, 15% of porosity on sintered samples could be observed.
In Figure 7, the red spots represent the pores, while the white spots represent the sintered duplex stainless-steel. As reported above, at the moment of the compaction and sintering process, if this ceramic is between two particles, it hinders the diffusion process of the samples, leading to an increase in porosity.  The microstructure analysis in Figure 8 shows images of duplex stainless-steel samples sintering at 1250 • C. The grain contours can be observed in regions with larger particles. The presence of pores with dimensions between 5 and 40 µm is identified and the presence of grain with an average size of 20 µm is observed inside the particles. In Table 2 it is observed that the values of microhardness in the steel as received decreased from 284.16 HV0.05 to 232.02 HV0.05 for the sample subjected to milling with the addition of 3% VC and is sintered to 1250 °C. The decrease in the microhardness of the sintered sample in relation to the sample of the material as received is probably due to the presence of pores in the sintered sample and the bad atomic diffusion that occurred during the sintering at 1250 °C. The hardness value obtained was 87% compared to the sample as received and it was verified that the carbides help in the milling process, increasing the process efficiency but reducing the hardness values, being related to the presence of pores and not to the dissociation of these matrix elements, which is concentrated in regions with pores.

Condition
Microhardness (HV) Sample sintering-1250 °C-1 h 232 ± 8 Material in condition as received 265 ± 4 Figure 9 shows the X-ray diffractogram under the following conditions: material as received, 50 h milled material after the tension relief treatment process, and the sintered material at 1250 °C. Note that X-ray diffraction peaks are enlarged due to the decrease in particle size, due to the intense plastic deformation that occurs during high-energy milling [43] as well as the high density of dislocations due to the high degree of deformation imposed by the process [16]. During milling, different processes such as structural defects, amorphization and reduction in grain size occur and this results in the broadening of diffraction peaks [11]. It can be seen in the diffractogram of Figure 8   In Table 2 it is observed that the values of microhardness in the steel as received decreased from 284.16 HV0.05 to 232.02 HV0.05 for the sample subjected to milling with the addition of 3% VC and is sintered to 1250 • C. The decrease in the microhardness of the sintered sample in relation to the sample of the material as received is probably due to the presence of pores in the sintered sample and the bad atomic diffusion that occurred during the sintering at 1250 • C. The hardness value obtained was 87% compared to the sample as received and it was verified that the carbides help in the milling process, increasing the process efficiency but reducing the hardness values, being related to the presence of pores and not to the dissociation of these matrix elements, which is concentrated in regions with pores. Table 2. Microhardness values of stainless-steel for different conditions.

Microhardness (HV)
Sample sintering-1250 • C-1 h 232 ± 8 Material in condition as received 265 ± 4 Figure 9 shows the X-ray diffractogram under the following conditions: material as received, 50 h milled material after the tension relief treatment process, and the sintered material at 1250 • C. Note that X-ray diffraction peaks are enlarged due to the decrease in particle size, due to the intense plastic deformation that occurs during high-energy milling [43] as well as the high density of dislocations due to the high degree of deformation imposed by the process [16]. During milling, different processes such as structural defects, amorphization and reduction in grain size occur and this results in the broadening of diffraction peaks [11]. It can be seen in the diffractogram of Figure 8 that after the milling process, there was a decrease in the intensity and the broadening of the peaks of the ferritic (110) and austenitic (111) phase. Note also the disappearance of the peaks of the austenite phase (200), (200), (311) and (222) and the relative peak intensity of the ferrite (220). The presence of deformation-induced martensite was also observed. Mendonça et al. (2017) verified the appearance of the martensite induced by deformation after the milling process, by increasing the values of saturation magnetization obtained by magnetic measurements [22].
The material was subjected to the stress-relief treatment at 1050 • C in the vacuum. After this process ferritic and austenitic phases could be observed. The presence of the 3% vanadium carbide was not observed on X-rays because it is below the detection limit of the apparatus. The material was subjected to the stress-relief treatment at 1050 °C in the vacuum. After this process ferritic and austenitic phases could be observed. The presence of the 3% vanadium carbide was not observed on X-rays because it is below the detection limit of the apparatus. Moreover, it can be seen in the diffractogram in Figure 10, in the regions of angles between 50° and 54°, that after the milling process there was a decrease in the intensity and the broadening of the peaks of the ferritic (110) and austenitic (111) phase, compared to the material as received (in the form of a sheet), this occurs due to the non-uniform plastic deformation (microtension) of the crystal line lattice resulting from the stacking faults, among other crystalline defects resulting from the grinding process and the reduction of particle size [44].
After heat treatment, there is a decrease in the width of the phase peaks and also an increase in the height of the phases. This behavior is associated with the reduction of micro-tensions caused by annealing. After the process of stress relief and sintering, the presence of the ferrite phase and the austenite phase is observed. The presence of the martensitic phase was not observed after the heat treatment process. The deformation of martensite induced is not stable at high temperatures and the reversion of martensite to austenite occurs by soaking at elevated temperatures [45][46][47][48]. Moallemi et al. (2017) [49] realized the cold rolling-reversion annealing was carried out at various temperatures in the range of 750-1100 °C. These authors verified that there was an increase in the reversion of the martensite fraction with the increase in temperature and a complete reversal occurred at 1050 °C. For the austenitic steels of the 304 family the total martensite reversion occurred at 750 °C [50]. Figure 10 shows a peak shift to the right for the ferrite phase (110) in the angle 52.14° for the treatment of tension relief of 0.08° and 0.12° after sintering and also a displacement of 0.09 after the stress relief treatment and 0.14° after sintering for the austenitic (111) phase, which characterizes the displacement of crystalline planes of the material after the sintering process.
In the industrial sequence for obtaining duplex stainless-steel by rolling includes two annealing steps: one intermediate after the hot rolling and the final one after the cold rolling [51,52]. In most applications, the duplex structures are manufactured from hot-rolled and annealed sheets. For this reason, its crystallographic texture results from the phenomena of deformation, recrystallization and Moreover, it can be seen in the diffractogram in Figure 10, in the regions of angles between 50 • and 54 • , that after the milling process there was a decrease in the intensity and the broadening of the peaks of the ferritic (110) and austenitic (111) phase, compared to the material as received (in the form of a sheet), this occurs due to the non-uniform plastic deformation (microtension) of the crystal line lattice resulting from the stacking faults, among other crystalline defects resulting from the grinding process and the reduction of particle size [44].
After heat treatment, there is a decrease in the width of the phase peaks and also an increase in the height of the phases. This behavior is associated with the reduction of micro-tensions caused by annealing. After the process of stress relief and sintering, the presence of the ferrite phase and the austenite phase is observed. The presence of the martensitic phase was not observed after the heat treatment process. The deformation of martensite induced is not stable at high temperatures and the reversion of martensite to austenite occurs by soaking at elevated temperatures [45][46][47][48]. Moallemi et al. (2017) [49] realized the cold rolling-reversion annealing was carried out at various temperatures in the range of 750-1100 • C. These authors verified that there was an increase in the reversion of the martensite fraction with the increase in temperature and a complete reversal occurred at 1050 • C. For the austenitic steels of the 304 family the total martensite reversion occurred at 750 • C [50]. Figure 10 shows a peak shift to the right for the ferrite phase (110) in the angle 52.14 • for the treatment of tension relief of 0.08 • and 0.12 • after sintering and also a displacement of 0.09 after the stress relief treatment and 0.14 • after sintering for the austenitic (111) phase, which characterizes the displacement of crystalline planes of the material after the sintering process.
In the industrial sequence for obtaining duplex stainless-steel by rolling includes two annealing steps: one intermediate after the hot rolling and the final one after the cold rolling [51,52]. In most applications, the duplex structures are manufactured from hot-rolled and annealed sheets. For this reason, its crystallographic texture results from the phenomena of deformation, recrystallization and phase transformation that occur in the austenite and ferrite phases [52,53]. These higher values obtained for phase quantification for the material as received (in the form of sheets) could be related to the crystallographic texture obtained in the processing of the material.
Likewise, Figure 10 exhibits an inversion of the intensity of the austenite peaks compared to that of the ferrite for the treatment carried out at 1050 • C. It can also be observed as an increase in the volumetric fraction of the austenite phase, the quantification value obtained by X-rays was 54% of austenite and 46% of ferrite. When the solution-treatment temperature is above 1050 • C but below the solvus line there is a progressive increase in ferrite content [54]. Gholami et al. (2015) [55] verified an increase in the amount of ferrite phase with increasing temperature of annealing. The volume fraction of ferrite for solution annealed at 1050 • C is 42% and increases to 52% and 69% for the sample solution annealed at 1150 • C and 1250 • C respectively. According to these authors, this phenomenon can be appropriately described by means of phase transformation diagrams [54,55].
The DSS is completely ferritic (δ) and the cooling of the steel enhances the precipitation of austenite at a temperature of 1300 • C [56]. On the other hand, fast water cooling can avoid the δ → γ. Then, in the sample sintering at 1250 • C and the sample being cooled to the furnace, a larger amount of austenitic phase may have occurred. Also, the formation of a larger amount of austenitic phase occurred in the sample submitted to the sintering treatment at 1250 • C and the sample cooled to the furnace. obtained for phase quantification for the material as received (in the form of sheets) could be related to the crystallographic texture obtained in the processing of the material. Likewise, Figure 10 exhibits an inversion of the intensity of the austenite peaks compared to that of the ferrite for the treatment carried out at 1050 °C. It can also be observed as an increase in the volumetric fraction of the austenite phase, the quantification value obtained by X-rays was 54% of austenite and 46% of ferrite. When the solution-treatment temperature is above 1050 °C but below the solvus line there is a progressive increase in ferrite content [54]. Gholami et al. (2015) [55] verified an increase in the amount of ferrite phase with increasing temperature of annealing. The volume fraction of ferrite for solution annealed at 1050 °C is 42% and increases to 52% and 69% for the sample solution annealed at 1150 °C and 1250 °C respectively. According to these authors, this phenomenon can be appropriately described by means of phase transformation diagrams [54,55].
The DSS is completely ferritic (δ) and the cooling of the steel enhances the precipitation of austenite at a temperature of 1300 °C [56]. On the other hand, fast water cooling can avoid the δ → γ. Then, in the sample sintering at 1250 °C and the sample being cooled to the furnace, a larger amount of austenitic phase may have occurred. Also, the formation of a larger amount of austenitic phase occurred in the sample submitted to the sintering treatment at 1250 °C and the sample cooled to the furnace. Figure 10. X-ray diffraction spectrum extended between 50° and 54° after milling and also subjected to stress relief and sintering treatment at 1250 °C for 1 h.

Conclusions
In this study, the results indicated that the powders obtained from the high energy ball mill exhibited a high volume of particles in the sub-micrometric scale. The milling process, with an addition of 3% VC, produced a particle size smaller than the initial chip size. The measurement of micrometric sizes obtained were between 26 and 132 μm. Due to the decrease in particle size, the network deformation and the intense plastic deformation, the X-rays diffraction peaks are enlarged after the high energy milling process. Diffractograms of the material subjected to the stress relief treatment at 1050 °C and sintering to temperature 1250 °C show the appearance of the ferritic, and the austenitic. Also, the peaks of the martensitic phase are observed only on samples submitted after the milling process and with no presence after treatment at 1050 °C.
The sintered material has a measurement of porosity evaluated at 15%. The obtained density of the material was 84% compared to the density of the material as stainless-steel duplex UNS S31803 Figure 10. X-ray diffraction spectrum extended between 50 • and 54 • after milling and also subjected to stress relief and sintering treatment at 1250 • C for 1 h.

Conclusions
In this study, the results indicated that the powders obtained from the high energy ball mill exhibited a high volume of particles in the sub-micrometric scale. The milling process, with an addition of 3% VC, produced a particle size smaller than the initial chip size. The measurement of micrometric sizes obtained were between 26 and 132 µm. Due to the decrease in particle size, the network deformation and the intense plastic deformation, the X-rays diffraction peaks are enlarged after the high energy milling process. Diffractograms of the material subjected to the stress relief treatment at 1050 • C and sintering to temperature 1250 • C show the appearance of the ferritic, and the austenitic. Also, the peaks of the martensitic phase are observed only on samples submitted after the milling process and with no presence after treatment at 1050 • C.
The sintered material has a measurement of porosity evaluated at 15%. The obtained density of the material was 84% compared to the density of the material as stainless-steel duplex UNS S31803 received. The value of the microhardness measurement was 232 HV. It is 13% exponentially less then stainless-steel duplex UNS S31803 received. The decrease in hardness and the increase in porosity values are related to the presence of the carbide on the surface of the particles making impossible contact between the metal-metal particles at the moment of compaction.
The milling process is an alternative route to reuse the UNS S31803 duplex stainless-steel chip. In addition, it allows for the production of duplex stainless-steel powders with vanadium carbide for the manufacture of parts in the industry, through the recycled material.
Thus, in conclusion, this study demonstrates the efficacy in the production of the metal-ceramic composite of a new method for recycling stainless-steel duplex UNS S31803 chips.