(Parte 1 de 2)

13º Congresso Ibero-americano de Engenharia Mecânica

13º Congreso Iberoamericano de Ingeniería Mecánica Lisboa, Portugal, 23-26 de Outubro de 2017


Maria Augusta Minguta de Oliveira1, Antônio Carlos de Araújo Santos1, Antonio Marinho Junior1

1. Department of Mechanical Engineering, Rio de Janeiro State University, Rio de Janeiro, Brazil, email:


Surfaces of UNS S32205 Duplex Stainless Steel were examined by Scanning Electron Microscopy (SEM) after erosion by impacts of alumina particles entrained in air flux, using an especially designed setup. Morphological observations were done on craters and singularities of the eroded surfaces. Images of craters in different stages of evolution were put in sequence; from those presenting exclusively plastic deformation to those evidencing mass loss. This sequence matches a well-accepted model on the behaviour of ductile metals under erosion by impact of solid particles. Cross sectional examination of erosion craters shows evidence of plastic deformation, strain hardening and cracks along Adiabatic Shear Bands (ASB). These bands are typically found in the microstructure of materials under dynamic compression loads, as in ballistic impacts and in some high-energy manufacturing processes. SEM images of morphological features allowed the correlation of fracture cracks in craters with the presence of ASB.

Keywords: Solid Particle Erosion, Duplex Stainless Steel, Adiabatic Shear Band.

XIII CIBEM – 2017 Lisboa 1. Introduction

Degradation of engineering materials is a matter of economics. Billions of dollars are expended annually to improve design, and to prevent and control the deleterious effects of this phenomenon [1-3]. Friction, wear, lubrication, erosion and corrosion are the main concerns of tribologists and many efforts have been made, in the last decades, to anticipate the effects of degradation in the performance of engineering materials [4,5]. Erosion of surfaces, especially in case of impact of hard particles entrained in moving fluids, is a relevant issue in design, operation and maintenance of pumps, turbines, separators, valves and tubes, as well as in structures. This phenomenon was object of some earlier studies [6- 8] and deserved detailed reviews [9,10]. Among other issues, the role of some variables was pointed out. For materials of the particle and of the target: Youngs modulus, Poissons ratio, plastic behavior and failure behavior; for the particles: specific gravity, size, shape, velocity and also the impingement angle. The more a particle is dense and hard, the greater its ability to cause erosion. The same applies to velocity, but there is a velocity threshold below which erosion is negligible. Within certain limits, wear is favored by irregular shape and by inhomogeneous size distribution of the particles. Only very low impingement angles can produce erosion by cutting (“ploughing”), regardless the brittle or ductile behavior of the target material. On the other hand, high angles up to 90° can enhance erosion, but only in brittle materials. In case of ductile metallic materials, intermediate angles are more effective to promote erosion, typically by mass losses related to craters [6-1]. Craters and ploughs have been considered the keys for explaining the mechanisms of mass loss in metallic materials under this type of erosion. In ductile metals, the craters play a relevant role in elucidating the phenomenon and there is a model based on the effects of successive impacts of single particles, on the same point of a surface. In a first impact, a particle can have sufficient energy to; at least, cause a localized plastic deformation in the target material, creating a crater. A second impact on this crater can increase deformation on the bottom of it, strain hardening the material and creating side borders (“lips”). From the following impact on, the target material could be submitted to strain conditions leading to fracture along a heavily sheared layer at the base of the lip. The detachment of the lip characterizes the loss of mass, i.e. erosion [9,12,13]. In most of experimental works cited, the erosion rate of a material, under specific conditions, was determined using mass loss measurements along the time. Through this procedure, it was possible to study the role of several variables in erosion process, although in general they did not consider morphological and topographical aspects of the surfaces. Observation of these aspects, by microscopy, could help in understanding the behavior of the target material surface and to discuss erosion models. Scanning Electron Microscopy (SEM), rather than Light Optical Microscopy (LOM), offered the possibility of observing eroded surfaces, detecting singularities and/or the number, size and distribution of craters, as well as their shape, depth and texture. Thus, it was possible to observe the morphological evolution of craters along the erosion process, as well as other aspects such as details of microstructural changes[14-17]. Degradation researches on ductile metals consider a variety of them, including traditional stainless steels (S), which have been studied for years, especially on their corrosion resistance. Presently, modern types of S like duplex stainless steels (DSS) are in evidence. Their applications by industry are spreading, not only to meet mechanical requirements but also to achieve a better corrosion resistance. They are having increasing applications by chemical, petrochemical, petroleum and power generation industries [18-21]. The use of DSS in structures, both outdoor or subsea, suggests the need to pay attention to their behavior under wear, as in the case of erosion by impact of solid particles entrained in air or water fluxes. Among others DSS, the UNS S32205 has been extensively used. Its satisfactory mechanical properties and good corrosion resistance are presently well known [2]. Other aspects, like its weldability, have been accurately examined in various studies, once heating conditions can lead to formation of some specific and deleterious metallurgical phases[23]. On the other hand, its behavior under erosion is an issue to be more explored, and this encourages efforts to get more knowledge on this material. The present work was undertaken aiming to study morphological remarks of the eroded surfaces of the DSS UNS S32205, by impact of alumina particles, which was chosen due to their effectiveness compared to other hard particles [24]. Erosive wear was produced by attacking the surfaces with particles entrained in air flux, through a specifically designed low cost installation. SEM images allowed observing, on eroded surfaces, the consequences of particle impacts in different conditions. In general, voids, cracks, pits and precipitates were revealed by the

XIII CIBEM – 2017 Lisboa images, and their possible significance and origins were discussed. In images of craters generated in the experiments, it was possible observing and analyzing particular aspects related to microstructural transformations attributed to impact of particles. It was discussed a possible association between the mechanisms of erosion by impact of solid particles and the effects of dynamic compression. Additionally, the observation of a set of craters, in the same image, allowed seeing them as a group where different shapes, depths and other aspects could indicate different stages of their evolution, according to conditions of erosive attack. Adiabatic shear band (ASB) is a narrow region of very large shearing that sometimes occurs in some materials under intensive dynamic load. The thickness may be of order of only a few tens of micrometers, whereas the lateral extent may be many millimeters. The presence of ASB alter the subsequent behavior and performance of the material. ASB commonly acts as sites for further damage and may act either a ductile or brittle manner [25] [26].

2. Methodology 2.1. Experimental approach

Samples of UNS S32205 duplex stainless steel were extracted from an as received commercial hot rolled plate, with 12.7 m thickness, produced by North American Stainless and supplied by Expander [27]. The chemical composition is [wt%][28]: 0.023%C, 1.752%Mn, 0.028%P, 0.001%S, 0.519%Si, 5.454%Ni, 2.010%Cr, 3.108%Mo, 0.178%N. The mechanical properties of this material were certified by Sandmeyer Steel Company [28] and is presented in Table 1.

Table 1: mechanical properties [28]

Ultimate tensile stress

Elongation in 50 m

Hardness HRC

The erodent particles were polyhedral alumina, with variable morphology and 150 μm average size (100 mesh). Compared to other hard materials, alumina particles with these characteristics are more effective to erode ductile metals [24]. To achieve the erosion of the surfaces examined, some steps were taken. First, a piece of 150 m x 200 m x 12.7 m was extracted from the original plate, by sawing. From that piece, 16 sticks of 10 m x 5 m x 3 m were machined by a linear planer. Then, from these sticks, 20 specimens of 5 m x 5 m x 3 m were cut by a grinding disc under water-cooling.

optical microscopy inspection

The specimens were embedded in acrylic resin, as in metallographic procedures, with the exposed surface (5 m x 5 m) corresponding to that of the original plate. The exposed surfaces of the specimens were grinding with SiC grit paper from 400 to 1500 grid. Then the specimens were polished with 3.0 μm, 1.0 μm and 0.5 μm diamond pastes. Both grinding and polishing procedures were performed in automated equipment, until the specimens surface be considered satisfactory by To do the erosive attack on the specimens’ surfaces, an installation consisting of a compact sand-blasting chamber, coupled to an air compressor, was set, see Fig. 1 (a). The internal diameter of the nozzle was 1.2 m. The velocity of the particles carried by the air flux was measured by the double disk method [29-31]. The particles velocity depends upon the line pressure of the air flux and therefore a calibration curve was plotted, to set the working point of the experiments. The line pressure was set to get an average speed of 50 m/s, assuring that the particles velocity was above 20 m/s, which is a velocity threshold for erosion by impact of particles [8].


Figure 1: Installation

Considering the objectives of this work, the mass flux of the particles was not measured, but it remained the same and stable along all the experiments. To ensure that, in each test, the particles feed rate was maintained constant. In each experiment, a specimen was placed in the locker “A” of the device shown in the Figure 1 (b). The nozzle “B” in the same figure was positioned to obtain an incident angle of 30°. This angle was chosen to maximize the erosive wear by impact of solid particles in ductile metals [10,1,17,20]. The distance from nozzle to the center of the target was set as 20 m, considering the exposed area available on the specimens’ surface and the full cone spray pattern of the air flux. In sequence, the device was placed inside the chamber and an adequate air pressure was set to achieve a mean velocity of 50 m/s for the particles. Then the specimens were submitted to the impacts

XIII CIBEM – 2017 Lisboa of particles, for different times of exposure, i.e. 3 s, 5 s, 8 s, 10 s, 20 s, 60 s, 120 s and 180 s. The eroded surfaces of the specimens were observed by Scanning Electron Microscopy, using a FEI Quanta 250 microscope, coupled to an image processor. To observe the effect of erosion in sub surfaces, the specimens corresponding to 5 s and 20 s of exposure were examined and manipulated by a TESCAN LIRA 3 FIB-SEM microscope [41]. After selecting a crater of interest, a trench milling was done along it, by FIB (“Focused Ion Beam”) and a cross sectional area was exposed. For protection, a platinum layer was deposited on the craters surface. Repeating this procedure, two more parallel cross sectional areas were examined, at different depths.

2.2. Results and discussion

SEM images of the specimen’s surfaces polished and after 3 seconds of erosive attack is shown in Fig. 2. Despite the short time of exposure to the erosive attack, it is evident the overall effects of the erosion. In an overview, it is possible to see many craters with different shapes, sizes, depths and orientations.

Figure 2: SEM micrographs of (a) polished surface and (b) after 3 s exposure to erosive attack.

manufacturing process of DSS [35,36]

There are also evidences of ploughing, pits and remains of the original surface. These features were expected and previously mentioned in other studies [16, 17], but they clearly pointed out to the amount and to the diversity of features available in the image of a experiment with a flux of particles. This fact could be attributed to the size distribution of particles and to the turbulent nature of the air flux, allowing impacts with different incident angles, orientations and energies. Some other images, representing the effects of larger time expositions, allowed a more detailed analysis of the specific features observed. Besides the diversity of shapes and sizes of the craters, it is possible to see spherical dark particles all over the surface. They were observed in different situations, as inside cavities, on deformed internal surface of a crater, on striations and in plastic flow dimples. An analysis of the causes of these localizations is beyond the objectives of this work. In Fig. 3, the image shows a massive occurrence of voids but also their coalescence into cracks (I) and (I) and in agglomeration as pits (I). Since the nitrogen content in DSS is higher than in common stainless steels, the occurrence of voids could be consequence of nitrogen introduction during the

Figure 3: SEM micrographs after 10 s exposure. (a) In details

(I) e (I) void coalescence into cracks and in (II) void agglomeration in pits.

As seen in the Fig. 4, the surface of the specimen exposed to a 120 s attack of the erodent particles, several situations can be observed at the same time. In the selected areas of the image, enlarged below it, the crater (I) appears to be consequence of one single impact, and shows exclusively plastic deformation, besides the evidence of the polyhedral shape of the particle. The shape of crater (I) could be consequence of at least two impacts, with formation of side borders, or lips, as seen in upper area, where it is also evident an

XIII CIBEM – 2017 Lisboa

incipient fracture. In the crater (I) it is seen that part of the border was detached, meaning some loss of mass. Probably, the previous impacts promoted the strain hardening of the material in the bottom of the crater and create local conditions of stress and strain leading to fracture and material discard. This sequence of events, related to the morphology of craters, match the Sundararajan- Shewmon model on erosion by solid particle impacts in ductile metals [9, 12, 13]. Such model contemplates also the mechanism of lips fracture, which is based on crack formation along adiabatic shear bands (ASB), find in dynamic compression processes [37-40]. Thus, observing crater (IV), it is possible to see a line of fracture surrounding it, suggesting the mechanism of fracture previewed by the model. This is highlighted in Fig. 5.

(Parte 1 de 2)