Tribos nro 5 junio 15 docx

e-TRIBOS Revista de la Asociación Argentina Tribología

Nro. 5 - Junio

2015

El desafío de las grasas de alta perfomance A scale effect in the mechanical and tribological coatings Plasma Nitriding and plasma Nitrocarburazing Tribologic behavior of thick and soft DLC coatings Wear regime in Incoloy 800 Spherical debris from roller bearings e-TRIBOS

Página 1

e-TRIBOS

Pรกgina 2

e-TRIBOS es una publicación de la ASOCIACIÓN ARGENTINA DE TRIBOLOGÍA. La distribución de esta revista se realiza sin cargo a los socios de la AAT y personas relacionadas con la TRIBOLOGÍA. Si Ud. desea ser incluido en el listado de distribución por favor envíenos sus datos a través de la página de www. aatribologia.org.ar Los editores no son responsables por lo expresado por los autores de los artículos publicados. Los datos, especificaciones y conclusiones son solo informativos.

Prohibida la reproducción total o parcial de los contenidos sin la expresa autorización del editor.

e-TRIBOS está abierta a la recepción de trabajos sobre cualquier aspecto de la disciplina TRIBOLOGÍA. Los autores son invitados a enviarnos los mismos los cuales de ser aceptados serán publicados sin cargo ni retribución.

Registro Nacional de la Propiedad Intelectual en trámite Editor: Roberto J. Leonetti

e-TRIBOS Junio 2015 Nro.5

Índice 3er INTERNATIONAL WORKSHOP OF TRIBOLOGY Página 4

El DESAFIO DE LAS GRASAS DE ALTA PERFOMANCE Página 5 a 8

TRABAJOS PRESENTADOS EN EL INTERNATIONAL WORKSHOP OF TRIBOLOGY SCALE EFFECT IN THE MECHANICAL AND TRIBOLOGICAL CHARACTERIZATION OF COATINGS Páginas 9 a 10 PLASMA NITRIDING AND PLASMA NITROCARBURIZING OF A LOW ALLOY STEEL SELECTED TO PRODUCE CAMSHAFTS FOR DIESEL ENGINES Páginas11 a 12 TRIBOLOGICAL BEHAVIOR OF THICK AND SOFT DLC COATINGS Páginas13 a 14 CHARACTERIZATION OF MIXED FRETTING WEAR REGIME IN INCOLOY 800 AT ROOM TEMPERATURE Páginas15 a 16 CHARACTERIZATION OF SPHERICAL DEBRIS FROM ROLLER BEARINGS UNDER BOUNDARY LUBRICATION Páginas17 a 18

COMISIÓN DIRECTIVA

Presidente: José L.Piña Vicepresidente: Esteban Lantos Secretario: Walter R. Tuckart Tesorero: Roberto J. Leonetti Vocal: Alfredo E. Eilenberger Vocal: Esteban P. Echeverría Vocal: Carlos L.Romano Vocal Suplente: Sonia P. Brühl Revisor Cuentas: Germán Prieto Revisor Cuentas: José A. Rossit Revisor Cuentas: Andrés R. Pereyra

e-TRIBOS

Asociación Argentina de Tribología

Av. Alem 1253 – Bahía Blanca (8000)- Buenos Aires-Argentina www.aatribologia.org.ar

Página 3

El El 13 y 14 de Mayo pasado se realizo el Tercer WORKSHOP Internacional de TRIBOLOGÍA en el Hotel 13 de Julio en la ciudad de Mar del Plata con el objetivo de “Reunir a especialistas de las distintas ramas de la tribología (lubricación, desgaste mecánico, fricción), ingenieros y técnicos especializados en mantenimiento de equipos rotantes e investigadores y estudiantes de posgrado de la comunidad universitaria, en un espacio de debate e intercambio de experiencias en temas vinculados con la tribología”.

La inauguración estuvo a cargo del Presidente de la Asociación Argentina de Tribología, el Ing. José Luis PIÑA y durante su desarrollo participaron más de 100 profesionales, proveniente de los más diversos sectores vinculados a la especialidad, tanto académico y/o empresarial.

El evento se desarrolló en dos jornadas, con un muy alto nivel de presentaciones y de participación del auditorio, lo cual transformo a este Taller en un particular ámbito de intercambio, tanto durante las exposiciones como en los distintos cortes.

Se presentaron trabajos que contemplaron los tres pilares en los que se sustenta la Tribología: Área Académica; aplicaciones de desarrollos académicos. Área Académica/Empresaria; equipos académicos universitarios que trabajan en conjunto con empresas en el desarrollo de aplicaciones y soluciones para la industria. Área de Aplicaciones Prácticas; sustentadas en las aplicaciones productivas y de mantenimiento.

La presencia de disertantes plenarios internacionales le dieron un marco superlativo a los excelentes trabajos locales presentados. Las Presentaciones Plenarias estuvieron a cargo de Esteban Broitman (Linköping University, Suecia) y de Julio Neves (Federal University of Technology - Paraná, Brasil) También se habilitó una sala para la exposición de trabajos sobre Poster, con la posibilidad de interactuar con los responsables de la preparación de cada trabajo.

Es muy importante destacar el apoyo que recibió la AAT para poder llevar a cabo este evento por parte de los patrocinadores.

e-TRIBOS

Página 4

El DESAFIO DE FORMULAR GRASAS DE ALTA PERFOMANCE Julieta Lois Milevicich – Servicio Técnico y Desarrollo para el Negocio de Soluciones Industriales de Dow Eduardo Lima – Especialista en Lubricantes para el Negocio de Soluciones Industriales de Dow

Tanto los aceites minerales como los sintéticos pueden ser utilizados como base para la formulación de grasas, representando entre el 50% y el 95% del peso total de la fórmula. Se utilizan distintas calidades de aceites para satisfacer la gran cantidad de aplicaciones. Varias propiedades de las grasas están muy influenciadas por el tipo y viscosidad de las bases. Por lo tanto, es importante seleccionar cuidadosamente el aceite más adecuado según la aplicación, antes de formular la grasa, para así asegurar que cumplirá su función. Generalmente, se utiliza una base sintética en aquellas aplicaciones en las que un aceite mineral no puede brindar el desempeño requerido. Algunas opciones funcionales disponibles en la industria son los ésteres, polialquilenglicoles (PAGs), poli-α-olefinas (PAOs), siliconas, perfluoro- y clorofluoroalquiléteres. Los PAGs tradicionales usados en la industria son solubles o insolubles en agua, pero incompatibles con aceites.

Recientemente se introdujo al mercado un nuevo tipo de PAGs solubles en aceite (OSP por su sigla en inglés). Estos nuevos productos tienen un alto índice de viscosidad, muy buenas propiedades de fricción y excelente solubilidad. Estas propiedades de los OSP son comparadas con los aceites base hidrocarburos (Grupos I, II y III). El equipo de Investigación y Desarrollo de The Dow Chemical Company decidió evaluar las propiedades de una grasa de alta performance formulada con una base PAG OSP y jabón de litio. Los aceites minerales nafténicos, a diferencia de sintéticos como las poli-α-olefinas,

e-TRIBOS

tienen un excelente poder de disolución de los aditivos; sin embargo, presentan como desventaja un bajo índice de viscosidad y un rango limitado de viscosidades entre 1,6 y 40 cSt cuando se los calienta a 100°C. Esta deficiencia de las bases de los grupos I a IV motivó la posibilidad de explorar formulaciones de grasas con bases lubricantes tipo PAG OSP [1].

Propiedades de los polialquilenglicoles solubles en aceite. Se realizaron diversos ensayos con el objetivo de evaluar las propiedades de los PAGs solubles en aceite y determinar si son apropiados para la formulación de grasas a base de jabón de litio. Los PAGs tradicionales, copolímeros de óxido de etileno y de propileno, se usan en contadas aplicaciones para fabricar grasas. Sin embargo, los OSP “Oil Soluble Polyglycol” no sólo dan la opción al formulador de ser utilizados como única base lubricante, sino que también pueden combinarse con aceites minerales y otros sintéticos, gracias a su compatibilidad con éstos, para lograr resultados superiores de lubricación, control y eliminación de depósitos. En la Figura 1 se muestran los perfiles tribológicos de los polialquilenglicoles, los cuales arrojan como resultados curvas de coeficiente de tracción de menor valor que los aceites minerales. En las Figuras 2 y 3 se muestran los resultados de un ensayo de fricción de bases minerales y sintéticas utilizando los OSP como aditivo. Los resultados muestran que con la adición de polialquilenglicol soluble en aceite se logra un mejor control de fricción. El mismo efecto se observa en el ensayo utilizando una poli-α-olefina.

Página 5

e-TRIBOS

Pรกgina 6

Una de las características más importantes de los polialquilenglicoles (PAG) es la capacidad de controlar la formación de depósitos. La molécula de los PAG es suficientemente polar para

solubilizar los productos de la oxidación, aún a altas temperaturas. La Figura 4 muestra una simplificación de los procesos de oxidación para un aceite mineral y para un PAG.

La Figura 5 muestra los ensayos de punto de anilina según la norma ASTM D611. Puede verse como la capacidad de disolución de los PAG es superior. Esta propiedad es fundamental al momento de formular con aditivos y lograr una

favorable interacción entre la base lubricante y el jabón. Hay una estrecha relación entre un bajo punto de anilina y la capacidad de solubilizar subproductos de degradación [1].

e-TRIBOS

Página 7

Formulación de una grasa de PAG OSP 220 En la tabla 1 se muestra la composición de la grasa, en la tabla 2 algunas de las propiedades evaluadas para la misma.

Tabla 1 - Composición de la grasa % en peso Componente 9,6 1,2-hidroxiestearato de litio 85,5 OSP 220 0,3 Antioxidante 2,0 Aditivo EP 0,5 Inhibidor de corrosión 0,1 Pasivador 2,0 Agente acomplejante

Tabla 2 - Propiedades de la grasa de OSP 220 Propiedad Método Color Visual Desgaste - Four Ball ASTM D2266 Carga de soldadura - Four Ball EP (kg) ASTM D2596 Test de oxidación ASTM D942 Grado NLGI ASTM D217 Torque a baja temp., -40°C, N.m ASTM D4693 Separación de aceite, 24h ASTM D6184 Po, mm/10 ASTM D217 Punto de caída (DP) ASTM D2265

Como conclusiones se observa que la adición de polialquilenglicoles solubles en aceite a las bases de los grupos API I a IV puede arrojar una mejora en la lubricidad, control y eliminación de depósitos, y control de la fricción.

El uso de grasas a base de OSP es una nueva herramienta para los formuladores a la hora de

e-TRIBOS

Resultado Beige claro 0,54 200 1,8 2 3,88 Ninguna 265 313

tener que afrontar algunos de los problemas actuales en esta industria. Pueden utilizar los OSP como bases, como co-bases o como aditivos.

[1] Paper #1115, NLGI. Dr. Govind Khemchandani, 2011, NLGI’s 78th Annual Meeting.

Página 8

Trabajo Presentado en el 3º INTERNATIONAL WORKSHOP OF TRIBOLOGY - Mar del Plata 2015

SCALE EFFECT IN THE MECHANICAL AND TRIBOLOGICAL CHARACTERIZATION OF COATINGS Esteban Broitman - Thin Film Physics Division, IFM, Linköping University, SWEDEN -esbro@ifm.liu.se

INTRODUCTION During the last years there has been considerable interest in the study of mechanical and tribological properties of coatings. At the macroscale (millimeter sizes), thin films are used to lubricate and protect parts of machines, as well as increase the life of tools. At micrometric and nanometric scales, there are many novel applications where coatings must be used to protect and/or lubricate surfaces operating under micro- or nano-contacts, like in hard disk drives, magnetic heads, micro-electromechanical systems (MEMS) and microsensors, etc.

Hundreds of experimental investigations are published every year showing the benefits of using thin films for improvement of mechanical and tribological properties. However, these studies are usually limited to characterizations done at one scale, and authors avoid to compare their results with measurements done at different scales. The main reason is the apparent contradictions that could appear when trying to compare different scales, as well as the lack of experimental methodologies allowing a correct comparison.

In this work, similarities and differences on mechanical (hardness and elastic modulus), and tribological (friction and wear) experimental results at macro- micro-, and nano-scales are reviewed and discussed.

MACRO- VS. NANO-INDENTATIONS In macroindentation, the calculation of the hardness H is: H = F/A where F is the applied force and A is the plastically deformed surface area left by the indent. The area A is obtained by optical microscopy because the distances are in the order of 100-1000’s of microns. In nanoindentation, however, the area of a nanoindent can be ~1 µm2 or even lower, which is only possible to observe and measure by scanning electron microscopy.

Nanoindentation become popular only after Oliver and Pharr [1] published a method to calculate the contact area at maximum applied load from the unloading curve of the load-displacement data

e-TRIBOS

(Figure 1). The method allows to calculate that area for perfect-symmetric tips as well as damaged ones 0 20 40 60 80 100 500 1000 1500 2000 Load ( N) Depth (nm) Fused Silica 0

Figure 1: Load-displacement curve of fused silica using a diamond Berkovich tip of 200 nm diameter. The nanoindentation hardness results 9.25 GPa and the relative Young’s modulus Er = 69.7 GPa.

without the use of a microscope. Once the area is obtained, the nanoindentation hardness, Young’s modulus E, and elastic recovery can be obtained without the use of a microscope. The calculation, however, is not always valid. In the case of thin films, the penetration depth must be lower than 10% of the total thickness because of possible substrate influence [2]. Also, to calculate E is necessary to know the value of the Poisson’s ratio.

As the nanoindentation deforms a very small area of the material, the hardness information is only local, and very difficult to compare with hardness values given by macroindentations like HRC, Vickers, etc because the tested material volume under the indent is considerable higher. For instance, for amorphous homogeneous materials, macro- and nano-indentation will give the same result. On the other hand, for polycrystalline materials with tenths of micrometer grain diameters, the macroindentation will give uniform mean values of the material, while the nanoindentation will reflect the mechanical behavior of each grain.

Página 9

Finally, it is necessary to stress that macroindentation measures the “true hardness”, i.e., the hardness from the plastic deformed area.

On the other hand, in nanoindentation, the hardness is calculated from the maximum deformed area which contains plastic and elastic deformation.

TRIBOLOGICAL CHARACTERIZATION AT MACRO-, MICRO-, AND NANOSCALES The macrotribology approximates the geometrical contact area of two bodies to the real contact area at atomic scale: the friction force depends only on the applied load N, and is independent of the surface area in contact. Also, at this length scale, friction is generally accompanied by elastic and plastic deformation, wear, and even fracture on the contact surfaces [2]. In microtribology, contact between two surfaces takes into consideration the real microscopic roughness through the interaction of asperities. The friction is considered to be originated from the plastic deformation and adhesion of interlocking spherical asperities. While in macrotribology the friction force is proportional to N, in microtribology is proportional to N2/3 [2]. In nanotribology, the surface interaction can be represented by a spring-like interaction that undergoes stretching and contraction as the atoms slide against each other.

There is also another important factor responsible for differences in measuring at different scales.

The sliding of materials in contact often involves wear, i.e., the generation and transfer of material from one surface to the other. This material, referred sometimes as the “third body”, influences the transient behavior of the sliding contact and can completely dominate the steady-state sliding behavior of many interfaces, especially for low friction coatings. At macro- and micro-scale, transfer films are formed during initial sliding, and these films determine the long-term frictional behavior of the interface. At nanoscale thirdbodies can also have a large impact on the contact properties; molecular dynamics simulations indicate that molecular intermediate species in asperity contacts have a dramatic effect on friction [2].

Macroscopic tribological characterization at laboratory scale is carried out typically by pin-ondisc or reciprocating instruments with applied loads in the order of N, pins of 1-7 mm dia and distances of several meters. In microtribology the loads are in the order of µN, pins of few µm dia, and distances of up to 100 µm. At the nanoscale, the tribometers are based on atomic force e-TRIBOS

microscopes, with loads in the order of 10-100 pN, distances of up to 1 µm, and tips of few nanometers diameter. Comparison of experiments at different scales are scarce because microscale experiments suffer of thermal drift errors.

However, we have recently published a new methodology that allows a precise determination of friction and wear at the microscale [3] (Fig 2 & 3).

0 5 10 15 20 25 30 35 0,10 0,11 0,12 0,13 0,14 0,15 Cycle Number Friction Force ( ) 01234Wear (105nm2)

Figure 2: Friction and wear of a diamond conical tip of 5 µm diameter oscillating over a carbon sample with an applied load of 1 mN. 0 2 4 6 8 10 -120 -80 -400 40 Pin Penetration (nm) Lateral Position ( m)

Figure 3: Wear profiles for sample of Fig. 2. All thermal drift influence has been eliminated.

FINAL REMARKS It is necessary to be careful when comparing mechanical and tribological data taken at different scales. A complete understanding of the limitations and approximations of each technique must be taken into account during the experiments.

REFERENCES 1 W. C. Oliver, G. M. Pharr, J. Mater. Res. 7(6), 1564 (1992).

2 E. Broitman and L. Hultman, in: “Comprehensive Materials Processing” Vol. 4, S. Hashmi (Ed.), Elsevier: Oxford (2014) Pages 389–412.

3 E. Broitman, Friction 2, 40-46 (2014). 4 E. Broitman, F. J. FloresRuiz, J. Vac. Sci. 1

Página 10

Trabajo Presentado en el 3º INTERNATIONAL WORKSHOP OF TRIBOLOGY - Mar del Plata 2015

PLASMA NITRIDING AND PLASMA NITROCARBURIZING OF A LOW ALLOY STEEL SELECTED TO PRODUCE CAMSHAFTS FOR DIESEL ENGINES A. Cabo1*, S. P. Brühl2, G. Prieto3,4, W. R. Tuckart3,4 1IONAR S.A. / 2Universidad Tecnológica Nacional, Concepción del Uruguay/ Universidad Nacional del Sur / 4 CONICET

INTRODUCTION Camshafts are a relevant part of diesel engines of extended use today. These components work under torsion and are also prone to fatigue and wear damage. Usually they are manufactured by casting, forging or machining from forged bars of low alloy steels. In most cases, the machined surfaces are quenched and tempered by induction heating. To withstand the efforts imposed on the active surfaces and improve tribological and fatigue properties, the industry used for decades thermochemical technologies such as: salt bath or gaseous nitriding and nitrocarburizing processes.

This work studied the effects of plasma nitriding and plasma nitrocarburizing, on the tribological behavior of the steel SAE 1045HM3 proposed to produce camshafts.

Results show that plasma nitrided samples present the best tribological behavior compared to the nitrocarburized and quenched and tempered ones. The influence of the roughness produced by the thermochemical processes also appears to be important.

MATERIALS AND METHOD The steel used was a SAE 1045HM3 in the form of annealed bar stock 57.10 mm in diameter. Its composition is presented in Table 1

Table 1: Chemical composition (%wt) of steel specimen

Disc samples 50 mm outside diameter, 12.6 mm inside diameter and 5 mm in thickness were machined. The samples were quenched and tempered to obtain a martensitic structure of 620 HV (56.2 HRC), similar to the case hardness obtained by induction hardening in the actual camshaft production.

Two groups of samples were selected, one for plasma nitriding (N) and the other for nitrocarburizing (NC). A third group, named “QT” is composed of blank samples, only conventionally heat treated. The parameters used in the plasma treatments are shown in Table 2

e-TRIBOS

Pin-on-disc tests were used for wear behavior characterization of the samples. Specimens were tested under two sets of conditions, namely, asreceived and polished. The as-received specimens were kept with the same roughness parameters that were obtained after the thermochemical treatments, while the second batch of specimens was polished using abrasive SiC papers up to 320 grit. All specimens were ultrasonically cleaned with toluene for 5 min before and after each test.

A 5 mm diameter tungsten carbide ball was used as the counter-body. Roughness parameters were Página 11

determined using an optical 3D roughness measurement system based on focus-variation.

The sliding speed in all cases was 0.06 m/s, while the sliding distance was set at 500 m for the asreceived stage and 28500 m for the polished stage.

The applied normal load was 10 and 40 N respectively.

Tests were performed using a low-viscosity, additive-free paraffinic Vaseline bath, with a cinematic viscosity of 17 cSt at 40 °C. All the experiments were performed under ambient laboratory conditions (25 °C, 65% relative humidity). Wear surfaces were analyzed by means of optical microscope while wear scar depth was determined in at least at 10 different positions using the aformentioned optical 3D measurement system.

RESULTS AND DISCUSSION Microhardness measurements shown that NC samples had an average white layer thickness of 15.6 µm and a diffusion layer thickness of 430 µm, while for the N samples the average white layer thickness was of 3.5 µm and the diffusion layer thickness of 330 µm.

QT specimens had an as-received roughness of 0.1 µm (Ra) and 1.40 µm (Rt), while in N and NC specimens it was of 0.2 µm (Ra) and 2.07 µm (Rt). Wear rate of QT specimens was so low that it could not be determined, while for N and NC specimens it was in the range of 1.0 x 10-5 mm3/Nm. The roughness parameters equated after polishing, thus results could be considered as independent of surface topography. Optical microscopy evaluation of polished specimens shown that the white layer of both N and NC specimens was not significantly diminished by polishing.

Roughness variations after thermochemical treatments had been reported by several authors, although in the majority of the cases, the analysis has been restricted to Ra values, such as the work of Çetin, et al. [2], who measured a Ra increase from 0.02 to 0.18 µm while studying plasma nitrided AISI 420 steel.

In our case, the higher value of Rt before polishing of N and NC specimens means that they had more prominent and singular peaks, which are prone to severe plastic deformation and subsequent break-off due to elevated contact stresses [3, 4]. Table 3 shows the results of the wear tests of the polished specimens

e-TRIBOS

CONCLUSIONS From the present study the following can be concluded: - The nitriding process produces less reduction of the hardness underneath the white layer; this is important from the stand point of the load bearing capacity of the case.

- Greater roughness after thermochemical treatments promotes a higher wear rate. However the friction coefficients are similar for all materials, ranging from 0.06 to 0.08.

- A low roughness level is more important than the thickness of the white layer. The nitrided samples, even though they had a thinner compound layer, had the same tribological perfomance than nitrocarburized specimens.

- Thermochemically treated materials showed wear rates one order of magnitude lower than quenched and tempered material.

REFERENCES 1 Bell, T. Heat Treat. Met. 2, 39-45. 1975.

2 Çetin A., Tek Z., Öztarhan A. & Artunç N., Surf. Coat. Technol., 201, 8127-8130. 2007.

3 Kapoor, A., Williams, J. A., Johnson, K. L., Wear, 175, 81-92.

1994.

4 Xie, Y., Williams, J. A., Wear, 196, 21-34.

Página 12

Trabajo Presentado en el 3º INTERNATIONAL WORKSHOP OF TRIBOLOGY - Mar del Plata 2015 Trabajo Presentado en el 3º INTERNATIONAL WORKSHOP OF TRIBOLOGY - Mar del Plata 2015

TRIBOLOGICAL BEHAVIOR OF THICK AND SOFT DLC COATINGS Eugenia L. Dalibon1, Daneil Heim2, Christian Forsich2, Sonia P. Brühl1* 1 Surface Engineering Group, Universidad Tecnológica Nacional (UTN-FRCU), Ing. Pereira 676, E3264BTD Concepción del Uruguay, Argentina / 2 University of Applied Sciences, Stelzhammerstr. 23, 4600 Wels, Austria

INTRODUCTION Austenitic stainless steels are widely used in industry due to their good corrosion resistance; however, they present poor mechanical properties.

Different coatings such as DLC “Diamond Like Carbon” can be used in order to improve surface properties. According to the sp3/sp2 ratio and the hydrogen content, DLC coatings can be classified in hard DLC or soft DLC. These coatings have low friction coefficient, good wear resistance and chemical inertia; however, they present adhesion problems when they are deposited on metallic substrates. For this reason, the plasma nitriding previous to the DLC coating deposition could be convenient. Although there are several publications about DLC coated and nitrided stainless steels, not many contributions have been found in the literature about soft and thick coatings. In this work, the tribological behavior and adhesion of thick and soft DLC coatings deposited on nitrided and non-nitrided austenitic stainless steels were studied.

The nitrided layer thickness was 12 m. The MATERIALS AND METHODS hardness was about (12 2) GPa and the Young’s Disk-type samples of 25 mm in diameter and 6 Modulus was (72 10) GPa, though it can be mm in height were cut from 316 L austenitic considered a soft DLC film. stainless steel bar. Plasma nitriding treatments were carried out in an industrial reactor by means of a pulsed DC discharge, for 14 hours at 400 ºC, with a gas mixture 20 % N2 and 80 % H2. The DLC coatings, which are in fact a:C-H-Si films (silicon containing amorphous hydrogenated carbon), were deposited by the Plasma Assisted Chemical Vapour Deposition technique (PACVD) in the same reactor used for nitriding with HMDSO and acetylene as precursor gas. The DLC coatings were deposited on austenitic stainless steel (named coated samples) and on nitrided austenitic stainless steel (named duplex sample) and the control group was the austenitic stainless steel (untreated samples). The films were characterized by EDS and Raman, hardness was assessed with nanoindenter and microstructure was analyzed by OM and SEM. The pin on disk tests were performed with alumina as counterpart and a hertzian pressure of 0.87 GPa. The abrasive Fig. 1: SEM Image of the coated sample The surface wear resistance was tested using the ASTM G65morphology of the coatings in the coated and the duplex 95 Dry Sand/Rubber Wheel test; where the sample was different (Figure 2). In both samples, the coating applied load was 45 N, and the duration of the test presented defects in the form of holes and protuberances as it was 8.5 min, using a mixture between parameters was reported by some of the authors2. suggested in Procedures A and D of this standard.

However, the protuberance size and the density of RESULTS AND DISCUSSION the defects were higher in the duplex sample than In the Raman spectrum, the films presented the D in the coated sample, probably due to the higher and G bands with an intensity ratio (ID/IG) of roughness produced by the nitriding process. The 1.08. The hydrogen content was about 43 % roughness (Ra) in the untreated sample was 0.040 which was calculated from the background of the m and in the nitrided sample was 0.070 m.

Raman spectrum1.

The coating thickness was about 36-37 m in both samples (Figure 1). e-TRIBOS

Página 13

Fig. 3: Friction coefficients registered in the pin-on-disc tests for different samples.

Fig. 2 SEM image of surface of the coated and duplex samples.

In the pin on disc tests, the friction coefficient was 0.09 for the only coated sample and 0.11 for the duplex sample (Figure 3), similar values as others have already reported 3. These values were considerable lower than the friction coefficient in the untreated sample. Wear tracks were undetectable, probably due to the low elastic modulus E of these films.

e-TRIBOS

In the abrasive test, the results were also very good for the coated samples, as they showed no damage, mass loss was undetectable and so were the tracks profiles. Anyway, as the wear tracks were visible, it could be observed that the defects were cracked and holes appeared in their place. It was determined that these holes did not pass through the coating thickness.

In the Scratch test, the critical load was higher in the duplex sample (13 N) than in the only coated sample (9 N). In the Rockwell C indentation, the behavior was similar in duplex and coated samples. In these thick coatings, the test did not evaluate the adhesion but the fracture toughness.

The nitriding as previous treatment allows improving the adhesion of the coatings but affect their topography, resulting with more defects, which affect the tribology behavior.

CONCLUSIONS These thick and soft DLC coatings had low friction coefficient and outstanding abrasive and sliding wear resistance. Probably a polishing process previous to the coating could be convenient in order to reduce the roughness.

REFERENCES 1 Casiraghi C., Ferrari A.C., Robertson J. Phys. Rev. B 72, 0854011-14, 2005.

2 Forsich C., Dipolt C., Heim D., Mueller T., Gebeshuber A., Holecek R., and Lugmair C., Surf. Coat. Technol. 241, 86–92, 2014.

3 Erdemir A. and Donnet C. J. Phy

PĂĄgina 14

Trabajo Presentado en el 3º INTERNATIONAL WORKSHOP OF TRIBOLOGY - Mar del Plata 2015

CHARACTERIZATION OF MIXED FRETTING WEAR REGIME IN INCOLOY 800 AT ROOM TEMPERATURE S.R.Soriaa,c*, M. Bergantb, A. Yawnya,c a División Física de Metales, Centro Atómico Bariloche /b Gerencia CAREM, Centro Atómico Bariloche / c CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas).

INTRODUCTION Fretting is a type of wear damage induced between two surfaces in contact under relative movement of small amplitude (1 to 300 µm) involving different wear mechanisms such as adhesion , abrasion and tribochemical reactions, among others [1]. The predominant surface damage varies depending on normal load and displacement amplitude and different fretting regimes are defined accordingly stick, stick-slip or mixed and slip [2]. They may evolve with the number of cycles and among them, the mixed fretting regime (MFR) has been identified to be the most critical in relation with crack nucleation, propagation and service failure [3].

Fretting is one of the main mechanisms responsible for tube degradation in steam generator tubes (SGTs) in nuclear reactors. It is originated in flow induced vibrations [4] which results in the relative movement between tubes and their supports at the positions were they contact each other.

In the present work, the influence of the normal and displacement amplitude on the characteristics of the MFR in Incoloy®800 SGTs were studied.

MATERIALS AND METHODS Incoloy®800 SGTs (I 800) with a diameter of 15.87 mm and wall thickness of 1.13 mm were used. The support plate was simulated with stainless steel AISI 304L and AISI 420 semicylindrical pads with the same external diameter as the I800 SGTs. Fretting tests were performed in a MTS 810 servo-hydraulic testing machine using a specially designed device. It allows applying the normal load between the SGT and the pad via an elastic cantilever beam supported from the upper grip of the testing machine. All tests were carried out in air at room temperature using the conditions established by ASTM- G204 [5], i.e., a normal load to 10 N and a frequency of 13 Hz. The displacement amplitudes used were 5, 10, 25, 35 and 50 µm and tests were conducted up to 1E6 cycles.

The fretting regime was determined by the evolution of the “tangential load-displacement” loops. Surface damage (scars) was characterized

by Scanning Electron Microscopy (SEM) and Optical Microscopy (OM). The local elemental composition in the different regions on the surface of the scars was determined by Energy Dispersive X-ray Spectroscopy (EDS).

RESULTS AND DISCUSSION The fretting map for I 800 SGT / AISI 304L pad can be seen in Fig. 1, where the imposed amplitude and normal load data points are classified from the appeareance of the respective tangential load vs. displacement loops. e-TRIBOS

Fig. 1: Fretting map by I 800 / AISI 304L pair.

For both pairs, i.e., I 800 / AISI 304L and I 800 / AISI 420, the MFR was detected for displacement amplitudes of 10 and 25 m. For an amplitude of 50 m, the slip regime (SR) was identified. An increase of the normal load to 20 N for a displacement amplitude of 10 m also resulted in MFR. The evolution of the tangential load vs. displacement loops with the number cycle is shown in Fig. 2 for the I 800 / AISI 420 pair for a normal load of 20 N and a displacement amplitude of 10 m. A variation in the shape of the loops, usually to elliptical (partial slip) and quasirectangular (gross slip) loops are typically encountered in the MFR.

Página 15

The change of material´s pad and the variation of normal load do not affect the damage intensity.

The occurrence of material transfer between both surfaces was confirmed by EDS. This was studied in the I 800 / AISI 420 pair since the AISI 420 does not contain Ni in its composition. Regions with Ni contents up to 13 wt.% were found in the contact area for this pad material.

Fig. 2: Fretting curves for I800/AISI 420 pair.

The analysis of the surface damage for displacement amplitudes of 10 and 25 m indicated that for the two STG / supports combinations studied, the tribopair was in MFR. In the case of the 10 m displacement amplitudes, an area of the contact region exhibiting no damage was detected.

Instead, for 25 m displacement amplitude, damage was observed in the whole extension of the contact region as can be appreciated from Fig 3.

CONCLUSIONS In this work, the mixed fretting regime (MFR) corresponding to Incoloy®800 steam generator tubes in contact with AISI 304L and AISI 420 support pads was studied. A change in the surface wear was found for displacement amplitudes between 10 and 25 m. For 25 m displacement amplitude, the entire contact area was damage while for the 10 m case, regions without surface damage inside the contact area were observed.

Keeping the displacement amplitude constant, no differences were found by varying the material pad or by changing the applied normal load between tube and pad.

Adhesive wear was predominant in MFR and the existence of material transfer between pair elements was confirmed by EDS. REFERENCES 1 Waterhouse R.B., Fretting Corrosion, Oxford, Pergamon Press1972.

2 Vingsbo O, Söderberg S, Wear 126, 131-147, 1988.

3 Zhou Z.R, Vincent L, Wear 181-183, 531-536, 1995.

4Guerout F.M, Fisher N.J, Journal of Pressure Vessel Technology 121, 304-310, 1999.

5 ASTM G-204, ASTM Internacional, 2010.

Fig. 3: Scar for I800/AISI 304L pair with 10 N to 10 um (A) and 25 um (B) of displacement amplitude. Arrows indicate the sliding directions.

e-TRIBOS

Página 16

Trabajo Presentado en el 3º INTERNATIONAL WORKSHOP OF TRIBOLOGY - Mar del Plata 2015

CHARACTERIZATION OF SPHERICAL DEBRIS FROM ROLLER BEARINGS UNDER BOUNDARY LUBRICATION L. Quinzani1, W. Tuckart2, A. Urrestarazu1,2* 1 Planta Piloto de Ingeniería Química, PLAPIQUI (UNS-CONICET/2 Departamento de Ingeniería, Universidad Nacional del Sur

INTRODUCTION The detection of wear particles in a lubricant provides a powerful diagnostic technique for the finding of distress in critical mechanisms and thus preventing catastrophic failure by allowing timely maintenance [0].

The presence of wear particles in a lubricant can be easily revealed by ferrography and confirmed by electron microscopy, and their composition can be determined by X-ray energy analysis. On machines where there are many elements with rolling motion and where vibration and ultrasound techniques do not allow a reliable diagnosis to early failure (such as in low-speed gear-boxes in cement mills), ferrography plays a most important role.

In general, the appearance of large numbers of spherical particles smaller than 10 m is associated with failures in rolling contact elements. However the physical process that generates these particles is yet not completely understood and leaves some questions [2,3]. In a previous work, we have found an appreciable number of metal microspheres with iron-oxygen and iron-chromium-oxygen elements on ferrograms made on oil samples from gear box in service. The question arised if these particles were from the bearing raceway that usually fails. To analyze this possibility, we began an experimental examination of wear debris from bearings under totally controlled laboratory conditions. The present work concentrates on examining the mechanism of formation and the source of generation of the microspheres and briefly describes the changes at various stages in the wear process MATERIALS AND METHODS A rolling contact home-made tribometer with marginal lubrication was designed using ¨51101 thrust ball bearings¨, SAE 52100 steel tracks, and Si3N4 balls. The tribodevice was made in aluminum alloy. The surface finish of the raceways were initially polished using diamond paste 6 m to better visualize the morphological changes.

The results presented in this work were carried out under axial load of 1000 N (102 kg) during 1800 minutes, and with tracks rotating at a speed of 2100 rev/min. For the applied load, 1540 and 1030 MPa of maximum and average contact pressures respectively, were calculated using hertz theory. The contact area was 0.98 mm2. The raceways were lubricated with an additive-free mineral oil, 13.6-15.4 cSt kinematic viscosity at 40°C (film thickness 0.018 m). The tests were performed at ambient temperature (21 to 24°C, 45-50% humidity).

The running surfaces and wear debris were observed using optical and scanning electron microscopy (SEM), coupled with chemical composition microanalysis (EDS).

RESULTS AND DISCUSSION

e-TRIBOS

Fig 1. Graph showing the number and composition of spherical particles detected at different stages

Figure 1 displays the amount of spherical particles detected using ferrography analysis of lubricant collected at different times during the test. As it can be observed, spherical particles are generated during the whole wear process. The two most important observations are that spherical particles begin to appear in the first few cycles of loading (running in), and that a large number of spherical particles appear in the last stage. Moreover, these particles present a more complex composition which can be associated with the degradation of the tribosurfaces and the occurrence of cracks RCF. Página 17

The first ferrograms were performed at 15, 30 and 45 minutes, corresponding to the 3.1, 9.4 and 6.3 x104 accumulated load cycles respectively. The presence of spherical particles smaller than 12 m in size during the running-in stage was observed.

The existence of these particles can be associated with the theory of particle generation by a rounding and smoothing process of adhesive wear particles.

Moreover, in the EDS spectrum, only peaks of iron and oxigen elements can be found. .

An intermediate stage, which extends from about 200 to 900 minutes, is observed, with uniform generation of spherical particles (approximately two to three spheres per ferrogram). These spheres have identical size and chemicals composition that the ones detected dutring the running-in. The raceways surfaces in this stage show no significant variation in their morphology.

After 900 minutes [1.9x106 cycles], a significant change in the wear process is observed. At that time, the first surface cracks are detected which show a significant increase in both, number and size, at 1500 minutes [3.1x106 cycles]. Figure 2 displays an image of the raceway obtained by SEM from the test at 1500 minutes where a large number of cracks can be observed.

Fig 2. Image obtained by SEM of the raceway from the test at 1500 min (magnification: 500x).

Additionally, the ferrogram of the sample at 1500 minutes shows a sharp increase in the number of spherical particles smaller than 12 m in size. In this last stage, most spherical particles display iron-oxygen composition like the ones found in previous stages. However, in this stage, a small

e-TRIBOS

amount of spheres are found in which iron, chromium and oxygen are detected. Figure 3 shows a SEM image corresponding to the final stage, displaying a large number of spherical particles. It can be concluded that a new forming mechanism appears in this last stage.

Fig 3. Spherical particles with iron-oxygen composition from a ferrogram at 1500 min [3.1x106 cycles] (magnification: 3000x).

CONCLUSIONS The results from the present study, in aggrement with other works from the literature, show that spherical particles appear on ferrograms from rubbing wear process. It is also concluded that: * spherical debris smaller than 12 m in size with iron-oxygen composition are generated at all stages of wear. They are produced by a process of agglomeration, rounding and polishing of oxides debris detached from the surface of the raceway. * degradation of the raceway surfaces by rolling contact fatigue, in the last stage, also develops spherical debris smaller than 12 m in size with iron, chromium and oxygen composition.

REFERENCES 1. D. Anderson, "Wear Particle Atlas (Revised)". Report NAEC. Naval Air Engineering Center, Advanced Technology Office, Support Equipment Engineering Department (1982). 92-163.

2. D. Scott and G. H. Mills, “Spherical debris - its occurrence, formation and significance in rolling contact fatigue”, Wear, 24 (1973) 235 - 242.

3. B. Loy and R. McCallum. Mode of formation of spherical particles in rolling contact fatigue. Wear, 24 (1973), p. 219-228.

Página 18