Wear resistance of vanadium-niobium carbide layers grown via TRD

Resistencia al desgaste de capas de carburo de vanadio-niobio crecidas por medio de TRD

Fabio Enrique Castillejo-Nieto ^a, Jhon Jairo Olaya-Florez ^b & José Edgar Alfonso ^c

^a Departamento de Ciencias Básicas, Universidad Santo Tomas de Aquino, Bogotá, Colombia. fabiocastillejo@usantotomas.edu.co
^b Departamento de Ingeniería Mecánica y Mecatrónica, Universidad Nacional de Colombia, Bogotá, Colombia. jjolayaf@unal.edu.co
^c Grupo de Ciencia de Materiales y Superficies, Departamento de Física, Universidad Nacional de Colombia, Bogotá, Colombia. jealfonsoo@unal.edu.co

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Abstract
Nb-V complex carbide coatings were produced on AISI D2 steel substrates using the thermo-reactive diffusion (TRD) process in order to improve the surface hardness and wear resistance of this tool steel. The carbide coating treatment was performed using molten borax with added ferroniobium, ferrovanadium, and aluminum at temperatures of 1223, 1293, and 1363 K for 2, 3, 4, and 5 h. The coating layers were characterized with scanning electron microscopy (SEM), X-ray diffraction (XRD) and energy-dispersive X-ray spectroscopy (EDS). The growth rates of the coatings were studied, and a kinetic model of the layer thickness was constructed as a function of the time and temperature treatment. The hardness and friction coefficient (COF) of the coatings was measured through nanoindentation and pin on disk test respectively. The carbide layers had a homogeneous thickness and a hardness of 37.63 GPa, which is close to values obtained in super-hard materials, and the COF was in the range of 0,3 for the coated steels.

Keywords: Thermo Reactive Diffusion, Niobium-Vanadium Carbide, Hardness, wear

Resumen
Recubrimientos mixtos de carburos de niobio - vanadio se fabricaron utilizando el proceso de difusión termo-reactiva (TRD) a fin de mejorar la resistencia al desgaste en aceros para herramientas. Los recubrimientos fueron depositados en un baño de bórax con ferroniobio, ferrovanadio y aluminio, con temperaturas de depósito de 1223, 1293 y 1363 K para 2, 3, 4 y 5 h. Los recubrimientos producidos se caracterizaron con microscopía electrónica de barrido (SEM), difracción de rayos X (DRX) y espectroscopia de dispersión de energía (EDS). Se estudiaron las tasas de crecimiento de los recubrimientos, y se construyó un modelo de la cinética del espesor de la capa como una función del tiempo y la temperatura del proceso. La dureza y el coeficiente de fricción (COF) de los recubrimientos se midieron a través de nanoindentación y bola sobre disco, respectivamente. Las capas de carburo presentaron un espesor homogéneo alcanzando durezas de 37,63 GPa que está cerca de valores obtenidos en materiales súper-duros, y el COF presento valores cercanos de 0,3 para los aceros recubiertos.

Palabras clave: Difusión Termo-Reactiva, Carburo de Niobio-Vanadio, dureza, desgaste.

Various surface treatments are used to produce coatings, mainly of carbides and nitrides of transition metals, in order to improve the tribological performance of materials that are subjected to conditions of high wear, such as forming dies and cutting tools such as drills and chisels in the metalworking sector. These coatings are generally produced using techniques such as physical vapor phase deposition (PVD) and chemical deposition in the vapor phase (CVD) [1,2]. These procedures require the use of complex equipment and high-vacuum conditions, which makes their industrial implementation costly. An affordable and competitive alternative has been the application of hard coatings with excellent wear resistance through deposition via the thermo reactive diffusion (TRD) process [3] on substrates containing a carbon percentage higher than 0.3% w [4]. The coatings deposited through this process exhibit good adhesion to the substrate, good thickness uniformity, low friction coefficients, and a high degree of hardness. [5]. In the TRD process, a bath of molten salts formed by borax, aluminum as a reducing element, and carbide forming elements (CFE's) such as titanium, niobium, chromium, and vanadium can be used [4]. The carbide is formed when the metal element dissolved in the salt bath has a relatively low energy of formation of carbide and an energy of formation of the oxide higher than that of boron oxide (B2O3) [6,7]. If this condition is not fulfilled, the boron atoms are not oxidized and are free to diffuse in the steel, where they combine with the iron to form layers of iron boride (FeB or Fe2B) [8,9].

The research reported to date has studied binary metal carbides deposited via the TRD process, and has characterized its wear resistance, microstructure, and mechanical behavior. For example, several studies have focused on producing VC and NbC coatings on substrates of AISI H13, AISI M2, and AISI D2 steel, reporting hardness values of up to 2300 HV [3,5]. Other investigations have studied the growth kinetics of niobium carbides on AISI 1040 [10], iron boride on AISI 4140 steel [11], and chromium carbide on AISI D2 steel [12| These papers reported the growth kinetics for binary systems such as NbC or VC, obtaining a layer thickness that depended on the time and the temperature of the treatment. Regarding electrochemical behavior, there are studies of iron boride and aluminum formed using borax on steels AISI S1, S2, and S3, varying the manganese content [13] and there are also studies about tribological and electrochemical performance of niobium-chromium carbides [14]. However, there are few micro-structural and electrochemical studies of possible ternary systems produced using two EFC's. Therefore, the aim of this work is to deposit ternary niobium-vanadium carbide coatings on a D2 steel substrate using the thermo reactive deposition technique and to study their structure, hardness, and wear resistance.

2. Experimental Methods

Coatings of ternary carbides were deposited on AISI D2 tool steel. The dimensions of the substrate were 15 mm in diameter and 4 mm in thickness. The preparation of the samples was performed through metallographic polishing with 220, 360, 400, 600, 1000, and 1200 sandpaper, ending with an ultrasonic cleaning in acetone. The chemical composition of the steel, as stated by the manufacturer, was 1.5% wt C, 11.5-12.5% wt Cr, 0.15-0.45% wt Mn, 0.8% wt Mo, 0.8 wt% V and Fe balance. The steel samples were supplied in the annealed condition with a maximum hardness of 2.6 GPa.

The coatings were produced using a salt bath formed by molten borax (Na2B4O7), 8% wt ferro-niobium (Fe-Nb), 8% wt ferro-vanadium (Fe-V) and 3% wt aluminum (Al). The aluminum that is added in the salt bath had the role of a reducing agent, i.e., it reduced boron oxide and oxidized it in order to prevent the added metal (niobium or vanadium) from oxidizing so that these metals are released and can bind with the carbon of the steel to form carbides. The growth rates of the coatings were studied through surface treatment with TRD at three different temperatures of 1223, 1293, and 1363 K for 2, 3, 4, and 5 hours, with preheating at 873 K.

The crystallographic structure of the coatings was analyzed using an X-PertPro Panalytical difractometer operating at 45 kV and 40 mA, with a configuration of q - 2q at an interval of 30° to 90° and steps of 0.02°, with the Ka line of copper (1.594 Å). The average thickness of the coatings was measured with a Philips scanning electron microscope, working at 40 KV and the EDX was used to determine the chemical composition of the coatings surfaces. The wear cracks were observed using the confocal microscopy. The hardness and the elastic modulus of the coatings were determined from measurements of nano-indentation using CSM Instruments equipment, with a Berkovich indenter at a speed of approach to the sample of 2000 nm/min with a load application rate of 10 mN/min and a maximum load of 30 mN every 15 s. The results obtained were results from an average of 6 measurements. The tribological properties of the coatings were measured at room temperature, via a CETR-UMC-2 tribometer, using the ball-on-disk technique with Al2O3 balls of 6 mm diameter and applying a load of 4 N at a speed of 50mm/s for 10 minutes with a humidity of 60%.

3. Results and Discussion

Fig. 1 shows the cross section (SEM) of two coatings of vanadium-niobium obtained at temperatures of 1363 and 1223 K for a deposit time of 5h for both coatings. The images allow it to be established that the coatings have thicknesses of 19.3±3.2 mm and 9.6±0.8 mm, respectively. Higher values of thickness were obtained at the higher temperature. Furthermore, the micrograph allows it to be seen that coatings were formed by splats with dimensions on the order of a micron.

Fig. 2 shows the coating thickness as a function of time, with the temperature as a study parameter. Fig. 2a shows that the thickness does not exhibit linear behavior with respect to the time of thermal treatment. Therefore, it can be assumed that the rate of diffusion of carbon in the layer determines the growth rate of the coatings [15], and if the layer growth is perpendicular to the substrate surface, we can assume that the film growth rate obeys an Arrhenius function (see eq. 1) [10]. From Fig. 2b, the coefficient of diffusion of carbon into the substrate was determined.

Where x is the thickness (cm), D is the diffusion coefficient (cm²/s) and t is the processing time (s). Given that the carbon that diffuses also reacts to form carbides, the diffusion coefficient calculated using Eq. (1) is an effective diffusion coefficient and, therefore, it does not correspond to pure diffusion of carbon in the substrate.

The relationship between diffusion coefficient and temperatures is given by [10]:

Where D₀ is the frequency factor, Q is the activation energy, R the gas constant and T is the absolute temperature. Linearizing this function, one obtains:

Fig. 3 shows the behavior of ln D as a function of the reciprocal of the temperature. In the inset, we show the values of the slope and the intercept, which allow determining that the activation energy Q = 137.688 kJ/mol (Ec. 3) and the value of D₀ is 3.88x10^-5. Substituting these numerical values into Equation 1, Equation 4 is obtained.

This expression allows calculating the thickness (t) of the coating in microns. In Table 1, the measured and calculated thickness values are shown.

Fig. 4 shows the XRD patterns of the coatings of the ternary vanadium-niobium carbides, obtained at a temperature of 1263 K and a time of 4 h of heat treatment. In order to establish the difference between the crystallographic phases in NbC and VC binary coatings [16] and in the ternary system, the three systems are included in the figure. Comparing XRD patterns allows establishing that the ternary vanadium-niobium carbides exhibit only the NbC crystallographic phase [17]. This result can be explained by considering that the free energy of the formation of NbC (-33.6 kcal) [18] is less than the free energy of formation of VC (-24 kJ / mol) [18].

Moreover, peaks with lower intensities present in the XRD pattern of the ternary system were assigned to the NbVC2 phase. The angular positions and the corresponding planes are shown in Table 2. The theoretical value was calculated with Powder Cell 2.4 software, in which values of the NbVC₂ crystallographic system were provided, such as the space group, the group number, and the Wyckoff parameters of this compound [19].

Table 3 shows the atomic percent measured through EDS analysis for every coating. The result shows that there is a greater percent of niobium than vanadium in the carbides; this can be explained based on the low formation energy of this carbide in comparison with the vanadium carbide.

Table 4 shows the values of hardness and elastic modulus of the coatings. Overall, the results show that coatings with ternary carbides increase their hardness value.

This increase in hardness may be associated with two factors: i) the crystal size reduction forming the coating, which is evidenced by the increased FWHM (Full Width at Half Maximum) of the peaks in the XRD, and ii) the macro-deformation, which is evidenced by the shift of the peaks with respect to their position in a binary carbide sample (see Fig. 4) [20-22]. The values of the hardness of the ternary coatings are close to those of the super-hard materials. [23].

3.3. Coefficient of friction

Fig. 5 shows the curve of the coefficient of friction as a function of the duration of wear testing for binary and ternary carbides systems deposited on AISI D2 steel. From the figure it can be established that the coefficient of friction (COF) of the binary carbides is less by about a factor two compared to the COF substrate. These results agree with those obtained in other studies [24-26]. In the ternary carbide systems, the COF values exhibit little change as a function of the percentage of vanadium, the coating with less vanadium having a lower COF value. The decrease of the COF in the coatings can be explained considering that on their surface there is an excess of aliphatic carbon atoms and graphite, which have Van der Waals type bonds. Calculation of the wear was not carried out because the depth of the wear scar was comparable to the roughness of the coating.

Figs. 6a to 6c show the EDS analysis, the morphology, and an image of confocal laser microscopy, respectively, of the NbC coatings. The EDS analysis shows signals of lines energy of Nb, Fe belonging to the substrate, and C. The SEM micrograph shows the plastic deformation of the coating, and the confocal image shows areas of maximum and minimum height of the coating and the line along which the roughness calculation was made. In Figs. 6d to 6f, we show the same information, but for vanadium carbide. The EDS analysis shows the presence of the K line of V, K line of Fe from the substrate, and K line of Al, possibly present in the salt bath. The confocal image allows the delamination of this carbide (dark green areas) to be observed, which appears mainly on the wear track.

Fig. 7 shows a SEM image for a wear track of NbVC₂ carbide deposited on AISI D2 steel. The cracking and delamination of the deposited carbide and the grooves or channels on the worn area is shown. The wear mechanism is based on the formation of cracks and parallel channels during the wear test. According to Adachi and Hutchings, this situation is known as a slotted wear, or an abrasion of two bodies [27]. These slots arise as a result of the cutting action of abrasive microparticles that remain essentially embedded in the ball during the wear test.

Fig. 8 shows the elemental mapping of the coating after the wear test. In the figure it is possible to see the reduction of niobium and vanadium in the wear area, and a large amount of iron is observed in the delaminated layer.

4. Conclusions

Ternary coatings of vanadium-niobium carbides were deposited via TRD; using XRD it was determined that the coatings were formed by binary compounds (NbC and VC) and a ternary one (NbVC2). The hardness value obtained was 38 GPa, which is close to the value obtained in super-hard materials. This increase in hardness can be explained by the formation of micro-deformations in the carbide and a decrease in crystal size, which is evidenced by a broadening of the XRD peaks that are exhibited in the ternary coatings. Additionally, the coating-substrate system was shown to have a lower coefficient of friction than that of the naked substrate. The binary systems show a good performance with hardness between 24 and 25 GPa and better wear performance than uncoated D2 steel.

Acknowledgements

References

[1] Marulanda, D. and Olaya, J.J., Unbalanced magnetron sputtering for producing corrosion resistance multilayer coatings. DYNA, 79 (171), pp. 74-79, 2012.         [ Links ]

[2] Su, Y.L. and Kao, W.H. Optimum multilayer TiN-TiCN coatings for wear resistance and actual application. Wear, 223, pp. 119-130, 1998. DOI: 10.1016/S0043-1648(98)00276-2        [ Links ]

[3] Castillejo, F.E., Marulanda, D., Rodriguez, O. and Olaya, J.J., Electrical furnace for producing carbide coatings using the thermoreactive deposition/diffusion technique. DYNA, 78 (170), pp. 192-197, 2011.         [ Links ]

[4] Arai, T. and Harper, S., ASM Handbook, 4 - Heat Treating. OH, USA: ASM International, Materials Park, 1991, 448 P.         [ Links ]

[5] Oliveira, C.K.N., Muñoz Riofano, R.M. and Castiletti, L.C., Evaluation of hard coatings obtained on AISI D2 steel by thermo-reactive deposition treatment. Surface and Coatings Technology, 201, pp. 1880-1885, 2006. DOI: 10.1016/j.surfcoat.2006.03.036        [ Links ]

[6] Arai, T., Fujita, H., Sugimoto, Y. and Ohta, Y., Diffusion carbide coatings formed in molten Borax systems. Journal of Materials Engineering, 9 (2), pp. 183-189, 1987. DOI: 10.1007/BF02833709        [ Links ]

[7] Child H.C., Plumb, S.A. and McDermott, J.J. Proc. Heat Treatment'84, The Metals Society, 310, 1984.         [ Links ]

[8] Sinha. A.K., ASM Handbook, 4 - Heat Treating. OH, USA: ASM International, Materials Park, 1991. pp. 437 P.         [ Links ]

[9] Sen, S., Ozbek, I., Sen, U. and Bindal, C., Mechanical behavior of borides formed on borided cold work tool steel. Surface and Coatings Technology, 135, pp. 173-177, 2001. DOI: 10.1016/S0257-8972(00)01064-1        [ Links ]

[10] Sen, U., Kinetics of niobium carbide coating produced on AISI 1040 steel by thermo-reactive deposition technique. Materials Chemistry and Physics, 86 (1), pp. 189-194, 2004. DOI: 10.1016/j.matchemphys.2004.03.002        [ Links ]

[11] Sen, S., Sen, U. and Bindal, C., The growth kinetics of borides formed on boronized AISI 4140 steel. Vacuum, 77 (2), pp. 195-202, 2005. DOI: 10.1016/j.vacuum.2004.09.005        [ Links ]

[12] Sen, S., A study on kinetics of Cr_xC-coated high-chromium steel by thermo-reactive diffusion technique. Vacuum, 79 (1-2), pp. 63-70, 2005. DOI: 10.1016/j.vacuum.2005.01.009        [ Links ]

[13] Tavakoli, H. and Mousavi Khoie, S.M., An electrochemical study of the corrosion resistance of boride coating obtained by thermo-reactive diffusion. Material Chemistry and Physics, 124 (2-3), pp. 1134-1138, 2010. DOI: 10.1016/j.matchemphys.2010.08.047        [ Links ]

[14] Castillejo, F.E., Marulanda, D.M., Olaya, J.J. and Alfonso, J.E., Wear and corrosion resistance of niobium-chromium carbide coatings on AISI D2 produced through TRD. Surface & Coatings Technology, 254, pp. 104-111, 2014. DOI: 10.1016/j.surfcoat.2014.05.069        [ Links ]

[15] Arai, T., Komatsu, N. and Mizutani, M., Diffusion layers on steel surfaces immersed in fused borax baths containing various kinds of additives. Journal of the Japan Institute of Metals, 39, pp. 247-255, 1975.         [ Links ]

[16] Castillejo, F., Marulanda, D., Olaya, J.J. and Rodriguez, O., Electrical furnace for producing carbide coatings using the Thermoreactive deposition/diffusion technique. DYNA, 78 (170), pp. 192-197, 2011.         [ Links ]

[17] Villars, P. and Calvert, L.D., Pearson`s Handbook of Crystallographic Phases. USA: ASM international, 1991.         [ Links ]

[18] Worrell, W.L. and Chipman, J., The free energies of formation of the vanadium, niobium, and tantalum carbides. Journal of Physical Chemistry, 68 (4), pp. 860-866, 1964. DOI: 10.1021/j100786a027        [ Links ]

[19] Paunovic, M. and Schlesinger, M., Fundamentals of electrochemical deposition, New York, USA: John Wiley & Sons, Inc., 1999.         [ Links ]

[20] Oliver, W.C. and Pharr, G.M., An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. Journal of Materials Research, 7 (6), pp 1564-1583, 1992. DOI: 10.1557/JMR.1992.1564        [ Links ]

[21] Niels, H., Hall-Petch relation and boundary strengthening. Scripta Materialia, 51 (8), pp. 801-806, 2004. DOI: 10.1016/j.scriptamat.2004.06.002        [ Links ]

[22] Veprek, S., Niederhofer, A., Moto, K., Bolom, T., Männling, H.-D., Nesladek, P., Dollinger, G., and Bergmaier, A., Composition, nanostructure and origin of the ultrahardness in nc-TiN/a-Si₃N₄/a- and nc-TiSi₂ nanocomposites with H_V=80 to ≥105 GPa. Surface and Coatings Technology, (133-134), pp. 152-159, 2000. DOI: 10.1016/S0257-8972(00)00957-9        [ Links ]

[23] Bourithis, L. and Papadimitriou, G.D., The effect of microstructure and wear conditions on the wear resistance of steel metal matrix composites fabricated with PTA alloying technique. Wear, 266 (11-12), pp. 1155-1164, 2009. DOI: 10.1016/j.wear.2009.03.032        [ Links ]

[24] Staia, M.H., Pérez-Delgado, Y., Sanchez, C., Castro, A., Le Bourhis, E. and Puchi-Cabrera, E.S., Hardness properties and high-temperature wear behavior of nitrided AISI D2 tool steel, prior and after PAPVD coating. Wear, 267 (9-10), pp. 1452-1461, 2009. DOI: 10.1016/j.wear.2009.03.045        [ Links ]

[25] Castro, G., Fernández-Vicente, A. and Cid, J., Influence of the nitriding time in the wear behaviour of an AISI H13 steel during a crankshaft forging process. Wear, 263 (7-12), pp. 1375-1385, 2007. DOI: 0.1016/j.wear.2007.02.007        [ Links ]

[26] Castillejo, F.E. and Olaya, J.J., Recubrimientos de carburos ternarios depositados con la técnica TRD, PhD. Thesis, Departmento de Ineniería Mecánica y Mecatrónica, Universidad Nacional de Colombia, Bogotá, Colombia, 2013.         [ Links ]

[27] Adachi K., and Hutchings I.M., Wear-mode mapping for the micro-scale abrasion test. Wear, 255 (1-6), pp. 23-29, 2003. DOI: 10.1016/S0043-1648(03)00073-5        [ Links ]

F.E. Castillejo-Nieto, received his BSc. Eng in Metallurgical Engineering in 1997, his MSc. in Physics in 2005, and his PhD degree in Engineering in 2014, all of them from the Universidad Nacional de Colombia. Medellin, Colombia. His research interests include: Coatings and thin films, wear and corrosion resistance of materials.

J.J. Olaya, is an associate professor at the Departamento de Ingeniería y Mecatrónica in the Universidad Nacional de Colombia, Bogotá, Colombia. He conducts research in the general area of development and applications of thin films deposited by plasma assisted techniques, corrosion and wear. He received his PhD in 2005 from the Universidad Nacional Autonoma de México, Mexico.

J.E. Alfonso, completed his BSc. degree in Physics in 1987 and his MSc. degree in Science - Physics in 1991, all of them being obtained from the Universidad Nacional de Colombia. In 1997 he completed his PhD in Science - Physics in the Universidad Autonoma de Madrid, Spain. He has taught at the Universidad Nacional de Colombia as a professor since 2000, where his research has been focused on material science, particularly on thin film processing as well as performance characterization, studying thin film optical, electrical and mechanical properties.