This article is one of eight papers to be presented exclusively on the web as part of the January 2000 JOM-e—the electronic supplement to JOM.
	The following article appears as part of JOM-e, 52 (1) (2000), http://www.tms.org/pubs/journals/JOM/0001/Agarwal/Agarwal-0001.html JOM is a publication of The Minerals, Metals & Materials Society

Functional Coatings: Research Summary

The Functionality of Laser-Surface-Engineered Composite Titanium Diboride Coatings

Arvind Agarwal and Narendra B. Dahotre

This study was undertaken to identify and develop a unique combination of coating technique and material to target properties similar to those of an ideal coating. An Nd:YAG laser was used to deposit ultrahard TiB₂ ceramic on AISI 1010 steel within a fixed envelope of processing parameters, such as laser power and traverse speed. A uniform, continuous, and crack-free coating with a metallurgically sound interface was obtained. The coating is composite in nature, comprising TiB₂ particles and iron from steel trapped in the laser-melt zone. The coating was evaluated for various properties, such as mechanical (hardness, elastic modulus, interfacial energy), chemical and structural (x-ray diffraction, scanning electron microscopy, high-resolution transmission electron microscopy) and high-temperature oxidation and corrosion properties. The coating is hard and tough in nature. It also has been concluded that a TiB₂ coating has a high interfacial energy, suggesting a stronger and adherent interface. The composite TiB₂ coating has significant oxidation resistance up to 800°C. Also, it has shown improved high-temperature corrosion resistance against liquid aluminum. Such features contribute significantly toward the goal of achieving an ideal coating.

INTRODUCTION

The ability to change the surface characteristics of a component/structure without changing the bulk material properties by use of surface coatings has opened up many new and diversified applications in several technological areas. Surface engineering or modification has also led to the development of various coating techniques and materials. Such coatings enable the use of the structure/component even in aggressive environments. Ceramic-based coatings have enormous commercial potential for applications, such as high-temperature coatings for turbine blades, conductive and erosion-resistant coatings for arc heaters, and protective and wear-resistant coatings for machine and tools. Moreover, they are also used in the electronic industry (high-temperature sensors), the sports industry (golf club heads, snow skis, shoe spikes, etc.), and medical equipment. An ideal coating could possibly be identified by its properties, such as high hardness, high-temperature strength, wear and erosion resistance, oxidation and corrosion resistance, and strong adherence to the substrate.

Conventionally, surface-modification processes like chemical vapor deposition (CVD), physical vapor deposition (PVD), and thermal spray are being used to deposit various ceramic coatings. However, the use of these techniques has been limited by several factors, such as the need for high input energy, the cost of a vacuum chamber, the size of the work piece, poor bond strength, and environmental hazards. Several new coating methods, such as pulse electrode surfacing, ion-beam assisted deposition techniques, and laser-assisted techniques, are being investigated.

In this study, hard TiB₂ ceramic was deposited over AISI 1010 steel substrate using a continuous wave Nd:YAG laser to produce a hard, wear-resistant coating to achieve the goals of an ideal coating. Titanium diboride (TiB₂) is a transition-metal-based refractory ceramic that has unique properties of extremely high hardness, a high melting point, wear resistance, corrosion resistance, and electrical conductivity.^1,2 It also has excellent corrosion resistance against metallic aluminum and cryolite, which makes it a suitable candidate as a coating material for inert cathodes used in the Hall-Héroult cell for aluminum electrolysis.^3,4

In spite of several attractive properties, TiB₂ is not widely used for commercial applications due to processing problems. Conventionally, TiB₂ ceramics are processed by hot pressing and sintering, which are very expensive and require high-temperature processing.⁵ Moreover, near-net-shape forming is also impossible.⁵ However, surface engineering is an alternative way to modify the surface properties similar to that of TiB₂ without altering the bulk properties of the engineering component or structure. Laser-surface engineering (LSE) is one of the surface-modification processes where a preplaced ceramic powder precursor is melted with a thin layer of the substrate to produce a laser-melt zone of desired composition.

Laser-surface engineering offers several advantages over other surface-modification techniques. The most important advantage stems from the fact that laser-surface modification is a nonequilibrium synthesis involving high cooling rates (10³-10⁸ K/s), which produce metastable phases by exceeding the solid-solubility limit beyond the equilibrium phase diagram.⁶ This leads to the development of a wide variety of microstructures with novel properties that cannot be produced by any conventional processing technique. Moreover, these coatings are metallurgically bonded, providing a sound and adherent interface between the coating and substrate. The surface structure can be tailored to the surface requirement of the application by varying process variables, such as laser traverse speed, power, and beam size and type and composition of the precursor.^6,7 Also, the laser beam has an excellent spatial resolution that makes it ideal for depositing coating on miniature-sized components, such as electronic sensors for high-temperature applications. Another advantage of laser-surface modification is that the laser beam can be transported to any remote location through fiber optics. This allows the deposition of coatings on components or parts that are remotely located.

CERAMIC/METAL INTERFACE

The prime criterion for a coating with an acceptable performance is a sound and adherent interface. The coating/substrate interface in the laser-engineered TiB₂ coating has been characterized at the microstructural, atomic, and mechanical level. Figure 1 shows the overall cross-sectional view of the composite TiB₂ coating on AISI 1010 steel substrate; a uniform, continuous, and crack-free coating is obtained. The coating is composite in nature, comprising TiB₂ particles and iron from steel trapped in the laser-melt zone. Needle/acicular-shaped boride particles are uniformly distributed in the melt zone. Figure 2 shows an adherent and metallurgically sound coating/substrate interface.

A bright-field TEM image of the composite coating zone is shown in Figure 3. TiB₂ particles of various shapes and sizes are present in the coating zone, and there is no crack at the ceramic/metal interface, suggesting a good bonding between TiB₂ particle and iron. A high-resolution transmission electron microscopy (HRTEM) image of the TiB₂ particle and iron interface is shown in Figure 4. A nanosize reaction zone occurs between TiB₂ particle and iron, suggesting the formation of metastable phase(s) in trace amounts.

The size of the reaction zone at the ceramic/metal interface is approximately 70-100 nm, and the reaction products are nanocrystalline in nature. It has been experimentally and theoretically estimated that iron does not react and acts as an excellent binder for TiB₂.^11,12 However, the existence of a small reaction zone between TiB₂ and iron is contrary to the above-mentioned fact, which is attributed to nonequilibrium synthesis during LSE.⁶ The laser beam has a high-energy density input of 15 MW/m² that leads to the formation of metastable phase(s) as a consequence of nonequilibrium synthesis. This fact is further corroborated by the x-ray diffraction spectrum of the TiB₂-coated surface (Figure 5). Apart from the major phases (TiB₂ and iron), some metastable phases(s), such as type Fe_aB_b and Ti_mB_n, are also observed.

HR-TEM provides very precise information about the ceramic/metal interface at the atomic level. These images can reveal many kinds of defects, which can be chemical or structural.¹³ Chemical defects occur if any deviation from the exact composition of two phases at the interface exists. Such deviation is often caused by the reaction product or interphase formation at the ceramic/metal interface, which is atomically smooth and sharp in nature (Figure 6). Also, there is no atomic-level amorphous phase or layer at the TiB₂/Fe interface, which suggests a good ceramic/metal bond at the interface. However, two-dimensional lattice fringes extend up to 5 nm into a one-dimensional iron lattice and bond directly with iron with a minimal mismatch (marked with an arrow in the figure). Such behavior was earlier observed in 6061Al/Si₃N₄ composites, where nanocrystalline MgO and MgAl₂O₄ particles grew at the interface and bonded with Si₃N₄ whiskers.¹⁴ Thus, it is very likely that some nanocrystalline phase or reaction product forms at the interface and grows into the iron matrix. Again, such observation is in accordance with the earlier hypothesis of the formation of metastable phase(s) during laser processing.

Figure 6 also shows the diffraction patterns obtained from the iron-rich side (Figure 6a) and the TiB₂ side (Figure 6b) of the interface. The selected-area diffraction (SAD) pattern corresponds to body-centered cubic (b.c.c.) iron along the [001] direction of the zone axis. However, there are additional diffraction spots that do not correspond to iron. Such spots (marked by an arrow) further correspond to the presence of a reaction product or precipitate or a metastable phase at the ceramic-metal interface. The SAD pattern obtained from the TiB₂ side of the interface corresponds to diffraction spots obtained from TiB₂ along the [0001] zone axis. Some of these diffraction spots have diffused intensity along with the presence of double-diffraction spots. Such diffraction patterns are often obtained from the presence of dislocations and stacking faults, which can lead to the volume distortion resulting in double-diffraction spots.¹⁵ The details of defects and precipitates within the TiB₂ particles are presented elsewhere.¹⁶

Several researchers have suggested that the formation of interfacial reaction products promotes wetting between the ceramic/metal interface, leading to a stronger and adherent interface.^17,18 Moreover, nanocrystalline reaction products and phase(s) formed at the ceramic/metal interface act as a stress absorber, further reducing the chances of crack propagation.¹⁹ Such microstructural features suggest a strong coating/substrate and ceramic/metal interface within the coating region. However, this coating would be highly functional only if the mechanical properties of the coating/substrate are also enhanced.

The adhesion of the coating is the most critical theme of the coating-substrate system. In most applications, the minimum criterion for acceptable performance is that the adhesion of the coating to the substrate should be sufficient so that the coating remains in place for the lifetime of the component in its operating environment. Thus, it is essential to measure the level of adhesion between substrate and coating in order to guarantee that the coating is fit for the purpose.

A four-point bend test is an appropriate method to evaluate the interfacial properties (critical-energy release rate) of a bimaterial.²⁰ The four-point flexure test is based on the storage of a well-known amount of elastic energy on bending and a release of this elastic energy on fracture. The critical-energy release rate has been computed using Equation 1 as²⁰

It should be emphasized here that prefix 1 denotes a composite TiB₂ coating, and prefix 2 stands for AISI 1010 steel substrate. The various parameters needed to compute the critical-strain-energy release rate are Poisson's ratio (n) of the coating and the substrate, elastic modulus (E) of the coating and the substrate apart from coating thickness (h₁), overall sample thickness (h), and critical-bending moment (M_c). E (477GPa) was obtained for the coating from the nanoindentation data; whereas, Poisson's ratio for the composite coating was computed using the rule of mixtures (ROM).²¹ Coating thickness is 200 µm, and the total thickness of the four-point bend sample is 1.5 mm.

Cracks in the composite TiB₂ coating are oriented vertically through the entire coating thickness and, as a consequence, do not cause delamination of the coating. Figure 7 shows a cross-sectional view of such a sample. No crack has initiated from the precrack/notch that was created using a diamond saw. However, the precrack has blunted and broadened, suggesting some plastic deformation. The vertical cracks are formed in the entire coating with 800-1,000 µm spacing between two successive cracks. Much of the crack propagates through a brittle ceramic-particle (TiB₂)-rich region, suggesting that crack energy was not dissipated fully, leading to a crack that runs through the coating and is deflected and blunted when it reaches the substrate (Figure 8). Cracks in the composite TiB₂ coating are oriented vertically through the entire coating thickness and, hence, do not cause delamination of the coating. The critical-energy release rate, G_c, computed for the point when the crack reaches the interface, was found to be 265 J/m². It should be noted that the energy-release rate for a perfectly brittle cleavage fracture is approximately 5 J/m²; whereas, the fracture energy for ductile steel is on the order of 300 J/m².²⁰ The critical-energy release rate in our case was found to be very high, suggesting a composite kind of strong interface.

WEAR RESISTANCE OF TiB₂ COATING

Such enhanced interfacial properties of a composite TiB₂ coating synthesized using a laser-based technique are expected to perform in a superior manner for wear applications. The most important application of ultrahard TiB₂ coatings seems to be as a wear-resistant coating for various machine tools and parts. The dry-sliding wear tests were conducted using a block-on-disk tribometer. Wear-test results are presented in Figure 9, which shows the weight loss over a 20 minute period. The comparative wear rates (obtained by the slope in Figure 9) for the LSE surface and AISI 1010 steel are presented as a bar diagram in the inset. Weight loss is drastically reduced for the LSE-coated surface in comparison to the AISI 1010 steel substrate. The wear rate for the LSE surface is decreased by a factor of 15 in comparison to the uncoated AISI 1010 steel substrate. Moreover, the wear rate for the LSE-coated surface tends to stabilize after an initial period of four minutes, suggesting the smoothening and/or breaking off of hard asperities and thereby providing a flat contact with the composite composition.

The wear mechanism in such multiphase composite materials is very complex and depends upon several factors, such as volume fraction, distribution, and morphology of the ceramic particles.²² Under ideal conditions, a monolithic ceramic such as TiB₂ is expected to fail in a brittle manner.¹ However, the presence of the softer matrix phase, iron, between the hard ceramic particles alters the wear nature of monolithic TiB₂. It has been mentioned and validated earlier that iron acts as an excellent binder for titanium-based refractory ceramics.^23-25 This prevents the debonding at the TiB₂ particle/iron matrix interface.¹⁸ The strength of the TiB₂/Fe interface has already been addressed at the microstructural, chemical, and atomic level. Under such materials conditions, iron-based matrix phases deform plastically to accommodate the high stresses experienced by TiB₂ particles. This prevents brittle fracture and fragmentation of TiB₂ ceramic particles. Hence, chipping off is prevented, and no loose debris is observed. A composite layer containing hard TiB₂ particles and soft iron phase(s) covers the entire substrate, which assists in reducing the wear rate. This suggests that wear occurs mainly by the plastic deformation of the softer matrix iron phase, resulting in very little loss of material. Figure 10 is the surface roughness profile of the worn surface of the LSE-coated sample. The wear surface clearly indicates the depressions (plastic deformation) caused by the rotating disk of the tribometer.

HIGH-TEMPERATURE OXIDATION IN AIR

The elevated-temperature oxidation behavior of TiB₂ coating in air was evaluated by thermogravimetric analysis (TGA) to study oxidation kinetics and long-term exposure to study oxide-scale growth and morphology.²⁶ The oxidation behavior of composite TiB₂ coating has been characterized through measurements of the mass change per unit surface area (Dm/a) during exposure at various temperatures and time. The oxidation rate for a composite TiB₂ coating was parabolic for all test temperatures of 600°C, 700°C, 800°C, and 1,000°C. The oxidation rate was very slow for 600°C and 700°C, whereas 800°C and 1,000°C showed higher oxidation rates. The activation energy, Q, was computed as 205 kJ/mol.

Long-duration exposure of composite TiB₂ coating at 600°C did not show significant oxide-scale formation. A protective layer of B₂O₃ forms, which evaporates at 775°C, causing an increase in the oxidation rate at 800°C. A porous and lamellar oxide scale forms on the composite coating and grows in thickness for a longer duration of exposure. Several complex mixed oxides of type Ti_aO_b, Fe_mO_n, and Fe_xTi_yO_z are formed within the scale, which further complicates the oxidation mechanism. Oxidation is even more severe at 1,000°C due to the rapid evaporation of B₂O₃. The attack is severe at the coating/substrate interface due to the formation of Fe₂O₃ that causes delamination of the coating layer from the substrate. The oxide surface gets rough in nature as complex oxides nucleate and grow upward on the surface.²⁶

A neural network has been successfully used to study the oxidation behavior of composite TiB₂ coating at intermediate temperatures of 750°C and 950°C.²⁷ It minimizes the number of experiments and predicts the trend curves with acceptable errors. There was no change in the oxidation mechanism of TiB₂ coating in the temperature range of 600°C to 1,000°C in comparison to earlier minimal experimental study. The activation energy for the oxidation of composite TiB₂ coating was 210 kJ/mol, which is similar to earlier studies. Evaporation of B₂O₃ was suggested by the simulated curves at 750°C and 950°C without performing the experiments.

HIGH-TEMPERATURE CORROSION RESISTANCE AGAINST LIQUID ALUMINUM

In addition to excellent wear resistance and significantly high oxidation resistance, TiB₂ has excellent corrosion resistance against metallic aluminum and cryolite, which makes it a suitable candidate as a coating material for the inert cathode used in Hall-Héroult cells for aluminum electrolysis.^3,4 Moreover, TiB₂ is easily wetted by liquid aluminum due to its low contact angle.²⁸ Such a unique combination of properties of TiB₂ makes it an ideal candidate to be used as mold coating material for the aluminum casting industry.

Laser-deposited composite TiB₂ coating was evaluated for its high-temperature corrosion properties against liquid aluminum by performing a simple dip test. Figure 11 shows the SEM micrographs of the cross section of samples dipped in liquid A356 aluminum for 4 h and 24 h, respectively. A sample immersed in liquid A356 aluminum for 4 h at 780°C showed better resistance in comparison to the sample immersed for 24 h. There was no attack/penetration of liquid aluminum in the coating layer. Few cracks through the thickness of the coating layer were observed as a consequence of thermal shock. The sample immersed in liquid A356 aluminum for 24 h at 780°C was attacked in a relatively severe way; liquid aluminum penetrated the top 50 µm of the coating, which was attributed to the dissolution of TiB₂ particles and its phase transformation to Ti_1.87B₅₀.²⁹ However, a laser-deposited ultrahard composite TiB₂ coating on AISI 1010 steel has shown significant corrosion resistance to liquid A356 aluminum.

CONCLUSIONS

The laser-engineered TiB₂/Fe composite coating demonstrated excellent functionality for various applications involving severe wear and high-temperature oxidation and liquid-metal corrosion environments. Such functionality is attributed to the evolution of a strong interface between both the TiB₂ particle and iron matrix within the composite coating and the composite coating (TiB₂/Fe) and the substrate (AISI 1010 steel) via formation of unconventional and nonequilibrium reaction products. Furthermore, the employment of an Nd:YAG laser with fiber-optic beam delivery provides limitless opportunities to synthesize and/or fabricate such a coating into various shapes and sizes, thereby making it even more functional for industrial and commercial applications.

ACKNOWLEDGEMENTS

The authors acknowledge financial support from the U.S. Air Force (contract no. F 40600-96-C-0004) for this work. The authors also thank Lalitha Reddy Katipelli for her assistance with image processing.

References
1. R. Cutler, Engineering Materials Handbook: Ceramic and Glasses (Materials Park, OH: ASM, 1995), p. 787.
2. A. Agarwal and N.B. Dahotre, Elevated Temperature Coatings: Science and Technology III, ed. J.H. Hampikian and N.B. Dahotre (Warrendale, PA: TMS, 1999), pp. 273-284.
3. K. Billehaug and H. Øye, Aluminum Verlag (Dusseldorf), 56 (1980), p. 642.
4. H. Øye and M. Sørlie, Aluminum Verlag (Dusseldorf), 2 (1994).
5. C. Mroz, Ceramic Bulletin, 78 (6) (1999), p. 77.
6. J. Mazumder, Metallurgical and Ceramic Protective Coatings, ed. Kurt H. Stern (London: Chapman and Hall, 1996), p. 74.
7. K. Komvopoulos, Materials Science Forum, 163-165 (1994), p. 417.
8. O.M. Akselsen, J. Mater. Sci., 27 (1992), p. 1989.
9. A. Agarwal and N.B. Dahotre, Int. J. Refractory Mater. and Hard Metals, 17 (4) (1999), p. 283.
10. K. Komvopoulos and K. Nagarathnam, J. Engg. Materials and Technology, 112 (1990), p. 131.
11. A. Agarwal and N.B. Dahotre, Surf. Coat. Tech., 106 (2/3) (1998), p. 242.
12. A.A. Ogwu and T.J. Davies, Mater. Sci. & Tech, 9 (1993), p. 231.
13. A.G. Evans and M. Rühle, MRS Bulletin, 10 (1990), p. 46.
14. X.G. Ning et al., Mat. Res. Soc. Symp. Proc., ed. W.A.T. Clark, U. Dahmen, and C.L. Briant (Pittsburgh, PA: MRS, 1991), p. 865.
15. D.B. Williams and C.B. Carter, Transmission Electron Microscopy, II: Diffraction (New York: Plenum Press, 1996), p. 237.
16. A. Agarwal, Ph.D. dissertation, "Laser Surface Engineering of Composite Titanium Diboride Coating on Steel: Synthesis and Characterization" (Knoxville, TN: University of Tennessee, December 1999).
17. Z.X. Guo and B. Derby, Composites, 25 (7) (1994), p. 630.
18. A. Agarwal, N.B. Dahotre, and L.F. Allard, Pract. Metall., 36 (1999), p. 250.
19. X.B. Zhou and J.Th.M. De Hosson, Acta Metall. Mater, 39 (10) (1991), p. 2267.
20. M. van de Burg and J.Th.M. De Hosson, Interface Science, 3 (1995), p. 107.
21. A. Agarwal and N.B. Dahotre, Metall. and Mater. Trans. A (to be published in February 2000).
22. N. Axen, I.M. Hutchings, and S. Jacobson, Tribology Intl., 29 (6) (1996), p. 467.
23. A. Agarwal, N.B. Dahotre, and T.S. Sudarshan, Surf. Engg., 15 (1) (1999), p. 27.
24. A. Agarwal and N.B. Dahotre, J. Mater. Engg. Performance, 8 (4) (1999), p. 479.
25. C. Raghunath, M.S. Bhat, and P.K. Rohatgi, Scripta Metall., 32 (4) (1995), p. 577.
26. A. Agarwal and N.B. Dahotre, Metall. & Mater. Trans. A (to be published in February 2000).
27. A. Godavarty, A. Agarwal, and N.B. Dahotre, communicated to Applied Surface Science (November 1999).
28. S.K. Rhee, Proc. of the Fall Meeting of the Ceramic-Metal Systems Division of the American Ceramic Society (1968), p. 386.
29. A. Agarwal and N.B. Dahotre, Lasers in Engg. 9 (3) (1999).

Arvind Agarwal and Narendra B. Dahotre are with the Department of Materials Science and Engineering, Center for Laser Applications, at the University of Tennessee Space Institute.

For more information, contact N.B. Dahotre, Department of Materials Science and Engineering, University of Tennessee Space Institute, Center for Laser Applications, MS 24, 411 B.H. Goethert Parkway, Tullahoma, Tennessee 37388-9700; (931) 393-7495; fax (931) 454-2271; e-mail ndahotre@utsi.edu.