Microwave assisted in situ composite coating using metallic glass precursor

Introduction

Composite structures have several unique advantages compared to monolithic materials. Composites have high specific strength and modulus, higher elevated temperature resistance, low coefficient for thermal expansion, and better resistance against surface degradation. ¹ Composite coatings have also shown promising results in tribological tests, ² biomedical applications, ³ as well as corrosion applications. ⁴ In contrast to ex situ composites, the in situ composites, wherein the reinforcing phase is synthesised within the matrix, are known to have better properties. ⁵ The improved properties of in situ composites are attributed to higher interfacial strength and improved wettability of the reinforcement with the matrix. Commonly used techniques for synthesising coatings include thermal spraying, physical vapor deposition (PVD), chemical vapour deposition (CVD), and laser assisted deposition. Laser cladding has been utilised for the fabrication of in situ composite coatings.^6,7 Its increased popularity arises from process flexibility, ease of operation and use of cladding design independent of the substrate. ⁸ However, laser cladding becomes economically infeasible for specimens with large surface area.

Most studies on the synthesis of alloy coatings involve the use of elemental powders comprising of a ductile material and harder reinforcements.^9–12 The elemental powders are mixed in the desired ratio followed by homogenisation in a mechanical mixer.^9,10 Since their mechanical properties are very different, complete homogenisation is a challenge. Moreover, contamination, agglomeration, and residual stresses in the individual elemental powders affect the homogenisation process. In the present study, we demonstrate a novel methodology for synthesizing in situ composite coatings using microwave processing and bulk metallic glass powder as precursor material. Utilising metallic glass powder as precursor eliminates the need for prior homogenisation. An earlier report on Fe based amorphous coating demonstrates superior properties by controlling crystallisation in these metastable systems. ¹³ Crystallisation pathways in multi-component metallic glasses can be controlled to achieve different functionality depending on the processing conditions. Microwave processing involves volumetric heating of the material, thereby minimizes thermal gradients and thermal stresses. ¹⁴ Compared to other commonly used coating techniques, microwave processing requires considerably less starting material. Further, the process utilises a microwave for coating synthesis, thereby eliminating the need for sophisticated and expensive equipments. It provides greater flexibility in terms of processing parameters and offers an economical route for coating synthesis. There are only a few studies on microwave assisted coatings^9,10 and the development of in situ composite coatings using metallic glass powders has not been reported so far.

Materials and methods

Fe based metallic glass powder with a nominal chemical composition of Fe₄₈Mo₁₄Cr₁₅Y₂C₁₅B₆ was selected as precursor material for synthesising the coating. Pure aluminium was used as substrate owing to its low hardness and inferior tribological properties. Deposition of hard coatings on aluminium substrate has been demonstrated in many earlier studies.^15–17 Amorphous powder was obtained by gas atomisation. The atomised powders were mechanically sieved to obtain powder within a narrow size range. Scanning electron microscope (SEM) image of the metallic glass powder is shown in Fig. 1a. The particles have predominantly spherical morphology and are in the size range of 1–20 μm. Differential scanning calorimeter (DSC) analysis of the powder is shown in Fig. 1b. The glass transition temperature T_g was 575°C, first crystallisation temperature (T_x1) was 653°C and second crystallisation temperature (T_x2) was 684°C.

OPEN IN VIEWER

1

a SEM image and b DSC of as received Fe based bulk metallic glass powder (Fe₄₈Mo₁₄Cr₁₅Y₂C₁₅B₆): powder has nearly spherical particles with 1–20 μm size range; DSC shows glass transition temperature (T_g), first crystallisation peak (T_x1) and second crystallisation peak (T_x2)

Schematic representation of the microwave processing is shown in Fig. 2. Metallic glass powder was placed on an aluminium substrate with approximately 1 mm uniform powder thickness. The substrate with metallic glass powder was then placed in a graphite crucible. The metallic glass powder was covered by a susceptor material (activated carbon) having high absorbtivity for microwave radiation. The powder was separated from the susceptor material using a thin graphite sheet in order to prevent its contamination. The crucible was placed in a microwave operating at 2·54 GHz frequency and 900 W power for a period of 6 min.

OPEN IN VIEWER

2

Schematic showing deposition of metallic glass powder on substrate material using microwave: substrate with metallic glass powder was placed in graphite crucible; metallic glass powder was covered by a susceptor material (activated carbon), separated using thin graphite sheet in order to prevent its contamination

X-ray diffraction (XRD) was done to determine different phases formed in the coating. Hardness and modulus of the coating was determined by nanoindentation, using a standard Berkovich indenter at a peak load of 20 mN. Scratch test across the substrate coating was done to determine the relative wear loss of the substrate, coating and the interface. Scratch test was done using 6 mm ceramic ball at a constant normal load of 100 N on a universal tribometer (RTEC Instruments). Velocity was kept constant at 0·03 mm s⁻¹ for a scratch length of 1 mm spanning across the aluminium substrate and coating through the interface. Scratch depth at different locations and wear volume loss for the substrate and coating was determined using optical profilometer.

Results and discussion

Microwave processing is a typical dielectric heating in which the material is subjected to alternating electric field. The microwave material interaction causes translation of free charges and dipole rotation. The resistance against this induced motion due to inertial, elastic and frictional forces results in heating of the material. ¹⁴ In contrast to conventional heating that result in steep thermal gradients, dielectric heating is a bulk heating process with volumetric heating of the material. The power absorbed during microwave heating is given as where f is the microwave frequency, ϵ₀ is the permittivity of the free space, tanδ is the loss tangent which quantifies the dielectric loss and E is the magnitude of internal field. ¹⁴ Activated carbon has a high dielectric constant as well as dielectric loss factor. It can significantly increase the fraction of the incident microwave absorbed, and there is a substantial rise in temperature. Most metallic materials show poor absorbtivity of microwave radiation at room temperature. In other words, the skin depth of most metallic materials at room temperature is small. However, the skin depth tends to increase with the rise in temperature, resulting in increased absorption of incident radiation. ¹⁸ Therefore, the temperature rise caused by carbon heating aids in establishing microwave-material interaction.

SEM images of the coating are shown in Fig. 3a through c. The coating surface is uniform and free from defects such as porosity and cracks. The coating thickness was measured at different places and average value was found to be nearly 1 mm. As shown in Fig. 3b and c, the coating microstructure comprises of three different phases: the white phase (marked by ‘A’), the dark phase (marked by ‘B’) and lamellar phase ‘C’. Energy dispersive spectroscopy (EDS) analysis of these phases is shown in Fig. 3c. The phases ‘B’ and ‘C’ have nearly similar composition. Both these phases are Fe rich and nearly devoid of Mo, whereas phase ‘A’ is Mo rich and devoid of Fe. XRD analysis for the coating is shown in Fig. 4a. The prominent phases have been marked in the figure which indicates the formation of Mo rich and Fe rich phases. XRD analysis also shows some Boron rich phases which are not shown in EDS because of its incapability to detect it. EDS elemental mapping was done for the coating substrate region, as shown in Fig. 4b. The elemental mapping indicates significant diffusion of elements across the coating-substrate interface. This diffusion signifies metallurgical bonding of the coating with the substrate, which is known to provide significant bond strength to the coating. ⁹

OPEN IN VIEWER

3

a low magnification; b high magnification SEM images of Fe based metallic glass coating on aluminium substrate; c EDS analysis of three different phases in coating: phases ‘B’ and ‘C’ are Fe rich and nearly devoid of Mo, whereas phase marked ‘A’ is Mo rich and devoid of Fe

OPEN IN VIEWER

4

a XRD analysis; b EDS mapping of Fe based metallic glass coating: XRD analysis indicates Fe rich and Mo rich phases; EDS mapping indicates significant diffusion of elements across coating substrate interface, which results in metallurgical bonding

Load–displacement curves obtained from nano-indentation for the harder and softer phases are shown in Fig. 5. Average values of hardness and modulus are marked alongside the figure. The hardness of ‘B’ and ‘C’ were found to be comparable. The hardness of phase ‘A’ is higher compared to phase ‘B’ by more than a factor of two. Thus, the coating comprises of a composite micro-structure, in which the hard phase is evenly distributed in a softer phase. These phases are formed in situ during processing. The average hardness of the coating's hard phase (∼15 GPa) is higher than the hardness of WC-Co coating (∼11 GPa) ¹⁹ and thermally sprayed Fe₄₈Mo₁₄Cr₁₅Y₂C₁₅B₆ coating (∼10 GPa). ²⁰ Coating's harder phase can act as strong reinforcement for the matrix and result in high strength composite material.

OPEN IN VIEWER

5

Load–displacement curves for Fe based metallic glass coating during nano-indentation: load–displacement curves for softer and harder phase of coating are shown separately; hardness (H) and modulus (E) values for harder and softer phase are marked along-side load–displacement curves

The results of scratch test are shown in Fig. 6. 3D image of the scratch across the substrate and coating is shown in Fig. 6a. The substrate shows significantly higher deformation compared to the composite coating. The wear volume loss for the substrate and coating, calculated from the profiler image, was found to be 3×10⁻³ and 1·57×10⁻³ mm³ respectively. Thus, the wear volume loss for the coating is nearly 50% lower compared to the substrate. Depth profile at different locations along the scratch is shown in Fig. 6b. Location 1-1 is on the aluminium substrate and location 4-4 is on the coating. The intermediate point (location 2-2) represents the transition from the substrate to the coating through the interface (location 3-3). The scratch depth in the substrate is roughly three times that of the coating and nearly two times larger than that at the interface which indicates higher hardness of the coating compared to the substrate. Variation of the frictional force (F_x) and coefficient of friction (COF) across the scratch length is shown in Fig. 6c. F_x and COF gradually rise on moving from substrate towards the substrate-coating interface. The average F_x and COF values are significantly higher in the interface region compared to the substrate. Both these values show a further increase in the coating (Fig. 6c). Higher friction coefficient for the coating is in contrast to its high hardness. This may be attributed to abrasive wear caused by fracture of the harder phase in the coating during scratch. Increase in friction coefficient by harder reinforcement in composites has been reported earlier during a single point scratch test. ²¹ Lower wear loss and high frictional force values indicate good interface strength and strong adhesion of coating with the substrate. To summarise, the coating demonstrated high hardness, low wear loss and higher frictional resistance.

OPEN IN VIEWER

6

a 3D image of scratch across substrate and coating; b depth profile of scratch across substrate coating; c variation of frictional force (F_x ) and coefficient of friction (COF) in scratch test across substrate coating: inset of b shows optical profiler image of scratch; scale bar shown in inset figure represents 250 μm; location 1-1 is on aluminium substrate and location 4-4 is on coating; intermediate points represent transition from substrate to coating through interface

Conclusion

A novel methodology for developing in situ composite coating using an iron based metallic glass powder precursor and microwave processing is presented. The coating microstructure comprises of molybdenum rich harder phase uniformly distributed in iron rich softer phase. The coating demonstrates high hardness and modulus as determined from nano-indentation. Scratch test across the substrate coating showed nearly 50% lower wear rate for the coating compared to aluminium substrate. The average frictional force and coefficient of friction increased from the substrate towards the coating. Higher friction coefficient for the coating is explained based on abrasive wear caused by fracture of the harder phase in the coating. This study establishes microwave processing as a promising route for the development of high strength composite coatings from a wide range of metallic glass precursors.