Damage tolerant functionally graded WC–Co/Stainless Steel HVOF coatings

  • Published on

  • View

  • Download

Embed Size (px)


<ul><li><p>C</p><p>ho</p><p>Heagio E</p><p>Functionally graded materialsHigh Velocity Oxygen-Fuel (HVOF) sprayingWear resistanceImpact testingResidual stress</p><p>ageng i10</p><p>ont</p><p>found by scanning electron microscopy and d</p><p>Ms) a</p><p>prole [1], designed in order to improve certain characteristics numerous synergistic factors leading to such cracking (including the</p><p>Surface &amp; Coatings Technology 205 (2010) 21972208</p><p>Contents lists available at ScienceDirect</p><p>Surface &amp; Coatin</p><p>l s(usually mechanical), over monolithic materials [25]. FGM manu-facturing is non-trivial, but is well-suited for thermal spray (TS)deposition methods, largely due to the ability of TS to process dis-similar materials concurrently, in sequential layers of differentcomposition [6]. Layer-by-layer compositional gradients can indeedbe easily manufactured by simultaneously feeding different powdersin the hot ame of the spraying torch [79], and gradually varying theratio of the two or more powders [1016]. This gradation can beobtained by the simultaneous control of the feed rate in various</p><p>mismatch in thermo-mechanical properties, but also the oxidation ofthe metallic bond coat material) [14]. Few studies have, up to now,addressed the subject of thermally-sprayed FGMs for wear andcorrosion protection [1722]. As traditional single-layer coatings forwear protection generally exhibit much higher hardness and elasticmodulus than the substrate (the typical combination being a cermet,like WCCo or WCCoCr, deposited onto steel or light alloy sub-strates), the stress distribution produced under localised contactconditions concentrates inside the coating and along the coating/powder feeders, by strategic injection of theplume, or by previously blending the powdproportions, and periodically replacing thesented here).</p><p> Corresponding author.E-mail address: alfredo.valarezo@gmail.com (A. Vala</p><p>0257-8972/$ see front matter 2010 Elsevier B.V. Aldoi:10.1016/j.surfcoat.2010.08.148icrostructure of whichding to a predetermined</p><p>(MCrAlY) / top coat (Y-PSZ) interface. As a result of those investiga-tions, very complex architectures have been proposed to address thecomposite materials, the composition and mare not spatially homogeneous but vary accor1. Introduction</p><p>Functionally graded materials (FGcurvature measurement technique reveals that during the deposition of the FGC, compressive stresses exist inthe lower, metallic layers owing to peening effect of successive impact, and these gradually evolve to hightensile, in the top layers. Tensile stresses appear to be due to quenching alone; thermal stresses are lowbecause of the gradation. All of this is benecial for the deposition of a thick coating.The FGC structure shows the ability to reduce cracking with increased compliance in the top layer duringstatic and dynamic normal contact loading, while retaining excellent sliding wear resistance (ball-on-disktests). Results are discussed in comparison to the behavior and properties of coatings of similar individualcompositions and thicknesses, as well as a thick monolithic WC12Co sprayed coating. Further improvementsin the processing are proposed to enhance the adhesion strength and avoid coating delamination under highload contact-fatigue conditions.</p><p> 2010 Elsevier B.V. All rights reserved.</p><p>re a special category of</p><p>The application of the FGM concept to TS has been extensivelystudied for thermal barrier coating (TBC) applications [1416], in anattempt to suppress the generation of cracks along the bond coatvarious powders in theer materials in differentfeedstock (as it is pre-</p><p>substrate interfa[2325]. Anti-coorder to minimihowever, duringsively large residmacrocracks or pre-examinationaddition, advancrezo).</p><p>l rights reserved.mposition and hardness through the coating thickness isepth-sensing microindentation, respectively. The in-situKeywords:Damage tolerant coatingsincrease in compliance with depth, and 2) an increase in fracture resistance by containment, arrest anddeection of cracks. A smooth gradation in the co12Co (top layer) depositedDamage tolerant functionally graded WC</p><p>Alfredo Valarezo a,, Giovanni Bolelli b, Wanhuk B. CLuca Lusvarghi b, Roberto Rosa b</p><p>a Center for Thermal Spray Research, Materials Science and Engineering Department, 130b Department of Materials and Environmental Engineering, University of Modena and Reg</p><p>a b s t r a c ta r t i c l e i n f o</p><p>Article history:Received 23 April 2010Accepted in revised form 31 August 2010Available online 8 September 2010</p><p>In this paper, effective damoxygen fuel (HVOF) sprayichange in composition from</p><p>j ourna l homepage: www.eo/Stainless Steel HVOF coatings</p><p>i a, Sanjay Sampath a, Valeria Cannillo b,</p><p>vy Engineering Bldg., Stony Brook University, Stony Brook, NY, 11794-2275, USAmilia, Via Vignolese 905, 41125 Modena (MO), Italy</p><p>tolerance of a functionally graded coating (FGC) deposited by high velocitys observed. The thick FGC (1.2 mm) consists of 6 layers with a stepwise0 vol.% ductile AISI316 stainless steel (bottom layer) to 100 vol.% hard WCo an AISI316 stainless steel substrate. Damage tolerance is observed via 1) an</p><p>gs Technology</p><p>ev ie r.com/ locate /sur fcoatce, where cracking and delamination become likelyrrosion coatings should be as thick as possible, inse or eliminate the interconnected porosity [26,27];the deposition of thick single-layer coatings, exces-ual stress is often built up [28,29] producing verticalremature delamination. These issues have led to theof thermal sprayed FGMs as potential solutions. Ines in process science, such as the in situmeasurement</p></li><li><p>of stress states, have enabled new insights into materials andprocesses providing new opportunities of layered and gradedmaterials design.</p><p>In this paper, the approach of damage tolerance, which has beenextensively used in the design of brittle components that withstandload without failure despite the presence of defects (e.g. cracks), isapplied to TS coatings for wear resistance. Given that TS coating layersare built by individual molten particles, the presence of defects (e.g.intersplat interfaces and pores) and, potentially, of cracksmakes themsingular systems that can prematurely fail at the weakest regions via</p><p>Cross-sectional samples of all of the coatings were prepared bycutting, hot-mounting in phenolic resin, grinding with SiC papers (upto 2500 mesh) and polishing with diamond slurry (up to 0.5 m). Thepolished sections were observed by scanning electron microscopy(SEM, Quanta-200 and XL30, FEI, Eindhoven, The Netherlands) andthe composition of the various layers was assessed by image analysis(NIH ImageJ version 1.37): at least 5 micrographs at 400 magni-cation were employed for each layer.</p><p>The residual stress build-up during the deposition of the FGC andof each single-layer coating was monitored by the in situ coating</p><p>2198 A. Valarezo et al. / Surface &amp; Coatings Technology 205 (2010) 21972208macro- or micro-cracking under applications of contact loads in wearsituations. Thus, damage tolerance applied to TS coatings, in thisresearch, considers the combination of process and microstructuraldesign concepts by grading structures to achieve enhanced perfor-mance. Here, i) crack inhibition is proposed to be achieved in hardmetal-based cermet coatings (e.g. usually embrittled WCCo, CrCNiCr due to decarburization) by exploiting the process conditions inorder to induce compressive residual stress by peening in the bottomlayers, and ii) crack arrest and deection is achieved by addition of atough metallic phase.</p><p>2. Experimental procedures</p><p>2.1. Coating deposition</p><p>The WC12 wt.%Co material (hereafter WCCo) used in theseexperiments, with particle size 62+42 m, was an experimentalbatch of ne carbide size produced by Osram Sylvania [30] withdesignation SX-432. The stainless steel 316 powder (hereafter SS316)was produced by ANVAL, particle size 46+24 m.</p><p>The WCCo/SS316 functionally graded coating (FGC) was depos-ited onto grit-blasted AISI316 stainless steel plates of228.625.41.6 mm size, using a Sulzer-Metco DJ 2700 torch withpropylene as fuel. The WCCo and SS316 powders were mixed invarious proportions to produce a FGC. Starting from a 100 vol.% steellayer, the amount ofWCCowas increased by 20 vol.% until a top layerconsisting of pure WCCo was deposited, thus obtaining a 6-layersystem of about 0.2 mm per layer. The parameters employed for thedeposition of each layer are listed in Table 1. For comparativepurposes, monolithic single-layer coatings corresponding to eachlayer of the FGC (and having identical thickness) were also depositedonto the stainless steel plates, using the same deposition parameterslisted in Table 1. They will be hereafter referred to as single-layercoatings. Additionally, a thick (0.6 mm) pure WCCo layer wasproduced, in order to compare the thick FGC to a homogeneouscoating of similar thickness. This layer could not be grown thickerthan 0.6 mm without debonding and severely distorting thesubstrate.</p><p>2.2. Microstructural and micromechanical characterization</p><p>X-ray diffractometry (X'Pert PRO, Panalytical, Almelo, TheNetherlands) was performed on the top layer of the FGC and on thesingle-layer coatings, using CuK radiation in the 30b2b90range.</p><p>Table 1Parameters employed in the HVOF-deposition of the various layers of the FGC.</p><p>Nominal powder blendcomposition (vol.%)</p><p>Layer thickness(mm)</p><p>Numberof passes</p><p>Fuel(l/min)</p><p>Oxygen(l/min)</p><p>100% SS316L 0.188 7 67 25180%SS316L-20%WCCo 0.184 8 78 25160%SS316L-40%WCCo 0.265 14 78 26640%SS316L-60%WCCo 0.164 9 78 26620%SS316L-80%WCCo 0.189 13 78 280100% WCCo 0.313 20 78 280property (ICP) sensor [3133]. ICP uses a laser displacement sensor tomeasure continuously the stress-induced curvature of a thin platesubjected to layer-by-layer deposition: quenching, peening, andthermal stress arise during spraying. A simultaneous measurementof temperature is carried out via multiple thermocouples in contact tothe back side of the sample.</p><p>The microhardness of each layer in the FGC was measured bydepth-sensing Vickers microindentation on the polished cross-section(15 indentations for each layer, 3 N load, 2.4 N/min loading andunloading rate, 10 s loading time), according to the OliverPharrprocedure [34]. The results were compared to the hardness of thecorresponding single-layer coatings, in order to verify whether theinclusion in the FGC structure could produce signicantmodications.</p><p>The coating elastic modulus for the SS316 and WCCo single-layers was obtained by instrumented microindentation (Micro-Materials Limited, Wrexham Technology Park, Wrexham, UK, WCCo Berkovich indentor at 5 [N] load). The modulus for the othercompositions was obtained by rule of mixtures.</p><p>2.3. Sliding wear testing</p><p>The dry sliding wear resistance of the FGCwas assessed by ball-on-disk testing, using a pin-on-disk tribometer (CSM Instruments,Peseux, Switzerland) with sintered WC6%Co spherical pins (diam-eter: 3 mm) as counterparts. Two different experimental congura-tions were employed: 10 N normal load, 0.20 m/s relative slidingspeed, 5000 m overall sliding distance, 5 mm wear track radius(hereafter referred to as test-1); 10 N normal load, 0.40 m/s relativesliding speed, 10000 m overall sliding distance, 8 mm wear trackradius (hereafter referred to as test-2). The friction coefcient wasmonitored during the test by a load cell attached to the pin-holdingarm, the wear rate of the sample was assessed by measuring thevolume of the wear scar using an optical confocal prolometer(Conscan Prolometer, CSM Instruments), and the wear scars wereobserved by SEM. The test was also performed on the 100 vol.% WCCo single-layer coating, to study whether the underlying layers in theFGC can modify its intrinsic sliding wear behaviour. Also, the 80 vol.%WCCo single-layer coating was tested to ascertain whether theaddition of a low amount of steel to the hard but brittle WCComaterial would improve or impair its wear performance.</p><p>It should be noted that, under these test conditions, the contactstress distribution mainly involves near-surface regions of thecoatings. An approximate computation of the Hertzian stressdistribution that occurs in the contact of a 3 mm-diameter sphereon a at surface having an elastic modulus of about 300 GPa and</p><p>Air(l/min)</p><p>Carrier gas(N2) (l/min)</p><p>Feed rate(g/min)</p><p>Average particletemperature (C)</p><p>Average particlevelocity(m/s)</p><p>280 12 35 1926 702306 12 39 1917 735350 12 39 1943 680350 12 40 1764 667350 12 38 1672 613350 12 41 1636 562</p></li><li><p>Poisson's ratio of 0.25 (typical properties of a HVOF-sprayed WC-Cocoating [35,36]) indicates that the maximum sub-surface shear stress(in both test conditions) is about 850 MPa and it is located about20 m below the surface, i.e. the dominant contact-induced stressdistribution occurs within the top most layer of the FGC.</p><p>2.4. High-load indentation and ball drop cyclic impact testing</p><p>To observe the effect of the deep compliant layers on themechanical behaviour of the top harder layers, high load indentationand cyclic impact tests were performed on the FGC as to probe itsload-carrying capability and damage tolerance. The results werecompared to those obtained on the 100 vol.% WCCo single-layer andon the 0.6 mm-thick WCCo coating as a reference.</p><p>The indentation tests were carried out using a Mitutoyo AVK-C2hardness tester with a load of 50 kg applied for 15 s, monotonicallyand cyclically (1, 10, and 100 times) in the same location. A WCCospherical indenter of 1/8 (3.125 mm) diameter was used in thesetests. The bonded-specimen technique was used to prepare thesamples. A detailed description of the sample preparation is describedelsewhere, thus a brief description is mentioned here [37]. In thistechnique, two 7 mm2.5 mm specimens were cut, and the cross-</p><p>2199A. Valarezo et al. / Surface &amp; Coatings Technology 205 (2010) 21972208sections were polished to 0.05 mwith Al2O3 solution. The two sliceswere then put together with a mechanical press and the top surfacewas polished. The assembly was indented from top at the interface.Afterwards, the faces were separated, and optical micrographs weretaken from the cross-section.</p><p>The ball drop cyclic impact test was carried out by dropping anX200Cr13 steel ball (diameter: 39 mm), attached to an overall weightof 12 N, at a frequency of 45 impacts/min on the sample's surface. Thedrop height was 95 mm, and the overall number of impacts was set at1000. The samples were inspected by optical and scanning electronmicroscopy after the test. To observe the cross-section of the coatingsafter the cyclic impact test, the sample was mounted in resin,dissectedwith ametallographic cuttingmachine and then ground andpolished with diamond slurries, as described above.</p><p>3. Results and discussion</p><p>3.1. Coating microstructures</p><p>Fig. 1 (ag) shows XRD patterns of the specimens. Signicantoxidation and phase transformations occurred during HVOF proces-</p><p>Fig. 1. XRD patterns of the single layers and of the FGC coating. Legend: a=100%stainless steel single layer; b=20% WCCo single layer; c=40% WCCo single layer;d=60%WCCo single layer; e=80%WCCo single layer; f=100%WCC...</p></li></ul>


View more >