Laser initiated Ti 3 SiC 2 powder and coating synthesis

Ceramics International xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Ceramics International journal homepage: www.elsevier.com/locate/ceramint Laser initiated Ti3SiC2 powder and coating synthesis Paweł Rutkowski , Jan Huebner, Dariusz Kata, Jerzy Lis, Adrian Graboś, Leszek Chlubny ⁎ AGH University of Science and Technology, Faculty of Materials Science and Ceramics, Department of Ceramics and Refractories, al. A. Mickiewicza 30, 30-059 Krakow, Poland A R T I C LE I N FO A B S T R A C T Keywords: A. SHS B. MAX phase C. Titanium silicon carbide D. Laser processing E. Coating In the work the SHS synthesis of MAX phases from Ti-Si-C system were carried and initiated with use of 30 W laser beam with 40 µm spot. That kind of initiation allows locally and rapidly start the SHS synthesis and avoid the contamination coming from heating wire present during conventional method. The reaction was monitored by high-accuracy radiation pyrometer and high quality optical camera. The recorded data, together with reaction bed thermal conductivity measurements allowed to correlate to obtained powders composition and reaction speed. The reaction bed morphology was investigated by scanning electron microscopy with element distribution (EDS). The second part of the paper concerns laser reactive deposition of SHS in-situ synthetized MAX phases layer on silicon carbide substrate. The paths of deposited layer were formed under argon overpressure of 2 bar using 120 W of laser power. 1. Introduction MAX Phases are new family of ceramic materials that recently attracted many scientists to study their unique properties. Due to properties combination of metals and ceramics, due to their specific, layered structure and strict chemistry, MAX phases became widely examined and large amount of new compounds were discovered [1]. Their high usefulness derives from high electrical [2] and thermal [3] conductivity in addition to good mechanical strength [4], corrosion resistance [5] and machinability [6]. Stoichiometry of such materials is stated as follows: M(n+1)AX. “M” is early transition metal, “A” is mainly 13- or 14-group element (Agroup element in older nomenclature) and “X” is either carbon or nitrogen. MAX Phases, as Nowotny phases, crystallize in layered arrangement of M6X tetragons and A-group element single atom layers. Such structure forms hexagonal unit cells with six-three screw axis and A-group mirror planes resulting in P63/mmc symmetry class. Unit cells differ between materials with different values of “n”, altering the number of M atom layers in their structure. Number of that layers equals “n”, thus 211, 312 and 413 MAX phases groups where distinguished, with their names referring to the numbers in stoichiometry [1]. While number of discovered and studied MAX Phases is constantly growing, with over fifty known materials up to date, there is simultaneously a problem of their synthesis in bulk. In the history of this carbides and nitrides, large number of new materials was only synthesized by very accurate, yet small-scale methods such as CVD [7]. ⁎ Although recent studies show many attempts to involve known methods, such as thermal explosion (TE) [8], combustion synthesis [9] or additive manufacturing (AM) [10], in MAX phases production, search for optimal and cost-efficient process continues. Ti3SiC2 is the most studied and described material from the group of MAX phases, up to date [1]. This compound is one of the first known MAX phases [11] and got common interest as relatively soft, easily machinable while having high fracture toughness for a carbide. Therefore wide range of studies were performed, including: oxidation resistance [12], thermal properties [13] and synthesis of bulk samples [14]. With that work and later studies of similar materials by Barsoum et al. MAX phases was described as a group of materials with shared properties [15]. Nowadays, Ti3SiC2 serves as reference for many new methods and approaches for the subject, as in this article. First attempts to synthesize its bulk form were difficult and resulted in large amounts of TiC and SiC phases within the microstructure [16]. This occurred mainly due to the significant differences in phase transition temperatures between Si and other elements. Those differences are now crucial in designing new, industry-oriented methods. Simultaneously, other MAX phases should be carefully examined and correlated with best methods to obtain. In order to achieve highly efficient and accurate method in terms of high purity, laser induced, self-propagating high-temperature sintering (SHS) was proposed. Due to the relatively fast reaction (about 3 s) and point induction of the process, creation of Ti3SiC2 phase was expected, before it could decompose into Si-rich liquid and solid phase as shown below: Corresponding author. E-mail address: pawel.rutkowski@agh.edu.pl (P. Rutkowski). https://doi.org/10.1016/j.ceramint.2018.03.143 Received 16 February 2018; Received in revised form 13 March 2018; Accepted 15 March 2018 0272-8842/ © 2018 Elsevier Ltd and Techna Group S.r.l. All rights reserved. Please cite this article as: Rutkowski, P., Ceramics International (2018), https://doi.org/10.1016/j.ceramint.2018.03.143 Ceramics International xxx (xxxx) xxx–xxx P. Rutkowski et al. Fig. 1. Initial powder morphology: a) titanium, b) silicon, c) graphite. powders were mixed with titanium carbide one. The synthesis was made in copper-cooled copper-mold laser melting furnace with use of 5 kW CO2 laser. The melting time was in the range 15–60 s. The authors obtained 88% purity product bed in short – 15 s process times. For longer process duration the MAX phase content decreased to 80%. In case of laser melting rapid synthesis of TiC/Ti3SiC2 the same authors synthesized Ti3SiC2 phase in quantity of 56–84 vol%, what depended on laser power ranging 3,5–5 kW and 15–60 s melting time. The reaction mixture was composed of titanium, silicon and titanium carbide powder. They confirmed that it is possible to produce bulk material with 4% porosity using laser melting process. The laser is also used in case of MAX pulse laser deposition technique [19]. It is possible to obtain in this way the Ti/Si/C film. The film was deposited from the pre-synthesized Ti3SiC2 target. Authors used Nd:YAG laser with 1064 nm wavelength at 6 ns pulse duration and frequency of 20 Hz. The laser energy did not exceeded 8 J/cm2. The oxidized silicon crystal was used as a film substrate and was heated up to 950 K. It was notice that too low temperatures result in lack of enough quantity of the silicon. The thickness of the obtain film was 750 nm. In the presents work Authors show the local initiation of Ti3SiC2 MAX phase SHS synthesis by the laser beam. Laser ignition allowed to start locally (40 µm laser spot diameter) contamination free synthesis in vacuum atmosphere. Specially constructed reaction chamber experiment station allowed to monitor precisely the temperature revealed during the reaction and visually record the synthesis propagation. The influence of the before-reaction bed compaction on the synthesis behaviour was explained by bed thermal conductivity examinations. In the second part of the paper the positive trial of MAX phase coating reactive deposition was made. Fig. 2. Laser apparatus set with control system for Ti3SiC2 synthesis. 2. Experimental procedure M(n + 1) AXn → M(n + 1) Xn + A To synthesize Ti3SiC2 the following commercially available powders were used: 44 µm grain size 99.7% purity TI-104 titanium metal powder of Atlantic Equipment Engineers, 1–5 µm 99 + % purity SI-100 silicon metal powder of Atlantic Equipment Engineers and < 50 µm 99.8% purity Merck no. 1.04206.9050 graphite. The morphology of these reactants were observed by NOVA NANO SEM 200 scanning electron microscopy of FEI EUROPE COMPANY and is illustrated in Fig. 1. Powders were set with an excess of silicon for following stoichiometry Ti3Si1,2C2. The reactants mixture was homogenized in isopropanol by ball mill for 2 h using cemented carbides (WC-Co) milling media. After drying the powders were put into the isolated graphite holder in form of: loose powder and 15, 30 MPa uniaxially pressed green samples. The holder with the powder mixture was placed into the Such reaction would result in high amounts of TiC and SiC inclusions in the microstructure, significantly influencing final properties of the material. The laser initiated self-sustaining high temperature syntheses becomes more important in case of reactive laser welding applications. The initiation of SHS by laser beam is local and provides higher system purity in comparison to the conventional SHS ignition by wire heating element. Thus evaporation of heating wire and unwanted reaction between reactants and igniter is avoided. In the literature there are negligible information concerning laser synthesis of MAX phase. D. Chen et al. obtained Ti3AlC2 and Ti3SiC2 based composites by use of laser melting technique [17,18]. In case of the Ti3AlC2 powder synthesis the titanium and aluminum metal 2 Ceramics International xxx (xxxx) xxx–xxx P. Rutkowski et al. infrared LT/2 M Optris® CSLaser pyrometer with response time of 2 ms and within temperature range from 800 °C to 2500 °C. Additionally the reaction course was recorded with high quality TV camera model UI1240LE-C-HQ. MAX layer formation was proceeded differently. The prepared powder mixture was deposited in the in the form of isopropyl alcohol dispersion on the silicon carbide substrate (Fig. 3). It was characterized by 98.5% dense polycrystalline material which was obtained at 2150 °C during pressureless sintering in form of rectangular tiles. The SiC tiles were cut into 20 × 20 mm square pieces with thickness of about 5 mm. Next, these substrates were placed into the reaction chamber as shown in Fig. 2. The argon was supplied into the chamber in order to obtain 2 bars of protective atmosphere overpressure. Titanium silicon carbide layer was formed during continuous work (CW) mode during laser scanning process of the mixture layer deposited on the SiC substrate. The idea of the layer formation and laser scanning process is presented in Fig. 4. Reactive formation of MAX phase layer was made at 120 W and 40 µm spot laser beam, 0.0025 m/s process speed in 2 bar of argon overpressure in order to prevent silicon evaporation. The “hurrySCAN 30″ head of SCANLAB with 1080 nm laser beam wavelength was used. The laser beam source was the same as in the case of above described SHS synthesis. In order to increase layer adhesion to the substrate, silicon carbide was etched by 80 W power laser beam. The etching process speed was 0.005 m/s. The bed powder morphology and layer microstructural observations were made with use of optical microscope of LEICA DM2500 M and NOVA NANO SEM 200 scanning electron microscope of FEI EUROPE COMPANY. The element distribution analysis were made by Energy Dispersive X-ray Spectroscopy (EDS) of EDAX. In order to analyse the correlation between compaction of the reaction bed and the SHS synthesis temperature the thermal conductivity of the reaction bed was investigated. Thermal conductivity was measured by laser flash analyser LFA 427 of Netzsch company in graphite crucible with use of triple-layer model. The measurement temperature was 25 °C to avoid the ignition of the powder mixture at higher temperatures. The laser pulse of the LFA measurements was 0.8 ms and voltage of 550 V. The thermal properties were find both for loose powder mixture and 15 and 30 MPa uniaxially pressed green bodies. Fig. 3. SHS powder mixture placement in the reaction chamber. Fig. 4. Ti3SiC2 layer laser formation in argon overpressure in the reaction chamber. 3. Monitoring of synthesis temperature reaction chamber shown in Fig. 2 equipped with special sapphire window. This allows to deliver the electromagnetic wave in form of laser beam to selected are on the powder surface. The JK200FL fiber laser beam source was used. The wavelength of laser beam was 1080 ± 10 nm and spot diameter on of 40 µm. The ignition line length was 5 mm and scanning speed was set to 0.0025 m/s. with laser power of30 W. Prior to SHS reaction the chamber was purified by air evacuation and then vacuumed in order to obtain 0.1 bar. The reaction of loose powder bed was initiated at 2.0 bars argon overpressure. In case of pressed green samples the SHS was initiated under vacuum. The process temperature was measured by high frequency recording The temperature recorded during the SHS synthesis of the loosefilled reaction mixture and uniaxial pressed green bodies is presented in Fig. 5. As shown in Fig. 5 the single modal temperature curve has similar shape for each sample. At the beginning, temperature have risen abruptly due to heat produced by reacting substrates. The maximum temperature was achieved in 2 s. The cooling down of the reaction bed to 800 °C took 8–10 s what allowed to freeze final product. The maximum reaction temperature reached 1700 °C in case of 30 MPa uniaxially pressed sample because of lower material porosity which Fig. 5. Reaction temperature versus: a) mixture compaction, b) atmosphere. 3 Ceramics International xxx (xxxx) xxx–xxx P. Rutkowski et al. Fig. 6. Recorded laser initiated SHS reaction step on loose-fill reaction bed, vacuum condition. Fig. 7. Recorded laser initiated SHS reaction step on 30 MPa uniaxially pressed reaction bed, vacuum condition. Fig. 8. Recorded laser initiated SHS reaction step on loose-fill reaction bed, 2 bars argon overpressure. 4 Ceramics International xxx (xxxx) xxx–xxx P. Rutkowski et al. higher temperature was recorded. The maximum temperature for 15 MPa pressed sample and loose powder mixture was measured at 1600 °C and 1500°C respectively. The slope of temperature curves was calculated and normalized to the value of loose reaction mixture. The normalized temperature rate is as following: 1 for loose powder, 1.4 for 15 MPa pressed sample and 1.7 for 30 MPa pressed one. The change of the atmosphere to argon resulted in an increase reaction temperature of about 75 °C in each case. The SHS process of loose reaction bed and 30 MPa green mixture were recorded by TV camera presented in Fig. 6 and Fig. 7 respectively. The largest differences in laser ignition was noticed at the beginning of SHS synthesis. For loose mixture laser heating caused the two step SHS reaction. First ignition took place at material surface Fig. 6. Afterwards, the reaction proceeded rapidly while exceed exothermic heat was transported to bulk material. Accumulated heat started heat wave propagation steady mode in whole bed volume. In case of 30 MPa pressed mixture sample, one-step reaction was observed (Fig. 7). The higher compaction of the mixture caused rapid ignition on bed surface followed by fast heat transfer. It is because of higher quantity of contacts between reacting grains. The higher reaction temperature is also confirmed by significant visual effect in case of pressed material. The change of the atmosphere in the reaction chamber leads to one-step SHS ignition of loose packed mixture like in the case of pressed material – Fig. 8. The confirmation of the above discussion is clearly visible in thermal conductivity results measured by LFA method – Table 1. The reaction temperature increases with material compaction, with agreement to thermal conductivity measurements. The lowest temperature of SHS process is 1500 °C for thermal conductivity coefficient equal to 2.5 W/mK. Sample obtained under 30 MPa of pressure decreased porosity which resulted in an increase of thermal conductivity to 3.92 W/ mK. After the SHS synthesis the bed was macroscale photographed showed in Fig. 10 and compared with those conventionally obtained presented in Fig. 9. It can be noticed that laser initiated reaction allowed to produce visually denser product with much less geometrical distortions. The conventionally initiated SHS bed showed in Fig. 9 is less structurally uniform and possess typical reaction wave-effect on the surface. Table 1 Thermal conductivity of SHS reaction mixture. Phase Reagents mixture compaction SHS temperature [°C] Thermal conductivity [W/mK] Loose-fill 15 MPa 30 MPa ≈ 1500 2.51 ≈ 1600 2.83 ≈ 1700 3.92 Fig. 9. Cross-section of the standard SHS reaction bed. Fig. 10. Uniform reaction bed obtained by laser initiated SHS synthesis – fragment of the bed. 4. Phase composition and microstructure The obtained reaction beds after laser initiated SHS synthesis were investigated by XRD Rietveld analysis. The qualitative and quantitative phase composition data are collected in Table 2. The results show that the quantity of titanium silicon carbide phase reaches maximum of 42 wt% in case of loose sample, where the reaction temperature was lowest (≈ 1500 °C). The content of Ti3SiC2 decreased with higher number of intergranular contacts which resulted in the elevated SHS temperature. The amount of other phases present in samples increased with the reaction temperature because of possible decomposition of Ti3SiC2during reaction. The phase composition of products obtained in argon atmosphere is similar to those in form of pressed samples. The synthetized powders were carefully examined by scanning electron microscopy including EDS element distribution analysis. The morphology of SHS products are presented in Fig. 11 and in Fig. 12. Generally there is no significant differences between those morphologies. However, it is observed that porosity decreased with higher initial compaction level. All particles are strongly agglomerated and partially melted with average agglomerate size of 10–20 µm. It is clearly visible that liquid phase in form of Ti-Si eutectic was formed during the process. This confirms that moving layer model mechanism of the SHS Table 2 Phase composition of the reaction bed versus material compaction. Phase SHS temperature [°C] Ti3SiC2 TiC Ti5Si3 TiSi2 SiO2 C Material compaction/phase content [wt%] 2 bars Argon Vacuum Loose-fill Loose-fill 15 MPa 30 MPa ≈ 1575 32 24 39 – – 5 ≈ 1500 42 20 18 20 – – ≈ 1600 33 29 36 – 2 – ≈ 1700 27 29 44 – – – resulted in higher thermal conductivity. The pressing of the reaction mixtures allowed to increase the quantity of contacts between particles of reagents. Better heat transport together with increased interphase contacts caused intensification in the reaction rate in whole bed, so 5 Ceramics International xxx (xxxx) xxx–xxx P. Rutkowski et al. Fig. 11. The SEM grains morphology of reaction beds: a) loose-fill in 2 bars argon a) loose-fill in vacuum, b) 15 MPa pressed green body in vacuum, c) 30 MPa pressed green body in vacuum. Fig. 12. SEM morphology observations of titanium silicon carbide (Ti3SiC2) phase for loose fill reaction beds: a) 2 bars argon overpressure, b) vacuum. 6 Ceramics International xxx (xxxx) xxx–xxx P. Rutkowski et al. Fig. 13. The microstructure of silicon carbide etched by laser ablation process. 20 µm. On the image there are visible etched are as with size close to the laser beam spot diameter (40 µm). Observed cracks were probably generated by thermal stresses and local material shrinkage. This way prepared surface was covered by the Ti-Si-C mixture layer. In accordance with the Fig. 4 the covering mixture was laser scanned in the atmosphere of 2 bar argon overpressure to prevent the evaporation of titanium and silicon. As the result of laser initiated SHS synthesis the paths of the MAX phase coating was obtained – Fig. 14. The distances between layers are formed due to programmed movement paths. In order to analyse the obtained coatings the material was investigated by scanning electron microscopy with elemental distribution analysis in areas marked in Fig. 14. As it is shown the path width of the formed MAX phase layers is around 330 µm. Fig. 15 presents EDS results that confirms deposited layer is reach in Ti and Si. In case of spaces between coating paths, silicon and carbon well detected which indicated etched SiC. The XRD analysis confirmed existence of titanium silicon carbide in the reactively formed layer but it was difficult to establish quantitatively the phase content. 6. Summary Fig. 14. The scanning electron microscope image of the laser paths deposited of MAX phase layer. • The laser ignition system allows to begin the reaction locally, and combustion took place [20]. However, compaction level and porosity of products are increased due to increasing pressing values of green samples. To additional confirmation of the existence of MAX phase, the laminate structure of Ti3SiC2 was illustrated in Fig. 12. Those layers are typical for hexagonal nanostructured of titanium silicon carbide. • • 5. Coating formation possibility In order to prepare Ti3SiC2 reactive laser deposition the silicon carbide surface was etched by laser beam. This allowed to increase material roughness for better connection between SiC and formed MAX phase. The details of SiC microstructure was presented in Fig. 13. By this method the oxide layer usually existing on the SiC surface was removed and additionally activity of grain boundary is improved to attract deposited layer. The etching process revealed elongated grains of SiC reaching • • 7 control precisely the energy supplied to the reaction bed and also avoid contamination coming from the ignition source. The configuration of reaction chamber with TV camera and pyrometer allows to control behaviour of the bed during SHS synthesis and measure the influence of material packing on reaction speed and reaction temperature. The first trials allowed to obtain maximum of 42% MAX phase but also to find the correlation between bed compaction, reactants bed thermal conductivity and reaction temperature. It was shown that too high material compaction lead to higher SHS synthesis temperature and lower content of Ti3SiC2 phase. 30 W laser power at 40 µm spot beam, 5 mm ignition line length with speed of 0.0025 m/s was enough to supply enough energy to begin SHS. It is possible to obtain stable MAX layer of Ti-Si-C system on silicon carbide substrate using 120 W power 40 µm spot laser beam and 2 bars argon overpressure Ceramics International xxx (xxxx) xxx–xxx P. Rutkowski et al. Fig. 15. EDS element distribution made on the created layer (point 1) and substrate (point 2) marked on the Fig. 13 SEM image. Acknowledgement [8] L. Yi, L. Yingxin, L. Fan, C. Han, Z. Lifeng, G. Shouwu, Synthesis and microstructure of Ti2AlN ceramic by thermal explosion, Ceram. Int. 43 (2017) 13618–13621. [9] T. Thomas, C.R. Bowen, Thermodynamic predictions for the manufacture of Ti2AlC MAX-phase ceramic by combustion synthesis, J. Alloy. Comp. 602 (2014) 72–77. [10] M. Krinitcyn, Z. Fu, J. Harris, K. Kostikov, G.A. Pribytkov, P. Greil, N. Travitzky, Laminated Object Manufacturing of in-situ synthesized MAX-phase composites, Ceram. Int. 43 (2017) 9241–9245. [11] W. Jeitschko, H. Nowotny, Die Kristallstructur von Ti3SiC2, EinNeuerKomplexcarbid, Typ. Mon. Chem. 98 (1967) 329–337. [12] T. Okano, T. Yano, T. Iseki, Synthesis and mechanical properties of Ti3SiC2, Trans. Met. Soc. Jpn. 14A (1993). [13] H. Nowotny, S. Windisch, High temperature compounds, Annu. Rev. Mater. Sci. 3 (1973) 171. [14] R. Pampuch, Advanced HT ceramic materials via solid combustion, J. Eur. Ceram. Soc. 19 (1999) 2395–2404. [15] M.W. Barsoum, T. El-Raghy, The MAX phases: unique new carbide and nitride materials, Am. Sci. 89 (2001) 336–345. [16] R. Pampuch, J. Lis, L. Stobierski, M. Tymkiewicz, Solid combustion synthesis of Ti3SiC2, J. Eur. Ceram. Soc. 5 (1989) 283. [17] D. Chen, X. Tian, H. Wang, Z. Huang, Rapid synthesis of TiC/Ti3SiC2 composites by laser melting, Int. J. Refract. Met. Hard Mater. 47 (2014) 102–107. [18] D. Chen, X. Tian, H. Wang, Z. Huang, Rapid synthesis of bulk Ti3AlC2 by laser melting, Mater. Lett. 129 (2014) 98–100. [19] C. Lange, M.W. Barsoum, P. Schaaf, Towards the synthesis of MAX-phase functional coatings by pulsed laser deposition, Appl. Surf. Sci. 254 (2007) 1232–1235. [20] R. Pampuch, J. Lis, Advanced methods for SHS of powders developed in Krakow, Int. J. Self-Propag. High.Temp. Synth. 17 (1) (2008) 85–91. The research work was financed from the National Science Centre as a project OPUS 6 No. UMO-2013/11/B/ST5/02275 titled „ Structure and basic physical properties of functional layered MAX phases”. References [1] M.W. Barsoum, MAX phases: properties of machinable ternary carbides and nitrides, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2013, pp. 1–59. [2] M.W. Barsoum, H.I. Yoo, T. El-Raghy, Electrical conductivity, thermopower and hall effect of Ti3AlC2, Ti4AlN3 and Ti3SiC2, Phys. Rev. B 62 (2000). [3] J.D. Hettinger, S.E. Lofland, P. Finkel, J. Palma, K. Harrell, S. Gupta, A. Ganguly, T. El-Raghy, M.W. Barsoum, Electrical transport, thermal transport and elastic properties of M2AlC (M = Ti, Cr, Nb and V) phases, Phys. Rev. B 72 (2005). [4] T. El-Raghy, A. Zavaliangos, M.W. Barsoum, S.R. Kalidindi, Damage mechanisms around hardness indentations in Ti3SiC2, J. Am. Ceram. Soc. 80 (1997) 513–516. [5] J.W. Byeon, J. Liu, M. Hopkins, W. Fischer, N. Garimella, K.B. Park, M.P. Brady, M. Radovic, T. El-Raghy, Y.H. Sohn, Microstructure and residual stress of alumina scale formed on Ti2AlC at high temperature in air, Oxid. Met. 68 (2007) 97–111. [6] S.S. Hwang, S.C. Lee, J.H. Han, D. Lee, S.W. Park, Machinability of Ti3SiC2 with layered structure synthesized by hot pressing mixture of TiCx and Si powder, J. Eur. Ceram. Soc. 32 (2012) 3493–3500. [7] J.J. Nickl, K.K. Schweitzer, P. Luxenberg, Gasphasenabscheidung im system Ti-C-Si, J. Less Common Met. 26 (3) (1972) 335–353. 8