It is well-established that (001)-textured α-Al₂O₃ coatings show substantially enhanced cutting performance - in some applications by several hundreds of per cents - over the other α-Al₂O₃ textures e.g. (012), (110) and (100). The introduction of the controlled deposition of the (001) texture on an industrial scale was thus one of the most important recent advances in the area of wear resistant CVD coatings [1, 2]. Several cemented carbide manufacturers can today deposit (001)-textured α-Al₂O₃ in a controlled way. However, there are no published works explaining the process conditions available.

This paper addresses this situation. The influence of experimental variables in combination with H₂S catalysis on the growth characteristics and especially on texture development of chemically vapor deposited (CVD) α-Al₂O₃ will be reported in detail. The experimental α-Al₂O₃ coatings were deposited from the AlCl₃-CO₂-H₂S-H₂ system at 1000 °C in a commercial, computer-controlled, hot-wall CVD reactor at a total pressures of 50-300 mbar. The α-Al₂O₃ coatings were deposited onto medium temperature (MTCVD) Ti(C,N) layers. The MTCVD Ti(C,N) layers were deposited onto cemented carbide substrates at a temperature of about 860 °C from the CH₃CN-TiCl₄-N₂-H₂ system. The coatings were characterized using X-ray diffraction, (XRD), scanning electron microscopy (SEM), electron back-scatter diffraction (EBSD) and optical microscopy. Based on a large amount of experimental data the process windows for deposition of α-Al₂O₃ coatings with (012), (110), (001) and (100) preferred growth directions were defined. The process data for depositing the above mentioned textures will be given and discussed in detail. In addition, the effects of H₂S, CO₂, CO and the total pressure on the growth characteristics of α-Al₂O₃ will be dealt with.

[1] S. Ruppi, International Journal of Refractory Metals & Hard Materials 23 (2005) 306–316.

[2] S. Ruppi, Surface & Coatings Technology 202 (2008) 4257-4269

In the area of cutting tools CVD-diamond coatings for machining graphite, printed circuit boards (PCB), green compacts or aluminum-silicon alloys are well established. New applications are machining of challenging composite materials such as carbon fiber reinforced plastics (CFRP) or finally sintered cemented carbides and ceramics. In other areas of technology the coating machine is capable to deposit films on large scale substrates. Examples are boron doped electrodes and self-supporting Diamond foils

Carbon fiber/epoxy composites are increasingly found in the aeronautical and space industry due to their lower density, high stiffness and low reactivity with regard to metallic alloys. However replacing metals by composites often results in the loss of required functions, e.g. electrical conductivity. The latter can be recovered by metallization. In the present contribution, a direct liquid injection metalorganic CVD (DLI-MOCVD) process is presented for the low temperature metallization of composites, ultimately aiming at the surface functionalization of 3D parts. The process involves the metalorganic precursor (hfac)Cu(MHY) (1). We determine chemical kinetics of the process and show the improvement of the adhesion of the Cu films by applying different pretreatments.

The Arrhenius plot established at 5 Torr between 165 °C and 255 °C on Si samples (poly-epoxy sustains T < 210 °C, only) reveals an activation energy at the kinetically limited regime equal to 53 kJ/mol. The diffusion limited regime prevails above 210 °C and is maintained at least up to 255 °C. Metallic, contamination-free Cu films are systematically obtained in the investigated conditions. Their adhesion to the epoxy surface is poor, requiring surface treatment to etch and/or activate the surface before deposition. To this purpose, gas phase and wet chemical pretreatments are used. Gas phase pretreatments consist either in the use of a remote plasma, or in the deposition of plasma-polymerized acrylic-acid layers (2). The liquid phase pretreatment is based on a commercial series of solutions that includes swelling, oxidation, and neutralization steps.

The adhesive strength of the Cu films on polyepoxy and on carbon fiber/epoxy composite surfaces is specifically investigated by peel and scratch testing, and is correlated with topological, chemical, and energetic characteristics of the surfaces prior deposition, investigated by profilometry, AFM, XPS and wettability through the sessile drop method. Preliminary results reveal that plasma-polymerized acrylic-acid layers result in surface functionalization with a significant increase of surface energy and improvement of the adhesion. In some cases the subsequent modification of the microstructure of the films is found to be beneficial to the electrical resistivity, resulting in values closely approaching that of bulk Cu.

1. Chen TY, et al. (2001) Chem. Mater. 13(11):3993-4004.

Austenitic stainless steels (300 series) are widely applied in the chemical, food processing, desalination and medical industries due to a combination of moderate cost, high general corrosion resistance, high formability and good mechanical properties. However, in aggressive corrosion environments austenitic stainless steels are susceptible to pitting corrosion. In response, high-cost pitting resistant alloys have been developed incorporating higher chromium, nitrogen and molybdenum contents. As an alternative to bulk alloying, austenitic steel may be surface modified. For example, chromising technique. Despite the potential corrosion benefits, the diffusion of chromium into the surface (chromising) of austenitic steels has not been extensively studied. In particular, the effect of alloy composition, cooling rate and isothermal holding on the resulting phase formation in the chromised layer has not been reported.

In this present study AISI 304L and AISI 316L austenitic stainless steels were chromised using fluidised bed reactor chemical vapour deposition (FBR-CVD) at 1050°C for 5 hours, and subsequently subjected to various cooling rates; furnace cooling (80°C/min), air cooling (47°C/min) and water quenching (347°C/min). Additionally, chromised samples were furnace cooled and held isothermally at 500, 700 and 900°C for 5 h. The resulting chromised layers were characterised by X-ray diffractometer, scanning electron microscopy, Glow-discharge optical emission spectroscopy and nano-indentation techniques.

This study has revealed that the phase composition of the chromised layer can be controlled by the cooling rate from the process temperature. The diffusion of chromium into the austenitic surface at 1050°C results in the stabilisation of ferrite. A rapid cooling rate favoured the retention of this BCC solid solution, forming a columnar grain structure across the 25 μm thick layer. A similar columnar structure consisting of BCC solid solution and sigma (FeCrNi) formed by moderate cooling rates. Further, by isothermally holding at s formation temperatures, layers composed entirely of sigma phase were formed exhibiting mm scale grain structure. The critical cooling rate required to retain a BCC solid solution was affected by the alloy composition with AISI 316L requiring a more rapid cooling rate. This was attributed to the sigma-stabilizing effect of molybdenum. The phase composition of the layer influenced the properties of the layer with the hardness of BCC solid solution and sigma phases being ~330Hv and ~1450Hv, respectively. Additionally, the pitting corrosion performance (Epit) of all chromised surfaces was superior to as-received austenitic steel.

There is a dual need to develop new industrial CVD processes especially to produce performant protective coatings either in continuous scrolling treatments or in large scale batch reactors: (i) find molecular precursors that lower deposition temperature and thus allow the growth on temperature sensitive substrates and (ii) feed the reactor with high vapor flow rates to cover large surface areas with a uniform and high deposition rate. The first item is satisfied for many years by the use of organometallic compounds as molecular precursors when they are sufficiently volatile. There is a wide range of organometallic compounds that can be selected according to their chemical and thermal properties. The limitation is often an insufficient vapor pressure and a low stability. To overcome these difficulties, more recently, pulsed direct liquid injection (DLI) of molecular precursors appeared as an emerging technology to deliver high vapor flow rates even for precursors with a low volatility. Liquid precursor can be directly injected for very high flow rates but it is generally diluted in a solvent for a better control of its vaporization and mole fraction at the inlet of the reactor.

DLICVD has been essentially developed for deposition of functional oxides thin films for optical and electronic devices. Oxygen-containing precursors are generally selected due to their good volatility, stability and solubility in organic solvents. Solution of precursor is injected and oxygen is frequently added to prevent the film contamination by carbon originating from the solvent which can be then consumed by combustion mechanism.

DLICVD processes for depositing non-oxide coatings have already being reported. In this paper, new DLICVD processes are described for depositing other non-oxide coatings. The role of solvent is discussed. Modeling of DLICVD process is also reported to optimize the thickness uniformity.

Carbon or glass fibers reinforced polymer composites show excellent stiffness, low weight, and low chemical reactivity, making them ideal for the replacement of metallic structural components. Sometimes composites must be coated to recover the original properties of the metal or to provide additional properties such as optical absorbance, magnetic screening, or aesthetical aspect. However, the formation of a coating on materials that exhibit very low surface energy is challenging when one targets durability (adhesion, resistance to fatigue), and a tailored microstructure.

In the field of aeronautics and space, many parts are subject to such a substitution with an objective of weight gain. Whereas many processes have already been developed for the coating of composites in different fields such as microelectronics, we present new challenges posed by aerospace applications. How to adapt surface pretreatments? Why CVD is a processing method of choice? What properties can we expect?

Unfortunately, an engineering approach based on a rational screening of process parameters is not sufficient to gather a general knowledge about the treatment of composites surfaces. Therefore, in an attempt to develop an integrated model for the whole process, we present an original methodology that aims to link surface science on a model sample with the real world.

A common method to functionalize highly porous materials is by grafting of aminosilanes (AS); either via solution phase grafting (SPG) or via vapor phase grafting (VPG). According to literature SPG may result in oligomerization of AS, causing inhomogeneous layers and pore blocking. In contrast, grafting via the vapor phase is claimed to result in homogeneous layers[3]. Several authors have compared functionalization of mesoporous silica by AS using SPG, VPG and atomic layer deposition (ALD)[4,5]. The deposition temperatures range from 110°C to 300°C, and the reaction times vary from 3 to 48 h. Also, the authors apply different pretreatments before AS deposition.

In this work we have deposited monolayers (ML) of APTES and APTMS utilizing a hot-wall flow type commercial F-120 ALD reactor[6] at a total pressure of 10 hPa, a reaction temperature of 80-180°C and pulse and purge times between 2-8 hours, where the long pulse and purge times were set to ensure precursor in- and out- diffusion into the porous matrix. The np-AMC pore width is around 4 nm and one APTES/APTMS molecule has a length of around 5 Å [7], substantially decreasing the np-AMC surface area. Hence it was important to limit AS coverage to 1 ML.

The functionalized and as-made np-AMC was characterized by IR-absorption, SEM-EDS and X-ray Photo-electron Spectroscopy (XPS). Water cycling tests were performed to establish the influence of functionalization on water uptake and release.

[1] A. Forsgren J, Frykstrand S, Grandfield K, Mihranyan A, Strømme M PLoS ONE 8(7) (2013).

[2] B I. Pochard et al., J. Phys. Chem. C. 119, 15680−15688 (2015).

[3] V.G.P. Sripathi, Barbara L. Mojet, Arian Nijmeijer, Nieck E. Benes, Microporous and Mesoporous Materials 172, 1–6 (2013).

[4] H. Ritter, M. Nieminen, M. Karppinen, D. Brühwiler, Microporous and Mesoporous Materials 121, 79–83 (2009).

[5] S. Ek, I. Eero I. Liskola, L. Niinisto, J. Keranen, A. Auroux, J. Vaittinen, T. T. Pakkanen; Langmuir, 19, 10601-10609 (2003).

[6] T. Suntola, Thin Solid Films 216, 82 (1992).

[7] M. Zhu, M. Z. Lerum, W. Chen; Langmuir 28(1): 416–423 (2012).

Appropriately tuned chemical vapor deposition (CVD) processes yield films with targeted microstructure and deposition rates. Moreover, they allow delineation of convenient operating windows for co-deposition or sequential deposition of metals, necessary for the development of complex intermetallics containing films and coatings. However, the predominance of chemical reactions with various kinetics pathways, increase the complexity of the process. Thus, the control of the latter is subjected to the comprehension and to the control of the involved chemistry. In this framework, the deposition of Al₁₃Fe₄ intermetallic coatings is examined, with the aim to benefit from its excellent catalytic properties; e.g. in the semi-hydrogenation of acetylene for the replacement of noble metals such as Pt and Pd [1] . Our investigation proceeds in two steps; first, aluminum [2] and iron depositions are performed separately using dimethylethylamine alane (DMEAA) and iron pentacarbonyl (Fe(CO)₅) as Al and Fe precursors, respectively. Possible reaction pathways and kinetic mechanisms are explored by macroscopic simulations [2], whereas the morphology of the films is simulated by microscopic simulations at the surface level. The modeling procedure enhances the control of the process as well as the control of the microstructure of the obtained films. Then, the two metals are co-deposited on silicon surfaces. The co-deposition process is performed (200^oC, 10 Torr) with a low hydrogen inflow rate. XRD and EPMA reveal that the as processed films do not contain any intermetallic phase. In contrast, they contain a high amount of oxygen, attributed to the decomposition of Fe(CO)₅. In order to circumvent oxygen contamination, the sequential deposition of the two metals is investigated. Aluminum and iron are deposited on glass substrates followed by in situ annealing at 575^οC, where the formation of the Al₁₃Fe₄ intermetallic phase has been reported [3]. Reactive diffusion of Fe through the Al layer results in the formation of the Al₁₃Fe₄ phase based on XRD and SEM analysis. Additional characterizations including TEM and GDOES analyses are performed to further investigate the peculiar structure of the developed crystals as well as the elemental composition profiles. Catalytic properties are currently under investigation.

References

[1] M Armbrüster, et al., Nat. Mater. 2012, 11, 690-693.

[2] I.G. Aviziotis, et al., Phys. Status Solidi C 2015, 12, 923-930.