Calcium–magnesium–aluminosilicate Glass Research Articles

CaO–MgO–Al2O3–SiO2 corrosion resistance of multi-rare-earth-oxide-doped zirconia thermal barrier coating

As a thermal barrier coating (TBC) material, 8 wt% yttria-stabilized zirconia (8YSZ) exhibits poor resistance against calcium–magnesium–aluminosilicate (CMAS) corrosion. This is caused by the penetration of molten CMAS glass and the depletion of Y3+, which limits the lifetime of 8YSZ. In this study, a multi-rare-earth-oxide-doped zirconia (15 wt% (Lu0.2Yb0.2Dy0.2Gd0.2Y0.2)–ZrO2, denoted as 15LYDGY–SZ) ceramic was prepared to examine its interaction with CMAS. The sluggish diffusion effect and the competition between various dopant elements toward reaction with CMAS effectively slow down the atomic diffusion and limit the transformation of tetragonal phase into monoclinic phase in 15LYDGY–SZ ceramic. Compared with the currently used 8YSZ ceramic, the corrosion resistance of 15LYDGY–SZ to CMAS and its retention rate of fracture toughness after CMAS corrosion are improved. The results show that the multi-rare-earth-oxide-doped zirconia ceramic prepared in this study is a promising candidate material for the development of future TBCs.

Interaction of ytterbium monosilicate environmental barrier coating material with molten calcium-magnesium-aluminosilicate (CMAS)

The corrosion behaviors and mechanism of CMAS on Yb2SiO5 (YbMS) ceramics were investigated. Results indicated that YbMS was corroded by CMAS to generate various apatite and garnet, whose stability depends on the amounts of doping cations. Noticeably, the Ca2+ acts as an obvious sign of the reaction front diffusing into YbMS along the grain boundaries and volume. The higher diffusion rate of grain boundary than volume results in discontinuous reaction layers. The YbO1.5 formed by the dissolution of YbMS diffuses along the CMAS glass grid until the solubility is saturated to form a cyclic apatite phase, above 1300 ℃.

Thermochemical degradation of HfSiO4 by molten CMAS

The thermochemical degradation of hafnium silicate (HfSiO4) was investigated with a molten calcium-magnesium-aluminosilicate (CMAS) glass relevant to gas turbine engine applications. Sintered HfSiO4 coupons were fabricated, within which wells were drilled and filled with CMAS glass powder at a loading of ∼35 mg/cm2. Samples were heat treated at 1200°C, 1300°C, 1400°C, and 1500°C for 1 h, 10 h, and 50 h. At 1200°C and 1300°C, slow formation of a Ca2HfSi4O12 cyclosilicate phase was observed at the HfSiO4-CMAS interface. At 1300°C and higher, rapid infiltration of CMAS into the material along the grain boundaries was observed. Initial conjecture into CMAS degradation mechanisms of HfSiO4 are presented herein.

Calcium–magnesium aluminosilicate (CMAS) interactions with ytterbium silicate environmental barrier coating material at elevated temperatures

Thermochemical interactions between a calcium-magnesium aluminosilicate (CMAS) glass and ytterbium disilicate (Yb2Si2O7), a candidate environmental barrier coating (EBC) material, have been investigated. Pellets of Yb2Si2O7 and CMAS glass powder were heat treated at 1200, 1300, 1400 and 1500 °C for 1, 10 and 50 h in air. Powder X-ray diffraction was employed to identify the resulting phases. In a second series of experiments, Yb2Si2O7 substrates were prepared by hot pressing, and cylindrical wells were drilled into the material surface. CMAS glass powder was added to the wells to achieve a loading of ~35 mg/cm2 and subsequently heat treated at 1200, 1300, 1400 and 1500 °C for durations of 1, 10 and 50 h in air. Sample cross-sections were characterized using scanning electron microscopy, X-ray energy-dispersive spectroscopy, X-ray diffraction, electron microprobe analysis and transmission electron microscopy to evaluate the resulting microstructure, phases and compositions at the CMAS/Yb2Si2O7 interface. Dissolution of ytterbium silicate into molten CMAS followed by precipitation of Yb2Si2O7 coupled with CMAS grain boundary penetration at elevated temperatures were the dominant mechanisms by which CMAS effectively infiltrated and altered the Yb2Si2O7 substrate microstructure.

High-Temperature thermochemical interactions of molten silicates with Yb2Si2O7 and Y2Si2O7 environmental barrier coating materials

The thermochemical behavior of EBC candidate materials yttrium disilicate (Y2Si2O7) and ytterbium disilicate (Yb2Si2O7) was evaluated with three calcium-magnesium-aluminosilicate (CMAS) glasses possessing CaO:SiO2 ratios relevant to gas turbine systems. Pellet mixtures of 50 mol% EBC powder to 50 mol% CMAS glass powder were heat treated at 1200°C, 1300°C, and 1400°C. The products of these interactions were evaluated using X-ray diffraction, scanning electron microscopy, and energy dispersive spectroscopy. Above glass melting temperatures, exposure of the disilicates primarily resulted in dissolution into the molten glass followed by precipitation of a Ca2RE8(SiO4)6O2 (RE = Yb3+, Y3+) apatite-type silicate and/or rare earth disilicate. In glasses with high CaO concentrations, apatite readily forms while the disilicate material is consumed by the reaction. As CaO content decreases, the disilicate phase becomes the main reaction product. Overall, reactions with yttrium disilicate favored more apatite crystallization than ytterbium disilicate. The viability of using these disilicates in various operating environments is discussed.

High-temperature interactions of desert sand CMAS glass with yttrium disilicate environmental barrier coating material

A calcium-magnesium aluminosilicate (CMAS) glass was prepared by melting a sample of desert sand to evaluate the high-temperature interactions between molten CMAS and yttrium disilicate (Y2Si2O7), an environmental barrier coating (EBC) candidate material. Cold-pressed pellets of 80 wt% Y2Si2O7 powder and 20 wt% CMAS glass powder were heat treated at 1200 °C, 1300 °C, 1400 °C and 1500 °C for 20 h in air. The resulting phases were evaluated using powder X-ray diffraction. In the second set of experiments, free standing hot-pressed Y2Si2O7 substrates with cylindrical wells were filled with CMAS powder to a loading of ~35 mg/cm2 and heat treated in air at 1200 °C, 1300 °C, 1400 °C and 1500 °C for 20 h. Scanning electron microscopy, energy-dispersive spectroscopy and electron microprobe analysis were used to evaluate the microstructure and phase compositions of specimens after heat treatment. An oxyapatite silicate (Ca2Y8(SiO4)6O2) phase was identified in all specimens after CMAS exposure regardless of heat treatment temperature. Apatite appeared to form by dissolution of Y2Si2O7 into molten CMAS, reacting with CaO in the melt according to the reaction 4Y2Si2O7 + 2CaO → Ca2Y8(SiO4)6O2 + 2SiO2, and followed by precipitation of the apatite phase.

Slow-crack-growth and indentation damage in calcium magnesium aluminosilicate (CMAS) glass from desert sand

Desert sand from a Middle East country was melted into calcium magnesium aluminosilicate (CMAS) glass. Its chemical composition was analyzed to be 25.2CaO-2.6MgO-8.2Al2O3-59.8SiO2-1.6Fe2O3-1.5K2O weight % using inductively coupled plasma-atomic emission spectrometry. The CMAS glass powder was hot pressed into billets. Slow-crack-growth (SCG) and indentation deformation/fracture of the CMAS glass was investigated. The SCG susceptibility parameter (n) was found to be 25 ± 3 which is within a range of n = 15–35 that has been observed in many silicate glasses and glass ceramics. A similarity in indentation hardness and toughness was found between the CMAS glass and the low-silica content (50–70%) glasses. However, an exception was that significant lateral cracking was typified in the CMAS glass, as quantified via stress analysis in the vicinity of an indent.

High-temperature corrosion mechanism of YSZ coatings subject to calcium–magnesium–aluminosilicate (CMAS) deposits: First-principles calculations

To study the degradation mechanism of yttria-stabilized zirconia (YSZ) coatings by CMAS glass attack at high temperature, density functional theory is used to calculate the energy of CMAS/YSZ(111) system. Over 1513K, the diffusion coefficients of atoms in CMAS are larger than those in YSZ (111) and the diffusion process of (Ca↔Y) and (Si↔Y) occurs, which induced to the formation of a sandwich structure and YSZ depleting in yttrium, allowing tetragonal to monoclinic phase transition of zirconia. The cracking and peeling off of the coating is attributed to the change of YSZ volume and the thermal stress.

Hot corrosion of RE 2 SiO 5 with different cation substitution under calcium–magnesium–aluminosilicate attack

Rare earth (RE) silicates have been applied as advanced environmental barrier coatings (EBCs) to protect silicon carbide fibers reinforced silicon carbide ceramic matrix from water vapor and molten salt corrosion in engines. This process, however, is limited by volcanic ash corrosion as assessment of ash-induced corrosion is anecdotal and quantitative data are insufficient. In this account, the corrosion behavior of RE monosilicates (RE2SiO5, RE = Y, Lu, Yb, Eu, Gd, and La) by calcium–magnesium–aluminosilicate (CMAS), with similar composition as volcanic ash, was comprehensively investigated. Results indicated that RE2SiO5 could react with CMAS at 1200 °C at the interface, where the products crystallized in CMAS glass. RE2Si2O7 was formed by the reaction between RE2SiO5 and silica (SiO2) in CMAS, which was followed by corrosion of RE2Si2O7 by CMAS. RE2SiO5 with Type B structure showed better resistance toward CMAS than RE2SiO5 with Type A structure. Moreover, RE2SiO5 with larger radii of RE3+ cations led to easy formation of oxyapatite phase; however, RE2SiO5 with smaller radii of RE3+ cations easily formed garnet phase. Besides, smaller radii RE3+ cations induced slower reactions. These findings can contribute to identifying, preventing, and minimizing the damage to matrix components with EBCs caused by volcanic ash.

La2(Zr0.7Ce0.3)2O7 thermal barrier coatings prepared by electron beam-physical vapor deposition that are resistant to high temperature attack by molten silicate

Calcium-magnesium-alumino-silicate (CMAS) attack is a significant failure mechanism for thermal barrier coatings (TBC). In this study, La2(Zr0.7Ce0.3)2O7 (LZ7C3) TBC prepared by electron-beam physical vapor deposition was found to be highly effective in resisting CMAS infiltration for prolonged durations (24h, 1250°C), while conventional Y2O3-stabilized ZrO2 TBC was completely penetrated by molten CMAS within 30min. This resistance was achieved through crystallization of CMAS as a result of high-temperature chemical interactions between LZ7C3 coating and CMAS glass. The crystalline products, mainly consisted of La-Ce apatite and C-ZrO2, tended to form a sealing layer and can effectively prevent CMAS from further infiltration.

High temperature viscosity of calcium-magnesium-aluminosilicate glass from synthetic sand

Viscosity of a calcium-magnesium-aluminosilicate (CMAS) glass, melted from a synthetic sand with composition replicating that of air-breathing turbine engine deposits, was experimentally measured between 1215°C and 1520°C using a rotating spindle viscometer. Chemical composition of the CMAS glass before and after viscosity measurements was nominally 23.3CaO-6.4MgO-3.1Al2O3-62.5SiO2-4.1Na2O-0.5K2O-0.04Fe2O3(mol.%) as determined using inductively coupled plasma atomic emission spectroscopy. Experimental viscosity values were compared with those estimated from composition-based calculators of Giordano et al., Fluegel and FactSage software. Although none of these models exactly predicted viscosity values, those determined by Fluegel and FactSage models were found to more closely match experimental viscosity of the CMAS glass.

Mechanical and thermal properties of calcium–magnesium aluminosilicate (CMAS) glass

Thermal stability of a synthetic sand composition, which was developed as a simulant for calcium–magnesium aluminosilicate (CMAS) turbine deposits, was characterized using thermogravimetric and differential thermal analysis (TG/DTA). The sand was melted into CMAS glass, which had a composition of 23.3CaO–6.4MgO–3.1Al2O3–62.5SiO2–4.1Na2O–0.5K2O–0.04Fe2O3 (mol.%), determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES). Bulk density of the glass was measured to be 2.63g/cm3, and the Young's and shear moduli were 84.3GPa and 33.6GPa, respectively, along with a Poisson's ratio of 0.26. Vickers microhardness and indentation fracture toughness were determined to be 6.14±0.1GPa and 0.70±0.05MPam1/2, respectively. Glass transition temperature, softening point and coefficient of thermal expansion of the glass were measured by dilatometry. Glass viscosities were estimated over a temperature range of 600–1500°C using dilatometric reference points of the glass and from composition-based methods. Times required for infiltration of molten CMAS glass into thermal or environmental barrier coatings were also estimated.

Properties of CMAS glass from desert sand

X-ray diffraction analysis of as-received desert sand from a Middle East country showed the presence of quartz (SiO2), calcite (CaCO3), gypsum (CaSO4.2H2O), NaAlSi3O8, Mg2(Al3.9Si5.1O18) and Mg3Al2(SiO4)3 phases. A batch of as-received desert sand was melted into calcium magnesium aluminosilicate (CMAS) glass at ~1500°C. From inductively coupled plasma-atomic emission spectrometry, chemical composition of the CMAS glass was analyzed to be 27.8CaO-4MgO-5Al2O3-61.6SiO2-0.6Fe2O3-1K2O (mole %). Various physical, thermal, and mechanical properties of the glass have been evaluated. Bulk density of CMAS glass was 2.69g/cm3, Young׳s modulus 92GPa, Shear modulus 36GPa, Poisson׳s ratio 0.28, dilatometric glass transition temperature (Tg) 706°C, softening point (Td) 764°C, Vickers microhardness 6.3±0.4GPa, indentation fracture toughness 0.75±0.15MPa.m1/2, and coefficient of thermal expansion (CTE) 9.8×10−6/°C in the temperature range 25 to 700°C. Temperature dependence of viscosity has also been estimated from various reference points of the CMAS glass using the Vogel-Fulcher-Tamman (VFT) equation as well as from the glass composition. The glass remained amorphous after heat treating at 850°C for 10h but crystallized into CaSiO3 and Ca2Mg0.5AlSi1.5O7 phases at 900°C or higher temperatures. Crystallization kinetics of the CMAS glass has also been investigated by differential thermal analysis (DTA).

2ZrO2·Y2O3Thermal Barrier Coatings Resistant to Degradation by MoltenCMAS: Part I, Optical Basicity Considerations and Processing

The higher operating temperatures in gas‐turbine engines enabled by thermal barrier coatings (TBCs) engender new materials issues,vizsilicate particles (sand, volcanic ash, fly ash) ingested by the engine melt on the hotTBCsurfaces and form calcium–magnesium–alumino–silicate (CMAS) glass deposits. The moltenCMASglass degradesTBCs, leading to their premature failure. In this context, we have used the concept of optical basicity (OB) to provide a quantitative chemical basis for the screening ofCMAS‐resistantTBCcompositions, which could also be extended to environmental barrier coatings (EBCs). By applyingOBdifference considerations to various majorTBCcompositions and two types of importantCMASs—desert sand and fly ash—the 2ZrO2·Y2O3solid solution (ss)TBCcomposition, with the potential for highCMAS‐resistance, is chosen for this study. Here, we also demonstrate the feasibility of processing of 2ZrO2·Y2O3(ss) air‐plasma sprayed (APS)TBCusing commercially developed powders. The resultingTBCs with typicalAPSmicrostructures are found to be single‐phase cubic fluorite, having a thermal conductivity <0.9 W·(m·K)−1at elevated temperatures. The accompanying PartIIpaper presents results from experiments and analyses of high‐temperature interactions between 2ZrO2·Y2O3(ss)APS TBCand the two types ofCMASs.

Air Plasma-Sprayed Yttria and Yttria-Stabilized Zirconia Thermal Barrier Coatings Subjected to Calcium-Magnesium-Alumino-Silicate (CMAS)

Yttria (Y2O3) and zirconia (ZrO2) stabilized by 8 and 20 wt.%Y2O3 thermal barrier coatings (TBCs) subjected to calcium-magnesium-alumino-silicate (CMAS) have been investigated. Free-standing Y2O3, 8 and 20 wt.%YSZ coatings covered with synthetic CMAS slurry were heated at 1300 °C in air for 24 h in order to assess the effect of Y2O3 on the corrosion resistance of the coatings subjected to CMAS. The microstructures and phase compositions of the coatings were characterized by SEM, EDS, XRD, RS, and TEM. TBCs with higher Y2O3 content exhibited better CMAS corrosion resistance. Phase transformation of ZrO2 from tetragonal (t) to monoclinic (m) occurred during the interaction of 8YSZ TBCs and CMAS, due to the depletion of Y2O3 in the coating. Some amounts of original c-ZrO2 still survived in 20YSZ TBCs along with a small amount of m-ZrO2 that appeared after reaction with CMAS. Furthermore, Y2O3 coating was found to be particularly highly effective in resisting the penetration of molten CMAS glass at high temperature (1300 °C). This may be ascribed to the formation of sealing layers composed of Y-apatite phase [based on Ca4Y6 (SiO4)6O and Y4.67(SiO4)3O] by the high-temperature chemical interactions of Y2O3 coating and CMAS glass.

CMAS-Resistant Plasma Sprayed Thermal Barrier Coatings Based on Y2O3-Stabilized ZrO2 with Al3+ and Ti4+ Solute Additions

The higher operating temperatures in gas-turbine engines made possible by thermal barrier coatings (TBCs) are engendering a new problem: environmentally ingested airborne silicate particles (sand, ash) melt on the hot TBC surfaces and form calcium-magnesium-alumino-silicate (CMAS) glass deposits. The molten CMAS glass degrades the TBCs, leading to their premature failure. Here, we demonstrate the use of a commercially manufactured feedstock powder, in conjunction with air plasma spray process, to deposit CMAS-resistant yttria-stabilized zirconia-based TBCs containing Al3+ and Ti4+ in solid solution. Results from the characterization of these new TBCs and CMAS/TBCs interaction experiments are presented. The CMAS mitigation mechanisms in these new TBCs involve the crystallization of the anorthite phase. Raman microscopy is used to generate large area maps of the anorthite phase in the CMAS-interacted TBCs demonstrating the potential usefulness of this method for studying CMAS/TBCs interactions. The ubiquity of airborne sand/ash particles and the ever-increasing demand for higher operating temperatures in future high efficiency gas-turbine engines will necessitate CMAS resistance in all hot-section components of those engines. In this context, the versatility, ease of processing, and low cost offered by the process demonstrated here could benefit the development of these new CMAS-resistant TBCs.

Air-plasma-sprayed thermal barrier coatings that are resistant to high-temperature attack by glassy deposits

Thermal barrier coatings (TBCs) used in gas-turbine engines afford higher operating temperatures, resulting in enhanced efficiencies and performance. However, at these high operating temperatures, environmentally ingested airborne sand/ash particles melt on the hot TBC surfaces and form calcium–magnesium–aluminosilicate (CMAS) glass deposits. The molten CMAS glass penetrates the TBCs, leading to loss of strain tolerance and TBC failure. Here we demonstrate the use of the commercial manufacturing method of air-plasma-spray (APS) to fabricate CMAS-resistant yttria-stabilized zirconia (YSZ)-based TBCs containing Al and Ti in solid solution. Results from thermal stability studies of these new TBCs and CMAS/TBC interaction experiments are presented, together with a discussion of the CMAS mitigation mechanisms. The ubiquity of airborne sand/ash particles and the ever-increasing demand for higher operating temperatures in future high efficiency/performance gas-turbine engines will necessitate CMAS resistance in all hot-section components of those engines. In this context the versatility, ease of processing, and low cost offered by the APS method has broad implications for the design and fabrication of next-generation CMAS-resistant TBCs for future engines.

Novel thermal barrier coatings that are resistant to high-temperature attack by glassy deposits

Airborne sand particles that deposit on thermal barrier coatings (TBCs) in gas-turbine engines melt and form calcium–magnesium–aluminosilicate (CMAS) glass, which attacks the TBCs. A new approach for mitigating CMAS attack on TBCs is presented, where up to 20 mol.% Al 2O 3 and 5 mol.% TiO 2 in the form of a solid solution is incorporated into Y 2O 3-stabilized ZrO 2 (YSZ) TBCs. The fabrication of such TBCs with engineered chemistries is made possible by the solution-precursor plasma spray (SPPS) process, which is uniquely suited for depositing coatings of metastable ceramics with extended solid-solubilities. Here, the TBC serves as a reservoir of Al and Ti solutes, which are incorporated into the molten CMAS glass that is in contact with the TBC. This results in the crystallization of the CMAS glass and the attendant arrest of the penetrating CMAS front. This approach could also be used to mitigate attack by other types of foreign deposits (salt, ash, and contaminants) on TBCs.

The crystallisation of chromite‐magnesiochromite spinels from a calcium magnesium aluminosilicate glass: Nucleation, crystal growth and final crystal sizes

AbstractThe crystallisation of chromite‐magnesiochromite spinels was studied from a calcium magnesium aluminosilicate glass (a simulated slag) containing 3 to 12 percent total iron oxides and 0.3 to 1.5 percent chromium(III) oxide, at temperatures from 1400° to 700 °C. – Spinel crystallisation occurred in glasses with 3–7 percent FeO and 0.7–1.1 percent Cr2O3. At temperatures 1100 °C and above, the nucleation was rapid and crystal numbers very high, at FeO contents above 3 percent and Cr2O3 contents above 0.7 percent; at 1056° and 1000 °C however, the crystal numbers reached some optimum values but then decreased as clinopyroxene crystals grew onto and enveloped the spinel microcrystals. In these glasses, the crystal lengths varied with growth time according to the relation, lt = 2 kg tα = Rg1 tα, where α = 0.7–1.0: this time dependence was a compromise between a relation for dendritic growth and one for facetted growth. The growth rates generally increased about five to seven times for 160 °C temperature rise: the energy of activation for the spinel crystal growth was then estimated as 180 ± 60 kJ mole−1. – No spinel crystals were observed in glasses with more than 7 percent FeO content, only clinopyroxene crystals. Probably, these latter had nucleated rapidly and grown onto spinel microcrystals, while the spinel microcrystals were still of < 0.1 μm size.