Recent CMAS related activities at IEK-1

Transcription

1 Mitglied der Helmholtz-Gemeinschaft Recent CMAS related activities at IEK-1 Daniel E. Mack, O. Helle, M. Mutter, R. Vaßen, O. Guillon Institute of Energy and Climate Research (IEK-1), Forschungszentrum Jülich GmbH, Germany PROVIDA Meeting, April 1st, 2015, Cambridge, UK

2 Summary of Interests Materials synthesis and processing for advanced (thermally sprayed) TBC systems Corrosion gradient burner rig testing TBC degradation under simultaneous load Degradation of structural integrity Impact of microstructure Impact of cycling frequency and deposition rate Impact of chemistry, mitigation in advanced ceramics Development of CMAS tolerant TBC systems Slide 2 Forschungszentrum Jülich GmbH

3 CMAS Burner Rig Testing Thermomechanical cycling Natural gas / oxygen burner Compressed air cooling (backside & front) Cyclic heating & cooling Typical lifetime w/o CMAS deposit for advanced YSZ TBC >2000 cycles at T surface = 1250 C >800 cycles at T surface = 1400 C with T bondcoat 1080 C and 5min heating, 2min cooling 2045 cycles w/o deposit Slide 3 Forschungszentrum Jülich GmbH

4 CMAS Burner Rig Testing Thermomechanical cycling Natural gas / oxygen burner Compressed air cooling (backside & front) Cyclic heating & cooling Continuous deposition simulated as during engine operation CMAS solution directly injected and atomized into the burner flame Orange discoloration visible (alkalides) Slide 4 Forschungszentrum Jülich GmbH

5 Variation of Load Conditions A number of parameters to be controlled Surface temperature Temperature gradient High temperature dwell time Cycling frequency CMAS composition Feeding rate to influence Melting, viscosity of deposit Reaction kinetics Phase formation Thermomechanical stresses Crack formation and growth Oxidation, sintering Slide 5 Forschungszentrum Jülich GmbH

6 CMAS Arrest & Mitigation Advanced TBC and Plant Deposits Psuedowollastonite Field Difficult to Crystallize SiO 2 Ca Mg Al Si Fe Na K CMAS CMAF V-ASH CMAS Glass Ca 3 Si 2 O 7 CaSiO 3 Psuedowollastonite Anorthite CaAl 2 Si 2 O 8 (Anorthite) Cation fractions in solutions (mol%) also variants with additions of sulphur, titania,.. Ca 2 SiO 4 Ca 3 SiO 5 Gehlenite Ca 2 Al 2 SiO 7 (Gehlenite) Al 2 Si 6 O 13 (Mullite) Ca CaO 3 Al 2 O 6 CaAl 2 O 4 CaAl 4 O 7 Al 2 O Ca 3 12Al 14O 33 CaAl 12O 19 Plant CMAF Deposit Anorthite & Gehlenite Field Easy to Crystallize (with Al 3+ /Ti 4+ /Gd 3+ ) Padture et al., Acta Mater., 2007 Slide 6 Forschungszentrum Jülich GmbH

7 Gradient Rig CMAS Invasion: APS Cycles to failure 160 cycles Time to failure 13,3 hours Test conditions: T surface : 1250 C Deposit: CMAS Rate: 0,12mg/min Dwell time: 1x (5/2) Top-down infiltration of porosity (along large cracks) Dissolution /Re-precipitation of YSZ Sparse invasion (preferred in large voids) Cathodoluminescence June 25th, 2014 Irsee, Germany Thermal Barrier Coatings IV Slide 7 Forschungszentrum Jülich GmbH

8 Gradient Rig CMAS Invasion: APS Test conditions: T surface : 1250 C Deposit: CMAS Rate: 0,12mg/min Dwell time: 1x (5/2) Cycles to failure 160 cycles Time to failure 13,3 hours Top-down infiltration of porosity (along large cracks) Dissolution/Re-precipitation of YSZ Sparse invasion (preferred in large voids) June 25th, 2014 Irsee, Germany Thermal Barrier Coatings IV Slide 8 Forschungszentrum Jülich GmbH

9 Gradient Rig CMAS Invasion: APS Test conditions: T surface : 1250 C Deposit: CMAS Rate: 0,12mg/min Dwell time: 1x (5/2) Cycles to failure 160 cycles Time to failure 13,3 hours Top-down infiltration of porosity (along large cracks) Dissolution/Re-precipitation of YSZ Sparse invasion (preferred in large voids) June 25th, 2014 Irsee, Germany Thermal Barrier Coatings IV Slide 9 Forschungszentrum Jülich GmbH

10 Impact of Different Dwell Times State of the art YSZ TBC Test conditions: T surface : 1250 C Deposit: CMAS Deposition rate: 0,45mg/min Dwell times: 1x (5/2) 5x (25/10) 10x (50/20) Degradation of TBC with flake like defoliation as often observed with CMAS No deposit T dwell = 5min cycles = 2045 T dwell = 25min cycles = 13 T dwell = 5min cycles = 50 T dwell = 50min cycles = 6 Slide 10 Forschungszentrum Jülich GmbH

11 Non-Cyclic Loading State of the art YSZ TBC Test conditions: T surface : 1250 C Deposit: CMAS Deposition rate: 0,5 mg/min) Dwell time: unlimited (only accidential interrupts) Delamination of TBC not flaky but en bloc near bondcoat Delamination occurs on heating ramp unlike usual in TGO derived failure total time = 6 hours interrupts = 2 total time = 7 hours interrupts = 2 Slide 11 Forschungszentrum Jülich GmbH

12 Summary on Dwell Time and Deposition Rate H=5min, C=2min, CMAS 0,45mg/min H=5min, C=2min, CMAS 0,45mg/min H=25min, C=10min, CMAS 0,45mg/min H=25min, C=10min, CMAS 0,45mg/min H=50min, C=20min, CMAS 0,45mg/min H=50min, C=20min, CMAS 0,45mg/min H=25min, C=10min CMAS 0,08mg/min H=25min, C=10min CMAS 0,08mg/min H>>60min, C=Accid. CMAS 0,50mg/min Cycle to failure H>>60min, C=Accid. CMAS 0,50mg/min Total deposition time (h) Accumulated deposit (g) Accumulated deposit apparently most important load factor! Slide 12 Forschungszentrum Jülich GmbH

13 # of cycles to failure High temperature corrosion with 1250 C 1. C Bond coat TBC HVOF-FS APS-FZJ HVOF-FS APS-FS Rene 80 VPS-FZJ APS-FZJ Coating Manufacture Hastelloy X HVOF-FS APS-FS CMAS composition (mol%) a 3 8 Mg Al Si Fe Na K Failure occured within the YSZ TBC and bondcoat interface Tests have been finished and characterization is ongoing Surface T / Substrate T, C 1261 / / / / 982 Porosity of TBC (as-sprayed), % No info. BC Roughness, um 10.3 / 13* * No info.

14 e.g. different porosity e.g. different bondcoat Chemistry but Micrographs after 400 burner rig cycles (5min/2min at 1250 C / 1080 C) e.g. different thickness YSZ A HVOF A Rene 80 System A YSZ B HVOF A Rene 80 System B e.g. different substrate YSZ C VPS A Rene 80 System C YSZ B HVOF B Hastelloy X System D

15 High temperature corrosion with 1250 C BC / TBC Manufacturer Surface T / Substrate T, C Lifetime Substrate HVOF-FS / APS- FZ 1261 / Rene 80 HVOF-FS / APS-FS 1256 / Rene 80 VPS-FZ / APS-FZ 1239 / Rene 80 HVOF-FS / APS-FS 1218 / Hastelloy X 300um 300um 300um 300um 25um 25um 25um Cross-section micrographs of tested samples 25um

16 System A ctf 160 cycles YSZ A HVOF A Rene 80 System A re-precipitation of YSZ sparse infiltration preferred in large voids

17 System B ctf 180 cycles YSZ B HVOF A Rene 80 System B re-precipitation of YSZ sparse infiltration preferred in large voids

18 Compare System B vs. D ctf 180 cycles YSZ B HVOF A Rene 80 System B ctf 25 cycles YSZ B HVOF B Hastelloy X System D Different degradation potentially due to different stress levels Layer thickness, interface temperature Substrate material Interface roughness

19 General comments: Major impact of microstructure (porosity, pore size) on performance and damage mode (shallow, spallation or in-depth delamination) Mechanism of in-depth degradation still insufficient understood Started some work on evolution of CMAS induced degradation by interrupted cycling tests same kind of YSZ TBC system same test conditions test not driven to failure for all specimen characterization of mechanical degradation by indentation testing started some modelling to describe

20 Vorversuche Last Gute Eindrück e Schlechte Eindrücke 5000mN 1000mN 500mN HEL0001 (TPro; CMAF) konf. Lasermikroskop höhere Last größere Eindrücke größeres Prüfvolumen besser für Statistik; aber auch mehr und evtl. ungewollte Defekte im Prüfvolumen (Poren) wichtig: höhere Last niedrigere gemessene Härte (Indentation Size Effect); auch strukturabhängig Vergleich versch. Lasten/Mikrostrukturen schwierig Idee: Prüfkraft an Mikrostruktur und gewünschter Defektgröße anpassen Frage: Welche Defekte sind für den Einfluss von CMAS/CMAF zu berücksichtigen? 17. April 2015 Oliver Helle 20

21 Serie 1 Indentation TBC: Standard YSZ APS BC: Amdry386 VPS Sub: IN738 Force: 250mN (Indent < Splat) HEL0028 (as-sprayed) konf. Lasermikroskop 17. April 2015 Oliver Helle 21

22 Serie 1 Failure and Degradation TBC delaminates near BC Horizontal cracks in central and upper levels HEL0023: 69 Zyklen, versagt konf. Lasermikroskop HEL0024: 46 Zyklen, intakt konf. Lasermikroskop 17. April 2015 Oliver Helle 22

23 Serie 1 Indentation Results expected trend of increased hardness in surface near region is there hardness drecrease near BC interface 17. April 2015 Oliver Helle 23

24 Serie 1 Indentation Results Similar trend for Young s modulus 17. April 2015 Oliver Helle 24

25 Youngs Modulus / GPa Density / kg/m³ Thermal Expansion Coefficient / ppm/k Thermal Conductivity / W/mK Rotation Axis Modelling: Geometry and Material Parameter y TC BC 0.45 mm 0.15 mm x CMAS Transition Zone 50 µm 100 µm Substrate 3 mm YSZ 300 µm TGO 5 µm 15 mm Material Parameter at Room Temperature: 4.00E E E E E E E E E+00 BC TGO YSZ Transition CMAS BC TGO YSZ Transition CMAS Position y / m Position y / m April 2015 Youngs Modulus Density Thermal Expansion Coefficient Thermal Conductivity

26 Modelling: Meshing and Boundary Conditions Thermal Boundary Conditions Case 1 Case 2 Isothermal 1000 C Cooling to 22 C Mechanical Boundary Conditions Heating: Convection (h = 2000 W/m²K, T = 1450 C) Cooling: Convection (h = 350 W/m²K, T= 22 C) Restriction of displacement at substrate bottom center 17. April

27 Happy for comments. Thank you for your attention Slide 27 Forschungszentrum Jülich GmbH

28 Advanced TBCs for CMAS Mitigation Assessment: ~80 cycles 0,12mg/min Deep exfoliation of YSZ reference YSZ A cycle 80 Gd YSZ B 2 Zr 2 O 7 (GZO) cycle 115 ter thickness Improved resistance YAG for advanced double layer ceramics with itial) mitigation within advanced ceramic surface cycle 80 YSZ-GZO -Comp cycle 80 near layer Slide 28 Forschungszentrum Jülich GmbH