Super-Hydrophobic Ceramic & Waterproof Nano Coatings _ What Are Nano Coatings
Environmental Barrier Coatings for SiC Ceramic Matrix ......Environmental Barrier Coatings for SiC...
1
Environmental Barrier Coatings for SiC Ceramic Matrix Composite Gas Turbine Blades Adithya Bhattachar, Mary Katherine O’Brien, Mitchell Rencheck, Gregory Scofield Faculty Advisors: Prof. Rodney Trice Industrial Sponsors: Dr. Kang Lee & Dr. Stephanie Gong Project Background Rolls-Royce has been developing silicon carbide (SiC) Ceramic Matrix Composites (CMCs) to be used in place of single crystal super alloys for gas turbine blades due to their lower weights and higher operating temperature limits. However, SiC degrades in a high-temperature combustion environment, creating a need for an environmental barrier coating (EBC) to protect the SiC. It has been hypothesized that an EBC containing mullite and barium strontium aluminosilicate (BSAS) can prevent the glassy phase from forming, while also displaying limited cracking when thermally cycled. During experimentation, this hypothesis was not confirmed, as adding BSAS to mullite caused the coating to bubble during thermal cycling. Experimental Procedure Slurry Production 1) Attrition milling was used to reduce particle size of mullite and BSAS powders. 2) Compositions of varying weight percentages of mullite, BSAS, polyvinyl butyral (PVB), and sintering aid (LiCO 3 or MgO) were added to a Darvan-C, ethanol mixture. 3) The slurry was mixed on a ball mill for 24 hours to uniformly mix ingredients for dip coating. Dip Coating 1) In preparation for dipping, air was removed from the slurries using an ultrasonic bath. 2) 1.27cm x 2.54cm (0.5” x 1”) SiC coupons were dipped in the slurry and spun at 600 RPM to remove excess slurry. 3) Coatings were air-dried for 24 hrs before being sintered at 1300°C for 5 hrs. Thermal Cycle Testing Results Coating Process Surface Characterization • Use dilatometer to quantitatively assess the changes in coating sintering temperature based on levels of sintering aid. • Investigate solvents that evaporate more slowly, in order to reduce mud cracking before sintering. Future Works Discussion MSE 430 - 440: Materials Processing and Design This work is sponsored by Rolls-Royce North America, Indianapolis, IN. • CMCs have the potential to increase operating temperatures inside gas turbine aircraft engines and thereby increase fuel efficiency. • SiC/SiC CMCs are desirable due to their low weight, low coefficient of thermal expansion (CTE), and high melting temperature. • SiC is problematic in high temperature environments due to interaction with water vapor, which causes oxidation and the formation of a low-T m glass phase [1] . • An EBC is needed to protect SiC from the water vapor present in a combustion environment. • BSAS has a CTE that is similar to SiC, but is easily penetrated by water. • Mullite prevents water penetration, but cracks when thermally cycled, due to a large difference between its CTE and the CTE of SiC. • A relatively crack free, impenetrable EBC with a porosity level of 20%, and a thickness of 25-75 μm, is required to meet Rolls-Royce's need. • A coating containing both mullite and BSAS has been hypothesized to resist both cracking and water penetration. [1] Lee, K. (2014). Environmental Barrier Coatings for SiC F /SiC. Materials, Modeling and Technology Ceramic Matrix Composites, 430-451. Retrieved November 23, 2015, from Wiley Online Library. Samples were thermally cycled by ablation testing [2] . Samples were ablated 3 times for 1 minute at 1400°C, with 1 minute of rest between cycles. [2] ASTM Standard E285-08, 2015, “Standard Testing Method of Oxyacetylene Ablation Testing of Thermal Insulation Materials,” ASTM International, West Conshohocken, PA, 2015. Results Continued SEM cross-sectional micrograph of mullite/BSAS coating 0 10 20 30 40 50 60 70 0.8 0.9 1 1.1 1.2 Coating Thickness (μm) Weight Ratio of Ethanol:Powder Image of single dipped (left) and double dipped (right) coating Single Dip Double Dip • Mud cracking appears in both single and double dipped coatings. • Double-dipping reduced number of through- thickness microcracks. • Double-dipping increased number of large through-thickness cracks. 20 μm 20 μm 2 μm 30 μm Coating SiC Constant Parameter w/ 2.5 wt. % MgO 4:1 wt. % Mullite:BSAS Wt. % Mullite:BSAS 9:1 4:1 7:3 Sintering Aid 0 wt. % 1.25 wt. % 2.5 wt. % 5 wt. % Sintered (Y/N) Y Y Y Sintered (Y/N) N N Y N Effective slurries contain • 1 wt. % Darvan-C • 4 wt. % PVB • 1:1 wt. % Ethanol:Powder Observations • Coatings adhered to SiC substrate after sintering. • When LiCO 3 was used as a sintering aid, coating uniformity was inadequate for continued testing. 200 μm 150 μm Most of the mullite/BSAS coating ablated off of the SiC substrate. Images of a 4:1 BSAS to mullite coating before thermal cycling (left) and after thermal cycling (right) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 20 30 40 50 60 70 80 90 Normalized Intensity 2θ 4:1 Mullite:BSAS Post Ablation 4:1 Mullite:BSAS Pre Ablation Thermal Cycle Testing Pinholes nucleated in regions that experienced increased densification during sintering. (a) (b) (c) SEM micrographs of coating containing 7:3 wt.% mullite:BSAS, with 2.5 wt% MgO. Sintering shown between particles in images (a) & (b). Increased densification shown around hole in image (c). Coating thickness became more variable with increased powder loading. Below 1.15 EtOH:Powder ratio, coating becomes too thin to meet project constraints. Investigation of binder • A polymerization reaction between PVB and water causes a skin to form on the surface of the slurry and reduces coating uniformity during dipping. • Binders that do not react in air would improve the repeatability of the dipping process. Thermal cycle testing • Sample with lowest mullite:BSAS ratio displayed bubbling in the coating, which suggests water penetration. Possible causes of coating failure • Based on XRD analysis, there is no evidence to suggest that a new phase formed in the coating when samples were thermally cycled. • Evidence therefore suggests that pressure from the ablation nozzle, the possible penetration of water, or differences in CTE between the coating and the substrate caused spallation to occur. Possible causes of regions with increased densification Formation of a silicate glass due to reaction occurring with mullite or BSAS during sintering. Conclusion • Single dipped coatings are more desirable than double dipped coatings due to uniformity of surface finishing. • A coating that mixed mullite and BSAS was insufficient to prevent the delamination that BSAS causes. • Greater ethanol:powder ratios are ideal for thinner coatings. SEM micrograph of double dipped coating with arrow showing stemming of through-thickness cracks 50 μm First dip SEM micrograph of single dipped coating with arrows pointing to through-thickness mud cracks 80 μm Mud cracks 25 mm 7:3 4:1 9:1
Transcript of Environmental Barrier Coatings for SiC Ceramic Matrix ......Environmental Barrier Coatings for SiC...
Environmental Barrier Coatings for SiC Ceramic Matrix Composite Gas Turbine BladesAdithya Bhattachar, Mary Katherine O’Brien, Mitchell Rencheck, Gregory ScofieldFaculty Advisors: Prof. Rodney TriceIndustrial Sponsors: Dr. Kang Lee & Dr. Stephanie Gong
Project Background
Rolls-Royce has been developing silicon carbide (SiC) Ceramic Matrix Composites (CMCs) to be used in place of single crystalsuper alloys for gas turbine blades due to their lower weights and higher operating temperature limits. However, SiC degrades in ahigh-temperature combustion environment, creating a need for an environmental barrier coating (EBC) to protect the SiC. It hasbeen hypothesized that an EBC containing mullite and barium strontium aluminosilicate (BSAS) can prevent the glassy phase fromforming, while also displaying limited cracking when thermally cycled. During experimentation, this hypothesis was not confirmed,as adding BSAS to mullite caused the coating to bubble during thermal cycling.
Experimental ProcedureSlurry Production1) Attrition milling was used to reduce particle
size of mullite and BSAS powders.2) Compositions of varying weight percentages
of mullite, BSAS, polyvinyl butyral (PVB),and sintering aid (LiCO3 or MgO) wereadded to a Darvan-C, ethanol mixture.
3) The slurry was mixed on a ball mill for 24hours to uniformly mix ingredients for dipcoating.
Dip Coating1) In preparation for dipping, air was removed
from the slurries using an ultrasonic bath.2) 1.27cm x 2.54cm (0.5” x 1”) SiC coupons
were dipped in the slurry and spun at 600RPM to remove excess slurry.
3) Coatings were air-dried for 24 hrs beforebeing sintered at 1300°C for 5 hrs.
Thermal Cycle Testing
ResultsCoating Process
Surface Characterization
• Use dilatometer to quantitatively assess thechanges in coating sintering temperature based onlevels of sintering aid.
• Investigate solvents that evaporate more slowly, inorder to reduce mud cracking before sintering.
Future Works
Discussion
MSE 430-440: Materials Processing and Design
This work is sponsored by Rolls-Royce North America, Indianapolis, IN.
• CMCs have the potential to increase operatingtemperatures inside gas turbine aircraft engines andthereby increase fuel efficiency.
• SiC/SiC CMCs are desirable due to their low weight,low coefficient of thermal expansion (CTE), and highmelting temperature.
• SiC is problematic in high temperature environmentsdue to interaction with water vapor, which causesoxidation and the formation of a low-Tm glass phase[1].
• An EBC is needed to protect SiC from the water vaporpresent in a combustion environment.
• BSAS has a CTE that is similar to SiC, but is easilypenetrated by water.
• Mullite prevents water penetration, but cracks whenthermally cycled, due to a large difference between itsCTE and the CTE of SiC.
• A relatively crack free, impenetrable EBC with aporosity level of 20%, and a thickness of 25-75 µm, isrequired to meet Rolls-Royce's need.
• A coating containing both mullite and BSAS has beenhypothesized to resist both cracking and waterpenetration.
[1] Lee, K. (2014). Environmental Barrier Coatings for SiCF/SiC. Materials, Modelingand Technology Ceramic Matrix Composites, 430-451. Retrieved November 23, 2015,from Wiley Online Library.
Samples were thermally cycled by ablation testing[2]. Samples were ablated 3 times for 1 minute at 1400°C, with 1 minute of rest between cycles.
[2] ASTM Standard E285-08, 2015,“Standard Testing Method ofOxyacetylene Ablation Testing ofThermal Insulation Materials,”ASTM International, WestConshohocken, PA, 2015.
Results Continued
SEM cross-sectional micrograph of mullite/BSAS coating
0
10
20
30
40
50
60
70
0.8 0.9 1 1.1 1.2
Coa
ting
Thic
knes
s (µ
m)
Weight Ratio of Ethanol:Powder
Image of single dipped (left) and double dipped (right) coating
Single Dip Double Dip• Mud cracking appears inboth single and doubledipped coatings.
• Double-dipping reducednumber of through-thickness microcracks.
• Double-dippingincreased number oflarge through-thicknesscracks.
20 μm 20 μm2 μm
30 μm
Coating SiC
Constant Parameter w/ 2.5 wt. % MgO 4:1 wt. % Mullite:BSAS
Wt. % Mullite:BSAS 9:1 4:1 7:3 Sintering
Aid0 wt.
%1.25 wt.
%2.5 wt.
%5 wt.
%Sintered (Y/N) Y Y Y Sintered
(Y/N)N N Y N
Effective slurries contain • 1 wt. % Darvan-C• 4 wt. % PVB• 1:1 wt. % Ethanol:PowderObservations• Coatings adhered to SiC
substrate after sintering.• When LiCO3 was used as
a sintering aid, coatinguniformity was inadequatefor continued testing.
200 μm150 μm
Most of the mullite/BSAS coating ablated off of the SiCsubstrate.
Images of a 4:1 BSAS to mullite coating before thermal cycling (left) and after thermal cycling (right)
00.10.20.30.40.50.60.70.80.9
1
20 30 40 50 60 70 80 90
Nor
mal
ized
Inte
nsity
2θ
4:1 Mullite:BSASPost Ablation
4:1 Mullite:BSASPre Ablation
Thermal Cycle Testing
Pinholes nucleated in regions that experienced increased densification during sintering.
(a) (b) (c)SEM micrographs of coating containing 7:3 wt.% mullite:BSAS, with 2.5
wt% MgO. Sintering shown between particles in images (a) & (b). Increased densification shown around hole in image (c).
Coating thickness became more variable with increased powder loading.
Below 1.15 EtOH:Powderratio, coating becomes too thin to meet project constraints.
Investigation of binder• A polymerization reaction between PVB and water
causes a skin to form on the surface of the slurryand reduces coating uniformity during dipping.
• Binders that do not react in air would improve therepeatability of the dipping process.
Thermal cycle testing• Sample with lowest mullite:BSAS ratio displayed
bubbling in the coating, which suggests waterpenetration.
Possible causes of coating failure• Based on XRD analysis, there is no evidence to
suggest that a new phase formed in the coatingwhen samples were thermally cycled.
• Evidence therefore suggests that pressure fromthe ablation nozzle, the possible penetration ofwater, or differences in CTE between the coatingand the substrate caused spallation to occur.
Possible causes of regions with increased densificationFormation of a silicate glass due to reactionoccurring with mullite or BSAS during sintering.
Conclusion• Single dipped coatings are more desirable than double
dipped coatings due to uniformity of surface finishing.• A coating that mixed mullite and BSAS was insufficient to
prevent the delamination that BSAS causes.• Greater ethanol:powder ratios are ideal for thinner coatings.
SEM micrograph of double dipped coating with arrow showing stemming
of through-thickness cracks
50 μm
First dip
SEM micrograph of single dipped coating with arrows pointing to through-thickness mud cracks
80 μm
Mud cracks
25 mm
7:3 4:1 9:1
