Computational investigation of porosity effects on fracture behavior of thermal barrier coatings

Jayaprakash Krishnasamy; Sathiskumar A. Ponnusami; Sergio Turteltaub; Sybrand van der Zwaag

doi:10.1016/j.ceramint.2019.07.031

Computational investigation of porosity effects on fracture behavior of thermal barrier coatings

Jayaprakash Krishnasamy^*, Sathiskumar A. Ponnusami, Sergio Turteltaub, Sybrand van der Zwaag

^*Corresponding author for this work

Research output: Contribution to journal › Article › Scientific › peer-review

19 Citations (Scopus)

43 Downloads (Pure)

Abstract

The influence of microstructural pore defects on fracture behaviour of Thermal Barrier Coatings (TBC) is analysed using finite element analysis involving cohesive elements. A concurrent multiscale approach is utilised whereby the microstructural features of the TBC are explicitly resolved within a unit cell embedded in a larger domain. Within the unit cell, a random distribution of pores is modelled along with three different layers in a TBC system, namely, the Top Coat (TC), the Bond Coat (BC) and the Thermally Grown Oxide (TGO). The TC/TGO and the TGO/BC interfaces are assumed to be sinusoidal of specified amplitude and frequency extracted from experimental observations reported in the literature. To simulate fracture in the TBC, cohesive elements are inserted throughout the inter-element boundaries in order to enable arbitrary crack initiation and propagation. A bilinear traction-separation relation with specified fracture properties for each layer is used to model the constitutive behaviour of the cohesive elements. Parametric studies are conducted for various pore geometrical features, porosity, fracture properties of Top Coat layer and Thermally Grown Oxide layer thicknesses. The results are quantified in terms of crack initiation and evolution. It is found that the presence of pores has a beneficial effect on the fracture behavior up to a certain value of porosity after which the pores become detrimental to the overall performance. Insights derived from the numerical results can help in understanding the failure behavior of practical TBC systems and further aid in engineering the TBC microstructure for a desired fracture behavior.

Original language	English
Pages (from-to)	20518-20527
Number of pages	10
Journal	Ceramics International
Volume	45
Issue number	16
DOIs	https://doi.org/10.1016/j.ceramint.2019.07.031
Publication status	Published - 2019

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Cohesive elements
Concurrent multiscale model
Fracture
Porosity
Thermal barrier coatings

Access to Document

10.1016/j.ceramint.2019.07.031

1_s2.0_S0272884219318504_mainFinal published version, 1.96 MB

Cite this

@article{78c5d99fe00040d6b0073ee8d7b032c4,

title = "Computational investigation of porosity effects on fracture behavior of thermal barrier coatings",

abstract = "The influence of microstructural pore defects on fracture behaviour of Thermal Barrier Coatings (TBC) is analysed using finite element analysis involving cohesive elements. A concurrent multiscale approach is utilised whereby the microstructural features of the TBC are explicitly resolved within a unit cell embedded in a larger domain. Within the unit cell, a random distribution of pores is modelled along with three different layers in a TBC system, namely, the Top Coat (TC), the Bond Coat (BC) and the Thermally Grown Oxide (TGO). The TC/TGO and the TGO/BC interfaces are assumed to be sinusoidal of specified amplitude and frequency extracted from experimental observations reported in the literature. To simulate fracture in the TBC, cohesive elements are inserted throughout the inter-element boundaries in order to enable arbitrary crack initiation and propagation. A bilinear traction-separation relation with specified fracture properties for each layer is used to model the constitutive behaviour of the cohesive elements. Parametric studies are conducted for various pore geometrical features, porosity, fracture properties of Top Coat layer and Thermally Grown Oxide layer thicknesses. The results are quantified in terms of crack initiation and evolution. It is found that the presence of pores has a beneficial effect on the fracture behavior up to a certain value of porosity after which the pores become detrimental to the overall performance. Insights derived from the numerical results can help in understanding the failure behavior of practical TBC systems and further aid in engineering the TBC microstructure for a desired fracture behavior.",

keywords = "Cohesive elements, Concurrent multiscale model, Fracture, Porosity, Thermal barrier coatings",

author = "Jayaprakash Krishnasamy and Ponnusami, {Sathiskumar A.} and Sergio Turteltaub and {van der Zwaag}, Sybrand",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.",

year = "2019",

doi = "10.1016/j.ceramint.2019.07.031",

language = "English",

volume = "45",

pages = "20518--20527",

journal = "Ceramics International",

issn = "0272-8842",

publisher = "Elsevier",

number = "16",

}

TY - JOUR

T1 - Computational investigation of porosity effects on fracture behavior of thermal barrier coatings

AU - Krishnasamy, Jayaprakash

AU - Ponnusami, Sathiskumar A.

AU - Turteltaub, Sergio

AU - van der Zwaag, Sybrand

N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2019

Y1 - 2019

N2 - The influence of microstructural pore defects on fracture behaviour of Thermal Barrier Coatings (TBC) is analysed using finite element analysis involving cohesive elements. A concurrent multiscale approach is utilised whereby the microstructural features of the TBC are explicitly resolved within a unit cell embedded in a larger domain. Within the unit cell, a random distribution of pores is modelled along with three different layers in a TBC system, namely, the Top Coat (TC), the Bond Coat (BC) and the Thermally Grown Oxide (TGO). The TC/TGO and the TGO/BC interfaces are assumed to be sinusoidal of specified amplitude and frequency extracted from experimental observations reported in the literature. To simulate fracture in the TBC, cohesive elements are inserted throughout the inter-element boundaries in order to enable arbitrary crack initiation and propagation. A bilinear traction-separation relation with specified fracture properties for each layer is used to model the constitutive behaviour of the cohesive elements. Parametric studies are conducted for various pore geometrical features, porosity, fracture properties of Top Coat layer and Thermally Grown Oxide layer thicknesses. The results are quantified in terms of crack initiation and evolution. It is found that the presence of pores has a beneficial effect on the fracture behavior up to a certain value of porosity after which the pores become detrimental to the overall performance. Insights derived from the numerical results can help in understanding the failure behavior of practical TBC systems and further aid in engineering the TBC microstructure for a desired fracture behavior.

AB - The influence of microstructural pore defects on fracture behaviour of Thermal Barrier Coatings (TBC) is analysed using finite element analysis involving cohesive elements. A concurrent multiscale approach is utilised whereby the microstructural features of the TBC are explicitly resolved within a unit cell embedded in a larger domain. Within the unit cell, a random distribution of pores is modelled along with three different layers in a TBC system, namely, the Top Coat (TC), the Bond Coat (BC) and the Thermally Grown Oxide (TGO). The TC/TGO and the TGO/BC interfaces are assumed to be sinusoidal of specified amplitude and frequency extracted from experimental observations reported in the literature. To simulate fracture in the TBC, cohesive elements are inserted throughout the inter-element boundaries in order to enable arbitrary crack initiation and propagation. A bilinear traction-separation relation with specified fracture properties for each layer is used to model the constitutive behaviour of the cohesive elements. Parametric studies are conducted for various pore geometrical features, porosity, fracture properties of Top Coat layer and Thermally Grown Oxide layer thicknesses. The results are quantified in terms of crack initiation and evolution. It is found that the presence of pores has a beneficial effect on the fracture behavior up to a certain value of porosity after which the pores become detrimental to the overall performance. Insights derived from the numerical results can help in understanding the failure behavior of practical TBC systems and further aid in engineering the TBC microstructure for a desired fracture behavior.

KW - Cohesive elements

KW - Concurrent multiscale model

KW - Fracture

KW - Porosity

KW - Thermal barrier coatings

UR - http://www.scopus.com/inward/record.url?scp=85068464939&partnerID=8YFLogxK

U2 - 10.1016/j.ceramint.2019.07.031

DO - 10.1016/j.ceramint.2019.07.031

M3 - Article

SN - 0272-8842

VL - 45

SP - 20518

EP - 20527

JO - Ceramics International

JF - Ceramics International

IS - 16

ER -

Computational investigation of porosity effects on fracture behavior of thermal barrier coatings

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this