Composites: Part A 39(2008)1829-1837 Contents lists available at ScienceDirect Composites: Part A ELSEVIER journalhomepagewww.elsevier.com/locate/compositesa Effects of steam environment on compressive creep behavior of Nextelm720 Alumina ceramic composite at 1200C M B. Ruggles-Wrenn N.R. Szymczak Department of Aeronautics and Astronautics, Air Force institute of Technology wright-Patterson Air Force Base, OH 45433-7765, US ARTICLE IN F O ABSTRACT Article history The compressive creep behavior of an oxide-oxide ceramic-matrix composite( MC) was investigated at Received 18 January 2008 1200C in laboratory air and in steam. The composite consis porous alumina matrix reinforced with Received in revised form 27 August 2008 ccepted 5 September 2008 laminated, woven mullite/ alumina(Nextel 720)fibers, has no interface between the fiber and matrix, and relies on the porous-matrix for flaw tolerance. The compressive stress-strain behavior was investi- gated and the compressive properties measured. The influence of the loading rate on compressive stress- strain response and on compressive properties was also explored. A change in loading rate by four orders A: Ceramic-matrix composites(CMCs) or magnitude had a profound effect on compressive properties in air and especially in steam Compres- ve creep behavior was examined for creep stresses in the -40 to-100 MPa range Minimum creep rate was reached in all tests In air, compressive creep strain magnitudes remained <0.4% and compressi reep strain rates approached -35 x 10-75-1. Creep run- out defined as 100 h at creep stress was achieved in all tests conducted in air. The presence of steam accelerated creep rates and significantly reduced creep lifetimes In steam, compressive creep strains approached -1.6%, and compressive creep strain rat es,-19x 10-2s-1 In steam, maximum time to rupture was only 3.9 h. Composite microstruc- ture, as well as damage and failure mechanisms were investigated. Published by Elsevier Ltd. 1 Introduction stable composites motivated the development of CMCs based on environmentally stable oxide constituents [6-11- Advances in power generation systems for aircraft engines The main advantage of CMCs over monolithic ceramics is their nd-based turbines, rockets, and, most recently, hypersonic mis- superior toughness, tolerance to the presence of cracks and defects, siles and flight vehicles have raised the demand for structural and non-catastrophic mode of failure. It is widely accepted that in materials that have superior long-term mechanical properties order to avoid brittle fracture behavior in CMCs and improve the and retained properties under high-temperature, high pressure, damage tolerance, a weak fiber/matrix interface is needed, which and varying environmental factors, such as moisture [1]. Cera- serves to deflect matrix cracks and to allow subsequent fiber pull- mic-matrix composites, capable of maintaining excellent strength out[12-14. It has been demonstrated that similar crack-deflecting and fracture toughness at high-temperatures are prime candidate behavior can also be achieved by means of a finely distributed materials for such applications. Additionally, lower densities of porosity in the matrix instead of a separate interface between ma- CMCs and their higher use temperatures, together with a reduced trix and fibers 15. This microstructural design philosophy implic need for cooling air, allow for improved high-temperature perfor- itly accepts the strong fiber/matrix interface. The concept has been ance when compared to conventional nickel-based superalloys successfully demonstrated for oxide-oxide composites 6,9, 11 [2]. Advanced aerospace turbine engines will likely incorporate fi- 16, 17]. Resulting oxide/oxide CMCs exhibit damage tolerance com- ber-reinforced CMCs in critical components, such as combustor bined with inherent oxidation resistance. An extensive review of walls [3-5]. Because these applications require exposure to oxidiz- the mechanisms and mechanical properties of porous-matrix CMCs ing environments, the thermodynamic stability and oxidation is given in [18, 19 resistance of CMCs are vital issues. The need for environmentally Porous-matrix oxide/oxide CMCs exhibit several behavio trends that are distinctly different from those exhibited by tradi- tional non-oxide Cmcs with a fiber-matrix interface For the non- oxide CmCs, fatigue is significantly more damaging than creep. al policy Contrastingly, Zawada et al. [20] examined the high-temperature Go Osition of the United States Air Force, Department of Defense or the Us mechanical behavior of a porous-matrix Nextel610/Aluminosili Corresponding author. Tel : +1 937 255 3636x4641: fax: +1 937 656 7053 cate composite Results revealed excellent fatigue performance at mail address: marina. ruggles-wrenn@afit.edu(M B. Ruggles-wrennl 000C. Conversely, creep lives were short, indicating low creep latter Published by Elsevier Ltd
Effects of steam environment on compressive creep behavior of NextelTM720/Alumina ceramic composite at 1200 C q M.B. Ruggles-Wrenn *, N.R. Szymczak Department of Aeronautics and Astronautics, Air Force Institute of Technology Wright-Patterson Air Force Base, OH 45433-7765, USA article info Article history: Received 18 January 2008 Received in revised form 27 August 2008 Accepted 5 September 2008 Keywords: A: Ceramic-matrix composites (CMCs) B: Creep B: Environmental degradation D: Mechanical testing abstract The compressive creep behavior of an oxide–oxide ceramic-matrix composite (CMC) was investigated at 1200 C in laboratory air and in steam. The composite consists of a porous alumina matrix reinforced with laminated, woven mullite/alumina (NextelTM720) fibers, has no interface between the fiber and matrix, and relies on the porous-matrix for flaw tolerance. The compressive stress–strain behavior was investigated and the compressive properties measured. The influence of the loading rate on compressive stress– strain response and on compressive properties was also explored. A change in loading rate by four orders or magnitude had a profound effect on compressive properties in air and especially in steam. Compressive creep behavior was examined for creep stresses in the 40 to 100 MPa range. Minimum creep rate was reached in all tests. In air, compressive creep strain magnitudes remained <0.4% and compressive creep strain rates approached 3.5 107 s1 . Creep run-out defined as 100 h at creep stress was achieved in all tests conducted in air. The presence of steam accelerated creep rates and significantly reduced creep lifetimes. In steam, compressive creep strains approached 1.6%, and compressive creep strain rates, 1.9 102 s 1 . In steam, maximum time to rupture was only 3.9 h. Composite microstructure, as well as damage and failure mechanisms were investigated. Published by Elsevier Ltd. 1. Introduction Advances in power generation systems for aircraft engines, land-based turbines, rockets, and, most recently, hypersonic missiles and flight vehicles have raised the demand for structural materials that have superior long-term mechanical properties and retained properties under high-temperature, high pressure, and varying environmental factors, such as moisture [1]. Ceramic-matrix composites, capable of maintaining excellent strength and fracture toughness at high-temperatures are prime candidate materials for such applications. Additionally, lower densities of CMCs and their higher use temperatures, together with a reduced need for cooling air, allow for improved high-temperature performance when compared to conventional nickel-based superalloys [2]. Advanced aerospace turbine engines will likely incorporate fi- ber-reinforced CMCs in critical components, such as combustor walls [3–5]. Because these applications require exposure to oxidizing environments, the thermodynamic stability and oxidation resistance of CMCs are vital issues. The need for environmentally stable composites motivated the development of CMCs based on environmentally stable oxide constituents [6–11]. The main advantage of CMCs over monolithic ceramics is their superior toughness, tolerance to the presence of cracks and defects, and non-catastrophic mode of failure. It is widely accepted that in order to avoid brittle fracture behavior in CMCs and improve the damage tolerance, a weak fiber/matrix interface is needed, which serves to deflect matrix cracks and to allow subsequent fiber pullout [12–14]. It has been demonstrated that similar crack-deflecting behavior can also be achieved by means of a finely distributed porosity in the matrix instead of a separate interface between matrix and fibers [15]. This microstructural design philosophy implicitly accepts the strong fiber/matrix interface. The concept has been successfully demonstrated for oxide–oxide composites [6,9,11, 16,17]. Resulting oxide/oxide CMCs exhibit damage tolerance combined with inherent oxidation resistance. An extensive review of the mechanisms and mechanical properties of porous-matrix CMCs is given in [18,19]. Porous-matrix oxide/oxide CMCs exhibit several behavior trends that are distinctly different from those exhibited by traditional non-oxide CMCs with a fiber-matrix interface. For the nonoxide CMCs, fatigue is significantly more damaging than creep. Contrastingly, Zawada et al. [20] examined the high-temperature mechanical behavior of a porous-matrix Nextel610/Aluminosilicate composite. Results revealed excellent fatigue performance at 1000 C. Conversely, creep lives were short, indicating low creep 1359-835X/$ - see front matter Published by Elsevier Ltd. doi:10.1016/j.compositesa.2008.09.005 q The views expressed are those of the authors and do not reflect the official policy or position of the United States Air Force, Department of Defense or the US Government. * Corresponding author. Tel.: +1 937 255 3636x4641; fax: +1 937 656 7053. E-mail address: marina.ruggles-wrenn@afit.edu (M.B. Ruggles-Wrenn). Composites: Part A 39 (2008) 1829–1837 Contents lists available at ScienceDirect Composites: Part A journal homepage: www.elsevier.com/locate/compositesa
M.B. Ruggles-Wrenn, N.R. Szymczak/Composites: Part A 39(2008)1829-1837 istance and limiting the use of that CMC to temperatures below resistance-heated furnace, and two temperature controllers was 0C. Ruggles-Wrenn et al. [21] showed that Nextel 720/Alu- used in all tests. An MTS TestStar ll digital controller was employed mina(N720 /A)composite exhibits excellent fatigue resistance for input signal generation and data acquisition. Strain measure- laboratory air at 1200C. The fatigue limit (based on a run-out con- ment was accomplished with an mts high-temperature air-cooled ition of 10. cycles) was 170 MPa(88% UTS at 1200C). Further- uniaxial extensometer of 125-mm gage length For elevated tem nore, the composite retained 100% of its tensile strength. perature testing, thermocouples were bonded to the specimens owever, creep loading was found to be considerably more dam- using alumina cement(Zircar)to calibrate the furnace on a peri aging. Creep run- out (defined as 100 h at creep stress) was odic basis. The furnace controllers(using non-contacting thermo- achieved only at stress levels below 50% UTS. Mehrman et al. couples exposed to the ambient environment near the test [22] demonstrated that introduction of a short hold period at the specimen) were adjusted to determine the power setting needed maximum stress into the fatigue cycle significantly degraded the to achieve the desired temperature of the test specimen. The deter- fatigue performance of N720/A composite at 1200C in air. In mined power settings were then used in actual tests. Tests in steam, superposition of a hold time onto a fatigue cycle resulted steam environment employed an alumina susceptor (tube with in an even more dramatic deterioration of fatigue life reducing it end caps), which fits inside the furnace. The specimen gage section to a much shorter creep life at a given applied stress is located inside the susceptor, with the ends of the specimen pass Because was shown to be considerably more damaging ing through slots in the susceptor. Steam is introduced into the than cyclic loading to oxide-oxide CMCs with porous-matrix susceptor(through a feeding tube) in a continuous stream with 20-22. high-temperature creep resistance remains among the slightly positive pressure, expelling the dry air and creating a near that must be addressed before using these materials 100% steam environment inside the susceptor. The power settings in advanced aerospace applications. This study investigates mono- for testing in steam were determined by placing the specimen nic compression and compressive creep behaviors of the Nex instrumented with thermocouples in steam environment and tel 720/ alumina(N720 A), an oxide-oxide CMc with a porous- repeating the furnace calibration procedure. Thermocouples were matrix, at 1200C in air and in steam environment. Results reveal not bonded to the test specimens after the furnace was calibrated that the presence of steam dramatically degrades compressive Fracture surfaces of failed specimens were examined using an opti creep lifetimes. cal microscope(Zeiss Discovery V12). All tests were conducted at 1200C. In all tests, a specimen was 2. Material and experimental arrang heated to the test temperature at a rate of 1C/s, and held at tem- perature for additional 30 min prior to testing. Straight-sided The material studied was NextelTM720/Alumina(N720/A), an 18 mm x 150 mm specimens were used in all compression tests xide-oxide CMC(manufactured by COl Ceramics, San Diego, CA) All N720 A test specimens used in this study were cut from a single consisting of a porous alumina matrix reinforced with NextelTM720 plate Monotonic compression tests were performed in stress con- fibers. There is no fiber coating. The damage tolerance of N720/A is trol with constant stress-rate magnitudes of 0.0025 and 25 MPa/s enabled by the porous-matrix. The composite was supplied in a form of 5.2-mm thick plates comprised of 240 /90 woven layers, tests the modulus of elasticity was calculated in accordance witi vith a density of 2.77 g/cm and a fiber volume of 44%. Com- the procedure in ASTM standard c 1358 as the slope of the com- posite porosity was w22%. The fiber fabric was infiltrated with pressive stress-strain curve within the linear region. In compres the matrix in a sol-gel process. The laminate was dried with a"vac- sive creep-rupture tests specimens were loaded to the creep uum bag"technique under low pressure and low temperature, stress level at the stress-rate magnitude of 25 MPa/s Creep run- then pressureless sintered [23]. Representative micrograph of the out was defined as 100 h at a given creep stress. In each test, untested material is presented in Fig. la, which shows 0o and 90 stress-strain data were recorded during the loading to the creep fiber tows as well as numerous matrix cracks. In the case of the stress level and the actual creep period. Thus both total strain as-processed material, most are shrinkage cracks formed during and creep strain could be calculated and examined To determine processing rather than matrix cracks generated during loading. the retained tensile(compressive)strength and modulus, speci Porous nature of the matrix is seen in Fig. 1b mens that achieved creep run-out were subjected to tensile(com- A servocontrolled MTS mechanical testing machine equipped pressive)tests to failure at 1200C. It is worthy of note that in all with hydraulic water-cooled collet grips, a compact two-zone tests reported below, the failure occurred within the gage section b 200um 0.5um Fig. 1. As-processed material: (a)overview showing shrinkage cracks, (b)porous nature of the matrix is evident
resistance and limiting the use of that CMC to temperatures below 1000 C. Ruggles-Wrenn et al. [21] showed that NextelTM720/Alumina (N720/A) composite exhibits excellent fatigue resistance in laboratory air at 1200 C. The fatigue limit (based on a run-out condition of 105 cycles) was 170 MPa (88% UTS at 1200 C). Furthermore, the composite retained 100% of its tensile strength. However, creep loading was found to be considerably more damaging. Creep run-out (defined as 100 h at creep stress) was achieved only at stress levels below 50% UTS. Mehrman et al. [22] demonstrated that introduction of a short hold period at the maximum stress into the fatigue cycle significantly degraded the fatigue performance of N720/A composite at 1200 C in air. In steam, superposition of a hold time onto a fatigue cycle resulted in an even more dramatic deterioration of fatigue life, reducing it to a much shorter creep life at a given applied stress. Because creep was shown to be considerably more damaging than cyclic loading to oxide–oxide CMCs with porous-matrix [20–22], high-temperature creep resistance remains among the key issues that must be addressed before using these materials in advanced aerospace applications. This study investigates monotonic compression and compressive creep behaviors of the NextelTM720/alumina (N720/A), an oxide–oxide CMC with a porousmatrix, at 1200 C in air and in steam environment. Results reveal that the presence of steam dramatically degrades compressive creep lifetimes. 2. Material and experimental arrangements The material studied was NextelTM720/Alumina (N720/A), an oxide–oxide CMC (manufactured by COI Ceramics, San Diego, CA) consisting of a porous alumina matrix reinforced with NextelTM720 fibers. There is no fiber coating. The damage tolerance of N720/A is enabled by the porous-matrix. The composite was supplied in a form of 5.2-mm thick plates comprised of 240/90 woven layers, with a density of 2.77 g/cm3 and a fiber volume of 44%. Composite porosity was 22%. The fiber fabric was infiltrated with the matrix in a sol-gel process. The laminate was dried with a ‘‘vacuum bag” technique under low pressure and low temperature, then pressureless sintered [23]. Representative micrograph of the untested material is presented in Fig. 1a, which shows 0 and 90 fiber tows as well as numerous matrix cracks. In the case of the as-processed material, most are shrinkage cracks formed during processing rather than matrix cracks generated during loading. Porous nature of the matrix is seen in Fig. 1b. A servocontrolled MTS mechanical testing machine equipped with hydraulic water-cooled collet grips, a compact two-zone resistance-heated furnace, and two temperature controllers was used in all tests. An MTS TestStar II digital controller was employed for input signal generation and data acquisition. Strain measurement was accomplished with an MTS high-temperature air-cooled uniaxial extensometer of 12.5–mm gage length. For elevated temperature testing, thermocouples were bonded to the specimens using alumina cement (Zircar) to calibrate the furnace on a periodic basis. The furnace controllers (using non-contacting thermocouples exposed to the ambient environment near the test specimen) were adjusted to determine the power setting needed to achieve the desired temperature of the test specimen. The determined power settings were then used in actual tests. Tests in steam environment employed an alumina susceptor (tube with end caps), which fits inside the furnace. The specimen gage section is located inside the susceptor, with the ends of the specimen passing through slots in the susceptor. Steam is introduced into the susceptor (through a feeding tube) in a continuous stream with a slightly positive pressure, expelling the dry air and creating a near 100% steam environment inside the susceptor. The power settings for testing in steam were determined by placing the specimen instrumented with thermocouples in steam environment and repeating the furnace calibration procedure. Thermocouples were not bonded to the test specimens after the furnace was calibrated. Fracture surfaces of failed specimens were examined using an optical microscope (Zeiss Discovery V12). All tests were conducted at 1200 C. In all tests, a specimen was heated to the test temperature at a rate of 1 C/s, and held at temperature for additional 30 min prior to testing. Straight-sided 18 mm 150 mm specimens were used in all compression tests. All N720/A test specimens used in this study were cut from a single plate. Monotonic compression tests were performed in stress control with constant stress-rate magnitudes of 0.0025 and 25 MPa/s in laboratory air and in steam environment. In all compression tests the modulus of elasticity was calculated in accordance with the procedure in ASTM standard C 1358 as the slope of the compressive stress–strain curve within the linear region. In compressive creep-rupture tests specimens were loaded to the creep stress level at the stress-rate magnitude of 25 MPa/s. Creep runout was defined as 100 h at a given creep stress. In each test, stress–strain data were recorded during the loading to the creep stress level and the actual creep period. Thus both total strain and creep strain could be calculated and examined. To determine the retained tensile (compressive) strength and modulus, specimens that achieved creep run-out were subjected to tensile (compressive) tests to failure at 1200 C. It is worthy of note that in all tests reported below, the failure occurred within the gage section Fig. 1. As-processed material: (a) overview showing shrinkage cracks, (b) porous nature of the matrix is evident. 1830 M.B. Ruggles-Wrenn, N.R. Szymczak / Composites: Part A 39 (2008) 1829–1837
M B. Ruggles-Wrenn, N.R. Szymczak/Composites: Part A 39(2008)1829-1837 1831 of the extensometer. In some cases one specimen was tested per 00 est condition. The authors recognize that this is a limited set of data. However, extreme care was taken in generating the data. ir. 25 MPa/s Selective duplicate tests have demonstrated the data to be very Steam, 25 MPa/s repeatable. This exploratory effort serves to identify the behaviora trends and to determine whether a more rigorous investigation hould be undertaken Air 0.0025 MPa/s UssIon 合 3. 1. Monotonic compression The n720/A specimens were tested in compression to failure at 1200C in air and in steam To investigate the effect of loading rate on compressive stress-strain behavior and on compressive proper ABS Strain(%) ties, the stress-controlled compression tests were conducted with Fig. 2. Compression stress-strain curves for N720/A composite obtained constant stress rate magnitudes of 0.0025 and 25 MPa/s Modulus, in air and in steam. Dependence of stress-strain behavior and comp strength and failure strain obtained in compression tests at various properties on stress-rate is evident. stress-rate magnitudes are summarized in Table 1. Compressive stress-strain behavior at 1200C is typified in Fig. 2, where stress magnitude vs strain magnitude curves are presented. rate from 25 to 0.0025 MPa/s resulted in a 65% decrease in modu lus and a 53% decrease in strength. At 0.0025 MPa/s, the stress In air, at 25 MPa/s, the average compressive modulus of 74 GPa strain curve departs from linearity at a low stress of-10 MPa. As is similar to the tensile modulus of 70 GPa reported earlier [24]. the compressive stress continues to increase, appreciable inelastic Conversely, the average compressive strength of 128 MPa is con siderably less than the corresponding tensile value of 181 MPa strain develops reaching -2. 12% at failure. [24]. Likewise the average compressive failure strain of-0.20% is 3.2. Creep-rupture significantly lower than the 0.38% failure strain obtained in tensile test. The compressive stress-strain e obtained at 25 MPa/s Results of the compressive creep-rupture tests are summarized consists of two nearly linear parts: the first part extends to a stress in Table 2, where creep strain accumulation and rupture time are of approximately -100 MPa, the second part with a slightly lower slope continues to failure. This stress-strain behavior suggests shown for each test environment and creep stress level. Creep composite damage at the change in slope. It is likely that as the curves obtained at 1200C in air and in steam are shown in Figs impressive stress approaches -100 MPa shear cracking in the 3a and b, respectively. Tensile creep data from prior work [21, 25 is included in Fig. 3 for comparison. At 1200C in air, tensile as well 90 fiber bundles and ply delamination take place. The damage re- as compressive creep curves exhibit primary and secondary creep form more readily by buckling, thus causing a lower slope in the regimes. In botn tension and compression transition Trom primary stress-strain curve. When the ressive stress reaches 130 MPa, the composite fails by buckling and shear fracture of creep continues to failure. Compressive creep curves in Fig 3a indi the oo bundles. As seen in Fig. 2. the compressive stress-strain cate that secondary creep is likely to persist for the duration of the behavior produced at the stress-rate of 0.0025 MPa/s is markedly nonlinear Departure from linearity occurs at fairly low stress of stress increases from 80 to 100 MPa, then decreases with increas- about-40 MPa. The decrease in loading rate by 4 orders of magn ing stress. All tensile creep strains accumulated in air significantly tude causes a 45% decrease in compressive modulus and a 16% de. exceed the failure strain obtained in the tension test. In compres- crease in compressive strength. sion, creep strain magnitude increases with increasing magnitude The presence of steam has a degrading effect on compressive of creep stress Compressive creep strain magnitudes accumulate the compressive strength was 115 MPa. Presence of steam dramat- to the failure strain magnitude obtained in compression test. How- ically amplified the effect of loading rate on compressive pro mately two times the failure strain magnitude obtained in ties and stress-strain behavior. In steam, the change in loading compression test. Note that all compressive creep tests conducted in air achieved a run-out, while the tensile creep run-out was achieved only at a low stress of 80 MPa(44% UTS mmary of compressive properties of the N720/ A ceramic composite at 1200C in laboratory air and steam environments Stress-rate Compressive modulus Compressive strength Failure strain Table 2 Summary of compressive creep-rupture results for the N720/A composite at 1200C in air and in steam environment 74.6 0.18 Environment Creep stress(MPa) 00025 Steam environmen 60000 640 0.52 14 Steam 00025 12 544 2.12
of the extensometer. In some cases one specimen was tested per test condition. The authors recognize that this is a limited set of data. However, extreme care was taken in generating the data. Selective duplicate tests have demonstrated the data to be very repeatable. This exploratory effort serves to identify the behavioral trends and to determine whether a more rigorous investigation should be undertaken. 3. Results and discussion 3.1. Monotonic compression The N720/A specimens were tested in compression to failure at 1200 C in air and in steam. To investigate the effect of loading rate on compressive stress–strain behavior and on compressive properties, the stress-controlled compression tests were conducted with constant stress rate magnitudes of 0.0025 and 25 MPa/s. Modulus, strength and failure strain obtained in compression tests at various stress-rate magnitudes are summarized in Table 1. Compressive stress–strain behavior at 1200 C is typified in Fig. 2, where stress magnitude vs strain magnitude curves are presented. In air, at 25 MPa/s, the average compressive modulus of 74 GPa is similar to the tensile modulus of 70 GPa reported earlier [24]. Conversely, the average compressive strength of 128 MPa is considerably less than the corresponding tensile value of 181 MPa [24]. Likewise the average compressive failure strain of 0.20% is significantly lower than the 0.38% failure strain obtained in tensile test. The compressive stress–strain curve obtained at 25 MPa/s consists of two nearly linear parts: the first part extends to a stress of approximately 100 MPa, the second part with a slightly lower slope continues to failure. This stress–strain behavior suggests composite damage at the change in slope. It is likely that as the compressive stress approaches 100 MPa shear cracking in the 90 fiber bundles and ply delamination take place. The damage relieves constraints acting on the 0 bundles allowing them to deform more readily by buckling, thus causing a lower slope in the stress–strain curve. When the compressive stress reaches 130 MPa, the composite fails by buckling and shear fracture of the 0 bundles. As seen in Fig. 2, the compressive stress–strain behavior produced at the stress-rate of 0.0025 MPa/s is markedly nonlinear. Departure from linearity occurs at fairly low stress of about 40 MPa. The decrease in loading rate by 4 orders of magnitude causes a 45% decrease in compressive modulus and a 16% decrease in compressive strength. The presence of steam has a degrading effect on compressive properties. In steam at 25 MPa/s the modulus was 64 GPa and the compressive strength was 115 MPa. Presence of steam dramatically amplified the effect of loading rate on compressive properties and stress–strain behavior. In steam, the change in loading rate from 25 to 0.0025 MPa/s resulted in a 65% decrease in modulus and a 53% decrease in strength. At 0.0025 MPa/s, the stress– strain curve departs from linearity at a low stress of 10 MPa. As the compressive stress continues to increase, appreciable inelastic strain develops reaching 2.12% at failure. 3.2. Creep-rupture Results of the compressive creep-rupture tests are summarized in Table 2, where creep strain accumulation and rupture time are shown for each test environment and creep stress level. Creep curves obtained at 1200 C in air and in steam are shown in Figs. 3a and b, respectively. Tensile creep data from prior work [21,25] is included in Fig. 3 for comparison. At 1200 C in air, tensile as well as compressive creep curves exhibit primary and secondary creep regimes. In both tension and compression, transition from primary to secondary creep occurs early in creep life. In tension, secondary creep continues to failure. Compressive creep curves in Fig. 3a indicate that secondary creep is likely to persist for the duration of the creep lifetime. In tension, creep strain increases as the applied stress increases from 80 to 100 MPa, then decreases with increasing stress. All tensile creep strains accumulated in air significantly exceed the failure strain obtained in the tension test. In compression, creep strain magnitude increases with increasing magnitude of creep stress. Compressive creep strain magnitudes accumulated at 60 and 80 MPa are, respectively, lower than and comparable to the failure strain magnitude obtained in compression test. However, creep strain magnitude accumulated at 100 MPa is approximately two times the failure strain magnitude obtained in compression test. Note that all compressive creep tests conducted in air achieved a run-out, while the tensile creep run-out was achieved only at a low stress of 80 MPa (44% UTS). Table 1 Summary of compressive properties of the N720/A ceramic composite at 1200 C in laboratory air and steam environments Stress-rate (MPa/s) Compressive modulus (GPa) Compressive strength (MPa) Failure strain (%) Laboratory air 25 74.6 130 0.19 25 73.8 132 0.18 25 73.8 122 0.19 0.0025 40.7 108 0.44 Steam environment 25 64.0 115 0.52 25 60.6 114 0.42 0.0025 14.8 55.6 1.32 0.0025 12.2 54.4 2.12 0 50 100 150 200 0.0 0.5 1.0 1.5 2.0 2.5 ABS Stress (MPa) ABS Strain (%) Steam, 0.0025 MPa/s Air, 25 MPa/s T = 1200 ºC Steam, 25 MPa/s Air, 0.0025 MPa/s Fig. 2. Compression stress–strain curves for N720/A composite obtained at 1200 C in air and in steam. Dependence of stress–strain behavior and compressive properties on stress-rate is evident. Table 2 Summary of compressive creep-rupture results for the N720/A composite at 1200 C in air and in steam environment Environment Creep stress (MPa) Creep strain (%) Time to rupture (s) Air 60 0.09 360,000a Air 80 0.15 360,000a Air 100 0.40 360,000a Steam 40 1.56 13,899 Steam 60 1.13 2,366 Steam 100 0.19 7 a Run-out. M.B. Ruggles-Wrenn, N.R. Szymczak / Composites: Part A 39 (2008) 1829–1837 1831
18 M.B. Ruggles-Wrenn, N.R. Szymczak/ Composites: Part A 39(2008)1829-1837 2.5 the high content of mullite, which has a much better creep resis- T=1200°c, tance than alumina [26 Conversely, in compression the creep- 2.0 100 MPa-Comilresswere rupture of the composite is largely dominated by an exceptionally eep weak porous alumina matrix. Both tensile and compressive creep 125 MPa are accelerated in the presence of steam. However, the degrading effect of steam environment is more pronounced in compression than in tension. While the tensile creep rate in steam is approxi mately an order of magnitude higher than that in air, the compres- 6 sive creep rate in steam can be as high as 10- times that obtained in for Stress-rupture behavior is summarized in Fig. 5, where creep stress magnitude is plotted vs time to rupture at 1200C in air and in steam. The tensile creep results from prior work [21, 2 100000 200000 300000 400000 are included for comparison. As expected, tensile creep life de Time(s) creases with increasing applied stress. However, in air compressive creep life(up to 100 h)appears to be relatively independent of ap- plied stress. All compressive creep tests conducted in air achieved a Tensile cree T=1200°c, Steam run-out. The presence of steam dramatically reduced creep life- Compressive creep times in both tension and compression In tension, the reduction in creep life due to steam was at least 90% for applied stress levels 60 MPa 40 MPa over 100 MPa, and -54% for the applied stress of 80 MPa. In com 125 MPa pression, creep lifetimes can be reduced by as much as 99.9% in the resence Retained compressive strength and modulus of the specimens that achieved a run- out in the -60 and -100 MPa creep tests con- ducted at 1200C in air are given in Table 3. Compressive stress- 80 MPa strain curves obtained for the N720/A specimens subjected to prior compressive creep are presented in Fig. 6 together with the com- pressive stress-strain curve for the as-processed material. Both 5000 10000 15000 specimens retained 100% of their compressive strength. However, prior compressive creep appears to have decreased compressive modulus. To evaluate the effects of compressive creep on tensile ir and(b)in steam. strength and stiffness, a specimen that achieved a run-out in Tensile creep data from Ruggles-Wrenn et al. [21, 25 are also shown. 80 MPa creep test was subjected to a tensile test to failure at 1200C Retained tensile strength and modulus are included in Ta- ble 3. Prior compressive creep caused a 30% decrease in tensile The compressive(tensile)creep curves produced in steam are strength and a 17% decrease in modulus. Tensile stress-strain qualitatively similar to the compressive(tensile)creep curves ob- behavior of the specimen subjected to prior compressive creep re- effect on creep strain and creep lifetime in both tension and com- Fig. 7). Note that the n720/ A composite subjected to 100 h of prior pression Tensile creep strain produced at 80 MPa in steam was x5 tensile creep at 80 MPa in air retained over 90% of its tensile times that obtained at 80 MPa in air. Yet at 100 and 125 MPa, the strength and over 86% of its tensile modulus [46]. Prior tensile tensile creep strains produced in steam were, respectively, 7% creep had no qualitative effect on tensile stress-strain behavior. and 30% lower than those produced in air. For compressive creep stresses of-40 and -60 MPa, creep strain magnitudes produced in steam were an order of magnitude higher than those in air. At 100 MPa, creep strain magnitude accumulated in steam was low- er than that in air. Note that in steam, creep strain magnitudes as 10E-01 well as creep lifetimes decrease with increasing creep stress mag- nitudes for both tension and compression. while the n720/A com- 10E-02 posite survived 100 h of tensile creep at 80 MPa in air, tensile creep 10E-03 run-out was not achieved in steam. Although all compressive creep tests conducted at 1200 C in air achieved a run-out in steam com 10E-04 pressive creep run-out was not achieved Minimum creep rate was reached in all tests. Creep strain rate 1.0E-05 magnitude as a function of applied stress magnitude is shown in Fig. 4, where results of previous work [21, 25 are also included. Re- 量1 sults in Fig. 4 show that at 1200C in air the compressive creep 1.0E-07 rate magnitudes are nearly an order of magnitude higher than 1.0E-08 the tensile creep rates produced at the same applied stress magni- T=1200°c tude. This result is hardly surprising, considering that in tension 10E-09 the creep-rupture of the N720/A CMc is likely dominated by 1000 creep-rupture of the Nextel 720 fibers. It is recognized that Nex tel M720 fiber has the best creep performance of any commercially available polycrystalline oxide fiber. The superior high-tempera- g. 4. Minimum creep rate magnitude as a function of applied at 1200"C in air and in steam Tensile creep data fro ture creep performance of the Nextel 720 fibers results from Ruggles-Wrenn et al [21, 25] are also showr
The compressive (tensile) creep curves produced in steam are qualitatively similar to the compressive (tensile) creep curves obtained in air. Nevertheless, the presence of steam has a noticeable effect on creep strain and creep lifetime in both tension and compression. Tensile creep strain produced at 80 MPa in steam was 5 times that obtained at 80 MPa in air. Yet at 100 and 125 MPa, the tensile creep strains produced in steam were, respectively, 7% and 30% lower than those produced in air. For compressive creep stresses of 40 and 60 MPa, creep strain magnitudes produced in steam were an order of magnitude higher than those in air. At 100 MPa, creep strain magnitude accumulated in steam was lower than that in air. Note that in steam, creep strain magnitudes as well as creep lifetimes decrease with increasing creep stress magnitudes for both tension and compression. While the N720/A composite survived 100 h of tensile creep at 80 MPa in air, tensile creep run-out was not achieved in steam. Although all compressive creep tests conducted at 1200 C in air achieved a run-out, in steam compressive creep run-out was not achieved. Minimum creep rate was reached in all tests. Creep strain rate magnitude as a function of applied stress magnitude is shown in Fig. 4, where results of previous work [21,25] are also included. Results in Fig. 4 show that at 1200 C in air the compressive creep rate magnitudes are nearly an order of magnitude higher than the tensile creep rates produced at the same applied stress magnitude. This result is hardly surprising, considering that in tension the creep-rupture of the N720/A CMC is likely dominated by creep-rupture of the Nextel 720 fibers. It is recognized that NextelTM720 fiber has the best creep performance of any commercially available polycrystalline oxide fiber. The superior high-temperature creep performance of the NextelTM 720 fibers results from the high content of mullite, which has a much better creep resistance than alumina [26]. Conversely, in compression the creeprupture of the composite is largely dominated by an exceptionally weak porous alumina matrix. Both tensile and compressive creep are accelerated in the presence of steam. However, the degrading effect of steam environment is more pronounced in compression than in tension. While the tensile creep rate in steam is approximately an order of magnitude higher than that in air, the compressive creep rate in steam can be as high as 105 times that obtained in air for a given stress. Stress-rupture behavior is summarized in Fig. 5, where creep stress magnitude is plotted vs time to rupture at 1200 C in air and in steam. The tensile creep results from prior work [21,25] are included for comparison. As expected, tensile creep life decreases with increasing applied stress. However, in air compressive creep life (up to 100 h) appears to be relatively independent of applied stress. All compressive creep tests conducted in air achieved a run-out. The presence of steam dramatically reduced creep lifetimes in both tension and compression. In tension, the reduction in creep life due to steam was at least 90% for applied stress levels over 100 MPa, and 54% for the applied stress of 80 MPa. In compression, creep lifetimes can be reduced by as much as 99.9% in the presence of steam. Retained compressive strength and modulus of the specimens that achieved a run-out in the 60 and 100 MPa creep tests conducted at 1200 C in air are given in Table 3. Compressive stress– strain curves obtained for the N720/A specimens subjected to prior compressive creep are presented in Fig. 6 together with the compressive stress–strain curve for the as-processed material. Both specimens retained 100% of their compressive strength. However, prior compressive creep appears to have decreased compressive modulus. To evaluate the effects of compressive creep on tensile strength and stiffness, a specimen that achieved a run-out in a 80 MPa creep test was subjected to a tensile test to failure at 1200 C. Retained tensile strength and modulus are included in Table 3. Prior compressive creep caused a 30% decrease in tensile strength and a 17% decrease in modulus. Tensile stress–strain behavior of the specimen subjected to prior compressive creep remained qualitative similar to that of the as-processed material (see Fig. 7). Note that the N720/A composite subjected to 100 h of prior tensile creep at 80 MPa in air retained over 90% of its tensile strength and over 86% of its tensile modulus [46]. Prior tensile creep had no qualitative effect on tensile stress–strain behavior. Fig. 3. Creep curves for N720/A composite at 1200 C: (a) in air and (b) in steam. Tensile creep data from Ruggles-Wrenn et al. [21,25] are also shown. 1.0E-09 1.0E-08 1.0E-07 1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 10 100 1000 ABS Creep Stress (MPa) ABS Creep Strain Rate (s-1) Tension, air Tension , steam Compression, air Compression, steam T = 1200°C Fig. 4. Minimum creep rate magnitude as a function of applied stress magnitude for N720/A ceramic composite at 1200 C in air and in steam. Tensile creep data from Ruggles-Wrenn et al. [21,25] are also shown. 1832 M.B. Ruggles-Wrenn, N.R. Szymczak / Composites: Part A 39 (2008) 1829–1837
M B. Ruggles-Wrenn, N.R. Szymczak/Composites: Part A 39(2008)1829-1837 uTs△ Tension,ar 口 Compression,a As-processed -Creep Predicted from Constant Stress Rate Data 150 a兰苏 UCS. Steam △口 100 h at -80 MPa △ T=1200°c T=1200°c 1E+001.E+011.E+021.E+031.E+041.E+051E+061.E+07 0.3 0.5 Time(s Strain (% Fig. 5. Creep stress magnitude vs time to rupture for N720/A ceramic composite at of constant stress-rate tests. Tensile creep data from Ruggles-Wrenn et al /21.25]. are inclu In the case of monolithic ceramics, the degradation of tensile Table 3 strength with decreasing stress-rate has been shown to proceed Retained tensile and compressive properties of the N720/A specimens subjected to by environmentally assisted subcritical crack growth [32-34] prior compressive creep at 1200C in laboratory air The subcritical(slow) crack growth rate can be described by the Creep stress Retained strength Retained modulus Failure strain empirical power-law [35-37 etained tensile prope 62.8 d -AIR Retained compressive properties where a is the crack size, t is time, KI is the mode I stress intensity 0.25 464 0.22 factor, Kic is the critical stress intensity factor (or fracture tough- ) under mode I loading, and a and n are the slow crack growth parameters. Under constant stress-rate loading, the fracture strength a can be derived as a function of applied stress-rate o Here d is a crack growth parameter associated with inert strength o, n and crack geometry, and given by 80 100 h at-100 MPa D=20n+1q2h AY(n-2) By taking logarithms of both sides Eg. (2)can be expressed in the following form: logar= n+1080+logD (4) Based on Eqs. (2)and(4), constant stress-rate testing has been established as a standard test method for determination of slo Fig. 6. Effects of prior compressive creep at 1200"C on compressive stress-strain behavior of N720/ A ceramic composite crack growth(SCG)parameters d and n for advanced monolithic ceramics at ambient(ASTM C1368 [ 36]) and elevated tempera tures(ASTM C1465 [37). Following procedure in 36, 37. experi 3.3. Prediction of creep lifetime from constant stress-rate test data mental data obtained for N720 A composite at 1200C in steam were plotted as log(UTS)vs log(applied stress-rate)[38 The The stress-rate dependence of the compressive strength exhib- parameters n and d were determined by a linear regression anal ited by the n720/ A composite at 1200C in steam is similar to the ysis from the slope and intercept, respectively, as n= 12 and rate dependence of ultimate tensile strength observed for several D=136. 42 ceramic matrix composites at elevated temperatures by Choi et The crack growth exponent n represe al. 27, 28 Recently, Choi and Bansal [29 reported that at bility of the material to subcritical crack growth. Typically for brit 1100C the shear strength of a 2-D Hi-Nicalon-fiber-reinforced tle materials, the susceptibility is considered high for n< 20, CMC also decreased significantly with decreasing test rate. Fur- intermediate for 30< n< 50, and low for n> 50. At 1200C in thermore, the dependence of shear strength on loading rate was steam the n720/A composite exhibits a significant susceptibility imilar to the rate dependency of tensile strength exhibited not to subcritical crack growth with n= 12 38 Similar results show only by the CMCs [25, 26, 30] but also by the advanced monolithic ing considerable susceptibility to delayed failure at temperatures ceramics such as silicon nitrides, silicon carbides and aluminas >1000C were reported for N720 fibers [39, 40, with the values [31l of n ranging from 9 to 18. Likewise Choi et al. [27, 28] found that
3.3. Prediction of creep lifetime from constant stress-rate test data The stress-rate dependence of the compressive strength exhibited by the N720/A composite at 1200 C in steam is similar to the rate dependence of ultimate tensile strength observed for several ceramic matrix composites at elevated temperatures by Choi et al. [27,28]. Recently, Choi and Bansal [29] reported that at 1100 C the shear strength of a 2-D Hi-Nicalon-fiber-reinforced CMC also decreased significantly with decreasing test rate. Furthermore, the dependence of shear strength on loading rate was similar to the rate dependency of tensile strength exhibited not only by the CMCs [25,26,30] but also by the advanced monolithic ceramics such as silicon nitrides, silicon carbides and aluminas [31]. In the case of monolithic ceramics, the degradation of tensile strength with decreasing stress-rate has been shown to proceed by environmentally assisted subcritical crack growth [32–34]. The subcritical (slow) crack growth rate can be described by the empirical power-law [35–37]: da dt ¼ A KI KIC n ð1Þ where a is the crack size, t is time, KI is the mode I stress intensity factor, KIC is the critical stress intensity factor (or fracture toughness) under mode I loading, and A and n are the slow crack growth parameters. Under constant stress-rate loading, the fracture strength rf can be derived as a function of applied stress-rate r as follows [35–37]: rf ¼ Dðr Þ 1 nþ1 ð2Þ Here D is a crack growth parameter associated with inert strength ri, n and crack geometry, and given by: D ¼ 2ðn þ 1ÞK2 ICrn2 i AY2 ðn 2Þ " # 1 nþ1 ð3Þ By taking logarithms of both sides Eq. (2) can be expressed in the following form: logrf ¼ 1 n þ 1 logr þ logD ð4Þ Based on Eqs. (2) and (4), constant stress-rate testing has been established as a standard test method for determination of slow crack growth (SCG) parameters D and n for advanced monolithic ceramics at ambient (ASTM C1368 [36]) and elevated temperatures (ASTM C1465 [37]). Following procedure in [36,37], experimental data obtained for N720/A composite at 1200 C in steam were plotted as log (UTS) vs log (applied stress-rate) [38]. The parameters n and D were determined by a linear regression analysis from the slope and intercept, respectively, as n = 12 and D = 136.42. The crack growth exponent n represents a measure of susceptibility of the material to subcritical crack growth. Typically for brittle materials, the susceptibility is considered high for n 6 20, intermediate for 30 6 n 6 50, and low for n > 50. At 1200 C in steam the N720/A composite exhibits a significant susceptibility to subcritical crack growth with n = 12 [38]. Similar results showing considerable susceptibility to delayed failure at temperatures P1000 C were reported for N720 fibers [39,40], with the values of n ranging from 9 to 18. Likewise Choi et al. [27,28] found that 0 50 100 150 200 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 Time (s) Stress Magnitude (MPa) Tension, air Compression, air Tension, steam Compression, steam Creep Predicted T = 1200 °C UTS UCS, Steam UCS, Air from Constant Stress Rate Data Fig. 5. Creep stress magnitude vs time to rupture for N720/A ceramic composite at 1200 C. The solid line represents prediction made based on Eq. (6) from the results of constant stress-rate tests. Tensile creep data from Ruggles-Wrenn et al. [21,25]. Table 3 Retained tensile and compressive properties of the N720/A specimens subjected to prior compressive creep at 1200 C in laboratory air Creep stress (MPa) Retained strength (MPa) Retained modulus (GPa) Failure strain (%) Retained tensile properties 80 133 62.8 0.29 Retained compressive properties 60 137 56.1 0.25 100 124 46.4 0.22 0 20 40 60 80 100 120 140 160 0.0 0.1 0.2 0.3 0.4 ABS Strain (%) ABS Stress (MPa) T = 1200 °C As-processed 100 h at -60 MPa 100 h at -100 MPa Fig. 6. Effects of prior compressive creep at 1200 C on compressive stress–strain behavior of N720/A ceramic composite. 0 50 100 150 200 0.0 0.1 0.2 0.3 0.4 0.5 Strain (%) Stress (MPa) As-processed T = 1200 °C 100 h at -80 MPa Fig. 7. Effects of prior compressive creep at 1200 C on tensile stress–strain behavior of N720/A ceramic composite. Results from [21] for the as-processed CMC are included for comparison. M.B. Ruggles-Wrenn, N.R. Szymczak / Composites: Part A 39 (2008) 1829–1837 1833