ournal 1. Am Ceram Soc, s1 [4] 1004-12 (1998) Crack Deflection and Propagation in Layered silicon Nitride/Boron Nitride ceramics Desiderio Kovar, .t.tM. D. Thouless. *s and John W. Halloran",t Materials Science and Engineering Department and Mechanical Engineering and Applied Mechanics Departme University of Michigan, Ann Arbor, Michigan 48109- Crack deflection and the subsequent growth of delamina- tion that occur at weak interfaces or within weak interphases tion cracks can be a potent source of energy dissipation between the strong layers. Although several models exist that during the fracture of layered ceramics. In this study, mu predict the conditions under which crack deflection should oc- tilayered ceramics that consist of silicon nitride(Si3N4)lay- cur, -o there is not much experimental data on all-ceramic ers separated by boron nitride/silicon nitride(BN/Si3N) port these models. Even more importantly, it has interphases have been manufactured and tested. Flexural been shown that, in some layered materials, delamination tests reveal that the crack path is dependent on the cor cracks kink out of the interface after propagating on the inter- position of the interphase between the Si,N, layers. Experi- face only a short distance. 10, II The result of such crack kinking mental measurements of interfacial fracture resistance and is that not much energy is absorbed during the fracture of these that both crease as the Si3N, content in the interphase increases. crack deflection and propagation along interfaces is needed to However contrary to existing theories, high energy maximize the energy dissipation capabilities of layered ceram- erbil ption capacity has not been realized in materials that exhibit crack deflection but also have moderately high in- In this the mechanical rties of Si3 N/BN multi- terfacial fracture resistance. Significant energy absorption layered ceramics are investigated. The properties of the inter- has been measured only in materials with very low inter ase are adjusted by ing the composition of the bn in- facial fracture resistance values. A method of predicting the terphase between the SiN layers. The strength and energy critical value of the interfacial fracture resistance necessary absorption of multilayered ceramics are measured, and the to ensure a high energy-absorption capacity is presented. crack path is characterized as a function of the composition of IL. Fabrication of Specimens T HAS previously been shown that it is possible to fabricate layered ceramics that have high strength in combination with with 2 wt% alumina(Al2O3)(HC-HP DBM, Reynolds Metals flexure -Because they can be manufactured from commer- Co., Bauxite, AZ) and 6 wt% yttria (Y20, )(99.g%, Johnson in ethanol. The slurry was dried and then compounded using ar these materials can potentially be significantly lower than instrumented high-shear-rate mixer(Model PL-2000, C.W those for fiber-reinforced composites. Thus, layered ceramics Brabender, South Hackensack, N)with a thermoplastic co- can provide a low-cost alternative to fiber-reinforced compos- polymer binder that was composed of equal parts ethylene- ites when strength and energy absorption capabilities are lim- vinyl acetate (Elvax 470, E I. DuPont De Nemours and Co iting factors in the component design. Silicon nitride(Si,N4) layered ceramics with weak boron Union Carbide Chemicals and Plastics Co, Cleveland, OH) in a novel layered structure known as a fibrous monolithic nation oro 32 nitride(BN)interphases have been previously manufactured in of the polymer/ceramic blend was controlled a conventional two-dimensional layered structure, as well as through addition of a lubricant that consisted of a combi- heral oil (white mineral oil-heavy, Mallinckrodt eramic b Impressive properties were achieved for both struc- Chemicals, Paris, KY) and methoxy-polyethyleI tures, with strengths of>600 MPa and work-of-fracture(WOF) (MPEG 550, Union Carbide, Danbury, CT). The tot values of -8000 J/m2. 7 These properties, as well as high- solids content in the compounds was varied from 37% temperature strength and oxidation resistance, make this sys which allowed the viscosity of the compounds to be tem attractive for commercial applications a factor of 2 monolithic ceramics result from crack deflection and propaga- compounds were chopped into blocks of material -1 mmono Many of the advantages that ceramic laminates have over To mold the materials into sheets, the polymer-ceram on each side and pressed between heated metal platens coated with aluminum foil and a lubricant( Carbowax 400, Union Carbide) at a temperature of 150 C under a pressure of 2.8 MPa. The resulting sheets could be varied in thickness from 100 um to 800 um, depending on the viscosity of the com- unds and the pressure at which the sheet was pressed. For the rrent study, the thickness of th sheets was fixed at po aa k ca th nder that had dimensions of ngi mm sxe6s mnre cut into rectangles amic Society To introduce weak interfaces between the Si,Na layers, the surface of each sheet was coated with a slurry that contained BN. The composition of the BN layers was varied through the nical Engineering and Applied Mechanics Department. addition of Si Na to the BN slurry. The slurries were made 1004
Crack Deflection and Propagation in Layered Silicon Nitride/Boron Nitride Ceramics Desiderio Kovar,*,†,‡ M. D. Thouless,*,§ and John W. Halloran*,† Materials Science and Engineering Department and Mechanical Engineering and Applied Mechanics Department, University of Michigan, Ann Arbor, Michigan 48109–2125 Crack deflection and the subsequent growth of delamination cracks can be a potent source of energy dissipation during the fracture of layered ceramics. In this study, multilayered ceramics that consist of silicon nitride (Si3N4) layers separated by boron nitride/silicon nitride (BN/Si3N4) interphases have been manufactured and tested. Flexural tests reveal that the crack path is dependent on the composition of the interphase between the Si3N4 layers. Experimental measurements of interfacial fracture resistance and frictional sliding resistance show that both quantities increase as the Si3N4 content in the interphase increases. However, contrary to existing theories, high energyabsorption capacity has not been realized in materials that exhibit crack deflection but also have moderately high interfacial fracture resistance. Significant energy absorption has been measured only in materials with very low interfacial fracture resistance values. A method of predicting the critical value of the interfacial fracture resistance necessary to ensure a high energy-absorption capacity is presented. I. Introduction I T HAS previously been shown that it is possible to fabricate layered ceramics that have high strength in combination with the ability to absorb large amounts of energy when tested in flexure.1–4 Because they can be manufactured from commercially available ceramic powders via conventional ceramic- and polymer-processing technology, the manufacturing costs for these materials5 can potentially be significantly lower than those for fiber-reinforced composites. Thus, layered ceramics can provide a low-cost alternative to fiber-reinforced composites when strength and energy absorption capabilities are limiting factors in the component design. Silicon nitride (Si3N4) layered ceramics with weak boron nitride (BN) interphases have been previously manufactured in a conventional two-dimensional layered structure,4 as well as in a novel layered structure known as a fibrous monolithic ceramic.6 Impressive properties were achieved for both structures, with strengths of >600 MPa and work-of-fracture (WOF) values of ∼8000 J/m2 . 7 These properties, as well as hightemperature strength and oxidation resistance, make this system attractive for commercial applications. Many of the advantages that ceramic laminates have over monolithic ceramics result from crack deflection and propagation that occur at weak interfaces or within weak interphases between the strong layers. Although several models exist that predict the conditions under which crack deflection should occur,8–10 there is not much experimental data on all-ceramic systems to support these models. Even more importantly, it has been shown that, in some layered materials, delamination cracks kink out of the interface after propagating on the interface only a short distance.10,11 The result of such crack kinking is that not much energy is absorbed during the fracture of these materials. Thus, an understanding of the factors that control crack deflection and propagation along interfaces is needed to maximize the energy dissipation capabilities of layered ceramics. In this paper, the mechanical properties of Si3N4/BN multilayered ceramics are investigated. The properties of the interphase are adjusted by varying the composition of the BN interphase between the Si3N4 layers. The strength and energy absorption of multilayered ceramics are measured, and the crack path is characterized as a function of the composition of the interphase. II. Fabrication of Specimens Si3N4 powder (M-11, H. C. Starck, Newton, MA) was mixed with 2 wt% alumina (Al2O3) (HC-HP DBM, Reynolds Metals Co., Bauxite, AZ) and 6 wt% yttria (Y2O3) (99.9%, Johnson Matthey Electronics, Ward Hill, MA) and ball milled for 24 h in ethanol. The slurry was dried and then compounded using an instrumented high-shear-rate mixer (Model PL-2000, C. W. Brabender, South Hackensack, NJ) with a thermoplastic copolymer binder that was composed of equal parts ethylene– vinyl acetate (Elvax 470, E. I. DuPont De Nemours and Co. Wilmington, DE) and ethylene–ethyl acrylate (DPDA-6182, Union Carbide Chemicals and Plastics Co., Cleveland, OH). The viscosity of the polymer/ceramic blend was controlled through the addition of a lubricant that consisted of a combination of mineral oil (white mineral oil–heavy, Mallinckrodt Chemicals, Paris, KY) and methoxy-polyethylene glycol (MPEG 550, Union Carbide, Danbury, CT). The total ceramic solids content in the compounds was varied from 37% to 51%, which allowed the viscosity of the compounds to be varied by a factor of 2. To mold the materials into sheets, the polymer–ceramic compounds were chopped into blocks of material ∼1 mm long on each side and pressed between heated metal platens coated with aluminum foil and a lubricant (Carbowax 400, Union Carbide) at a temperature of 150°C under a pressure of 2.8 MPa. The resulting sheets could be varied in thickness from ∼100 mm to 800 mm, depending on the viscosity of the compounds and the pressure at which the sheet was pressed. For the current study, the thickness of the green sheets was fixed at ∼200 mm. After molding, the sheets were cut into rectangles that had dimensions of 51 mm × 76 mm. To introduce weak interfaces between the Si3N4 layers, the surface of each sheet was coated with a slurry that contained BN. The composition of the BN layers was varied through the addition of Si3N4 to the BN slurry. The slurries were made F. W. Zok—contributing editor Manuscript No. 191231. Received February 4, 1997; approved July 21, 1997. Supported by DARPA, administered by the U.S. Office of Naval Research under Contract No. N0014-95-0302. *Member, American Ceramic Society. † Materials Science and Engineering Department. ‡ Current address: Mechanical Engineering Department, The University of Texas at Austin, Austin, TX 78712–1063. § Mechanical Engineering and Applied Mechanics Department. J. Am. Ceram. Soc., 81 [4] 1004–12 (1998) Journal 1004
April 1998 Crack Deflection and Propagation in Layered silicon Nitride/Boron Nitride Ceramics Bulk si n I00 1.2 I N, in Interphase (% Crosshead Displacement(mm Fig. 1. You (Eof the ic. measured the impulse ensile stress(o), plotted versus crosshead displace. ine is the rule- of- odulus. The value ment, for specimens containing 10, 25, 50, and 80 vol% Si,N4 in the en taken from Kovar et al. interphase four-point bending tent in the interphase in Fig. 1. The value of E seems to increase from hexagonal BN(HCP, Advanced Ceramics Corp, Cleve- land, OH), Si3N4, water, and ethanol. Individual billets were linearly as the Si3N4 content in the interphase increases, and E follows the voigt rule of mixtures 13 The e value measured manufactured using interphases made from 0, 10, 20, 50, and from the load-deflection plots followed a similar trend, and 80 vol% SiaNa(the remainder was BN). After coating, the moduli measured using both techniques agreed within 6% sheets were dried, stacked, and pressed at a temperature of 130C under a pressure of 6.9 kPa to mold them into a solid (2) Strength and Energy Absorption billet Four-point flexural tests were performed using a screw- After forming the billet, the polymer binder was pyrolyzed driven machine operated in displacement control( Model 4483 by heating it slowly in a flowing nitrogen atmosphere. The Instron, Danvers, MA). All tests were performed using a fully heating rates were60°C/htol50°C,2°C/hto250°C,4°C/hto articulating testing jig with free-rolling pins using an outer span 370°C,andl8°C/hto700°C. A slow heating rate was neces of 40 mm and an inner span of 20 mm. Data were collected ary to minimize bloating and cracking during pyrolysis, which using a computerized data-acquisition system at a rate of 5 can result in distortion of the layers. After pyrolysis, the billets points per second. Strength and WOF were measured on un- were placed in a BN-coated graphite die and hot pressed at notched specimens at a crosshead displacement rate of 0.5 1750 C for 2 h under a pressure of 25 MPa. Specimens for mm/min. Prior to testing, the specimens were polished to a 3 flexural tests were cut and ground from the billets to um finish using resin-bonded diamond wheels (TBw, Furlong, 4mm×50mm PA)on the tensile surface and on one side surface. The edges of the specimen on the tensile surface were also chamfered IlL Results Tests were interrupted when the specimen fractured com pletely, the retained load dropped below 5 N, or the crosshead After hot pressing, the thicknesses of the layers were mea- displacement exceeded I mr hichever came first. The sured on a polished surface of representative specimens using strength of the specimens was calculated using standard elastic- optical microscopy. The layer thicknesses were 116 34 um beam equations, whereas the WOF value was calculated by and 36+ 18 um for the SiaN4 layers and the BN-containing dividing the total area under the load-deflection curve by twice he cross-sectional area of the specimen. For specimens that cated that all of the Si3Na transformed to B-SisNa during hot fractured catastrophically, the WOF value was reported as zero pressing Hexagonal BN and a very small amount of tetragonal The nominal stressft on the tensile surface for representative irconia(t-ZrO2)were also detected The 0, contamination specimens is plotted versus crosshead deflection for unnotched resulted from the media used during the ball milling of the specimens in Fig. 2. In general, the load remains linear up to the peak load for all the materials. After the peak load, some of ( Young's Modulus the specimens continue to retain load at specimen deflections as large as I mm. the greatest degree of load retention is The Youngs modulus of the specimens was measured using observed in the materials with the lowest SiN, content in the n impulse-excitation technique(Grindo-Sonic MK4x,JW interphase; no load retention is observed following the peak Lemmens, St. Louis, MO), according to ASTM Method C load when the SiaN4 content in the interphase exceeds 25% 1259-94. To verify that these results were valid for layered The nominal strength and wOF are plotted in Fig. 3 as a ceramics, the stiffness of selected specimens were also mea- function of the Sia Na content in the interphase. Although there sured from the slope of load-deflection curves taken in the is scatter in the nominal strengths, there does not seem to be a elastic regime in four-point bending Specimen deflection at systematic change in strength with increasing Si3, content in the center of the span was monitored using a linearly variable the interphase. However, the WoF value decreases precipi displacement transducer(LVDT) and corrected for the compli tously as the SiaN4 content in the interphase increases. The ance of the machine, which had been determined previously slight decrease in strength and WOF for the specimens that The Youngs modulus (E), determined using the pulsed excitation technique, is plotted as a function of the Si3 N4 con- nd that the stress state is American Society for Standards and Testing, Philadelphia, PA when cracking occur
from hexagonal BN (HCP, Advanced Ceramics Corp., Cleveland, OH), Si3N4, water, and ethanol. Individual billets were manufactured using interphases made from 0, 10, 20, 50, and 80 vol% Si3N4 (the remainder was BN). After coating, the sheets were dried, stacked, and pressed at a temperature of 130°C under a pressure of 6.9 kPa to mold them into a solid billet. After forming the billet, the polymer binder was pyrolyzed by heating it slowly in a flowing nitrogen atmosphere. The heating rates were 60°C/h to 150°C, 2°C/h to 250°C, 4°C/h to 370°C, and 18°C/h to 700°C. A slow heating rate was necessary to minimize bloating and cracking during pyrolysis, which can result in distortion of the layers. After pyrolysis, the billets were placed in a BN-coated graphite die and hot pressed at 1750°C for 2 h under a pressure of 25 MPa. Specimens for flexural tests were cut and ground from the billets to nominal dimensions of 3 mm × 4 mm × 50 mm. III. Results After hot pressing, the thicknesses of the layers were measured on a polished surface of representative specimens using optical microscopy. The layer thicknesses were 116 ± 34 mm and 36 ± 18 mm for the Si3N4 layers and the BN-containing interphases, respectively. X-ray diffractometry (XRD) indicated that all of the Si3N4 transformed to b-Si3N4 during hot pressing. Hexagonal BN and a very small amount of tetragonal zirconia (t-ZrO2) were also detected. The ZrO2 contamination resulted from the media used during the ball milling of the powders. (1) Young’s Modulus The Young’s modulus of the specimens was measured using an impulse-excitation technique (Grindo-Sonic MK4x, J. W. Lemmens, St. Louis, MO), according to ASTM Method C 1259-94.¶ To verify that these results were valid for layered ceramics, the stiffness of selected specimens were also measured from the slope of load–deflection curves taken in the elastic regime in four-point bending. Specimen deflection at the center of the span was monitored using a linearly variable displacement transducer (LVDT) and corrected for the compliance of the machine, which had been determined previously.12 The Young’s modulus (E), determined using the pulsedexcitation technique, is plotted as a function of the Si3N4 content in the interphase in Fig. 1. The value of E seems to increase linearly as the Si3N4 content in the interphase increases, and E follows the Voigt rule of mixtures.13 The E value measured from the load–deflection plots followed a similar trend, and moduli measured using both techniques agreed within 6%. (2) Strength and Energy Absorption Four-point flexural tests were performed using a screwdriven machine operated in displacement control (Model 4483, Instron, Danvers, MA). All tests were performed using a fully articulating testing jig with free-rolling pins using an outer span of 40 mm and an inner span of 20 mm. Data were collected using a computerized data-acquisition system at a rate of 5 points per second. Strength and WOF were measured on unnotched specimens at a crosshead displacement rate of 0.5 mm/min. Prior to testing, the specimens were polished to a 3 mm finish using resin-bonded diamond wheels (TBW, Furlong, PA) on the tensile surface and on one side surface. The edges of the specimen on the tensile surface were also chamfered. Tests were interrupted when the specimen fractured completely, the retained load dropped below 5 N, or the crosshead displacement exceeded 1 mm, whichever came first. The strength of the specimens was calculated using standard elasticbeam equations, whereas the WOF value was calculated by dividing the total area under the load–deflection curve by twice the cross-sectional area of the specimen. For specimens that fractured catastrophically, the WOF value was reported as zero. The nominal stress†† on the tensile surface for representative specimens is plotted versus crosshead deflection for unnotched specimens in Fig. 2. In general, the load remains linear up to the peak load for all the materials. After the peak load, some of the specimens continue to retain load at specimen deflections as large as 1 mm. The greatest degree of load retention is observed in the materials with the lowest Si3N4 content in the interphase; no load retention is observed following the peak load when the Si3N4 content in the interphase exceeds 25%. The nominal strength and WOF are plotted in Fig. 3 as a function of the Si3N4 content in the interphase. Although there is scatter in the nominal strengths, there does not seem to be a systematic change in strength with increasing Si3N4 content in the interphase. However, the WOF value decreases precipitously as the Si3N4 content in the interphase increases. The slight decrease in strength and WOF for the specimens that ¶ American Society for Standards and Testing, Philadelphia, PA. ††The nominal stress is calculated using standard elastic-beam theory, assuming elastic isotropy. It is recognized that the true stress is dependent on the local microstructure (i.e., the stiffer Si3N4 bears higher stress) and that the stress state is altered when cracking occurs anywhere in the beam. Fig. 1. Young’s modulus (E) of the layered ceramic, measured using the impulse-excitation technique, plotted versus Si3N4 content in the interphase; the solid line is the rule-of-mixtures modulus. The value for bulk Si3N4 has been taken from Kovar et al.11 Fig. 2. Nominal tensile stress (s), plotted versus crosshead displacement, for specimens containing 10, 25, 50, and 80 vol% Si3N4 in the interphase tested in four-point bending. April 1998 Crack Deflection and Propagation in Layered Silicon Nitride/Boron Nitride Ceramics 1005
Joumal of the American Ceramic Sociery-Kovar et al. Vol 8I. No 4 6000 abdz 4,000 3,000 SiN, in Interphase( Fig 3. Nominal strength(o)and work-of-fracture(WOF), plotted versus the Si3N4 content in the interphase contain no Si3 NA in the interphase is probably due to manu- crack deflection is apparent between Si3N4 layers, the lengths facturing defects that were present in this billet of the delamination cracks are extremely short(<100 um) The lengths of the delamination cracks in the other materials () Crack Deflection and Delamination Cracking also are dependent on the composition of the interphase be- SEM micrographs of the side surfaces of representative tween the Siy N4 layers. For example, long delamination cracks specimens after testing are shown in Figs. 4(aH(d). Cracks are are observed between almost every SigNa layer in the materials deflected between almost every layer until the Si3N4 content in the the interphase is increased to 80 vol%, no crack deflection is observed in the specimen that contains 80 vol% SiN4 in the content in the interphase is increased to 25 and 50 vol% SigN interphase. In Fig. 5, a higher-magnification micrograph of the in the interphase. The delamination cracks in these materials side surface is shown for a specimen that contains 50 vol% are observed to kink out of the interphase after propagating Si3N4 in the interphase. This micrograph shows that, although only a short distance. Unfortunately, it is difficult to quantify 4 Fig 4. SEM micrographs of the side surface of flexural specimens containing(a)10,(b)25, (c)50, and (d)80 vol%Si, N4 in the interphase(after testing). Crack deflection is observed for specimens containing up to 50 vol% Si, N4 in the interphase
contain no Si3N4 in the interphase is probably due to manufacturing defects that were present in this billet. (3) Crack Deflection and Delamination Cracking SEM micrographs of the side surfaces of representative specimens after testing are shown in Figs. 4(a)–(d). Cracks are deflected between almost every layer until the Si3N4 content in the interphase is increased to 80 vol%; no crack deflection is observed in the specimen that contains 80 vol% Si3N4 in the interphase. In Fig. 5, a higher-magnification micrograph of the side surface is shown for a specimen that contains 50 vol% Si3N4 in the interphase. This micrograph shows that, although crack deflection is apparent between Si3N4 layers, the lengths of the delamination cracks are extremely short (<100 mm). The lengths of the delamination cracks in the other materials also are dependent on the composition of the interphase between the Si3N4 layers. For example, long delamination cracks are observed between almost every Si3N4 layer in the materials that contain 0 vol% and 10 vol% Si3N4 in the interphase. However, the delamination distances decrease rapidly as the Si3N4 content in the interphase is increased to 25 and 50 vol% Si3N4 in the interphase. The delamination cracks in these materials are observed to kink out of the interphase after propagating only a short distance. Unfortunately, it is difficult to quantify Fig. 3. Nominal strength (s) and work-of-fracture (WOF), plotted versus the Si3N4 content in the interphase. Fig. 4. SEM micrographs of the side surface of flexural specimens containing (a) 10, (b) 25, (c) 50, and (d) 80 vol% Si3N4 in the interphase (after testing). Crack deflection is observed for specimens containing up to 50 vol% Si3N4 in the interphase. 1006 Journal of the American Ceramic Society—Kovar et al. Vol. 81, No. 4
April 1998 Crack Deflection and Propagation in Layered silicon Nitride/Boron Nitride Ceramics %05 50%S Delamination Distance, &(mm) Fig. 7. Spacing between through-thickness cracks in the Si,N4 lay ers. measured for each of the materials the cumulative fraction of the delamination cracks shorter than a given value are shown for each the materials 500um interphase is shown in Fig 8(a). It is clear from this micrograph hat crack deflection occurs within the bn interphase near the g magnification SEM micrograph of the side surface of Si3 N,/BN interface, rather than at the interface between the two ens containing 50 vol% Si3N4 in the interphase materials. As shown in Fig. 8(b), subsequent delamination lection at many of the BN-containing interphases, cracking also occurs within the BN interphase. The crack often t wever, the length of the delamination cracks are limited by cracking meanders within the bn interphase, and no systematic trend with respect to the crack path, could be discerned. The location of the crack within the BN-containing interphase did not seem Is not easy to to change as SiN was added to the interphase vever, a measure (4) Interfacial Fracture Resistance Interfacial fracture resistance was measured using notched etween through-thickness cracks in adjacent SiaN4 layers flexure tests, following the analysis of Charalambides et al, 14 chematic illustrations that show how these distances were from the steady-state load necessary to propagate a delamina- measured are shown in Fig. 6. A cumulative distribution plot of tion crack. One advantage of performing this test on multilayer delamination crack lengths is shown in Fig. 7 for each of tI materials. The delamination distances are longest in the mate specimens rather than on simple bilayer specimens is that re- sidual stresses present due to thermal mismatch between the als that contain 0 vol% and 10 vol% Si,N4 in the interphase. BN and SisN, do not influence the measurement of the inter- Consistent with the micrographs shown in Fig. 4, the delam facial fracture resistance. 15 The applied phase angle (y nation distances decrease markedly as the Si3N4 content is tan- [Ku/kid) was calculated assuming that there was a suffi- cient number of layers so that the elastic properties of a single A higher-magnification SEM micrograph of a through interphase did not influence the overall elastic properties of the hickness crack in a Si3N4 layer that is impinging on a BN specimen. Thus, the measured Young s modulus(E)of the composite Ised to calculate y. For the current experi- ments was cut to approximately the center of the N resulted in a y value of 42 The interfacial fracture resi T was calculated from14 3P2(S-L)(1 Si3N4 H2(1-n)3H3 where P is the applied load at which the delamination crack extends, v the Poisson's ratio of the composite, E the in-plane Youngs modulus of the composite, b the width of the speci- men, H the height of the specimen, and m the distance from the tensile surface of the beam to the delamination crack divided SiaN by the total height of the beam. S and L are the outer span and he inner span in the four-point test fixture, respectively BN Representative load-detlectometer-displacement curves are shown for notched specimens tested in four-point flexure Figs. (aHd) for materials that contain 10, 25, 50, and 80 vol% Si, N, in the interphase. For materials with <50 volSINSA8 the interphase, the crack paths are generally similar. The le increases linearly until a crack is initiated from the notch and propagates into the closest BN-containing interphase, where the distance between the crack is deflected and arrests. Subsequent specimen deflec- througl tion causes the delamination cracks to propagate stably in the interphase to either side of the notch at an almost-constant load
the length of delamination cracks, because it is not easy to discern the crack tip in the BN interphase. However, a measure of the delamination distances can be obtained from the distance between through-thickness cracks in adjacent Si3N4 layers. Schematic illustrations that show how these distances were measured are shown in Fig. 6. A cumulative distribution plot of delamination crack lengths is shown in Fig. 7 for each of the materials. The delamination distances are longest in the materials that contain 0 vol% and 10 vol% Si3N4 in the interphase. Consistent with the micrographs shown in Fig. 4, the delamination distances decrease markedly as the Si3N4 content is increased. A higher-magnification SEM micrograph of a throughthickness crack in a Si3N4 layer that is impinging on a BN interphase is shown in Fig. 8(a). It is clear from this micrograph that crack deflection occurs within the BN interphase near the Si3N4/BN interface, rather than at the interface between the two materials. As shown in Fig. 8(b), subsequent delamination cracking also occurs within the BN interphase. The crack often meanders within the BN interphase, and no systematic trend, with respect to the crack path, could be discerned. The location of the crack within the BN-containing interphase did not seem to change as Si3N4 was added to the interphase. (4) Interfacial Fracture Resistance Interfacial fracture resistance was measured using notched flexure tests, following the analysis of Charalambides et al., 14 from the steady-state load necessary to propagate a delamination crack. One advantage of performing this test on multilayer specimens rather than on simple bilayer specimens is that residual stresses present due to thermal mismatch between the BN and Si3N4 do not influence the measurement of the interfacial fracture resistance.15 The applied phase angle (C 4 tan−1 [KII/KI ]) was calculated assuming that there was a sufficient number of layers so that the elastic properties of a single interphase did not influence the overall elastic properties of the specimen. Thus, the measured Young’s modulus (E) of the composite was used to calculate C. For the current experiments, the notch was cut to approximately the center of the specimen, which resulted in a C value of 42°. The interfacial fracture resistance, Gi , was calculated from14 Gi = 3P2 ~S − L! 2 ~1 − n2 ! 2Eb2 F 1 H3 ~1 − h! 3 − 1 H3G (1) where P is the applied load at which the delamination crack extends, n the Poisson’s ratio of the composite, E the in-plane Young’s modulus of the composite, b the width of the specimen, H the height of the specimen, and h the distance from the tensile surface of the beam to the delamination crack divided by the total height of the beam. S and L are the outer span and the inner span in the four-point test fixture, respectively. Representative load–deflectometer-displacement curves are shown for notched specimens tested in four-point flexure in Figs. 9(a)–(d) for materials that contain 10, 25, 50, and 80 vol% Si3N4 in the interphase. For materials with <50 vol% Si3N4 in the interphase, the crack paths are generally similar. The load increases linearly until a crack is initiated from the notch and propagates into the closest BN-containing interphase, where the crack is deflected and arrests. Subsequent specimen deflection causes the delamination cracks to propagate stably in the interphase to either side of the notch at an almost-constant load. Fig. 5. Higher-magnification SEM micrograph of the side surface of one of the specimens containing 50 vol% Si3N4 in the interphase, showing crack deflection at many of the BN-containing interphases; however, the length of the delamination cracks are limited by cracking kinking. Fig. 6. Schematic illustration showing how the distance between through-thickness cracks, d, was measured in materials that exhibited (a) delamination cracking and (b) crack kinking. Fig. 7. Spacing between through-thickness cracks in the Si3N4 layers, measured for each of the materials; the cumulative fraction of the delamination cracks shorter than a given value are shown for each of the materials. April 1998 Crack Deflection and Propagation in Layered Silicon Nitride/Boron Nitride Ceramics 1007
Joumal of the American Ceramic Sociery-Kovar et al. Vol 81. No 4 deflection curve for this material exhibited a peak in load ha aox aterials that contained less Si3N4 in the interphase the lo the first delamination crack propagated; subsequent cra S growth occurred at lower loads. Specimens that contained 80 ol% SiaN in the rophically with crack defl he interfacial fracture resistance (Ti )is plotted as a function of SinA content in the hase in Fig. 10. The interfacial fracture resistance increases linearly from -30 J/m2 to 90 J/m2 as the Si,N content in the interphase is increased from 0 vol% to 50 vol%. Because no crack deflection occurred in the speci that contained 80 vol% Si3 N4 in the interphase, th terfacial fracture resistance could not be determined using the four-point delamination test Figures 11(a) and(b)show SEM BN micrographs of the interfacial fracture surfaces for specimen hat contain 10 and 50 vol% Si,N in the interphase. Because delamination cracking occurred within the weak interphase both Bn and Si,N are visible on the fracture surfaces. Quali- 20 a tatively, the ratio of bn to SiNa visible on the fracture sur ly equal to the ratio of bn to Si3N4 in the interphases themselves, which may explain why the interfacial fracture resistance seems to follow a rule of mixtures; the energy required to fracture the interphase should be the sum of the energies required to separate the constituent (b) Si3N4 (5) Frictional Sliding Resistance Because frictional sliding can be a potent source of energy dissipation in fiber-reinforced composites, 6 the frictional slid- ng resistance, Ts, was assessed in these layered ceramics as a BN developed by Kovar and Thouless. 2 This test was performed using the same specimen geometry as that in the flexural strength measurements. The side of the specimen was notched, which allowed a wedge to be inserted. The wedge was driven into the notch until the specimen split completely through a BN weak interphase. The specimen was then reassembled an loaded in three-point flexure. When the shear stress along the flection using an LVDT that was placed in contact with the specimen, the onset of slipping and, hence, Ts was determined from the point where a change in compliance is observed dur- 4 15u lated from the hysteresis area for a series of load-unload cycles taken over a range of loads. The hysteresis area, W, is related Fig 8. Path of (a)a crack impinging on a Si3 N,/BN interface an a delamination crack after crack deflection has occurred. Not EbHYTIT+3lIn(n-1)1 crack deflection and crack propagation both occur within the BI terphase. The arrow in Fig. 8(a) indicates the direction of 12[n(n-1)1-3n+3n2) where 2 is the normalized span between the outer loading points(2= SH) and II is the normalized load range(l (Pmax -Pmin)(EbH). Because the solution of Eq(2)for T Eventually, the delamination crack arrests when the crack yields two real roots, the physically correct root must be de- reaches the end of the inner loading span As specimen deflec- termined by examination of the experimental data. The normal tion is continued, the load again begins to increase linear pressure applied to the interface during the test has been cal- until the uncracked portion of the beam cannot support the culated by dividing the mean load during a given load-unload applied load anymore. A crack then initiates in the Si3 N4 layer cycle by the area of the interface that is sliding(the width of the closest to the delamination crack and propagates until it pecimen multiplied by its length ). Because of the high inter deflected in the next BN-containing interphase. This process facial fracture resistance in the material that contained 80 vol% repeated until the through-thickness cracks propagate com- Si N4 in the interphase, specimens made from this material did pletely through the specimen. not split cleanly through the interphase during precracking. As For specimens that contain 50 vol% Si N4 in the interphase, a result, frictional sliding resistance could not be measured in when cracks initiated from the notch, they were deflected only to one side of the notch before being arrested. Subsequent All the materials exhibited some degree of hysteresis energy ecimen deflection caused the delamination crack to grow dissipation during testing. Representative hysteresis loops stably only a short distance before kinking out of the interpha taken over different load ranges are shown for the material that and through the neighboring SiN, layer. This kinking process contains 50 vol% Si3N4 in the interphase in Fig. 12. As was was repeated through successive layers as loading continued observed in all the materials, the hysteresis loops have a ten- which resulted in a zig-zag crack patch similar to that shown dency to be wider at higher loads, which implies that the slid- Fig. 5 for an unnotched bar of the same material. Unlike the ing resistance increases as the normal pressure on the interface
Eventually, the delamination crack arrests when the crack reaches the end of the inner loading span. As specimen deflection is continued, the load again begins to increase linearly until the uncracked portion of the beam cannot support the applied load anymore. A crack then initiates in the Si3N4 layer closest to the delamination crack and propagates until it is deflected in the next BN-containing interphase. This process is repeated until the through-thickness cracks propagate completely through the specimen. For specimens that contain 50 vol% Si3N4 in the interphase, when cracks initiated from the notch, they were deflected only to one side of the notch before being arrested. Subsequent specimen deflection caused the delamination crack to grow stably only a short distance before kinking out of the interphase and through the neighboring Si3N4 layer. This kinking process was repeated through successive layers as loading continued, which resulted in a zig-zag crack patch similar to that shown in Fig. 5 for an unnotched bar of the same material. Unlike the materials that contained less Si3N4 in the interphase, the load– deflection curve for this material exhibited a peak in load when the first delamination crack propagated; subsequent crack growth occurred at lower loads. Specimens that contained 80 vol% Si3N4 in the interphase failed catastrophically with no crack deflection. The interfacial fracture resistance (Gi ) is plotted as a function of Si3N4 content in the interphase in Fig. 10. The interfacial fracture resistance increases linearly, from ∼30 J/m2 to 90 J/m2 , as the Si3N4 content in the interphase is increased from 0 vol% to 50 vol%. Because no crack deflection occurred in the specimens that contained 80 vol% Si3N4 in the interphase, the interfacial fracture resistance could not be determined using the four-point delamination test. Figures 11(a) and (b) show SEM micrographs of the interfacial fracture surfaces for specimens that contain 10 and 50 vol% Si3N4 in the interphase. Because delamination cracking occurred within the weak interphases, both BN and Si3N4 are visible on the fracture surfaces. Qualitatively, the ratio of BN to Si3N4 visible on the fracture surfaces for all the materials is approximately equal to the ratio of BN to Si3N4 in the interphases themselves, which may explain why the interfacial fracture resistance seems to follow a rule of mixtures; the energy required to fracture the interphase should be the sum of the energies required to separate the constituent phases. (5) Frictional Sliding Resistance Because frictional sliding can be a potent source of energy dissipation in fiber-reinforced composites,16 the frictional sliding resistance, ts, was assessed in these layered ceramics as a function of the composition of the interphase using a technique developed by Kovar and Thouless.12 This test was performed using the same specimen geometry as that in the flexural strength measurements. The side of the specimen was notched, which allowed a wedge to be inserted. The wedge was driven into the notch until the specimen split completely through a weak interphase. The specimen was then reassembled and loaded in three-point flexure. When the shear stress along the cracked interface exceeded the sliding resistance ts, slipping along the interface occurred. By measuring the specimen deflection using an LVDT that was placed in contact with the specimen, the onset of slipping and, hence, ts was determined from the point where a change in compliance is observed during loading or unloading. The sliding resistance ts was calculated from the hysteresis area for a series of load–unload cycles taken over a range of loads. The hysteresis area, W, is related to the normalized sliding resistance, T (equal to ts/E), by W = EbH2 S3 T @T + 3Ph~h − 1!# 12@h~h − 1!~1 − 3h + 3h2 !# (2) where S is the normalized span between the outer loading points (S 4 S/H) and P is the normalized load range (P 4 (Pmax − Pmin)/(EbH)). Because the solution of Eq. (2) for ts yields two real roots, the physically correct root must be determined by examination of the experimental data. The normal pressure applied to the interface during the test has been calculated by dividing the mean load during a given load–unload cycle by the area of the interface that is sliding (the width of the specimen multiplied by its length). Because of the high interfacial fracture resistance in the material that contained 80 vol% Si3N4 in the interphase, specimens made from this material did not split cleanly through the interphase during precracking. As a result, frictional sliding resistance could not be measured in this material. All the materials exhibited some degree of hysteresis energy dissipation during testing. Representative hysteresis loops taken over different load ranges are shown for the material that contains 50 vol% Si3N4 in the interphase in Fig. 12. As was observed in all the materials, the hysteresis loops have a tendency to be wider at higher loads, which implies that the sliding resistance increases as the normal pressure on the interface Fig. 8. Path of (a) a crack impinging on a Si3N4/BN interface and (b) a delamination crack after crack deflection has occurred. Note that crack deflection and crack propagation both occur within the BN interphase. The arrow in Fig. 8(a) indicates the direction of crack growth. 1008 Journal of the American Ceramic Society—Kovar et al. Vol. 81, No. 4