Disclaimer for FAA Research Publication

Transcription

1 Disclaimer for FAA Research Publication Although the FAA has sponsored this project, it neither endorses nor rejects the findings of the research. The presentation of this information is in the interest of invoking technical community comment on the results and conclusions of the research.

2 DEVELOPMENT AND EVALUATION OF FRACTURE MECHANICS TEST METHODS FOR SANDWICH COMPOSITES Daniel O. Adams, Jeffery A. Kessler, Brad Kuramoto, Josh Bluth, Chris Weaver, Andy Gill Department of Mechanical Engineering University of Utah Salt Lake City, UT ABSTRACT The increased use of sandwich composites in applications where high stiffness-to-weight and strength-to-weight ratios are required has increased the need for mechanical properties to predict structural performance. One particular area of need is the fracture mechanics properties associated with delamination of the facesheet and core of the sandwich structure. The goal of this research investigation is to identify test methods for characterizing both the Mode I and Mode II energy release rate associated with facesheet/core delamination growth in sandwich composites. A literature review of previously performed work along with modified and newly created tests led to the identification of candidate test methods for evaluation. Initial evaluation was performed using finite element analysis. Follow-on testing was performed to investigate whether stable interlaminar crack growth occurred at the facesheet/core interface. Based on these initial evaluations, the selection of a recommended test configuration was made for both Mode I and Mode II testing. Follow-on testing and analysis is currently underway to determine the range of composite sandwich materials and geometries for which the test is valid. Further work will involve the final test fixture design, specimen configuration, data analysis methodology, and determination of a range of acceptable sandwich materials and geometries for use in the preparation of draft ASTM standards for Mode I and Mode II fracture toughness testing of sandwich composite materials.

3 INTRODUCTION Although the development of test methods for fracture mechanics of monolithic composite laminates has reached a high level of maturity in recent years, relatively little attention has been given to the development of fracture mechanics test methods for sandwich composites. Further, a majority of the efforts to date has focused on determining the fracture toughness of a particular sandwich material or the effects of specific environmental conditions. As a result, there is no consensus on a suitable test configuration or specimen geometry for either Mode I or Mode II fracture toughness testing. Thus, the objective of this research investigation is to develop test methods for the determination of both the Mode I and Mode II fracture toughness of sandwich composite materials that are suitable for ASTM standardization. To date, the first two phases of a proposed three-year research investigation have been completed. Accomplishments associated with the first two phases of this research investigation are highlighted in this paper. CANDIDATE TEST METHODOLOGIES This project began with a comprehensive literature review to identify and assess candidate test methodologies for both Mode I and Mode II fracture toughness testing. An initial assessment focused on identifying the strengths and weaknesses associated with each candidate test configurations for both test types.

4 Mode I Test Methods A total of five Mode I test configurations were identified for initial investigation. These configurations were based on three basic loading configurations: the double cantilever beam (DCB), the single cantilever beam (SCB), and the three point bend (TPB) configurations. A schematic of the selected DCB test setup for sandwich composites is illustrated in Figure 1. An initial delamination is created by placing a Teflon film insert between the facesheet and core materials during fabrication of the sandwich composite panel. Piano hinges are adhesively bonded to the facesheets and a vertical load is applied along these hinges in opposite directions. The free end of the specimen is not constrained and will rotate as the crosshead displacement increases due to asymmetric placement of the delamination. Applied Load Delamination Piano Hinge Crack Tip FIGURE 1. SANDWICH DCB CONFIGURATION INVESTIGATED

5 A variety of modified DCB configurations have been explored by researchers for sandwich composites. One such modification to the double cantilever beam test configuration was developed by Shivakumar and Smith [1]. This Modified DCB (MDCB) configuration was referred to by the authors as the Modified Mode I Cracked Sandwich Beam test. A schematic of this test configuration is illustrated in Figure 2. To eliminate the rotation of the DCB specimen due to asymmetric placement of the delamination, a block support was placed at the free end of the specimen. This configuration was used by the authors for testing a variety of foam core sandwich composite specimens. Support Block Applied Load Delamination Piano Hinge Crack Tip FIGURE 2: MDCB CONFIGURATION INVESTIGATED Further modification to the DCB test for sandwich composites has resulted in the Single Cantilever Beam (SCB) configuration. To prevent bending

6 deformation in the core and bottom facesheet, additional constraints are placed on the specimen. Thus, only the upper facesheet is considered a cantilever beam, and thus the name single cantilever beam. Several variations of the SCB test configuration have been investigated. One SCB test configuration selected for investigation is illustrated in Figure 3. A vertical upward load is passed through a single piano hinge bonded to the upper facesheet on the delaminated end of the specimen. The opposite end of the specimen is clamped to prevent rotation. Cantilever Blocks Applied Load Delamination Piano Hinge Crack Tip FIGURE 3. SCB CONFIGURATION WITH CANTILEVER END SUPPORT Another variation of the SCB Configuration utilizes a plate support to prevent bending deformation in the core and bottom facesheet. This platesupported SCB test configuration is shown in Figure 4. By affixing the bottom of the specimen to a lower support plate, bending deformation is eliminated in the bottom facesheet. Similar to the SCB configuration with cantilever end support (Figure 3), the specimen is loaded using a single piano hinge adhesively bonded

7 to the top facesheet above the delamination. Variations of this plate-supported SCB test configuration have been researched by Cantwell and Davies [2], and Carlsson et al. [3-7]. Applied Load Piano Hinge Delamination Crack Tip Plate Support FIGURE 4. PLATE SUPPORTED SCB CONFIGURATION A modified Three Point Bend (TPB) test configuration was developed by Cantwell et al. [8]. The authors determined that this test configuration was Mode I dominant. A schematic of this TPB configuration that was selected for evaluation in this investigation is shown in Figure 5. To directly load the top facesheet in the delaminated region of the specimen, a section of the core and bottom facesheet below the delamination is removed. A modified three point bend fixture is used to support the bottom facesheet of the left end and the top facesheet of the right end. A vertical downward load is applied at the center support located at approximately the midpoint of the specimen s top facesheet.

8 Left Support Rod Center Support Rod Applied Load Delamination Crack Tip Right Support Rod FIGURE 5. TPB CONFIGURATION In summary, five candidate Mode I test configurations were identified for use in determining the Mode I critical energy release rate associated with facesheet/core interface fracture in sandwich composites. The goal of this research was to identify the best suited test configuration for measuring the Mode I critical energy release rate associated with facesheet/core debonding of sandwich composites. Mode II Test Methods A total of ten candidate Mode II test configurations methods were identified for initial evaluation. A literature review led to the discovery of six concepts that could be evolved into Mode I test configurations for sandwich composites. Additionally, four other test configurations were developed in this research program in an attempt to maximize the shear stresses occurring at the crack front during loading and create crack front opening. These candidate methods identified for initial evaluation were:

9 End Notch Flexure Test Mixed Mode Bending Test Four Point Bend Delamination Test Cracked Sandwich Beam With Hinge Test End-Loaded Split Test Double Cantilever Beam With Uneven Bending Moments Test Modified Crack Sandwich Beam With Hinge Test Tensile Facesheet Test Three Point Cantilever Test Double Sandwich Test Due to the number of candidate Mode II test methods identified, they are not profiled individually. Following initial evaluations, however, only two test methods appeared to satisfy two fundamental requirements: a) a predominantly Mode II crack growth; and b) crack opening during loading such that frictional forces were not created at the interface which could produce artificially high energy release rate values. These two test methods were the Mixed-Mode Bending (MMB) test and a hinged Cracked Sandwich Beam (CSB) test. Note that the mixed mode bending test (MMB) was created as an ASTM standard (ASTM D6671) as a means to determine the interlaminar energy release rate, G C, under a mixture of Mode I and Mode II loading on composite laminates. Shown in Figure 6, the MMB test setup varies the amount of mode mixture by varying the loading lever arm length.

10 Applied Load Lever arm Half span length Artificial Pre-crack Span length FIGURE 6. MIXED MODE BENDING TEST The specimen is designed similar to a Mode I DCB specimen. Load is applied to the piano hinges producing crack growth. For this investigation, this test method was transitioned from a composite laminate test to a sandwich composite test in which a Mode II dominant loading is used (at least 90% Mode II). As a result, the crack front is held slightly open to reduce frictional forces while shearing forces produce stable crack growth. While not an ASTM standard, a common method used to determine Mode II energy release rate in composite laminates is the three point end notch flexure test (3ENF). The test is performed by simply-supporting the bottom side of a laminate in two locations. The distance between these two supports is referred to as the span length. A load is applied on the top of the specimen half way between each bottom support. The specimen, which is manufactured with an artificial pre-crack, undergoes flexure, which creates large shear stresses at the

11 crack front. Mode II interlaminar crack growth occurs as the shear stress becomes large enough to propagate the crack. The 3ENF (three-point loading) test was modified and adapted to sandwich composites by Carlsson and investigated as a means to determine the bond strength of sandwich composites under shear loading [9]. As shown in Figure 7, the cracked sandwich beam (CSB) test is performed in the same fashion as the 3ENF test. Applied Load Artificial Pre-crack Span Length Half Span Length FIGURE 7. CRACKED SANDWICH BEAM TEST Shipsha et al. [10] modified the basic CSB test setup to address problems associated with unstable crack growth and frictional forces between the core and facesheet. The cracked sandwich beam hinge specimen (CSB-h), shown in Figure 8, is created in a similar manner to the CSB specimen with the exception that a section of core is removed and a hinge is mounted to the top and bottom facesheet directly above the bottom support where the section of core once was. The mounted hinge holds the crack front open slightly, alleviating the frictional forces between the core and facesheet and also redistributes the stresses in the crack tip vicinity.

12 Applied Load Artificial Pre-crack Span Length Aluminum Hinge Half Span Length FIGURE 8. CRACKED SANDWICH BEAM WITH HINGE TEST ASSESSMENT OF CANDIDATE TEST METHODOLOGIES Preliminary assessments of the candidate test methodologies identified were performed using two-dimensional plane-strain finite element analysis. These analyses were used to evaluate whether a suitable fracture mode mixity (i.e. Mode I vs. Mode II) was present in each candidate test configuration. Additionally, finite element simulations were used to assess whether other undesirable failures would be predicted to occur within the sandwich specimen during testing. As a result of these finite element simulations, the number of candidate Mode II test configurations was reduced to seven for preliminary testing. However, all five Mode I test configurations were evaluated via mechanical testing. Additionally, these simulations were used to establish suitable geometries for prototype test fixturing and test specimens.

13 Once the initial test configurations were identified from finite element analyses, two sandwich composite configurations were selected for initial testing; woven carbon-epoxy facesheets/polyurethane foam core and carbon-epoxy facesheets/nomex honeycomb core. All composite sandwich panels were fabricated at the University of Utah. Initial testing was performed using the selected Mode I and Mode II test configurations. Emphasis was placed on producing stable, self-similar delamination growth at the facesheet/core interface without producing secondary failures at other locations in the specimen. From these tests, the most promising Mode I and Mode II test configuration was selected. SELECTED TEST METHODS Mode I Test Methods Both experimental and finite element analysis was performed for each of the five candidate test configurations using the two composite sandwich configurations. Results from finite element analysis showed that all five test configurations were Mode I dominant. The SCB beam displayed the lowest percentage of Mode II energy release rate, whereas the modified DCB displayed the greatest percentage. Results from mechanical testing showed crack growth away from the interface and into the core using both DCB configurations and one SCB configuration, an undesirable result. Based on the results of analysis and testing, the plate-supported SCB test configuration, shown in Figure 9, was selected as the best suited Mode I test configuration for sandwich composites.

14 FIGURE 9. PLATE-SUPPORTED SINGLE CANTILEVER BEAM (SCB) TEST CONFIGURATION SELECTED FOR MODE I TESTING. Mode II Test Method Based on results from finite element analysis, the number of candidate Mode II test configurations was reduced from ten to seven. This reduction was based on examining the mode mixity for predominantly Mode II crack growth and crack opening during loading. Mechanical testing of the seven candidate Mode II configurations led to only two which produced the stable crack growth at the facesheet/core interface: the Mixed-Mode Bending (MMB) test and a hinged Cracked Sandwich Beam (CSB) test. In follow-on testing with aluminum honeycomb cores, however, the MMB test produced core crushing whereas the

15 hinged CSB test produced stable interface crack growth. Thus, the hinged CSB test, shown in Figure 10, was selected as the best suited Mode II test configuration for sandwich composites. FIGURE 10. HINGED CRACKED SANDWICH BEAM TEST CONFIGURATON SELECTED FOR MODE II TESTING DEVELOPMENT OF TEST FIXTURING Following the preliminary selection of both Mode I and Mode II test configurations, further development of the Mode I and Mode II test fixturing and data analysis procedures was performed. For Mode I, a prototype test fixture was constructed to perform the plate-supported Single Cantilever Beam (SCB) test. This prototype fixture, shown in Figure 11, was designed to accommodate specimen widths ranging from 1 in. to 3 in. The fixture features a sliding base plate to which the sandwich specimen is secured. To minimize time and cost associated with test preparation, a clamping device was developed to secure the

16 lower facesheet of the sandwich specimen to the base plate, thus eliminating the need for adhesive bonding. As loading increases and the delamination along the upper facesheet propagates, the sandwich specimen is allowed to translate (sliding base plate), ensuring that the applied load remains normal to the sandwich specimen for Mode I testing. Results from initial testing indicated stable crack growth behavior such that several crack length measurements were possible. Additionally, the delamination remained at the facesheet/core interface during propagation. FIGURE 11. PROTOTYPE TEST FIXTURE FOR PLATE SUPPORTED SINGLE CANTILEVER BEAM (SCB) TEST For Mode II testing, a test fixture was developed for the Modified Cracked Sandwich Beam (MCSB) test. This test fixture is a modification to the standard three-point flexure fixture used for testing composite laminates. As shown in Figure 12, however, one of the outer loading supports was modified such that as the base of the sandwich specimen is loaded from the bottom, the upper facesheet is displaced such that it remains separated from the core at this end of

17 the specimen. Through numerical simulations, this end separation was shown to result in a Mode II dominant energy release rate while preventing the delaminated surfaces from contacting. Initial evaluation focused on the use of carbon-epoxy/ Nomex honeycomb sandwich specimens. The specimen was found to be significantly stiffer than those tested initially because the core was not removed from the specimen in the delaminated region (to simplify specimen preparation). As a result, the pre-crack was lengthened from the previous 1/2 inch length to 1 inch. Testing performed using carbon/epoxy facesheets and Rohacell foam core produced stable crack growth, with the delamination remaining at the facesheet/core interface. Additionally, the Mode II test fixture appeared to keep the crack faces separated in the delaminated region during testing, thus eliminating friction effects. FIGURE 12. PROTOTYPE TEST FIXTURE FOR CRACKED SANDWICH BEAM TEST

18 CURRENT AND PLANNED RESEARCH Current research is focusing on assessing the range of usage for the proposed Mode I and Mode II sandwich fracture test methods. Of interest are both the geometric parameters (ex: core and facesheet thickness) and material parameters (ex: elastic properties of facesheet and core) of the sandwich specimen. Although efforts have focused on commonly used materials, emphasis has been placed on evaluating a range of key geometric and material parameters to identify limitations that may be applied to these test methods. Finite element analysis is being used to investigate ranges of such parameters. Of particular interest are the acceptable ranges of facesheet flexural stiffness (EI) as well as the core stiffness (E) of candidate composite sandwich specimens. Specimen width effects are also under investigation using the prototype platesupported Single Cantilever Beam (SCB) test fixture. In the upcoming phase of this project, efforts will focus on the advanced development of the proposed Mode I and Mode II fracture toughness test methods for sandwich composites in preparation for standardization. Further Mode I and Mode II fracture toughness testing and analysis will be performed using the current test configurations to establish the range of materials and sandwich geometries from which stable, self-similar delamination propagation is obtainable. Based on results obtained from testing and analysis, a final test configuration and analysis methodology will be selected for both Mode I and Mode II fracture toughness testing. Additionally, acceptable ranges of sandwich materials (facesheet and core), material parameters (facesheet fiber orientations,

19 core densities), and sandwich geometries (facesheet and core thicknesses) will be proposed for inclusion in both the Mode I and Mode II test methods. After the final test configuration has been determined, a final set of sandwich panels will be selected for validation testing. Using these sandwich panels, the final phase of testing will explore in more detail the full range of sandwich materials, material parameters, and sandwich geometries proposed for inclusion in the proposed test standards. The final validation phase of Mode I and Mode II fracture toughness testing will be performed using the final proposed test fixture and analysis methodology. The goal of this final testing phase is to arrive at a final test fixture design, specimen configuration, data analysis methodology, and a final range of acceptable sandwich materials and geometries for both Mode I and Mode II fracture toughness testing. The final task of this investigation will be the preparation of draft ASTM standards for determining both the Mode I and Mode II fracture toughness of sandwich composite materials. REFERENCES 1. K.N. Shivakumar and S.A. Smith, Influence of Core Density and Facesheet Properties on Debond Fracture Strength of Foam Core Sandwich Structures, Proceedings of the American Society for Composites 17 th Technical Conference. 2. W.J. Cantwell and P. Davies, A test technique for assessing core-skin adhesion in composite sandwich structures, Journal of Materials Science Letters, Vol. 13, pp , Xiaoming Li, and Lief Carlsson, The Tilted Sandwich Debond (TSD) Specimen for Face/Core Interface Fracture Characterization, Journal of Sandwich Structures and Materials, Vol., 1, 1999, pp

20 4. Xiaoming Li, and Lief Carlsson, Fracture Mechanics Analysis of Tilted Sandwich Debond (TSD) Specimen Journal of Composite Materials, vol.35, 2001, pp X. Li and L.A. Carlsson, Elastic Foundation Analysis of Tilted Sandwich Debond (TSD) Specimen, Journal of Sandwich Structures and Materials, Vol. 2, January Gilmer M. Viana, and Lief Carlsson, Mode Mixity and Crack Tip Yield Zones in TSD Sandwich Specimens with PVC Foam Core, Journal of Sandwich Structures and Materials, Vol.4, 2002, pp G.M. Viana and L.A. Carlsson, Influences of Foam Density and Core Thickness on Debond Toughness of Sandwich Specimens with PVC Foam Core, Journal of Sandwich Structures and Materials, Vol. 5, April W.J. Cantwell et al., Interfacial fracture in sandwich laminates, Composites Science and Technology, Vol. 59, pp , Carlsson, L.A., On the Design of the Crack Sandwich Beam (CSB) Specimen Journal of Reinforced Plastics and Composites, Vol. 10, pp , Shipsha, A., Burman, M., and Zenkert, D., Interfacial Fatigue Crack Growth in Foam Core Sandwich Structures Fatigue and Fracture of Engineering Materials and Structures, Vol. 22, pp , 1999.