18 TH INTERNATIONAL CONFERENCE ON COMPOSITE MATERIALS VOIDS CHARACTERISATION IN CARBON FIBRE/EPOXY COMPOSITE LAMINATES J. E. Little 1 , X. W. Yuan 1* , M. I. Jones 2 1 Department of Mechanical Engineering, 2 Department of Chemical and Materials Engineering, University of Auckland, Auckland, New Zealand *xw.yuan@auckland.ac.nz Keywords : voids, porosity, micro-CT, composite laminates the samples void volume fraction. Fibre 1 Introduction degradation and convection currents affect the Increased use is being made of advanced composite accuracy of this technique, because fibre mass materials for the critical structure of both civilian can be degraded or removed. This error is and military aircraft. Such materials often have a especially prevalent for carbon fibre reinforced higher strength-to-weight ratio compared to composite (CFRC) laminates, because the conventional materials, are corrosion resistant and temperatures reached when achieving matrix are able to be tailored to meet specific mechanical burn are above carbon fibre oxidation [5]. criteria [1]. Problems can arise during the (3) Matrix digestion, a method that is similar to manufacture and handling of these materials due to Matrix burn-off, except that acid is used to their non-homogeneous nature, which result in digest and remove the matrix from around the defects, some of which can be hidden or fibres instead of using thermal degradation. uncharacterised [1]. Void content/porosity, Toxic fumes are produced during the chemical delamination, matrix cracks and impact damage are break down of the matrix [6], rendering this just some of the composite material defects. These process messy and dangerous. defects need to be identified and characterised for (4) Microscopy, which is a visual and destructive determination of the ‘quality’ of a composite analysis technique that allows the void size, structure, as these parameters have a significant shape and distribution to be analysed [7]. The effect on the mechanical properties of the material method is widely used in industry due to its and can act as failure initiation points [2]. visual characterisation advantage, but also has Conventional composite void characterisation an inherent section-bias error associated with techniques provide results of limited accuracy and/or analysing two dimensional (2D) cross-sections reliability due to inherent testing errors. Such of the sample. techniques include: The above-mentioned conventional void analysis (1) Archimedes test, for which the samples actual techniques are limited in their accuracy and density is calculated using the Archimedes reliability, and also have limitations regarding principle. Then, the void volume fraction is extraction of information from their data. The first calculated using the theoretical density of the three techniques, which are non-visual, give only sample with no voids present [3]. The technique total void fraction and cannot analyse individual relies on accurate knowledge of the fibre and void size, shape and distribution. Microscopy can matrix density and weight fraction values of the detect this important visual information, but it is laminate. The method can give a negative void limited by the information being 2D and not fraction, which is theoretically impossible [4]. representative of the whole sample. (2) Matrix burn-off, which utilises a furnace to degrade and remove the matrix from around the X-ray Computed Tomography is a novel non- fibres, facilitating measurement of the fibre destructive and visual technique that can be used for weight fraction. Archimedes density testing is three dimensional (3D) analysis of internal initially required to measure the samples actual structures. The technique has been used for about 5 density [3], with accurate knowledge of the decades, predominately for medical diagnosis [1]. constituent densities still required to calculate
Micro-computed tomography (micro-CT), a more % � ������ % � ����� � � � 100 � � ������ � � ������ � � ����� � (1) recent development, allows for wider application, including the qualitative and quantitative non- destructive evaluation of composite materials [1]. The material properties must be accurately known, The technique is based on recording the change in because minor variations can significantly affect the attenuation along the path of an X-ray beam fired calculated void volume fraction. Using Archimedes through a sample [8]. This allows a cross-sectional principle, the actual density ( ρ sample ) of the sample is image to be reconstructed mathematically from determined, with the sample first being weighed in attenuation measurements made at different angles air and then while submerged in water. These steps by the detectors. Because the thickness of each 2D should be carried out accurately, with no air bubbles slice is known, the images can be stacked virtually present on the sample surface while submerged. to create a 3D model. Binary conversion (black and An Environmental Scanning Electron Microscopy white) of the image then allows the voids to be (ESEM) microscope was used for the microscopy separated from the fibre and matrix, which make up analysis due to its superior contrast and digital the laminate. imaging capabilities compared to that of an optical microscope. The technique is destructive, because a The objectives of this study are a comparison of the sample cross-section is required. The testing Archimedes technique and microscopy with micro- preparation required for microscopy analysis of CT for analysis of voids in CFRC materials. composite materials is time consuming and sometimes difficult, due to the non-homogenous composition of composites. The sample is set in 2 Experimental Procedures resin after which a flat cross-section is prepared to a Two 24-ply oven cured carbon fibre reinforced mirror-like finish, with no visible preparation thermoset plastic (epoxy) (CFRP) laminates were damage, such as scratches and void distortion. manufactured. One panel was made from Sequential grinding of the surface with smaller sized unidirectional (UD) preimpreginated (prepreg) particle paper, from 35 to 2.5 microns, achieves a carbon fibre, while the other was a twill weave high quality surface finish. The process is difficult, carbon fibre prepreg. Oven curing of the laminates because the brittle fibres regularly break free and under vacuum was used to ensure that voids were gouge large scratches in the specimen surface. The present. Samples of size 15 x 7 x 5 mm were cut finished surface should have few or no scratches, from the middle of the panels for testing. This size because these appear as voids when using image was a compromise between requirements of the analysis software. different techniques, with micro-CT requiring small Image analysis software is used to process the samples to reduce data file size. Carbon fibre/epoxy microscopy images. Analysis is undertaken of the laminates were used for the samples due to the individual pixels that make up the image cross- particular interest in characterisation of this material. section. The process requires the microscopy cross- The determination of the void volume fraction using section image to be converted to binary form, with the Archimedes Theoretical versus Actual Density the voids remaining black and the matter (fibre and technique was carried out in accordance with ASTM matrix) converted to white. The image is then D3171. The samples were smaller than the ‘cleaned’, to remove unwanted noise and outliers. recommended testing standard volume of 1cm 3 , due The image is then analysed using built-in tools, to size restrictions of the micro-CT technique. This giving the total number of voids, their overall area is assumed to have negligible impact on the fraction, as well as individual void size, shape and accuracy of the Archimedes test results. 2D location. The void volume fraction (V V ) is given by equation The micro-CT scanning and image reconstruction 1, in terms of the mass fraction (% m ) and density ( ρ ) requires little specimen preparation, because the of the primary constituents technique is non-destructive. The scanning, image reconstruction and the data acquisition processes can
Recommend
More recommend
