NDT.net - March 1999,
Vol. 4 No. 3
The interlaminar strength of carbon fibre reinforced composites has been improved by adding amorphous thermoplastic polyamide particles at the prepreg interface. This process gives a material with high compressive strength after impact and its use is becoming more widespread in the aerospace industry. Because the use of this reinforced material composite for aircraft primary structures is relatively recent, there has been very few non-destructive analyses studying the effect of low energy impact damage on such material. Impact damage may cause delamination and reduce the compressive strength of composite laminates; thus such defect must be efficiently detected and accurately sized. The effect of ply orientation on the impact resistance of thermoplastic toughened thermoset resins polymeric matrix composites has been studied using multiple non-destructive testing (NDT) techniques. Non-destructive evaluation of impacted [0/45/-45/90] and [0/90] thermoplastic toughened thermoset resin polymeric matrix panels was carried out. Experimental results of these non-destructive inspections were used to determine which of the two ply orientations provides higher impact resistance. A comparison of impact damage propagation between these two ply orientations was made. In addition, the performance of the NDT techniques used in function of ply orientation was assessed and the most appropriate NDT method for the inspection of carbon fibre reinforced material with different ply orientations proposed.
Carbon fibre reinforced composites are becoming more increasingly used by industries and more particularly by the aerospace industry. As material specifications become more strict and demands for high performance materials increases, novel advanced reinforced composites are being developed. The thermoplastic toughened thermoset resin polymeric matrix composite developed by Toray Industries and manufactured under the reference T800H/3900 is a perfect example of such improvement in the development of advanced carbon fibre reinforced plastics (CFRPs). Such material offers higher impact resistance than conventional CFRPs and can be used for aircraft primary structure. Although the impact resistance of this material is higher compared with conventional CFRP composites, very little analyses comparing the effect of stacking sequence on the impact resistance of this newly reinforced material have been published. In terms of non-destructive testing (NDT) analyses, the study presented in this article appears to be the first for this kind of material and for impact energies of 10 Joules or less.
Analysis of the stacking sequence on the impact resistance was based on non-destructive measurements obtained by ultrasonic C-scan testing. Several CFRP panels were impacted at different energies, the detectability of the damage analysed, and impact damaged area calculated. It was showed that impact damage detection is easier in [0/45/-45/90] specimens than in [0/90] specimens for low impact energy, that delamination occurred a bit deeper in [0/90] materials than in [0/45/-45/90] composites, and that impact damage tends to propagate quicker and extend further at higher energies in [0/45/-45/90] samples than in [0/90] samples.
III. Thermoplastic toughened thermoset resin polymeric matrix composite
The thermoplastic toughened thermoset resin polymeric matrix composite tested is developed by Toray Industries in Japan and consists of Torayca T800H high strength carbon fibres and 3900 high toughness resin with improved impact resistance. An improvement in strength through the increase of elastic moduli with higher crystallinity was used to obtain high performance carbon fibres. In order to limit delamination caused by impact, a combination of thermoplastic resin between layers and epoxy resin for each layer was used. The thermoplastic resin offering high toughness while epoxy resin gives high heat resistance and high elastic moduli to the material. A more detailed description of this thermoplastic toughened thermoset resin polymeric matrix composite can be found in the literature . This material finds applications in the manufacture of the Boeing B777 aeroplane for the fin torque, stabiliser torque box and floor beams .
The specimens tested were square panels of 150 mm in length and 1.55 mm thick. In this article we will consider the effect of impact of energies of 4, 6, 8, 9 and 10 Joules. Two different stacking sequences were analysed [0/90] and [0/45/-45/90]. Although conventional fracture analyses have been carried out by several researchers , there has been very little study using NDT methods of the effect of impact damage on thermoplastic toughened thermoset resin polymeric matrix composites.
IV. Impact damage
Impact damage is considered as the primary cause of in-service delamination in composites and can be brought about by dropping a tool during maintenance, hailstones, bird strikes, stone, runway debris, thrown tire treads or baggage loading. In addition to delamination, impact damage can also lead to matrix cracking and fibre fracture which cause reductions in static residual strength. Assessing accurately the extent of impact damage is necessary in order to evaluate the residual strength of a structure.
Fibres absorb part of the energy during an impact, but they also distribute some of the load internally. This excess energy can induce cracking and delamination. Impact damage detection and quantification can be a difficult task, especially with low energy impacts where most of the damage is not readily visible on the external surface. Barely visible damage (also referred as blind side impact), produces visible cracking of the matrix solely on the back side of the component, and cannot be detected with conventional rapid visual examination
Impact damage, even of small energy, may affect the structural integrity of a material. Although very low energy impact may not be critical for the safety of a structure, such damage may propagate during use due to stress and strain. Thus, impact damage, no matter how small, should be detected and accurately quantified as soon as possible in order to facilitate monitoring of its growth during use.
V. Non-destructive evaluation
Several NDT methods exist for the testing of CFRP materials. These include, but are not limited to, visual analysis, ultrasonic testing, eddy current testing, radiography, infrared thermography and shearography. Among these, very few are capable to detect and accurately quantify low energy impact damage in composites. Ultrasonic testing, eddy current testing and shearography constitute probably the top end of NDT methods for such purposes. Emerging technologies such as SQUID magnetometry may prove to be a more sensitive technique for the detection and characterisation of flaws in composite materials.
Because investment from industries greatly favoured ultrasounds, such systems are widely used for non-destructive examination (NDE) of composite materials at the manufacturing level. In the experiments presented here ultrasonic C-scan examination was carried out using an automated scanning system developed by Krautkramer Japan. The specimens tested were immersed into water, a 25 MHz frequency probe was selected as well as a sound velocity of 3000 m/s. Figure 1 shows ultrasonic C-scan images obtained from the examination of a [0/45/-45/90] panel subject to an impact of 6 and 10 Joules. Figure 2 shows ultrasonic C-scan images obtained from the examination of a [0/90] panel subject to an impact of 6 and 10 Joules. It can be seen that, for a 6 Joules impact, no damage was detected on the [0/90] sample and only a slight indent appears on the [0/45/-45/90] panel. With a 10 Joules impact, damage area appears more or less similar for both specimens. Delamination propagating along the fibre direction can be clearly seen on the [0/45/-45/90] panel with the defect spreading in a diagonal manner at 45(. An analysis by Fuoss et al.  showed a similar damaged area pattern of the damage propagating along the direction of the first fibre lay-up orientation.
Fig 1. |
Ultrasonic C-scans of [0/45/-45/90]
toughened thermoset resin polymeric matrix composites subject
to an impact of 6 Joules (top) and 10 Joules (bottom)
Ultrasonic C-scans of [0/90] toughened thermoset resin polymeric matrix composites subject to an impact of 6 Joules (top) and 10 Joules (bottom)
V. Results and discussion
Ultrasonic examination of [0/45/-45/90] composite subject to an impact energy of 4 Joules did not revealed the presence of any damage. Small indents were revealed for impacts of 6, 7 and 8 Joules. There is little variation in damaged area, either due to the high impact resistance of the material or simply due to the limitation in sensitivity of the ultrasonic apparatus. The damaged area increased more rapidly for a 9 Joules impact, and become rapidly evident as impact energy reaches 10, 11 and 12 Joules. At 10 Joules, delamination occurred around 0.8 mm in depth within the material.
The non-destructive examination of the [0/90] samples did not detect any damage caused by the 4 and 6 Joules impacts. Delamination was first detected for impact energies of 8 Joules or greater. The 10 Joules impact caused delamination located around 1.0 mm in depth within the material, thus slightly deeper than with the [0/45/-45/90] sample.
A comparison of impact damage area on [0/90] (3900C) and [0/45/-45/90] (3900A) composites as a function of impact energy is shown on figure 3 (note that only detected impact damage is indicated on this graph). One can se that at equal energy (i.e. 8, 9 and 10 Joules) there seems to be very little differences in impact damage area depending on the stacking sequence (cf table 1 for additional details). Also, it appears that damaged area remains low for impacts of less than 8 Joules and then extends sharply with a little increase in impact energy.
|Table 1: Damaged area observed on [0/90] and [0/45/-45/90] thermoplastic toughened thermoset resin polymeric matrix composites for increasing impact energy
Fig. 3. Plots of impact damage area on [0/90] (T800H/3900C) and [0/45/-45/90] (T800H/3900A) materials against impact energy
|9.0||62.0 mm2||62.0 mm2
|10.0|| 88.0 mm2||93.0 mm2
Ultrasonic C-scan examination of thermoplastic toughened thermoset resin polymeric matrix composites revealed that for low impact energies (up to 10 Joules), the stacking sequence has very little effect on the extent of the damaged area. Indeed, for a 9 Joules impact no difference in damaged area was observed. This NDE also showed that in the case of low impact energies, this composite material helps limiting impact damaged area. However, above a threshold of 8 Joules, impact damaged area increases sharply before reaching another plateau.
Additional work would be required to analyse the effect of stacking sequence when such material is subject to higher impact energies (i.e. above 14 Joules). Also, the use of other NDT techniques such as infrared thermography, eddy currents, shearography and SQUID magnetometry could provide additional information and help in identifying low energy impact damage that was not detected with ultrasonic C-scan.
The authors would like to thank M. Matsushima from the National Aerospace Laboratory for his help in ultrasonic testing. Toray Industries Inc. is also acknowledged for kindly supplying CFRP specimens. The help of K. Hiraoka is also acknowledged.
- N. Odagiri, H. Kishi and T. Nakae, "T800H/3900-2 Toughened epoxy prepreg system: toughening concept and mechanism", Proc. of the 6th Conf. of the American Society for Composites, 43 (1991).
- N. Odagiri, H. Kishi and M. Yamashita, "Development of TORAYCA prepreg P2302 carbon fiber reinforced plastic for aicraft primary structural materials", Adv. Composite. Mater., 5, 249 (1996).
- K. Kageyama, I. Kimpara, I. Ohsawa, M. Hojo and S. Kabashima, "Mode I and mode II delamination growth of interlayer toughened carbon/epoxy (T800H/3900-2) composite system", Proc. of the Conf. on Composite Materials: Fatigue and Fracture, 5, 19 (1995).
- E. Fuoss, P.V. Straznicky and C. Poon, "Effects of stacking sequence on the impact resistance in composite laminates - Part 1: parametric study", Composite Structures, 41, 67 (1998).
Copyright © NDT.net,
/DB:Article /SO:ICAC /AU:Gros_X_E /AU:Takahashi_K /CN:JP /CT:NDT /CT:UT /CT:composite /ED:1999-03