Investigating the effect of ultrasonic consolidation on shape memory alloy fibres

This research was driven by the capability of the Ultrasonic Consolidation (UC) manufacturing process to create smart metal matrix composites for use within high value engineering sectors, such as aerospace. The UC process is a hybrid additive/subtractive manufacturing process that embeds fibres into metal matrices via the exploitation of a high plastic flow, low temperature phenomenon encountered at ultrasonic frequency mechanical vibrations. The research concerned an investigation of the use of the UC process for embedding Nickel-Titanium alloy (NiTi) shape memory alloy (SMA) fibres into Aluminium (Al) matrices which could potentially be used as vibration damping structures, stress state variable structures, as well as other future smart material applications. It was hypothesised that the fibre volume fraction within a UC matrix was limited due to a reduction in foil/foil bonding, caused by increased fibre numbers, as opposed to the total level of plastic flow of the matrix material being insufficient to accommodate the increased fibre numbers. This hypothesis was tested by increasing the NiTi SMA fibre volume fraction, within an Al 3003 (T0) metal matrix, beyond that of previous UC work. The metal matrix and the fibre matrix interface of these samples was then microscopically analysed and the overall UC sample integrity was tested via mechanical peel testing. It was found that a fibre volume fraction of ~9.8% volume (30 X Ø100 µm SMA fibres) was the maximum achievable using an Al 3003 (T0) 100 µm thick foil material and conventional UC fibre embedding. A revised hypothesis was postulated that the interlaminar structure created during UC was affected by the process parameters used. This interlaminar structure contained areas of un-bonded foil and the increase of UC process parameters would reduce this area of un-bonded foil. Areas of this interlaminar structure were also thought to have undergone grain refinement which would have created harder material areas within the structure. It was suggested that maximum plastic flow of the matrix had not been reached and thus the use of larger diameter NiTi SMA fibres were embedded to increase the effective SMA fibre volume fraction within Al 3003 (T0) UC samples. It was suggested that the embedding of SMA fibres via UC had an abrasive effect on the SMA fibres and the SMA fibres had an effect on the Al 3003 (T0) microstructure. It was further suggested that the activation of UC embedded SMA fibres would reduce the strength of the fibre/matrix interface and the matrix would impede the ability of the SMA fibres to contract causing a forceful interaction at the fibre to matrix interface, weakening the UC structure. The investigation to test the revised hypothesis was broken down into three sections of study. Study 1 was a methodology to determine the characteristics of the interlaminar surface created via UC and how this surface affected the nature of the consolidated sample. UC samples of Al 3003 (T0) were manufactured using a range of process parameters. The analysis involved optical microscopy to determine the UC weld density and the interlaminar surface; mechanical peel testing to quantify the interlaminar bond strength; white light interferometry to measure the interlaminar surface profile and microhardness measurements to determine the hardness of the interlaminar material. Study 2 was a methodology to allow the analysis of the microstructural and mechanical interactions at the fibre/matrix interface, post-UC. Al 3003 (T0) samples were manufactured via UC using a range of process parameters with various NiTi SMA fibre diameters embedded. The analysis involved using mechanical peel testing to determine the interlaminar bond strength; optical microscopy to determine the level of fibre encapsulation; scanning electron microscopy and focussed ion beam analysis to analyse the fibre and matrix grain structures and microscopic interactions. Study 3 was a methodology to investigate the fibre usage as would be expected from envisaged applications of an SMA containing metal matrix composite. Samples were manufactured using a range of UC process parameters with various SMA fibre diameters embedded and the embedded SMA fibres were subjected to different extension/contraction cycle numbers. The analysis involved using mechanical peel testing to determine the interlaminar bond strength; optical microscopy to determine the level of fibre encapsulation and the interlaminar effect of fibre activation; fibre pullout testing to measurement the strength of the fibre/matrix interaction and load rate testing of the activated SMA fibres to monitor performance. The interlaminar surface was found to affect the strength and density of interlaminar bonding during the UC process and the use of higher UC process parameters affected this interlaminar structure. Levels of un-bonded material were found within the interlaminar structure and these levels were found to decrease with increasing sonotrode amplitude and pressure with reducing speed. It was suggested that a specifically texture sonotrode could be developed to modify the interlaminar structure to the requirements of the intended sample application. The measurement of the interlaminar material hardness was unsuccessful and future work would likely require a different methodology to measuring this. The work identified a grain refining effect of the embedded SMA fibres on the Al 3003 (T0) matrix material, (grain sizes were reduced from ~15 µm to <1 µm within 20 µm of the SMA fibres), as well as localised damage caused by the UC process to the SMA fibres. The performance of the activated SMA fibres established that this damage did not prohibit the ability of the SMAs to function however the compressive nature of the Al 3003 (T0) matrix was identified as reducing the ability of the SMA fibres to contract. Additionally it was found that the activation of SMA fibres within an Al 3003 (T0) matrix resulted in a reduction of the fibre/matrix interface strength which allowed fibres to be pulled from the composite with greater ease (a loss of ~80% was encountered after a single activation and extension cycle). The use of larger SMA fibre diameters allowed for the fibre volume fraction to be increased however the activation of these SMA fibres had a delaminating effect on the Al 3003 (T0) structure due to the size of the radial expansion of the SMA fibre. The work furthered the understanding of the effect of UC on SMA fibres and highlighted the importance of the interlaminar surface in UC and that to increase the SMA fibre volume fraction to a useable level (25-50%) then an alternative fibre embedding method within UC is required. The fibre/matrix interface interactions during SMA activation have implications in the ability of UC SMA embedded smart metal matrix composites to function successfully due to weakening effects on fibre matrix interface strength and the ability to achieve SMA fibre activation within the metal matrix.