Final Report Summary - INTERFACE (Interfacial Engineering in Copper Carbon Nanofibre Composites (Cu-C MMCs) for high thermally loaded applications)
Carbon-based nanomaterials (carbon nanofibres or carbon nanotubes) are promising candidate materials for reinforcing metallic matrices such as Al, Cu or Ti. The attractive mechanical properties (high mechanical strength) combined with interesting thermophysical properties (low coefficient of thermal expansion and high thermal conductivity) are of interest for many applications. However, most of the time, the beneficial properties of these nanomaterials are only exploited to a few percent. To take advantage of the excellent thermal and mechanical properties of carbon nanofibres or carbon nanotubes several challenges have to be solved at the same time.
The main goal of INTERFACE has been to address these critical areas and to work on possible solutions to overcome these limitations and fabricate a carbon nanofibre reinforced Cu composite combining high thermal conductivity (> 400 W/mK) and a CTE in the range of 8- 12 ppm/K. Such a material, which can be easily machined, has been tested as a heat sink in laser bars, LEDs and high power modules.
As an important outcome of the project, UBO, in collaboration with the University of Reims, has developed novel metrology techniques to assess the interfacial contact resistance in flat systems, using various photothermal methods. The task has resulted specially challenging for the glassy carbon used as substrate material in the flat model systems, but investigations carried out in parallel on diamond substrates have shown that the high sensitivity of the methodology. By investigating different metal-substrate systems with different measurement devices available at Bochum (from 10 Hz to 10kHz) and Reims (100 to 100kHz), supported by phase / amplitude simulations for Cu films on glassy carbon substrates and diamond substrates, it was possible to establish the measurement sensitivity of the thermal contact resistance (TCR). For the glassy carbon substrate, TCRs below 10 -7 m2W/K could not be measured, while for diamond substrates, the technique can measured TCRs as low as 10-8 m2 W/K could be found.
Motivated by the relatively low thermal conductivities obtained with respect to expectations, a thorough investigation was carried out by UCAM. Generally, in the literature, the view is that this is caused by the poor interfacial thermal conductance. While this effect is no doubt important, it was observed that the VGCFs lose their graphitic structure during processing and that this is the predominant reason for the fall in thermal conductivity to that of copper containing an equivalent volume fraction of pores. Extensive transmission electron microscopy has been carried out to characterise this behaviour. A technique based on Raman spectroscopy has also been developed to quantify this effect. This allows a study of many of the important details much more rapidly and this is being used elsewhere in the programme. It is found that similar behaviour occurs in K1100 8 µm carbon fibres. However, being larger the amorphous region does not penetrate throughout the fibre. Although some amorphisation occurs on heating, the primary effect appears to be associated with the wetting aids that are added to improve the interface conductance diffusing into the nanofibre.
The main goal of INTERFACE has been to address these critical areas and to work on possible solutions to overcome these limitations and fabricate a carbon nanofibre reinforced Cu composite combining high thermal conductivity (> 400 W/mK) and a CTE in the range of 8- 12 ppm/K. Such a material, which can be easily machined, has been tested as a heat sink in laser bars, LEDs and high power modules.
As an important outcome of the project, UBO, in collaboration with the University of Reims, has developed novel metrology techniques to assess the interfacial contact resistance in flat systems, using various photothermal methods. The task has resulted specially challenging for the glassy carbon used as substrate material in the flat model systems, but investigations carried out in parallel on diamond substrates have shown that the high sensitivity of the methodology. By investigating different metal-substrate systems with different measurement devices available at Bochum (from 10 Hz to 10kHz) and Reims (100 to 100kHz), supported by phase / amplitude simulations for Cu films on glassy carbon substrates and diamond substrates, it was possible to establish the measurement sensitivity of the thermal contact resistance (TCR). For the glassy carbon substrate, TCRs below 10 -7 m2W/K could not be measured, while for diamond substrates, the technique can measured TCRs as low as 10-8 m2 W/K could be found.
Motivated by the relatively low thermal conductivities obtained with respect to expectations, a thorough investigation was carried out by UCAM. Generally, in the literature, the view is that this is caused by the poor interfacial thermal conductance. While this effect is no doubt important, it was observed that the VGCFs lose their graphitic structure during processing and that this is the predominant reason for the fall in thermal conductivity to that of copper containing an equivalent volume fraction of pores. Extensive transmission electron microscopy has been carried out to characterise this behaviour. A technique based on Raman spectroscopy has also been developed to quantify this effect. This allows a study of many of the important details much more rapidly and this is being used elsewhere in the programme. It is found that similar behaviour occurs in K1100 8 µm carbon fibres. However, being larger the amorphous region does not penetrate throughout the fibre. Although some amorphisation occurs on heating, the primary effect appears to be associated with the wetting aids that are added to improve the interface conductance diffusing into the nanofibre.
