IeMRC Conference – Henry Ford College, University of Loughborough

Citation

Ling, J.H. (2007), "IeMRC Conference – Henry Ford College, University of Loughborough", Circuit World, Vol. 33 No. 4. https://doi.org/10.1108/cw.2007.21733dac.002

IeMRC Conference – Henry Ford College, University of Loughborough

IeMRC Conference – Henry Ford College, University of Loughborough, 5 September 2007

In the first paper, of an extremely well- attended 2nd Annual IeMRC Conference, Professor Chris Bailey from the University of Greenwich looked at developments and trends in thermal management technologies as part of his involvement in the Power Electronics Flagship project. Thermal management applies to many different industries, where heat dispersal is important; materials include graphite, diamond, and pyrolytic materials, all with thermal conductivity. The key is cost, and ways to make thermal management materials in a cost effective manner. He illustrated ways in which vapour deposition was used in the lay-up of conductive layers of graphite, and vapour-grown carbon fibres, which have five times the conductivity of copper, and also metal graphite bonding. There is metal graphite foam, giving 12,000W/m.K. and metal diamond composites, which are isotropic in conductivity,

Carbon nanotube composites offer very high-thermal conductivities, 3,000W/m.K. whilst the equally efficient but lower cost graphite nanoplatelets are relatively new. Boron nitride nanotubes have also been considered. Additionally, hyper- conductive graphite is being used, mainly in aerospace applications. Nanoscale concepts will be important in the next generation.

Passive thermal management is used in plasma TVs, as a spreader shield applied to the glass, at relatively low cost. It is also in printed circuit boards (PCBs), where it gives improved thermal conductivity with a good CTE match. For cooling, passive cooling, nowadays it is aluminium, copper and carbon that are the prime materials; in the future it may be carbon or boron nanotube composites.

Chris also looked at active soldering; describing the S-Bond system that can solder a variety of materials, including ceramics to copper, graphite to aluminium, copper to alumina, even two pieces of silica can be joined together. He also spoke on active cooling technologies, with Purdue University in the USA doing the most work in this field; there is micro-channel cooling, created by the powder extrusion process, which can give 5- 10mm channels. There is spray cooling, phase change impingement, but not yet jet impingement. Thermo-electric and thermionic devices can lower temperatures by 78 and whilst active cooling gives high heat transfer rates, it comes with complicated designs (and higher cost) and there are potential reliability problems.

Thermally induced failure still dominates in electronics, and many SMEs and large companies are active in this field; it is the use of high heat flux applications that is driving research forward.

Another collaborative project is in the field of power electronics, presented by Professor Mark Johnson from Nottingham University. The efficient and flexible control and conversion of electrical energy are studied. Power electronics is about energy efficiency and this work is looking at longer life, better performance and control, better flexibility, and lower environmental impact. What is driving power electronics? The increasing use in transmission and distribution systems, and the call for smaller, lighter and more functional mobile products. Common themes include power densities, lower EMC emissions, extreme operating environments, and the need for “plug and go” systems.

In the field of packaging, passive devices and semiconductor devices all have performance limitations, and power electronics aims to enhance the competitiveness of the UK power electronics industry through improvements to the design and manufacturing capability for high- power density. Projects cover road- mapping, reliability, thermal management and partial discharge effects, and looking at reliability and the physics of failure. There are five partners in the Power Electronics Flagship project; Oxford, Manchester, Newcastle, Greenwich and Nottingham Universities, and their report on Road Mapping will be given on 31 October at PERA, Melton Mowbray.

The power electronics presentation was added to by Paul Taylor from Dyconex, one of the partners, who said that his company was very active in R&D, manufacturing semiconductor devices, IGBT power electronics, assemblies and integrated circuits. Demand for power modules is growing at 15 per cent per annum, and research into power module manufacturing is going well.

Summing up, Mark said that the IeMRC Flagship project was now well established, with £9 million funding of the cluster, and is providing substantial advantages for UK manufacturing.

“Cost estimation for low volume long life products in electronic defence systems” was the title of a paper given by Richard Marsh. It is a project being run by the University of Bath and the University of the West of England, and funded by the IeMRC. For complex defence systems, conventional approaches are unreliable, so they needed to develop aims and objectives and a methodology to perform through life costing (TLC), giving greater detail to thereby improve the accuracy of TLC. There are four TLC standards at least, and there is a great complexity of terminology. Software evaluation was exhaustive, so they took the modular approach – with top down conceptual design, and bottom up detailed design, allowing a joint approach to this complex subject. TLC costing methods have been identified, terms, boundaries and elements have been defined, an exemplar involved in the first tier, an industrial collaborator has been selected, and a library of electronics has been established. As a result they are well on the way to providing a new approach to parametric sot prediction.

Results for chemically amplified molecular resists for e-beam lithography were given by Dr Alex Robinson of the University of Birmingham. Given that there will be a drift towards line and space patterning of 20nm by 2020, the present system of defining or patterning has many disadvantages. Etch resistance is low, there is pin holing, and this leads to pattern collapse, so molecular resists have been proposed to address the problems of polymeric resists. Triphenylenes were favoured, but as they cross link under light, (5W/cm²) this is not practical. The fullerene resists have very high-etch resistance, but require vacuum sublimation for coating. However, triphenylene resists come in many forms, and both fullerene and triphenylene resists are capable of sub 20nm patterning. The problem is sensitivity – they are too slow. However, with assistance from the IeMRC, new studies have resulted in 12nm spare patterns, and 20nm dense patterns, now down to 14nm. Further progress in stability and pattern size is anticipated.

Dr Changqing Liu from Loughborough University presented a paper on a Micro Interconnections project. Here the issue is smallness, which causes problems, especially with the joints. Each joint will have different properties, with a small number of grains with several orientations. To overcome this, the work has looked at producing a single crystal tin whisker, or a single crystal nano wire, or single crystal copper column by electro- deposition. The latter system is the best, so they are pursuing the deposition of copper columns to produce microstructures which conform to normal processing.

See, though conductive plastics were discussed by Dr Constantina Lekakou of Surrey University, where she is deeply engaged on work with transparent conductive and flexible plastics. Conductivity of 10-100V/sq. can be obtained in electrically conductive materials produced by dispersing electrically conductive nanoparticles, such as nanotubes, or nanorods, in an elastomeric polymer matrix. The application is for transparent electrodes, and in optoelectronics, and solar cells. They have worked with several polymers, nanocomposites, polyanilines, and with a non-conducting emeraldine base foam. Spin-coated nanocomposites, and nanopowders are dispersed using solvents, and once the solvent has evaporated the residue of conductive material remains within the polymer. Polythiophenes are one option for conductive transparent polymers – CNT nanocomposites are the best for medium electrical conductivity, and high relativity for superconductors. Further, work focuses on improving the dispersion of CNTs to improve transparency. There is also a proposal for the formulation of electrically conductive transparent inks for electroprinting.

Dr Andy Cobley from the Sonochemistry Centre at Coventry University gave a paper on his work to modify materials surfaces. Sonochemistry is the acoustic cavitation of a surface. He looked at the desmear process for drilled PCBs using the traditional process with permanganate etchant at high temperatures, viz 75-808C. There is a lot of rinsing, which means a lot of water, and as water is becoming increasingly metered, costs are higher than they need to be. The VOCs of the etchant have to be exhausted, and permanganates is not nice to work with, and there are high-waste treatment costs. The use of hydrofluoric acids for ceramics and glass are also not good.

Ultrasound can bring about a physical change. Micro-jetting occurs at a solid surface, the micro-jet hits the surface at 200m per second, and destroys the boundary layers, chemicals break down in an cavitation bubble, this is a physical attack on the surface, generating a scrubbing or cleaning action and the destruction of the boundary layers provides a movement of debris way away from the surface. He stated that by such a process a high Tg epoxy laminate could be sonochemically cleaned in 10min against a conventional chemical process taking 47min. It is a lean and green process, and could be useful. However, the presentation did raise some interesting questions. Dr Cobley did not elaborate on how the debris in the drilled holes would be removed, and a British company called Kerry Ultrasonics (now part of Guyson International) have been making ultrasonic cleaning systems for circuit board desmear and pre-assembly cleaning for many years now, so where the novelty was in this process remained unclear.

Dr Alan Dinsdale of the National Physical Laboratory came to tell us about European COST Action MP602. MP602 is a new cost project on high temperature lead free solders. Alan explained what COST is and how it operates, and what it plans to do. There are 27 EU member states involved in COST. This new action follows on from COST531 action on lead-free soldering, in which 45 research organisations in 17 countries had been involved. There were now comprehensive databases on lead-free soldering materials, and an atlas of microstructures of lead-free solders and solder joints. MP602 was looking at advanced solder materials of high temperature considerations for the next four years, with the main focus on replacement materials for high temperature lead-free soldering. Likely candidates include zinc-aluminium, tin- antimony; gold-tin, or zinc-tin based solders, and so far four group projects had been identified, one of which was the modelling of microstructural changes occurring in the interdiffusion zone. E-mail: alan.dinsdale@npl.co.uk

The optical PCB Flagship was described by Dr David Selviah from University College, London, who are the lead partner in a project with Xyratex, Heriot-Watt University and Loughborough University plus several other industrial partners. One of the major downsides of the conventional PCB is that copper corrupts high-speed signals, and cross talk, caused by electro-magnetic interference, is less than desirable. So this project is working on the application of optical laser beams as communications, running down optical wave guides.

He was joined by Dr David Hutt, of Loughborough University, who explained that his team are looking at wave guide formation, created by efficient excimer laser ablation and inkjet printing, and how to terminate the ends of waveguides.

Ink jet deposition was also discussed, using PMMA in solvent. He covered ink formulation, drying effects, looked at wall roughness caused by multiple droplets, and wetting or droplet spread. He explained how they might control surface wetting through control of contact angle of polymer droplet on the surface.

Professor Andy Walker from Heriot- Watt University talked about the direct laser writing of polymer waveguides. They hope to increase writing speeds and manufacturability, using a photo- polymer formulation, and working on large areas. The polymer recipe is from Excellis, and is a multifunctional acrylate polymer, with photo-initiator, and cured by UV. They use a Gaussian laser beam split into three, and the substrate moves on an XY-axis below the laser. They are planning to produce a fully operational optical PCB for the end of the year.

Professor Rachel Thomson from Loughborough University talked about a project to use a phase-field method as a modelling technique to simulate the formation and growth of intermetallics during lead-free soldering processes. They hope to create an understanding of how lead-free materials behave by modelling interactions between the solders and the substrate materials. These interactions include the formation of interfacial compounds, called intermetallics, and these can lead to joint failure. A full supply chain is in place, with modelling, design rules, layout solutions, they have a fabrication department, and can transfer this technology to a PCB manufacturer/assembler. Phase field modelling can effectively simulate the growth of IML grains, and her illustrations showed that Cu₆Sn₅ grains maybe need a third element to obtain a “scallop” like grain, which would go a long way towards preventing failure.

Professors Andrew Richardson of Lancaster University and Chris Bailey of The University of Greenwich presented a paper on “Design for manufacture methodology for SiP technology” – they have two projects running; reliability modelling, and embedded test. The focus is on electrical reliability testers, as the industry is moving towards wafer layer packaging (WLP), in fact no less than 70 per cent of WLP was applied to integrated passives in 2005. With larger WLP modules, there is a question of board level reliability during assembly flow to customer acceptance. Chris Bailey spoke on the behalf of the other partner in the project, the University of Greenwich, who are modelling to capture uncertainty. This covered design steps flow, finite element calculations, (warpage and fatigue life) and Virtual DoE generate reduced order models; Prof Bailey covered the rules of design for reliability, based on probabilities, and design for robustness. Applications include embedded health monitoring, strategies for non-electrical functions, and reliability simulation. They have also investigated underfills.

Nigel Rix spoke about the work of the Electronics Knowledge Transfer Network (KTN), which is all about disseminating knowledge throughout the electronics industry. They encourage networking, which means getting people to talk together, and then work together. There 22 KTNs. There are 10,000 British companies involved in electronics. The Electronics KTN offers networking, a capability directory, web based delivery of knowledge pool, collaboration and networking tools. Web site: www.electronics.ktn.com

“Molecular junctions – made to measure by electrodeposition” was the title of a paper given by Dr Frank Marken from the University of Bath. Here he talked about junctions between two substances being grown by electrodeposition. Their aim is to enable the cost effective production of molecular gap junctions. They also want to implement a bipotentiostatic electrochemical methodology, with junction formation by tunnel control. The work has resulted in a simple protocol, using templating agents to form molecular junctions. Liquid phase experiments will also be considered.

Dr Darren Southee, from Brunel University, spoke with engaging enthusiasm about lithographically printed integrated circuits, which are close to commercial realisation. They are working with Arjo Wiggins, Gwent Electronic Materials, and Hallmark Cards in the UK. They use offset lithography to print conductive inks, which has proved successful, and which had led to such things as RFID tags, printed transducers, humidity sensors, strain sensors, all possible by printed electronics. These “printed” circuits are now powered by voltaic cells, also printed by lithography. They produce anode and cathode inks, from zinc and carbon, and also employ manganese dioxide paste, using blotting paper as an insulator, ammonium chloride, have produced cells onto membranes, using polyart paper, and now have an integrated power source generating 6V. Shelf-life problems were traced back to the anode, and the sheet resistivity of active material ink films.

His collaborator, Dr Gareth Hay talked about ink rheology and the problems with solids content, using zinc as a conductive medium. They have now produced an ink yielding 10V, steady over 24h. This was a very interesting talk about overcoming problems with inks, a printing medium that has exercised the minds of many chemists for over 100 years.

The conference was closed by Professor Paul Conway, the IeMRC Academic Director, who brought delegates up to date with the work done by the Centre, and the work taking place between now and March 2008 to prepare for Phase 2. Much to do, but much of great interest and application has been done already, and IeMRC are to be congratulated for the untiring efforts that go into helping the country maintain technological leadership and innovation, and on holding an annual conference so that a great number of people can hear about it.

J.H. LingAssociate Editor, Circuit World.