CitationDownload as .RIS
Emerald Group Publishing Limited
Copyright © 2012, Emerald Group Publishing Limited
IeMRC 6th Annual Conference Loughborough University 21 September 2011
Article Type: Exhibitions and conferences From: Soldering & Surface Mount Technology, Volume 24, Issue 1
Held at the excellent Holywell Conference Centre on the university campus, this year’s event was a popular venue for many delegates who filled the hall to a point of repletion. Once they could be persuaded into the room, it was Professor Paul Conway, Academic Director of IeMRC who welcomed the many. He provided a quick snapshot of what the IeMRC is about, which is funding research in 27 institutes, through 90 investigators, with 42 research associates, and, happily, funding runs through to 2015.
The first speaker was Professor James Morris (Figure 8) of Portland State University, who had delayed his long-overdue return to Oregon just so that he could address this conference. We were lucky to have him, as his paper was a reflection on nanoparticles – particles of less than 100 nm. Their properties were described, including Ostwald ripening, where the particles do not come into contact; and sintering, a thermal diffusion effect, and coalescence, based on random charging. His laboratory had successfully printed nanoparticle interconnects using Ag nanoparticle paste applied by sintering at room temperature on a flexible substrate, and found that 60 °C was the optimum temperature. With nanoparticle sintering, surface treatment of the particles is needed to avoid agglomeration. Silver nanoparticles are used to fill vias on inner layers. Carbon nanotubes have a major advantage of conductivity and high-current densities that they will carry, and a CTE of 0. Using them as interconnects has advantages, too, with no resistance, and as TSVs, (through silicon vias) they are still a concept, but makes 3D interconnect possible.
Hazel Assender (Figure 9) from the Department of Materials at the University of Oxford spoke about roll-to-roll manufacture of organic transistors for low-cost circuits, a flagship project that is now 15 months old, and they are working with their fellow universities in Leeds, Manchester, and Bangor. Is the concept realisable? So far, so good. They are using evaporation AS opposed to sputtering, which is slower, and looking at a belt speed of 50 m/m. Circuit design has to be taken into account when printing these transistors, and at that sort of speed. Certain materials have been developed especially for this project, including semiconductor and gate insulator layers. The equipment is specific; at Oxford they have a vacuum web coater with a web width of 350 mm, and a web speed of 5 m/s. The substrate is PET, with a modified surface to smooth it off, and they then evaporate the gate metal, applying an acrylic insulator, and an organic semiconductor on top for a possible interlayer.
They have started working with DNTT material which has better stability, yet retains the high-mobility needed; depositing the insulator where the material is atomised into a liquid, which is cured with e-beam. But with electron-beam curing you do have to anneal, and you cannot do that in line as the web speed is too high. So, they have tried plasma curing, and this works, curing in a single pass as the polymer is deposited. Their molecular semiconductors being produced with plasma curing are showing reasonably stable performance in air over ten cycles, but they need to be stored in a vacuum. Improved pentacene morphology gives better devices, they have found, with greater mobility and better off-current behaviour. Device encapsulation could be fully integrated into the production process, she suggested, but overall the picture painted reflected good progress.
Tin whiskers occupy the attention of Geoff Wilcox (Figure 10) of the Department of Materials at Loughborough University, and he came to relate his observations on their formation. What are tin whiskers? There are three areas of growth – filaments, nodules, or odd-shaped eruptions (OSE). They vary from 1 mm to a world record whisker of 37.5 mm long, and are a serious problem. A growing whisker can cause shorting, by arcing, or falling off and shorting out the PCB. The general consensus is that stress is the root cause, either residual electroplating stress in a deposit, or an externally applied mechanical stress or CTE mismatch, or even the intermetallic formation. He spoke about the WHISKERMIT programme; a recent IeMRC funded project examining two main strategies to combat such types of growth. They are looking at the tin-plating process in terms of modification to the plating process which produces fewer whiskers. Bright tin is a known culprit. They are also looking at surfaces most likely to whisker, and then putting a conformal coating over to prevent it forming. No single variable controls whisker growth. It is lead that is showing the most reduced effect on tin, as it stays in formation in terms of composition, and with 8 per cent lead content, the whisker growth is very much smaller. Conformal coatings have been looked at, of various resin types, and with alkyd polymeric plating, there were no tin whiskers after 50 days. But, alas, after a further period of time, whiskers did break through. More work continues.
Professor Marc Desmulliez (Figure 11) of Heriot-Watt University fame is working with his team there on additive technologies, specifically the patterning of fine metallic tracks onto flexible substrates. It is the development of economical, high through-put processes that will allow low-cost electronics to meet the mass markets. His work is on such low capital cost, reel-to-reel, systems, with an additive process that includes thermally induced deposition using lasers, with surface modification of the substrate, and ink-jet printing, as well as non-vapour phase direct metallisation, but with substrate surface modification for subsequent electroless deposition. The base substrate is Kapton, immersed in a solution of KOH. This is rinsed with de-ionised water and dried and stored. Silver, platinum or copper can be used as an additive metal.
In one process an MPEG coating is applied to reduce the silver ions, and then they use a laser beam to UV pattern at 690 J · cm−2 at a process speed of 0.5 m/s. Electroless plating of the printed tracks follows. Thus, far they have achieved 50-100 μm lines and spaces but down to 15 μm will be possible. The need for annealing exists, but this is not part of any high-speed process.
The second approach was with chemical reduction, using a water developable photoresist from MacDermid, the process being UV exposure, develop, ion reduction, resist strip and etch.
The view from an end-user perspective was given in a paper from Odette Valentine (Figure 12) from Brunel University, the topic being soft-wear. She works on design for the use of meaningful technology in wearable technology. This includes conductive circuits woven into textiles, or knitted with conductive yarns, or encapsulating micro devices directly into the material itself, and ones incorporating light technologies.
Her research looks at the relationships between designers and consumers, with wearable technology. The initial approach has been to look at user needs and user acceptance of products – which, thanks to social media, can be available online, on the peg, or even in the shop window. The channels of expression that wearable systems could deliver are numerous, the effect function, i.e. body performance, mental stimulation; communication – to whom and when and why. The potential for incorporating non-body systems, energy harvesting by the body, generated for the body; Odette warmed to her presentation in such a way that one imagined all sorts of fascinating fashion statements being available, something really new for the catwalk.
The morning sessions were rounded off very neatly by Professor Yannis Vardaxoglou (Figure 13) from Loughborough University whose task is exploring ways in which flexible fabric electronics can be employed for megahertz frequency communications. Essentially this is an IeMRC-funded project, running over three years, on the creation of a textile antenna and its associated electronics, and the integration of this antenna into textiles. A joint project between industry and academia, other partners include Nottingham Trent University, Defence Marine Systems, and Cash’s.
Normal antennas are bulky, and have wires sticking out, but a woven antenna has considerable advantages, and not only in the military field. The products need to be low cost, there has to be good conductivity of the yarns, good definition, repeatability in manufacture, and interconnection with traditional systems. Minimising body interference, consistency of performance in harsh environments and scalability to cost-effective mass-manufacture are some of the criteria involved.
Given the knowledge of the dangers facing our troops in Afghanistan from IEDs, the paper given by Robert Blue (Figure 14) of Strathclyde University was of particular resonance. He is working on new organic semiconductor base sensors for nitro compounds, essentially sensors designed for detecting explosives through vapours. Sensors to detect the presence of explosives need to be low cost, small, portable and available. Explosives nowadays are detected by either spectroscopic methods, or by metal detectors, or by trained canine teams. Whilst the latter are highly effective, dogs cost a lot to train, and can only be used for a medium period. Metal detectors require slow and thorough scanning of the ground, and the operators are exposed to attack.
Their research involves mass sensors, optical sensors, and capacitance sensors. Sensors are made from different polymers applied by inkjet onto a substrate, and each polymer demonstrates different properties and sensitivity. They can be made more nitro sensitive. The sensor measures capacitance change, the bigger the response the more likely the detection is of a nitro-propane compound at less than 100 ppm. The arrival of such smart sensors as this on the battlefield will doubtless save many lives.
Professor Bill Drury (Figure 15) works with Bristol University as a consultant, he has 30 years experience in the field of power electronics, and came to inform the assembly on how they could be employed in the low carbon UK economy. What is power electronics? Is it any good? Are we any good at it? Power electronics is managing power/energy. Power electronics is for industrial processes or control and energy efficiency; it is used in transportation, in building environment and access. The UK is good in this area, with companies such as Alstom Grid, Control Techniques, and Converteam, etc. His own company, Emerson-Control Techniques, is a leading manufacturer of motor control and power conversion technology for commercial and industrial applications, whose products are used in the most demanding applications requiring performance, reliability and energy efficiency, and specialise in high-efficiency power conversion systems for renewable energy applications.
The UK has a strong supply chain, and world class universities. Power electronics is a £70 billion direct global market, and many of the Top 20 global manufacturers on our shores, as is his own company. Six out of ten international companies have design and/or manufacturing here in the UK, so it is a real success story. Whilst reminding the delegates that the UK electronics industry is worth £356 billion to the economy, it is the matter of skill shortages that plagues us. Only 3,000 people applied to universities this year to read electrical and electronic engineering, and it is a fact that 33 per cent of engineering graduates take non-engineering related jobs. That is our biggest challenge. He will be involved with launching a Power Electronics strategy in October in London.
“Thermosonic-adhesive flip-chip assembly” was the topic of the final paper of the day, from Professor Andrew Holmes (Figure 16) of Imperial College. His team are working on an adhesive-based flip chip for flip chip assembly, in partnership with Sony, Henkel, and GE Aviation Systems. Describing the conventional ways in which chips are bonded, Dr Holmes compared adhesive based to solder based, and ACA packaging, which is cheap but unreliable. So what he has been doing is combine the best of thermosonic bonding with adhesive and soldering bonding. There are many advantages, and hardly any disadvantages; in fact only one, which is limited chip size.
In the IeMRC project, they introduced the thermosonic bond step into the ACA NCA assembly in order to replace the mechanical contacts by metal-metal thermosonic bonds. Thus, far they have first trials with chip to glass substrate and chip to flex substrate, and TA bonding materials with a process for ACA with embedded gold pillars. Professor Holmes described the IR laser heating of the pick-up tool for the heating of the chip, with 30 W on each side. Results of initial trials are most encouraging. Further work will continue with existing NCF material (non-conductive film), and they are looking at making their own bonding adhesives, and then running reliability tests with GE Aviation Systems.
Bringing the day to an end, Professor Martin Goosey, Industrial Director of the IeMRC reflected upon the excellence of the papers which we had listened to, and thanked the speakers for their time and professional counsel. He reminded everyone that on the subject of power electronics, there is a two-day conference – iPOWER – being held at the University of Warwick on 30 November and 1 December, in conjunction with IeMRC, iMAPS-UK and NMI (www.nmi.eventwax.com/ipower/register).
Attendance this year at this excellent IeMRC Annual Conference was very much higher than last year, which may be timely support of the view of the IMF that economic recovery for the UK will be technology led. What the Department of Education might well be doing is making it possible for young men and women to be the many E&E engineers that we will need to assist in this. Doubtless IeMRC are making this urgent need known at the right levels.
John LingAssociate Editor
1. International Monetary Fund, not to be confused with the Institute of Metal Finishing, who would doubtless concur anyway.