MicroTech 2013 IMAPS-UK Annual Conference TWI, Granta Park, Abington, Cambridge, 21 March 2013

Citation

Ling, J. (2013), "MicroTech 2013 IMAPS-UK Annual Conference TWI, Granta Park, Abington, Cambridge, 21 March 2013", Microelectronics International, Vol. 30 No. 3. https://doi.org/10.1108/mi.2013.21830caa.011

MicroTech 2013 IMAPS-UK Annual Conference TWI, Granta Park, Abington, Cambridge, 21 March 2013

Article Type: Conferences and exhibitions From: Microelectronics International, Volume 30, Issue 3

Andrew Holland, the IMAPS-UK Chairman, welcomed the delegates to a sold-out event held on the campus of The Welding Institute in the spacious Granta Park, just south of the city of Cambridge.

Dr Paul Woolin is the Research Director of TWI who were hosting the conference. He is running a £45 million fund for core research projects, and is always looking for areas of collaboration – whilst the TWI activities mainly hinge around materials, welding and joining, and how things behave in service, and how long will they survive, much of the work encompasses microelectronics and he was happy to see so many people on his turf, and looked forward to meeting as many of them as he could.

The first session was chaired by Andy Longford of PandA Europe, with the first paper given by Nick Mair of Ultra Electronics entitled the UK Manufacturing is OK. We have a lot to be proud of in the UK he suggested. His company is one of 29 companies in the group, aerospace and defence primarily, operating in the CEM field. With three facilities in the UK, building electronic systems, and they also have a microelectronics branch in Porchester. It is a fast growing business, with Cambridge being their main growth point for MEMS, and 3,500 m² of manufacturing facility is now complete. The first roots of a microelectronics business in Cambridge.

The culture we live in today can be traced back to the Victorian era, the industrial revolution, which was financed by wealthy individuals who had vision and took risks, and in those days we made things, but when our customers became our competitors, we saw manufacturing slipping away to low cost areas, and not just lower costs than the UK, not just cheaper labour, but where their governments were prepared to take risks, and become involved with economic development.

What we have is versatility, we have the same time zone, the same language, our workforce is versatile and adaptable, and we are inventors – the world knows that and camps on our doorstep.

We are innovators, with strong support at the R&D end; we are great at lateral thinking, at genetic engineering, at civil engineering, electronic engineering, biochemical engineering, mechanical engineering, software engineering, semiconductor engineering, the list goes on.

Nick looked at functionality, which has rocketed, and demands more silicon with higher connectivity, but many of the obvious methods have disadvantages, but what if we could put the chip IN the board, and this something his company is working on with Stevenage Circuits, they are getting much better interconnectivity with this system. Moulded circuits, additive layer manufacturing (ALM) but the ability to make a real 3D object from CAD data is attractive. Another area is 45-axis milling, and sliding head techniques, and now the Raspberry Pi, a £25 computer made in Wales.

Ian Sturland runs the advanced technology centre at BAE SYSTEMS and took a look at the UK supply chain in MEMS devices. There is no standardisation in MEMS manufacture; it is not anything like as mature as the semiconductor industry. Development time is long slow and tedious, and it takes many years to get an idea into production. However, the market for MEMS is estimated to be worth $20 billion by 2016. Commercial success is in the hands of the few foundries who have aimed at the high volume requirements, but they are not in the UK. In the UK we have modest volumes, aimed at niche markets, such as pressure sensors, and this is what his company specialises in. In their product range they have gyroscopes and accelerometers,aerodynamic sensors and actuators, optical devices, RF devices,and smart corrosion detectors; the common factor is the cleverness, the ability,the functionality, the uniqueness, but not the volume. Having a good supply chain in the UK is enormously helpful, and by and large they find that they have these sources, complimented by very strong process equipment manufacturing base.

Tecan sent along Shaun Mason to talk about the three processes used down in Weymouth. They are Laser Cutting, Photo Chemical Etching, and Photo Electroforming. They are working with LPKF on laser-cutting stencils,creating stencils with tolerances of between 6-9 μm. Photo chemical etching is a burr and stress free process, with tight tolerances, high yields, used on a variety of metals, and suitable for tightly controlled production. There is low cost tooling, set-up and optimisation, and it is perfect for metal finishing and RFI-screening cans.

Photo-electro forming isa process of synthesizing a metal object by controlling the electrodeposition of metal passing through an electrolytic solution onto a metal or metallized form. The LIGA process was described, which can give aspect ratios of up to 12:1, with 3:1 as normal. It comprises exposure, development, electroforming, stripping and replication.

Refined advances in photo-resist technology enables the production of 3D parts where “holes” and “lands” can be produced with up to 12:1 aspect ratios. The LIGA process also provides extremely beneficial strength to size ratios – very fine features with both great accuracy and mechanical strength, which is critical for the production of parts such as high-strength micro-meshes, screens, embossing tools or micro-mould tools.

Shaun concluded with the facts that Tecan can produce Inkjet nozzles, with jets 4 μm apart, and by vacuum patterning can produce masks and stencils used for OLED manufacture, filters and mesh for medical applications and microfiltration. Also micro-parts, used in the production of optical components, with 5 μm accuracy. In any one year they will produce 100 million components.

There followed a series of short but fascinating presentations from members of the academic world.

3D micro patterning was the subject of a paper from Dr Gavin Williams of the University of Sheffield which was all about the work they have been doing with patterning on grossly non-planar substrates. Their 3D lithography technique uses computer-designed holographic masks within a custom-built exposure unit which generates non-planar light distribution which exposes a photoresist that has been conformably-deposited (by spraying or electrophoresis) over the substrate surface. Using this technique they have successfully produced cylindrical and conical antennae, undertaken vertical routing on silicon, and MEMS devices.

Miss Anne Vongerichten from UCL spoke about the work being done on electrode requirements for electric impedance recordings. They are looking for partners to join them in pursing the notion that the use of titanium foil would be preferable for the chronic implantation used in electrical impedance tomography, rather than the costly platinum-silicone electrode arrays, which also were prone to failure and poor mechanical properties during manual preparation stages. Initial work looks encouraging, and with alternative designs for 3D electrode placement, the collection of fast neural EIT data at lower cost seems possible.

Miss V. Giagva is also from UCL, their Analogue and Biomedical Electronics Group, where she is working on a project that will provide epidural electrical stimulation for paralysed patients. She described how arrays made from laser-cut platinum foil sandwiched in silicone rubber ceased to function after one week, due to cracks on the platinum tracks, noted on sites where the arrays bent sharply, or where the amount of metal changed significantly, or on the angle of the tracks. She suggested that a de-multiplexing integrated circuit embedded on the array might be the way forward in overcoming the problem.

Mark Sugden of the Wolfson School of Mechanical and Manufacturing Engineering Faculty at Loughborough University informed us that the use of mono-sized metal coated polymer based microspheres is emerging as a method of achieving more reliable and robust interconnects for BGAs and chip-scale packages. One important advantage of using such micro-spheres is that their mechanical properties can be tailored to suit the needs of the application, in addition to the relatively low usage of metals. He described how 9.8 μm ENIG particles were charged before being deposited using electrophoretic techniques to deposit them where required at the correct rate.

Stephen Taylor of the Institute of Orthopaedics and Musculoskeletal Sciences at UCL introduced us to the wonders of PEEK. PEEK is polyetheretherketone polymer used in orthopaedic implants, because of its favourable mechanical properties, and its biocompatibility. It can be injection moulded, it is inert, incompatible, radiation and chemical resistant, and has very low heat tolerance. However, it does absorb water, so it has to be sealed. Thin film coatings are barriers to moisture ingress and are being investigated as are desiccants.

Nick Donaldson from the Implanted Devices Group at the Department of Medical Physics and Bioengineering at UCL had wafer-bonded micro packages as his topic. An implantable IC package is needed for neural prosthetics; something that is very small yet can be reliable for decades in the body. He and his colleagues at UCl are exploring bonding a cap wafer to the device wafer before singulation, which exposes the pads. Thus, the chip is itself the base of the micro package and the active area of the chip is within a tiny gas-filled volume. They have been testing humidity sensors that are for4me during the CMOS process and will be inside the gas cavity and can take the device out of use should moisture penetrate.

Gilles Simon works with CEA Leti in Grenoble, France, and is involved with bio-medical packaging. He is particularly concerned with 3D silicon, which offers many opportunities, such as ultra-miniaturisation, better performance, the mixing of many functionalities and technologies, standardisation, and production in volume from microelectronics Fabs. However, there are some challenges to be faced, such as reliability, biocompatibility/in-vitro and in-vivo tests/regulations, as well as the need for new supply chains to be set up; the availability of chips in low volume, and low volume production planning. However, the market in the USA for implantable devices will be $50 billion by 2015. The need for high integration is in diagnosis – such as high performance endoscopy: miniaturization; Biopsy tools: miniaturization, and in surgery tools (catheter), smart surgery tools embedding sensors,wireless communication, and in implants -higher and new functionalities, etc.

Mr Simon emphasised that the key factor in any devices is hermeticity. Advanced packages will comprise very small sizes, they will need to be soft, ultra-thin, flexible, and will use new materials, such as silicon, metal, organic and flex. There will be more connections, more build-up structures, and there will need to be both biocompatibility and bio-stability, but most of all there needs to be hermeticity.

Tony Winster from Henkel and Sharona Sente from Henkel in Belgium delivered a paper on materials for printed electronics. These have been developed at Henkel, who mainly make adhesives and detergents, as well as encapsulants and solder pastes, and now they have materials for medical printed electronics, made with conductive inks. Here the price of silver is a major factor, so Henkel are using copper powder instead, where the copper is silver plated (SPC). They first used this in die-attach material, and the results were good, so they looked at using it in a printing ink, it proved to be very stable, they printed on various foil substrates, and with SPC a resistance of 20 milliohms/γ was achieved, and adhesion was fine, too. The cost of SPC was 1.5 times cheaper than silver products, and they have been playing around with some dynamic DSC, too, which has proved stable at 30°C for weeks and months. There is a growing need for medical diagnostics where high speed and low temperature processing is imperative. Controlling costs is therefore important, and it seems that Henkel have been successful in doing this.

Dr Ian Mayes of Nordson Dage described the role of materials testing in failure analysis. Like buildings and bridges, electronic circuits are structures, and predictable and reliable behaviour depends on understanding their properties and the strains resulting from temperature change, vibration and shock. It is the dimensional changes that occur as components heat up and cool down, will ultimately determine the reliability of a PC board. In microelectronics we encounter many materials and most types of material behaviour – the brittle fracture of semiconductors and ceramics, the fatigue of metals and polymers, the creep of solder, polymers and adhesives, and the plastic deformation of metals and polymers. Dr Mayes looked the testing of laminates, where higher temperatures needed for lead-free solder place thermal strains, and cracks can extend if the board is subjected to high strains, shock or thermal cycling, with eventual failure of connections. He very clearly and comprehensively illustrated the tests carried out on the effects of abrasion, creep,brittle fractures, pad cratering and solder fatigue.

Stewart McCracken of MCS Limited is also familiar with failure. His company thrives on it, and has developed a new process which allows deformation and smear-free cross-sectioning without obscuring the sample surface, and this process uses a broad ion beam system can be used for the non-destructive examination of wire bonds, solderable coatings and solder joints. Masked Broad Ion Beam (BIB) milling uses three large (∼1 mm) ion beams to strip (sputter) away exposed sample, where a sputter resistant shield masks half the ion beam, creates a lateral sputteringplane that strips away a layer of the sample surface.The resulting planarised surface is topographically flat and free of artefacts, and there is no deformation. By such a process can nanometre scale information be obtained, but large samples can be prepared, be they from metallic, ceramic, polymeric or composite materials. The ion beam etching reveals the interface, phase, and edge details with incredible accuracy. It is suitable for all solderable finishes, and direct linear measurement of plated layer thickness can be measured, too. His company can tackle all manner of problems, be they.

Solderability, solder joint failure, metallurgy, conformal coatings, PCB issues, black-pad, popcorning, gold content, plating thickness, device failure, EOS, ESD, corrosion, contamination, field-failure, cracking, wire bonding defects, component authenticity, and hermeticity.

This was a very-attended seminar and as the only event in the UK dedicated to micro-assembly and related technologies in the UK, it was heartening to see so many people taking advantage of not only the excellent presentations, but also the time allowed for meeting up with colleagues, and not least, many of the key industry suppliers who were there in strength. IMAPS-UK knows how to put on a good event, and this was no exception. Their membership is increasing, and they are a financially strong society, well-placed to continue their support of the microelectronics industry.

John LingAssociate Editor