The Marketing Minute: When Everyone Is a Thought Leader, No One Is
The Chemical Connection: More Solutions to Common PCB Etching Questions
It’s Only Common Sense: How to Demonstrate Grace Under Pressure During Tough Times
“Innovative Development of PCB Technology and Design” was the theme of the second session of the 2024 EIPC Summer Conference, June 4-5, at the European Space Centre, Noordwijk, The Netherlands. The session was moderated by Dr. Michele Stampanoni and his first speaker was Dr. Evelyne Parmentier, research and development engineer with Dyconex in Switzerland. Her topic was the functionalisation of printed circuit boards by introducing alternative metals through sputtered layers.
“There are 93 metals in the periodic table,” she began. “But traditional PCB technology is focused on just a few—mainly copper, with nickel, gold, and silver occasionally incorporated.” She asked the rhetorical question, “Why not include more of the 93, since each has interesting properties: electrical, thermoelectrical, corrosion resistant?” The audience pondered this while she compared and contrasted PCB technology with thin film technology for making circuits.
Thin film related to thicknesses mostly less than 1 micron, deposited by vacuum evaporation, magnetron sputtering, or other process methods requires specialised equipment and expertise. Metals such as copper and aluminium are commonly deposited, as well as dielectrics like silicon dioxide and some semiconductors. Typical applications are for high temperature resistance, corrosion resistance or high insulation.
She described sputtering as a deposition technique in which a solid target is bombarded by energetic particles of plasma or gas. Material is ejected from the target surface and deposited onto a substrate. Extremely fine layers can be generated, and the advantages of sputtering over other thin film deposition techniques are that materials with very high melting points are easily sputtered, sputter-deposited films have a composition close to that of the source material, and sputtered films typically have a better adhesion on the substrate than evaporated films.
It is preferable not to deposit a thin film as an overlay on top of copper but to deposit it on the substrate after etching a pattern, but before stripping the etch resist. (Her schematics explained things more clearly than my description.) A technique has been developed to overcome undesirable edge effects, whereby the desired thin-film pattern is first etched in the copper with features slightly over-size. The sputtered metal is deposited, followed by a thin deposit of copper. Then, the whole area is electroplated with copper before final photo-imaging at the correct size and etching to reveal the thin film neatly and reliably edge-connected at the bottom of the of the etched copper sidewall.
Parmentier showed a series of examples of metals and applications: gold, titanium, niobium, constantan, and nichrome. Her presentation demonstrated how thin film technology can be combined with classical PCB technology to achieve properties such as biocompatibility, built-in resistors, built-in thermocouples, built-in thermistors and superconductivity.
Joan Tourné from NextGIn technologies discussed tackling the future PCB demands. “Silicon is not the problem; it’s the interconnect. We need to re-think the challenges,” he said in his opener.
Under his heading, “Higher data rates and frequencies,” the challenges he listed were: lower loss materials with less reflections, crosstalk reduction, higher power with higher conductivity, and cooling. The challenges in assembly reliability are flatness, smaller devices, smaller pitches, and space for routing. Because product lifespan is short, green interconnects demand less materials, indirect and direct, and there is a need to reduce energy for fabrication.
His schematic cross-section of an advanced package is an effective illustration of its complexity, and the greater the number of substrate interconnects, the more potential for failure. Integrated substrates and interposers will mean less solder for other joints. He commented that HPC processors and network switches are trending toward over 8,000 solder joints and these joints caused signal disruptions and reflections. Not least, the assembly was expensive.
So what could the PCB fabricator, looking for added value, do to help? Tourné suggested re-thinking the process: integrating power solutions in the PCB with more copper for high amperage, more cooling bodies, creating cost effective high data rate solutions and perfect connections and perfect vertical transmission, with shielding for suppressing crosstalk.
And how could these attributes be achieved? Tourne advocated considering the VeCS concept of vertical conductive structures as an alternative to the traditional plated-through-hole approach. VeCS-2 shielded slot technology employs blind slots that can stop at any particular layer and offers many advantages in electrical performance as well as in a less complicated manufacturing process. He showed many examples of VeCS-2 in a range of applications, demonstrating zero stub length capability, and the benefits of extended shielding and impedance tuning. The principle also offers solutions for high-power applications, enabling the creation of wider channels for signals and ground and power, using heavy copper under the surface to connect direct from the silicon to the board on the opposite side, and connecting front-to-back using heavy slots to minimise ground shift.
Tourné emphasised that PCB shops all have the tools. There are many things that can be achieved without large investments. It’s simply a matter of thinking differently about how to use existing tools and processes.
Looking at the Environment
Session 3 focused on environmental aspects of PCB development and was moderated by Martyn Gaudion. His first speaker was Dr. Maarten Cauwe, representing IMEC, the Interuniversity Microelectronics Centre based in Belgium, whose topic was a parametric approach to quantifying the environmental impact of PCB manufacturing.
It is estimated that of the annual global carbon emission of 36 billion tons of CO2, the ICT industry is responsible for 1.2 billion tons, of which electronics manufacturing accounts for a large share. For a specific product, life-cycle assessment (LCA) is a process of evaluating the effects that a product has on the environment over the entire period of its life, thereby increasing resource-use efficiency and decreasing liabilities. Life cycle impact assessment (LCIA) is the point where the potential impacts of a product or process are identified and turned into quantifiable units. LCA of electronics is based on matching components to available datasets and scale by weight or size. Cauwe commented that because LCI data (energy, water, material, waste, etc.) for electronic components is often incomplete, outdated or not available, environmental concerns, regulatory obligations and market demands push end users to provide an impact assessment of their electronic products. Component, IC, and PCB suppliers are requested to provide LCI data for their respective manufacturing processes. Although the ultimate solution would be an eco-passport database, an efficient methodology to exchange relevant information was missing.
Cauwe described the steps required to facilitate progressing data collection via parametric LCI, from static and aggregated datasets to design-dependent production LCI models. He explained that the general principle for any system is to develop parametric LCI models for all existing manufacturing and assembly processes, and gave examples both of simple industrial products and of complex printed circuit board assemblies.
He made reference to the U.S. Environmental Protection Agency’s Design for the Environment Printed Wiring Board project and gave examples of PCB manufacturing datasets in commercial LCA databases, comparing aggregated datasets compiled from annual environmental reporting and averaged by the yearly production, with data sets based on a parameterised model that allowed adjustment of further variables. The latter represented full coverage of the manufacturing processes, relying on primary industry data, completed where necessary by secondary data. He acknowledged the collaborative contributions of industry associations iNEMI and HDPUG.
Cauwe described the methodology for building a PCB manufacturing LCI model, discussing material content, how process data is collected and the impact of the source of the process information on the calculation, noting that the LCI model does not consider production yield and assumes 100% for all steps.
He admitted that the absolute accuracy of life cycle inventory data is very difficult to determine, benchmarking with other datasets is challenging, and confidentiality of datasets prevents sharing absolute numbers. The current model is intended as a proof of principle.
The world’s first biodegradable PCB substrate was described by Steve Driver from Jiva Materials in the UK. His philosophy is, “Think differently, change what is changeable.” The problem, he said, is that existing PCB laminate technologies are unsustainable and toxic. He noted that 57 million tonnes of waste electronics are produced every year, increasing by 3-4% annually, that only 17% of EU e-waste is tracked for recycling at end-of-life, and that current PCB recycling is inefficient, energy intensive, costly and polluting. He reminded delegates that PCB production waste, in the form of panel frames, scrap and overproduction is also a contributor, especially in high-mix quick-turn shops.
Driver explained that Jiva’s Soluboard has been developed as a PCB substrate that tackles some of the environmental issues presented at end of life. Differing from the incumbent thermoset materials, it is an environmentally friendly composite material that uses natural fibres and a thermoplastic polymer. At end-of-life, the bonding can be disrupted with hot water to facilitate a more efficient and less harmful recycling process. Not intended to be a replacement for the incumbent materials, Soluboard is offered as an alternative material with comparable properties at a lower carbon footprint and a material that can be processed in PCB factories using standard machines and processes without any need for additional equipment.
He showed an array of 12 PCB materials available for the designer or OEM to choose when creating a new product, FR-4 being the most common default choice. But there are other choices when FR-4 doesn’t fit. “When would you consider an alternative? In a single use or low life expectancy application the driving factor could be ESG or regulatory take-back schemes,” he said. One of his many case-study examples was of a refrigerator board, which demonstrated Soluboard’s fitness for purpose. It was also being evaluated in drivers for LED lighting and in end-to-end tracking sensors. Jiva had proven that the industry is receptive to change. The company accepted that Soluboard will never replace all the incumbent materials, but had worked with customers who wanted an environmentally friendly alternative that fitted their application. He closed by stating, “I would urge you to have an open mind and try to change where change is possible. The impact could be huge.”
Page 1 of 2
