Search

The Annual Technical Report this year is focused on infrastructure reliability. Infrastructure constitutes those things that are apparent only in their absence. We take the infrastructure for granted, assuming it will always be there. We turn on our water facet, and drinkable water has always flowed out, for most of us, most of the time. Our infrastructure is subject to environment breakages (e.g., earthquakes), accidents (e.g., dig ups of cables), sabotage, intrusion, and compromise. Also everyday component, software or system failures can bring our infrastructure down. Our global connectivity and communications, as well as our world wide distributed development and maintenance systems, increase our productivity and efficiency, but can also increase our vulnerabilities. Our critical infrastructures can be found in many places.

Today, most of the microelectronics packaging needs are met by semiconductor devices in plastic surface mount (SM) packages. Microelectronics packaging of the future will be either bare chip or chip size/scale packaging (CSP). Of the 45 billion SM packaged ICs to be manufactured in 2000, CSPs will be a small 3.4% but growing at 62% (compound annual growth rate). The use of direct bonded chip-on-board and flip chip (FC) technology for custom solutions may not match the growth of CSPs. The popcorn problem of existing plastic packages has been solved in many ways including the use of hydrophobic composite encapsulants as the best solution and thorough bake-out and storage as the long-standing practical solution. The popcorn problem which was more severe with the smaller and thinner encapsulations of CSPs is also solved with modern hydrophobic materials and new non-paddle package designs. Further, there is good evidence that reliability is not impaired even by delaminations in the bulk of the encapsulations – small delaminations being an inevitable consequence of stress relaxation following transfer moulding. CSP and FC bump joint reliability is safeguarded both by good soldering practice and by effective underfill. High reliabilities are achievable with the range of new packages built from modern materials, with random failure rates down to 10 failure units, infant mortalities controlled to low levels by six sigma manufacturing processes and wearout lifetimes exceeding 100 years even in tropical operation.

The opportunity for mutual benefit across Europe to develop low‐cost MCM technologies arose from recognition of the scientific skills and design and prototyping capabilities in organic and inorganic circuits in countries of Central Europe. As a result, the leading research institutions and small/medium‐size enterprises of Hungary, Romania and Slovenia together with relevant institutions of the UK and Belgium proposed and received approval for a European Union INCO‐Copernicus project “Cheap multichip models” to establish fast prototyping low cost multichip module (MCM) technology facilities. The project commenced in May 1997.

Plastic encapsulation is now a highly reliable method of packaging microelectronics and can be significantly more reliable than military specified “hermetic components” in field deployments of electronics in humid tropical climates. MIL883 hermeticity is an inadequate safeguard against penetration by high levels of moisture in the atmosphere. In this new work, tropical climates are analysed and shown to be significantly more severe than those safeguarded by current MIL883 standards. Field failure returns have revealed that MIL 883 hermetic CerDIP (ceramic dual-in-line) packaged ICs installed in digital switching systems failed catastrophically with a failure rate of 1755 FITs. The cause of failure was severe ingress of moisture, resulting in dewpoints up to 30°C. Alternative indigenous developments of modular digital switches for widespread rural use in India have incorporated plastic encapsulated components selected according to the criteria developed from earlier extensive and successful reliability work by British Telecommunications (BT). Such criteria include the use of HAST (highly accelerated stress technique, invented at BT Labs). The BT Labs pioneering work demonstrated that commercial transfer-moulded epoxy packaged devices from certain sources were more reliable than their hermetic counterparts. Evidence from other published pioneering work has confirmed that plastic packaging now endures as a high reliability method of packaging microelectronics. Such pioneering work is also reviewed in this paper. In the USA, the IEEE Gel Task Force followed the earlier BT Laboratories initiative and concluded that silicone gels are indeed suitable for high reliability applications. AT&T undertook comprehensive materials analyses to show that silicone gels possess remarkably useful properties to safeguard high reliability integrated circuits. CALCE, University of Maryland, conducted studies of simple logic circuits and showed that the achieved reliability in temperature climatic conditions is three times greater than that predicted for military parts. BT Labs continued its work and extended its applicability to optoelectronics. Analyses of MCM applications for satellites by the author also show that plastics are a preferred option. The evidence is that plastic encapsulation is now a very reliable and cost-effective option.

Safeguarding the reliability of electronics exposed to the severe conditions occurring in tropical climates, such as India, has not been adequately understood or dealt with in international specifications. Consequently, components and equipment supplied to such specifications have failed in such climates. The vulnerability of supposedly hermetic components has been suspected as a result of a full analysis of moisture ingress, and this has been borne out by comprehensive analyses of the actual failures of components supplied to international specifications, and the climatic conditions actually occurring in India. Understanding of reliability assessment and achievement has been significantly advanced by such analysis and has shown the necessity and applicability of the highly accelerated stress testing technique invented at British Telecom Laboratories more than two decades ago. Comparison has shown the significant cost benefit from the use of plastic encapsulated devices of proven reliability.

Technology's impact on human health has been a concern for at least 200 years. These concerns ranged from the waste and pollution due to mining and heavy manufacturing to emissions from the automobile and fossil fuel power plants. Now the proliferation of electronics products, their manufacture and disposal are posing additional threats to the fragile balance of the Earth's eco systems. Nevertheless, electronics can also be the solution to many of our current and future environmental and energy related problems. The environmental threats from electronics can be significantly reduced or eliminated through consciously environmentally responsible choice of materials, technology, design, manufacture, distribution, usage, and end-of-life disposal. In a similar way the application of electrical and electronics technologies and devices for power generation, electric and hybrid automobiles and solid-state lighting can significantly positively contribute to reducing mankind's carbon footprint and improve our environment. Thus, consciously and demonstrably, electronics technologies, devices and applications are making major contributions through the removal of toxins and pollutants; innovative materials enable reduction in use of other materials; new electronics energy sources will become the long-term major sustainable sources; solid state lighting is decreasing the consumption of energy whilst providing improved illumination. Nano-electronics products now enable more efficient use of energy in many walks of life

Electrical, Electronic and Electromechanical (EEE) components may represent only 5% of a total space system cost, but at an individual equipment level they can be 50%. This can seriously disadvantage equipment suppliers. Commercial pressures, largely due to the massive increases in demand from other sectors, are reducing the number of manufacturers interested in supplying the specialised space market. Both of these factors represent a challenge for the space industry. This paper identifies component industry trends, examines the traditional methods of achieving "space quality", addresses how the space industry can adapt to this changing scenario, looks at the options for space standardisation/qualification, challenges some established practices and, finally, considers performance specifications.

EUROPRACTICE is an European initiative on technology transfer to benefit European industry, which started in 1995. The focus was on the three technologies: Integrated Circuits (ICs), MultiChip Modules (MCMs) and Microsystems together with two supporting activities which were CAD software, and Continuing Professional Development (CPD) Training. EUROPRACTICE Training was established under the title “Training and Best Practice Service (TBPS)”. The TBPS goals are to identify the training needs of European industry and academia in ICs, MCMs and microsystems, and to identify related good calibre training programmes provided by reputable Course Providers (CPs) to match the identified training needs. TBPS has set up an extensive Course information base together with the marketing, handling and helpline infrastructure to reach those seeking training. The user friendly web site (http://cie.it.dtu.dk/tbps) has course data and the electronic Training Registration On-Line (TROL) system. Over four years, TBPS has penetrated through to many Course Providers in Europe and enabled training directly for over 900 attendees from industry, plus over 230 attendees at customised courses delivered at company premises. Where gaps in training have been identified, the TBPS has organised training at two levels: Awareness Training to raise awareness amongst novices to the field, followed by Advanced Training to help build up the skills of those who then seek in-depth knowledge. Further help is provided on a consultancy basis. This enables more rapid learning than following a conventional route (Figure 1) and has proved to be very popular.

The microelectronics markets around the world have undergone major shifts as some technologies have receded and others have gained ground1. Consequently, engineers and managers in industry must continually update their professional skills by means of continuing education, to keep up with the fast changing markets and opportunities.

The use of Electronics Technologies in Space pose interesting but manageable challenges and significant opportunities for the use of Multi-Chip Modules (MCMs) and low cost packaging technologies — whose properties offer significant advantages in space and locations involving exposure to radiation. Communication via satellites located in space require performance at high frequencies in the microwave range (delivered by silicon or GaAs MMICs- monolithic microwave integrated circuits — and controlled by MCMs), hardware technologies which are compact and lightweight and will withstand the acceleration forces during launch and manoeuvre into orbit, semiconductor device and packaging and interconnection technologies which will withstand prolonged exposure to radiation, and an extremely high reliability because of the high telecommunications traffic carried and the enormous cost of repair if a failure should occur.

With the advent of surface-mounting technology, an apparent problem that has received much attention is the potential failure of solder joints between microcomponents and printed circuit boards. One tequnique used to assess such joints is to subject the assemblies to thermal cycling over a wide temperature range. The validity of the method has, however, not been proven and there are serious reservations about its relevance when considering: the mechanical and thermal properties of solders and other materials associated with the assemblies, the temperature excursions and their rates likely to be encountered in real life, and the more probable problems due to intermittent power dissipation in surface-mounted components. Thus, thermal cycling may prove to be useful only in process development and quality testing. In the event, if the object is to subject the joint to cycling strain, then a more rapid and therefore more efficient test has been devised whereby mechanical cycling is employed to simulate the effects of thermal cycling.

A systematic evaluation of a range of plastic coatings, to identify those suitable for use on bare microcircuit dice for high reliability applications, has been conducted by materials analyses and by damp heat overstress of coated special test vehicles. Some promising junction coatings were first selected and then used in subsequent tests with added top coatings in order to evaluate a range of composite coatings. Over 4,000 hours of overstress have been accumulated from tests on some 500 test vehicles and over 15 different coatings examined. The materials analyses provided a coarse screen and roughly correlated with the damp heat test results, which proved to be more discriminating. While some junction coatings gave excellent results, it was very apparent that their performance could be severely impaired by the addition of poor top coatings. Nevertheless the outcome from this exercise has been the identification of a number of promising pairs of coatings, with the caution that evaluations with more sensitive active elements have yet to be done.

The once highly publicised Porcelain Enamelled Steel (PES) substrates seem to have disappeared from the public gaze, or have they? They certainly have not. If anything, they have consolidated their place in electronics applications and are growing in use at a remarkable pace in particular applications supplied by the major USA source, namely Ferro‐ECA.