packagings. The paper discusses the environmental life cycle aspects of the related technology development. Life cycle management has to provide appropriate ...
IS SMALL GREEN? LIFE CYCLE ASPECTS OF TECHNOLOGY TRENDS IN MICROELECTRONICS AND MICROSYSTEMS Karsten Schischke1, Hansjoerg Griese2 1
Berlin Center of Advanced Packaging (BeCAP), c/o Technical University of Berlin, Gustav-Meyer-Allee 25, 13355 Berlin, Germany, Phone: ++49 30 46403-156, Fax: ++49 30 46403-131 2 Berlin Center of Advanced Packaging (BeCAP), c/o Fraunhofer IZM, Gustav-Meyer-Allee 25, 13355 Berlin, Germany
Abstract - The technology development in the field of microelectronics and micro integration is the backbone of the highly innovative information and communication technology, creating products with steadily increasing functionality and completely new applications for electronics. This paper exemplifies future trends in electronics, especially ubiquitous electronics, grain-size components and merging of electronics with textiles and packagings. The paper discusses the environmental life cycle aspects of the related technology development. Life cycle management has to provide appropriate measures to guide this development. Research for advanced electronics specifically has to consider environmental aspects but faces certain constraints, which will be discussed. However, life cycle management for technology R&D faces weak databases on future technologies and vague scenarios on later use patterns – thus, further LCA methodology development is required. For a pragmatic approach to life cycle oriented eco-design in the highly complex field of electronics, screening assessment tools are preferable to time-consuming life cycle analysis. 1. INTRODUCTION Goal of the paper is to draw attention to the complexity, dynamic development, and short innovation cycles of electronics products, and the consequences for LCA. There are several fields of application of Life Cycle Analysis, among them • policy making, including preparation of environmental legislation • strategic product development • design work flow integrated eco-design • technology development for future applications of electronics Each of these fields sets certain LCA requirements, such as “real-time decision support”, use scenario development, trend analysis. 2. CASE STUDIES AND EXEMPLARY TRENDS 2.1 Personal Computers For one decade the 1993 MCC report on life cycle aspects of a computer workstation [1] was a landmark for life cycle management in the information and communication technology sector. However, nowadays this basic study is outdated and using its data neglects the immense technology progress in the ICT sector. A new study by the Technical University of Berlin and Fraunhofer IZM now calculated the current primary energy consumption of PC manufacturing [2]: Whereas the MCC report stated 2.125 kWh as primary energy consumption for manufacture of a computer configuration of the late 1980s – excluding monitor –, TU Berlin/Fraunhofer IZM report an energy consumption of 535 kWh for a 1999 PC, see figure 1. This fourfold increase in efficiency demonstrates how important up-to-date data are in the field of assessment of electronics products. Yet, this improvement does not imply a “greening of electronics” and reduction of environmental burdens in general: From 1989 to 2003 the number of globally sold PCs increased from 21 Mio to more than 150 Mio units. Thus, the efficiency benefits have been overcompensated by mass production. The main conclusion of the PC case study is:
• Up-to-date data for highly innovative electronic products is essential for life cycle management in this sector.
primary energy consumption for manufacturing one PC in MJ in kWh
sold PC world wide per year 140 Mio
MCC - report
8.000 2.000
120 Mio
7.000
6.000
100 Mio
1.500 5.000
80 Mio
4.000 1.000
60 Mio 3.000
Technical University of Berlin data
40 Mio
2.000
EU - eco-label - report (component assembly, final assembly , HDD excluded)
20 Mio
1986
1988
1990
1992
1994
1996
1998
500
1.000
2000
2002
Figure 1 – Primary Energy Consumption for PC Manufacturing Compared with Number of Units Sold World wide 2.2 Mobile Phones A main indicator of the environmental performance of advanced electronic devices is the chip area, as processing of semiconductor chips contributes significantly to the overall energy consumption throughout a product’s lifetime. Fig. 2 shows the average energy consumption per chip area multiplied with the specific chip area for several mobile phones of different manufacturers [3]. CDMA (Code Division Multiple Access) technology represents the second generation (2G) mobile communication standard, whereas CDMA2000 and W-CDMA are standards for the third generation (3G), such as UMTS. The main results from this analysis are: Innovation rates for established technologies show a steady decrease in energy consumption for manufacturing. Thus, also in this case generic LCI data for electronic products are outdated within short time cycles – if not time-dependent correction factors can be applied. However, this requires thorough understanding of the timedependency of a parameter and the underlying effects of the parameter itself. 80,0 Primary Energy Consumption ICs [kWh/device] 70,0 60,0 50,0 40,0 30,0 CDMA
W-CDMA
20,0
CDMA2000
10,0 CDMA+Analog 0,0 1996
1997
1998
1999
2000
2001
2002
2003
Figure 2 – Mobile Phones: Innovations vs. Energy
Introduction of new technologies might also cause a major change regarding environmental impacts. In the above mentioned example, the step from CDMA to W-CDMA comes along with a significant increase of energy consumption. On the other hand, the CDMA2000 as another 3G technology comparable to W-CDMA resumes the trend of decreasing power consumption for chip manufacturing. The main conclusion of the mobile phone case study is: • Life cycle impacts vary significantly from one product generation to the next; hence generic product data should incorporate a “technology development factor” for main parameters. 2.3 Major Trend: Shrinking dimensions raise production environment requirements Miniaturization in electronics promises increased functionality and decreased footprint by application of tinier structures on microchips, more bare chips assembled, and less conventionally packaged chips. However, handling of bare chips needs an extremely clean production environment to avoid yield losses. This involves energy intensive heating, ventilation and air conditioning systems (HVAC). Shrinking interconnection pitches also require advanced printed circuit boards with increased interconnection densities. Consequently board manufacturing also takes place in a clean room environment: The IPC National Roadmap [4] predicts state of the art manufacturing environments to be • Clean room class 10.000 in 2002/03, • Clean room class 1.000 in 2004/07, • Clean room class 100 in 2008/12. This will cause increased energy consumption, as a HVAC system requires per m² of clean room facility • 2.280 kWh/(m².a) for class 10.000 • 4.330 kWh/(m².a) for class 1.000 • 8.440 kWh/(m².a) for class 100 according to Tschudi [5]. This is also relevant for supply materials like process chemicals and gases: The demand for higher purity levels implies more technical efforts for chemical purification, e. g. additional energy consumption. As generic LCI data sets by now only refer to standard qualities, the data gaps are significant – especially in the case of semiconductor manufacturing. Plepys [11] researches the life cycle implications of purification processes for semiconductor chemicals. It is assumed, that the purification processes are remarkably power consuming as e.g. distillation processes are used. Latest findings and methodologies of Plepys will be presented at the Electronics Goes Green conference in Berlin, September 6-8. 2.4 Ubiquitous electronics Three examples from current research at Fraunhofer IZM illustrate the future trends in electronics [7]: • Small electronic devices, such as keyboards and displays, will be integrated into clothes, interconnected via conductive textile fibers and connected with miniaturized devices, such as cameras, PDAs, and mobile phones. • Smart tags with silicon microchips or as transistors made of polymers – so called "Polytronics" – could replace barcodes in the near future, see fig. 3. • Grain-sized electronic modules – so called "eGrains" –, including sensors, data storage, antenna, and energy supply, that interact with each other, are on the technology roadmap for the 2010s (fig. 4).
Figure 3 – Smart Tags (photos by: Fraunhofer IZM) 1999
2001
2005
2010
thinner
Standard Dimensions: 80 mm x 50 mm x 10 mm Volume: 40.000 mm³
Miniaturised Dimensions: 30 mm x 30 mm x 2 mm Volume: 1.800 mm³ Flexible, thin, laminated, folded Dimensions: 10 mm x 10 mm x 1 mm Volume: 100 mm³ Cubic 3D Integration „e-grain“ Dimensions: 4 mm x 4 mm x 2 mm Volume: 32 mm³
Figure 4 – e-grain Roadmap (active transponder) Thus, in the near future we will have a merge of ICT and “conventional products” – electronics really become ubiquitous. These technology trends need to be addressed by life cycle management, as they will influence tomorrow’s products and services significantly. On the other hand, there are a couple of obstacles for life cycle management by now: • Product follows technology: The product ideas where new technologies might be used for, are vague, e.g. application of electronics in textiles will depend on costs, acceptance, functionality and many other parameters. Consequently a life cycle is hard to analyse as long the product, and thus the life cycle, is widely unknown. • Environmental impacts will depend on efficiency of processes – which is vice versa goal of R&D: improvement of process efficiency for competitiveness. Thus, the environmental impacts are known when the technology is developed – but that’s too late to start with life cycle management. • Electronics will shrink to nano-scale; a level, where material behavior is different compared to the macro world. Possible environmental impacts have been addressed already by some studies, but scientific data is still missing to a large extend. This topic needs much more attention in the near future. Despite these obstacles and the need for research, some preliminary statements on life cycle management aspects of ubiquitous electronics are possible: • Autonomous micro systems need advanced power supply: Parallel development of efficient regenerative energy supply is essential. The amount of data to be processed will steadily increase, meaning also more energy consumption. • Software controlled self-organizing systems will replace physical interconnects: This dematerialization will result in a rebound effect – with the net balance still unknown. • Replacement of material interconnects by wireless communication means an increase in electromagnetic radiation. Health and environmental impacts of this radiation has to be analyzed more in detail for optimization of future wireless systems.
• Nano-scale products are respirable. Nano particles have to be assessed carefully with regard to health effects and/or securely bound to a matrix. • Wide-spread autonomous micro systems cause new problems in electronics recycling: Takeback will hardly be possible. Consequently, micro systems for ubiquitous applications have to be designed bio-compatible. • Electronics will be integrated into other products, affecting established recycling technologies. Example: Smart tags on paper packages. Electronics has to be compatible with established recycling paths [8]. • New technological possibilities have to be used for better efficiency of logistics, process control, environmental and health protection. Thus, ubiquitous electronics might be able to result in a net benefit for the environment – if life cycle management of these new technologies is properly done. 3. MAIN OBSTACLES FOR LCA FOR MICROELECTRONICS The unique characteristics of microelectronics and microsystems mean serious obstacles for LCA application. In short, main methodology and data acquisition improvement potentials to be tackled by future research are: • Electronics are too complex for full scale LCAs of high data quality, if detailed decision support is intended • Innovations, innovations, innovations: Time frame for LCA in electronics is extremely short, if eco-design should be supported • Lack of generic data: tens of thousands of different electronics components in use; “umbrella LCA data” might be needed, provided that data quality of such “umbrellas” is understood • Huge amount of electronic specific inputs (e.g. high purity chemicals) – another data gap • Huge amount of electronics specific outputs – impact assessments have to be complemented • Currently, toxicity assessment is a weak point of LCA methodology – but of major interest for ICT products, as the public discussion strongly focuses on toxicity issues • Further impacts – such as electro-magnetic radiation – need to be addressed by life cycle impact assessment • Modeling of disposal is very complex: electronics are composed of several hundred substances • Use patterns of electronic devices are constantly changing (new features, new applications, new lifestyle), making modeling of the use phase difficult without reliable statistic data • Rapid shifting of functionality towards integration of even more features raises obstacles concerning the definition of the “functional unit” and comparability of product systems • Global supply chains hinder availability of specific LCA data 4. SCREENING TOOLS FOR RAPID DECISION SUPPORT Life cycle management for information and communication technology devices faces serious obstacles, including complexity of products and product-service-systems, extremely short innovation cycles, and lack of specific data sets. For mid-term policy making and long term decision support Life Cycle Analysis is - within the mentioned constraints – applicable. For a real eco-design of electronic products LCA is hardly implementable into design work flows. Therefore specifically designed screening assessment tools, such as the IZM/EE-Toolbox in the electronics sector, are necessary for a pragmatic approach. These tools address significant aspects of a product life cycle instead of the whole life cycle. For a survey of screening tools for the electronics sector see Hagelüken and Schischke [12]. As stated in the explanatory memorandum of the draft EuP (energy-using products) directive "the major environmental aspects are focused around the use of materials, the consumption of
energy and the toxicity of some of the constituents” [9]. Screening tools focusing on these major environmental aspects complement the LCA methodology where a full-scale LCA is not applicable for life cycle management. Fig. 5 gives an overview of the tools of the IZM/EE-Toolbox. Further details on the methodology can be found, e.g., in [10]. life cycle focus
raw materials acquisition
environmental aspect
tool
!
energy
!
ERM (energy raw materials)
!
energy
!
EMfg (energy manufacturing)
!
toxicity
!
!
material contents / toxicity
!
!
material contents / emissions
!
!
energy
!
!
emissions
!
!
material contents / recycling feasibility
!
"
manufacturing
ProTox (process toxicity screening)
"
product
TPI (Toxic Potential Indicator)
TEP (Toxic Emissions Potential)
EUse (energy use phase)
"
end-of-life
SES (Simple Emissions Screening)
RPI (Recycling Potential Indicator)
energy ! ! ERec Figure 5 – Screening the life cycle of electronic products: The IZM/EE-Toolbox
Nowadays eco design is applied frequently in the electronics sector, but mostly as an “add on” activity or a separate case study - rarely as an integrated aspect of electronics design. The key document for eco design is the ISO Technical Report “Environmental management Integrating environmental aspects into product design and development” published in 2001 [6]. ISO/TR 14062 links the workflow for product design and development with life cycle thinking, see fig. 6. The generic model of integrating environmental aspects into the product development process generally is applicable for all kinds of products, but does not reflect the complexity of certain fields of application: Especially ISO/TR 14062 guidance for the stages conceptual, detailed design, testing / prototype does not address the design workflow for micro systems adequately. For the stage conceptual design the TR recommends application of tools, such as # “guidelines and checklists, e.g. regarding materials ecological impact, assembly/disassembly, and recycling, # manuals, e.g. describing strengths and weakness of design options, # material data bases”. Further details, where to apply which tools are not given. To make eco design feasible as an integrated procedure needs more detailed guidance: Currently research is going on at Fraunhofer IZM with support of Technical University Harburg, Hamburg, to analyse the design work flow for micro systems and to develop an integrated approach.
More specific guidance is needed to spread life cycle thinking among the product developers community. The Design for Environment Task Force of the UNEP/SETAC Life Cycle Initiative elaborates guidance documents currently.
Figure 6 - Generic model of integrating environmental aspects into the product development process according to ISO/TR 14062
5. CONCLUSIONS For the state-of-the-art in Life Cycle Analysis concerning the information and communication sector, following statements can be derived from the presented case studies: 1. Life Cycle Analysis is applicable for assessments with a broader scope and timeframe, such as policy-making and trend analysis. Main difficulties with LCA arise when short innovation cycles, rapid process improvements, rapidly changing use patterns, etc. come into effect. 2. For short-term decision making, such as “real time eco-design”, LCA methodology requires too much data to be applied efficiently. Screening tools in these applications are more efficient as they focus on specific environmental aspects – though the complete life cycle has to be kept in mind. Research and development in the field of microelectronics and micro systems has to be accompanied by life cycle management. The remaining serious methodological problems have to be addressed by future research, involving both screening and full-scale LCA methods, but also following sometimes a more pragmatic approach. The final answer to the question “is small green?” cannot be given now, although there are encouraging trends to observe. But as there are rebound effects compensating efficiency gains, life cycle thinking in the microelectronics sector still needs more attention and specific guidance. 6. REFERENCES [1] The Microelectronics and Computer Technology Corporation (MCC): Environmental Consciousness: A Strategic Competitiveness Issue for the Electronics and Computer Industry. Comprehensive Report: Analysis and Synthesis, Austin, 1993 [2] Schischke, K.; Kohlmeyer, R.; Griese, H.; Reichl, H.: Life Cycle Energy Analysis of PCs – Environmental Consequences of Lifetime Extension through Reuse, 11th LCA Case Studies Symposium, Lausanne, December 3-4, 2003
[3] Chip area data by Portelligent: Key Metrics Report on Cellular Phones and Wireless Handsets; March 2003 [4] IPC National Technology: IPC National Technology Roadmap – Electronic Interconnections - 2002/2003 Overview [5] Tschudi, W.: Energy Benchmarking in Cleanroom Facilities, LBNL, Oct. 2000 [6] ISO/TR 14062 - Environmental management - Integrating environmental aspects into product design and development, 2001 [7] Michel, B. (Ed.): Fraunhofer-Institut für Zuverlässigkeit und Mikrointegration IZM Annual Report 2002/2003, Berlin 2003 [8] Müller, J.; Griese, H.; Hagelüken, M.; Reichl, H.: Environmental Risks of Mass produced, Small and Cheap Products Regarding End-Of-Life Scenarios, Proceedings of the EcoDesign 2003 3rd International Symposium on Environmentally Conscious Design and Inverse Manufacturing, Tokyo, Japan, 2003 (in preparation) [9] Commission of the European Communities: Proposal for a directive on establishing a framework for the setting of Eco-design requirements for Energy-Using Products and amending Council Directive 92/42/EEC, Brussels, 01.08.2003 [10] Stobbe, I. et al.: IZM/EE-Toolbox – effiziente Screening-Bewertungsmethoden für ein Design umweltverträglicher elektronischer Produkte und Prozesse, in: Birkhofer, H.; Spath, D.; Winzer, P.; Müller, D.: Umweltgerechte Produktentwicklung, Kapitel 3.4.6.2, Beuth Verlag GmbH, Berlin, 2004 [11] Plepys, A.; Schischke, K.: Beyond the walls of semiconductor fabs - energy intensity of high grade chemical manufacturing, Electronics Goes Green 2004+, Berlin, September 6-8, 2004 (in preparation) [12] Hagelüken, M.; Schischke, K.: „Welcome to the Jungle“ – Survival of the Fittest Environmental Screening Indicators? Electronics Goes Green 2004+, Berlin, September 6-8, 2004 (in preparation)
