research priorities - inemithor.inemi.org/webdownload/ri/inemi_2005_research...•...

2005Research Priorities

2005 iNEMI Research Priorities

In order to help ensure the ongoing competitiveness of the electronics industry, iNEMI continues to focus on improving the manufacturing capability of its member companies. Because the infrastructure required to efficiently provision the customer base is so broad and diverse, the founders of iNEMI wisely established a planning methodology to insure that its members could focus on high impact areas that can truly make a difference in the marketplace. At its fundamental level, that process includes technology roadmapping and technology deployment. While we continue to refine that process to improve the results, the underlying principals remain unchanged:

1. Create industry roadmaps by drawing upon the expertise of a broad cross section of individuals from industry, academia, and government. The results of this work are open to the electronics industry worldwide.

2. Identify the major areas that iNEMI will focus on based on need, participation, and ability to make a business impact.

3. Conduct a gap analysis on these major areas that identifies the major challenges and opportunities facing the industry.

4. Create five-year plans for the major areas that identify the projects and activities deemed necessary to close the identified gaps. These plans become the basis for the formation of iNEMI projects.

5. Develop a ten year vision of research needs to insure a vibrant, innovative electronics industry by prioritizing the research needs identified in the iNEMI Roadmap. The prioritized research needs and options are the core of this document and are being distributed as resource materials for the industry, research institutes, and funding agencies

The 2004 iNEMI Roadmap identified a number of key technology and business challenges facing the electronics industry over the next decade. Forecasts indicate that most of today’s key semiconductor and electronic packaging technologies are not capable of meeting the needs of the industry in 2015. If these challenges are not promptly addressed with innovative solutions, the continued viability of the electronics manufacturing industry will weaken over the next decade.

With the downsizing of industrial research it is critical to encourage academic and institutional R&D centers to focus on technology needs identified in the iNEMI Roadmaps. Recent concern in high cost regions over the loss of manufacturing capabilities has stimulated discussion that our industry must place greater emphasis on innovation and the development of disruptive technologies. iNEMI is committed to improving the roadmapping process to better identify disruptive technologies and research priorities.

On behalf of the iNEMI Board of Directors, I congratulate all who are helping to create this Research Priorities document. Your contributions will assure that we continue to focus on those areas that can leverage our industrial base for the benefit of member companies as well as the entire electronics industry.

Sincerely,

Jim McElroy Executive Director and CEO

National Electronics Manufacturing Initiative, Inc.

2


Table of Contents

INTRODUCTION ............................................................................................................... 7

Background..................................................................................................................................................................... 7

Overview ......................................................................................................................................................................... 8

Summary ......................................................................................................................................................................... 8

CHAPTER 1: PRODUCT TECHNOLOGY NEEDS BY PRODUCT SECTOR.................. 9

Portable/ Consumer Product Sector ............................................................................................................................. 9

Office / Large Business Systems Product Sector.......................................................................................................... 9

Netcom Systems Product Sector.................................................................................................................................. 10

System in Package (SiP) Product Sector .................................................................................................................... 11

Medical Product Sector................................................................................................................................................ 11

Automotive Product Sector.......................................................................................................................................... 12

Defense / Aerospace Product Sector ........................................................................................................................... 13

CHAPTER 2: RESEARCH PRIORITIES SUMMARIZED BY RESEARCH AREA.......... 14

Manufacturing Processes............................................................................................................................................. 14 Top Research Priorities.............................................................................................................................................. 14 Additional Research Needs........................................................................................................................................ 14

System Integration........................................................................................................................................................ 14

- Information Integration............................................................................................................................................. 14 Top Research Priorities.............................................................................................................................................. 14

-Technology Integration............................................................................................................................................... 15 Top Research Priorities.............................................................................................................................................. 15 Additional Research Needs........................................................................................................................................ 15

Energy and the Environment ...................................................................................................................................... 15 Top Research Priorities.............................................................................................................................................. 15

Materials & Reliability................................................................................................................................................. 15 Top Research Priorities.............................................................................................................................................. 15 Additional Research Needs........................................................................................................................................ 15

Design ............................................................................................................................................................................ 16 Top Research Priorities.............................................................................................................................................. 16 Additional Research Needs........................................................................................................................................ 16

3


CHAPTER 3 - SIGNIFICANT ISSUES FROM ROADMAP ............................................. 17

Strategic Gaps............................................................................................................................................................... 17 Active Device Technology ........................................................................................................................................ 17 Thermal Management................................................................................................................................................ 17 Increased Communications Bandwidth ..................................................................................................................... 17 Next Generation Packaging Technology ................................................................................................................... 17 Design and Simulation Tools..................................................................................................................................... 17 Sustainability Metrics ................................................................................................................................................ 17

Evolutionary Gaps........................................................................................................................................................ 17 Manufacturing Challenges from Miniaturization ...................................................................................................... 18 Materials Reliability .................................................................................................................................................. 18

CHAPTER 4 - OPTIONS FOR INNOVATION ................................................................. 19

Active Device Technology ............................................................................................................................................ 19 1. Implementation of advanced, non-classical CMOS devices with enhanced drive current .................................... 19 2. Identification, selection and implementation of advanced devices (beyond CMOS) and new system architectures................................................................................................................................................................................... 19 3. Printable active devices ......................................................................................................................................... 19

Thermal Management.................................................................................................................................................. 20

Increased Communications Bandwidth...................................................................................................................... 20

Next Generation Packaging Technology .................................................................................................................... 20 1. Innovative packaging for giga-function system in package (SiP).......................................................................... 20 2. Materials and reliability ......................................................................................................................................... 20 3. Sensors................................................................................................................................................................... 21

Design and Simulation Tools ....................................................................................................................................... 22

Sustainability Metrics .................................................................................................................................................. 22

Nanotechnology ............................................................................................................................................................ 22

APPENDIX A: 2004 INEMI ROADMAP OVERVIEW ...................................................... 25

Introduction .................................................................................................................................................................. 25

iNEMI Roadmap Process............................................................................................................................................. 25 Portable/Consumer Product Sector Drivers ............................................................................................................... 26 Office/Large Business Product Sector Drivers .......................................................................................................... 26 Netcom Product Sector Drivers ................................................................................................................................. 27 Automotive Product Sector Drivers........................................................................................................................... 27 Defense/Aerospace Product Sector Drivers............................................................................................................... 27 SIP Product Sector Drivers ........................................................................................................................................ 27 Medical Product Sector Drivers................................................................................................................................. 28

APPENDIX B: RESEARCH PRIORITIES INDEXED BY RESEARCH AREA AND TWG......................................................................................................................................... 29

4


Manufacturing Processes............................................................................................................................................. 29

System Integration........................................................................................................................................................ 30

Energy and the Environment ...................................................................................................................................... 30

Materials & Reliability................................................................................................................................................. 30

Design ............................................................................................................................................................................ 32

APPENDIX C: R&D NEEDS SECTIONS FROM 2004 INEMI ROADMAP...................... 33

Board Assembly............................................................................................................................................................ 33

Connectors .................................................................................................................................................................... 33

Energy Storage Systems............................................................................................................................................... 33

Environmentally Conscious Electronics ..................................................................................................................... 34

Final Assembly.............................................................................................................................................................. 34

Interconnect Substrates - Ceramic.............................................................................................................................. 34

Interconnect Substrates - Organic .............................................................................................................................. 34

Mass Data Storage........................................................................................................................................................ 35 NAND Flash Storage (Semiconductor Storage) ........................................................................................................ 36

Modeling, Simulation, and Design Tools .................................................................................................................... 37 Tools and Methodology Research and Development Needs in the Manufacturing Systems/Supply Chain Management Areas: ................................................................................................................................................... 37 Emerging Areas for Electronics Packaging ............................................................................................................... 37

Optoelectronics ............................................................................................................................................................. 40 Device Technology.................................................................................................................................................... 40 Optoelectronics Packaging (JISSO Level 2).............................................................................................................. 40 Optoelectronic Subsystems (JISSO Level 3&4)........................................................................................................ 40

Packaging ...................................................................................................................................................................... 42 Materials .................................................................................................................................................................... 42 Technology ................................................................................................................................................................ 42 Recommendations on Potential Alternative Technologies ........................................................................................ 43

Modeling and Simulation............................................................................................................................................. 44 Research and Development Needs............................................................................................................................. 44

Passive Components ..................................................................................................................................................... 45

Product Lifecycle Information Management ............................................................................................................. 46

Rf Components & Subsystems .................................................................................................................................... 48 Recommendations on Priorities and Alternative Technologies ................................................................................. 48

5


Sensors........................................................................................................................................................................... 49

Test, Inspection, and Measurement ............................................................................................................................ 50

Thermal Management.................................................................................................................................................. 51 Research Needs.......................................................................................................................................................... 51 Recommendations ..................................................................................................................................................... 52

APPENDIX D: EXTRACT OF COMMENTS FROM THE 2004 INEMI ROADMAP RELEVANT TO NANOTECHNOLOGY........................................................................... 54

Design Technologies ..................................................................................................................................................... 54

Manufacturing Technologies ....................................................................................................................................... 54

Component/Subsystem Technologies.......................................................................................................................... 54

iNEMI Technical Projects............................................................................................................................................ 54

Materials Development ................................................................................................................................................ 54

Technology Development............................................................................................................................................. 55

APPENDIX E: INEMI OVERVIEW................................................................................... 56

iNEMI History .............................................................................................................................................................. 56

iNEMI Mission.............................................................................................................................................................. 56

iNEMI Organization .................................................................................................................................................... 56

iNEMI Organization Chart ......................................................................................................................................... 57

iNEMI Deliverables ...................................................................................................................................................... 57

Standards ...................................................................................................................................................................... 57

Further Information..................................................................................................................................................... 58

APPENDIX F: GLOSSARY ............................................................................................. 59

APPENDIX G: CONTRIBUTORS.................................................................................... 62 NOTE: This document is copy protected. Please contact the iNEMI secretariat at 703-834-0330 or [email protected] for authorization to use text or graphics.

6

mailto:[email protected]


Introduction

Background The iNEMI Research Priorities document presents the consensus on R&D needs identified in the 2004 iNEMI Roadmap. The 2004 Roadmaps were developed by nineteen Technology Working Groups (TWGs), in response to inputs from representatives of OEMs in seven product sectors or Product Emulator Groups (PEGs). The nineteen TWGs can be classified into five categories:

• Business Technologies – Product Lifecycle Information Management (PLIM) TWG

• Design Technologies - Modeling, Simulation, and Design; Thermal Management; and Environmentally Conscious Electronics TWGs

• Manufacturing Technologies - Board Assembly; Final Assembly; and Test, Inspection, and Measurement TWGs

• Component/Subsystem Technologies - Semiconductor Technology; Packaging; Interconnection Substrates – Organic and Ceramic; Passive and RF Components; Optoelectronics; Displays; Mass Data Storage; Energy Storage Systems; Sensors; and Connectors TWGs

The 2004 iNEMI Roadmap consists of over 1200 pages, with contributions from more than 470 participants representing over 220 companies, universities, and associations from 11 countries. Further details on the roadmapping process and contents are contained in Appendix A: 2004 iNEMI Roadmap Overview.

The iNEMI Research Priorities document is intended to serve as a resource for all who are tasked with directing R&D funding and execution. Research needs are developed for seven product sectors:

• Consumer/portable,

• Office/large business systems,

• Automotive,

• Aerospace/defense,

• Medical,

• Netcom,

• System in Package (SiP)

We believe that these sectors are the major drivers of technology and the associated R&D needs. The technology needs for each of these sectors are summarized in the next chapter.

The R&D needs identified in the 2004 iNEMI Technology Roadmap are organized into five categories:

• Manufacturing Processes;

• Energy & Environment;

• Systems Integration;

• Materials & Reliability;

• Design.

The iNEMI process is a bottoms up Delphi process relying on numerous technology experts to give their vision of the technology needs that must be developed to meet their view of the products of the future. The roadmapping process does not explicitly identify disruptive technologies, but by identifying needs, particularly those for which there are no known solutions that meet the performance and cost requirements, members of the iNEMI roadmapping team implicitly identify areas for innovation and the utilization of disruptive technologies. As an example, the first NEMI Roadmap in 1994 determined that the introduction of area array packaging created a need for a new substrate technology. Recent concern in high cost regions over the loss of manufacturing capabilities to China has stimulated

7


discussion that we must place greater emphasis on innovation and the development of disruptive technologies to ensure the continued growth of the global electronics industry. iNEMI is committed to improving the roadmapping process to better identify disruptive technologies, find matches to current needs, and help anticipate new applications and products.

Overview The first chapter highlights the technology needs for each of the product sectors.

The second chapter presents the Research Committee’s consensus of the top five priorities in each research area. Additional research needs are listed, but in no specific order. Further detail on research priorities is provided in Appendix B, which indexes research priorities by the five research areas and the seventeen TWGs. The seventeen Roadmap Chapters' sections referencing R & D needs are included in Appendix C in their entirety. The complete chapters can be accessed in the 2004 iNEMI Roadmap available through the iNEMI Headquarters website (www.inemi.org).

The third chapter highlights Significant Technology Issues from the 2004 Roadmap:

• Potential end of commercially viable CMOS scaling

• Manufacturing challenges from miniaturization

• Design and simulation tools

• Increased communication bandwidth.

• Advanced materials and materials reliability

• Continued cost, size, and performance expectations

The fourth chapter examines how newly discovered phenomena may be exploited in future technologies and could lead to meeting the R&D needs defined in the iNEMI Research Priorities. In particular, this section describes possible future areas in nanotechnology resulting from the current federal R&D research funding investment.

As we move beyond the digital convergence of electronic products, we anticipate the merger of micro and nano chemical, mechanical, and biological sensors with micro and nano electronics for disruptive innovations in many areas. In some cases the disruptive technologies may also find application by being embedded in conventional product embodiments. As an example, nano-particle fillers may enhance select properties of existing polymeric materials. These applications will likely result in new opportunities to extend the life of current materials and manufacturing infrastructure, enabling them to deliver enhanced device or component functionality. Breakthroughs may take the form of disruptive technologies that supplant existing technologies. Examples of these are quantum computing systems, molecular electronics, and spintronics replacing CMOS semiconductor technology. Others may be radically new applications, such as sensor and drug delivery systems that detect emerging disease in the body or treat existing diseases. Such personal healthcare systems would give an added dimension to the iNEMI consumer and portable product sectors.

Summary It is only through effective prioritization of limited R&D programs that the electronics industry will maintain the technology leadership that it has enjoyed for the past fifty years. This 2005 iNEMI Priorities document has been prepared to aid this prioritization, facilitating more focused investments and an improved rate of return. With a steady stream of research results to harvest, the electronics industry can continue to enjoy growth and prosperity driven by our society’s adoption of break-through products that increase productivity and improve lifestyles.

The 2004 Roadmap has clearly identified strategic technology issues that must be addressed over the next decade. In addition to publishing the iNEMI Research Priorities, iNEMI sponsored an Innovation Leadership Forum to develop a top-down vision of the evolution of electronic applications against the strategic technology issues, to provide focus areas for future coordinated R&D between the government, industry and academia. The results of this Forum will be presented in a separate document. The following chapters and appendices served as resource material for the forum.

8


Chapter 1: Product Technology Needs by Product Sector This chapter provides a synopsis of the technology needs by product sector as developed for the 2004 iNEMI Roadmap. Experts from major OEMs described the market changes that they anticipated for their industry and the technical needs that would be required to meet the market expectations.

Portable/ Consumer Product Sector As the name implies, portable devices are meant to be carried on the person, battery powered, and designed to be as small and light as is technically and commercially possible. Typical products are mobile phones, palmtop computers, PDA’s (personal digital assistants), and wireless email or short messaging devices. Portable and consumer electronics production are expected to reach $183Bn in 2004, and increase at an average rate of 5.1% per year to reach $223Bn by 2008. The market is driven by advancements in display technologies, digital image capture, video compression, data transfer, and next-generation digital recording media. Mobile phone shipments are also on the rise. After two lackluster years, mobile phone shipments increased by 21% in 2003 to reach record shipments of 499M units. Asia accounts for about 48% of global production, followed by Europe with 21%, Japan with 17%, and the Americas with 14% share. About 25% of assembly value is expected to be outsourced in 2004.

Although audio players, camcorders, digital cameras and medical electronic devices such as glucose monitors are included in this sector, the domination of wireless devices in this category over-shadows the overall analysis. Wireless has come to be seen by the design and marketing community as a sure way to enhance the functionality and market desirability of the entire range of hand held products. This has resulted in the development of Bluetooth, 802.11 and 3G mobile phone wireless communications standards to support design efforts. Additionally, the convergence of camera, music, GPS, TV, video, etc into mobile phones is masking the impact of these individual units in our analysis.

As more functionality continues to be incorporated into hand held products, the need intensifies to establish a more thorough communication process at the design architecture level. The past decade was a period of focus on the supply chain and developing partnerships to support an agile and asset light structure. The present landscape suggests that this business level cycle will be repeated except with respect to the design chain. The design partnerships required will be between enterprises with specialized design expertise in order to support an agile and low cost product design process. The emerging trend is tilted towards modular sub-system component suppliers and ODM manufacturing to bring these building blocks together. The push is towards a System on Chip device to meet major functional needs. MEMs and sensors will be key to many electronic products over the next 15 years. Switching, filtering, signal generation for wireless applications, numerous sensing applications for personal health and environmental monitoring are just a few of the early examples of MEMs now finding their way to commercial success.

Office / Large Business Systems Product Sector The Office / Large Business System Product Sector (OLBSPS) incorporates both the back office servers and desktop/ notebook PCs. Both sub-segments are affected by the same technology drivers, the only differences are the priorities assigned to the drivers. Specifically, the main technology driver for the desktop office products is the cost, while for the backend servers and notebooks it is the cooling and power management of the system. Even though the backend servers are characterized by very high contact counts, high clock frequency, high reliability, high power, and scalability, this sub-sector is experiencing continued price pressure from the Office Systems sub-sector. This pressure creates the need for this sub-sector to adopt technology that provides product differentiation through significant cost performance improvements or features that reduce system complexity for the user. As a result, it is expected that the design of ASICs (Application Specific ICs) and packages will differentiate the high performance products from those of the office systems’ products. On the other hand, cost reduction will continue to drive the Office System product sub-sector.

Worldwide production of computers and office equipment is expected to reach $379Bn in 2004, and grow at an average rate of 6.2% per year to reach $483Bn in 2008. This is the largest segment of the electronics industry, accounting for about 37% of overall production, and is driven by both business and individual consumer spending. Asia accounts for the majority of computer and office equipment regional production—41%. This region is followed by North America, Europe, and Japan with 34%, 16%, and 9% share, respectively. Thirty-one percent of assembly value is outsourced to EMS companies.

9


The end of the traditional semiconductor scaling that looms large on the tactical horizon is generating significant reverberations to the package approaches and structures of all Office and Large Business systems. The first consequence is the gradual but certain reduction of emphasis on the frequency of microprocessor metric, and the corresponding increase in importance of the system’s throughput metric. This shift in the system’s performance metric generates an increased demand for higher bandwidth to and from the microprocessor in any system. Current designs have already utilized 3-6 Gbps data rates and designs in the near future are expected to increase this bandwidth capability to 10-12 Gbps for electrical interconnects. In both Office and large business sub-sectors circuit design workarounds to achieve this goal will be introduced prior to development of more expensive package technologies or the introduction of all together different interconnect technology like optical interconnects.

Another consequence of the potential demise of the traditional scaling of the semiconductors is the increased need for improved cooling and operating junction temperature reduction due to the expected large leakage currents and the corresponding increase in chip power. Processor chip operating power from 150- 300 Watts is expected in the tactical future in order to maintain acceptable processor chip performance. The difficulty with such levels of power dissipation will be even more prevalent in the desktop and mobile devices due to their limited space available for the cooling of their microprocessors. It is widely recognized that technology advancements alone cannot achieve the required improvements, and the actual logic design of the microprocessors has to be enhanced to help attain this goal.

Netcom Systems Product Sector The telecommunication and data communication industry’s next generation hardware requirements are collectively represented by a Product Emulator Group called NETCOM (Network, Telecommunication, and Data Communication). This product emulator group spans a hardware product portfolio from consumer access to long-haul transport and addresses system capacities ranging from kilobits per second to terabits per second. Despite the breadth of the product portfolio being addressed here, common themes for next generation development have emerged. Of most significance is the need to develop common design tools, manufacturing and test capabilities, and component specifications that can be leveraged across the communications hardware industry. These needs reflect the inevitable maturation of a relatively young industry, and advancements in those areas are viewed as essential for enabling the future success and return to prosperity in the communications industry.

Global production of netcom equipment is expected to be $180Bn in 2004—about 18% of the electronics industry. Driven primarily by Internet use, this segment is expected to increase at an average rate of 6.3% per year to reach $230Bn in 2008. In 2004, approximately 39% of communications system assembly value is expected to be outsourced.

This industry provides the backbone over which global communication (voice and data) are made possible. It is fundamental to the global economy and developments in this industry are continuously changing peoples’ way of life. Challenging requirements exist for those companies intending to serve the communications industry. Namely, communication hardware is expected to always be operating and always accessible to the end customer, independent of how many users are simultaneously placing demand on the network. The end customer expects connectivity to anywhere in the globe. Finally, the end customer has a finite willingness to pay for these services, and generally has come to expect 24/7 access and usage of the network with continuously improving access speeds for a flat monthly fee. The industry has responded to the global communication needs with a diverse set of technologies: wireless, wired over copper, and wired over optical. Standards regulating the usage and deployment of these fundamental technologies vary by region across the globe and even within a country. Significant trends exist in the industry, towards convergence of the requirements and needs of the telecommunication and data communication infrastructure. Opportunities for converging data and voice networks will likely dominate next-generation developments in the global communications industry.

An uncertainty in this segment is what bandwidth of communications will be needed in different applications and for each application, i.e. which technology will be cost effective: wireless, copper, or optical. The packaging of high speed and high-density optical and electrical components is an important cost contributor to the component and the system. Continued development in packaging technologies, materials, and design is essential in order to allow Netcom to meet its next generation requirements. Efficient transportation and management of data and communication traffic within a system is an increasingly important subject, and at the heart of this is the continued development of electronic and optical interconnection technologies at the chip level, at the board-to-backplane level, at the board-to-board level, and at the board-to-customer interface.

10


System in Package (SiP) Product Sector A SiP is defined as a combination of one or more semiconductors or passive devices integrated into an industry standard semiconductor package. The broadest adoption of SiP has been for stacked memory/logic devices and small modules (used to integrate mixed signal devices) and passives for mobile phone applications. Both these markets have driven large enough demand to establish a broad base of suppliers and a very cost competitive market over the last four years. Numerous concepts for 3D packaging are now starting to emerge and this area is presently a growing focus area for research to help address the performance and functional integration challenges of future systems. In 2004, 1.89Bn SiPs were assembled. By 2008, this number is expected to reach 3.25Bn, growing at an average rate of about 12% per year. Altogether, about 90% of SiP products are currently consumed in mobile phones.

System in Package (SiP) technology has rapidly evolved from a specialty technology used in a very narrow set of markets to a broad market base, high volume technology that has a wide-ranging impact on the electronics market. As with most emerging markets, there remain a number of critical infrastructure issues to be resolved to improve market penetration, cost structure, reliability, and performance. These include the need for low cost higher density substrates, high-speed simulation tools for electrical and mechanical analysis, wafer level packaging, lower cost assembly equipment, and improved materials for encapsulation.

It is very apparent that a new generation of low-cost packaging technology will be required during the next decade. This new generation will need to interface with the interconnect structures of the new nano-devices, provide the required thermal cooling and electrical performance, and be compatible with a number of devices and applications including rf, optical, sensors, MEMs, and biological. Thus a packaging technology that can support mechanical, acoustic, thermal, chemical, seismic, environmental, biological, sensors, and optical and rf communications will be required. This next generation packaging technology will require materials with enhanced properties and advanced manufacturing processes to address the rapidly decreasing dimensions and cost expectations.

Medical Product Sector The medical product sector group encompasses implantable medical devices, information technology used for patients’ records, medical diagnostic tools, and monitoring devices.

The United States health care system is currently the largest sector of the gross domestic product and is increasing as those born between 1946 and 1964 (baby boomers) approach retirement age. There will be roughly 1.2 billion retirees world wide by 2012; the largest group in history. Even within the United States there will be only 3 workers for every one retiree in that timeframe. According to the Bureau of Labor Statistics, the first and third fastest growing employment sectors are nursing (59% growth rate) and physician’s assistants (49% growth rate). These two careers lead to the need for information technology workers (47% growth rate) since roughly one third of the national expense of healthcare is related to the movement of medical information and billing. Increasing the role of electronics technology through medical sensing and patient monitoring is one solution that could deliver the needed healthcare while curbing ever increasing costs.

Research in the medical electronics industry is emerging as two separate areas; monitoring and implantable devices, and biological and silicon integration. Monitoring and implantable research is seen as one of the growing areas of research in the short term. In this model of data gathering the implantable contains a small radio. This radio will signal a cellular phone which in turn will collect data or forward it onward to caregivers or medical personnel. These types of devices require research to determine a material suitable for implantation that will allow the appropriate radio signal to reach the receiver. Research to promote greater life in implantable or wearable sensors is needed. Power scavenging is one strategy being investigated and relies on the thermal energy of the patient’s body (up to 30 micro Watts/cm2) or mechanical motion of the patient’s body (up to 10 micro Watts/cm2). Powering sensors in this manner is needed to extend the battery life and reduce the need for higher capacity batteries. Research in this area has been focused on using MEM’s (Micro Electromechanical Systems), Mechanical Technology, and Pelletier type devices.

Biological and silicon integration is best explained as establishing the interconnection between the biologic device and the silicon. One example is the recently announced growth of neurons on silicon by the non-profit Belgium research consortium IMEC. In this research, neurons are grown on silicon to make a device that enables the transistor to monitor brain activity of a patient with epilepsy. Once the sensor detects the onset of a seizure, the silicon transistors counteract the brain’s erratic electrical impulses thereby curbing the severity of or averting the seizure. Other research in this field conducted at the University of Arizona has calcium phosphate backed sensors that are implanted on the

11


surface of bone. The calcium phosphate stimulates the growth of oestoblasts and oestocasts, producing a sensor that becomes an integral component of the structure of the bone. These two examples are only the beginning of the interconnection between the biological and electrical worlds. With the tools of nano-technology and design, one could easily foresee the interaction of silicon devices with muscle tissue to reconstruct the neurological damage associated with spinal cord injury or muscular dystrophy. A greater focus in this research area is needed to make progress toward solving many of the crippling diseases currently impacting the third world.

In the Medical Products Group there is no greater opportunity for development than that presented by the issue of total stack integration driven by roughly one-third of the national health care budget going to IT management of the data associated with patient care. Even with $300 billion dollars as a market, no single infrastructure is established such that data may go seamlessly from any in-vivo sensor to recording device to a patient’s records. Even the movement of patient records from one section of a hospital to another may cross two systems without a compatible means of moving the data seamlessly.

Automotive Product Sector The main factor that distinguishes the Automotive Product Sector from the other iNEMI Product Sectors is the vehicular environment in which the products must perform reliably. Many of the attributes, such as cost, density, and components, overlap into the other products. Increasing density is important for these applications due to cost, size, and weight reductions. The assembly and manufacturing/test equipment requirements are also critical to ensure that the products meet reliability requirements. Automotive applications are extremely cost sensitive and therefore require cost targets similar to those of consumer products. The automotive electronics industry is approximately 7% of global electronics production, or $65 billion in 2004. Growth is expected to increase about 4% per year through 2008 due to an increase in average electronic content per vehicle. The Automotive Product Sector must adapt other technologies to meet the high temperature, environmental, and reliability requirements cost effectively.

The major challenges for automotive products are cost effectiveness and reliability. Reliability requirements have changed dramatically, especially for the automotive electronics in the harsh environment. The harsh environment is fast becoming harsher, with attendant demands on increased reliability. The infrastructure to support the requirements comes partly from the other product sectors and needs to be adapted with agility and modification. There are four major challenges currently facing automotive harsh-environment electronics:

1) The automotive manufacturers are offering extended warranties to consumers for 100,000 miles/10 years or even more. This has necessitated increased reliability requirements. However, correlation information between field data and screening tests is difficult to obtain and almost non-existent. To meet this challenge, reliability test stipulations have become stringent and time consuming; having an adverse effect on the electronic product development cycle, while fast-to-market has become the key to winning.

2) The arrival of HEV (Hybrid Electric Vehicles) for fuel economy and reduced emissions is creating new opportunities for automotive electronics. Many automotive customers are offering HEV vehicles today and others have already announced plans for HEV equipped vehicles in the future. These vehicles will have higher voltage systems (42V to 500V) for key components, keeping current levels low for efficient thermal management. HEV’s use an AC motor (with a small internal combustion engine) to provide acceptable torque and horsepower with improved fuel economy. These systems also have continuously variable transmissions, regenerative brakes, and nickel-metal hydride battery packs.

3) The third challenge is a paradigm shift from the practice of considering automotive electronics and consumer electronics as separate compartments. The automotive electronics market is growing in many dimensions. In addition to the traditional engine management and comfort and entertainment systems of the past, there is a growing market for office and entertainment systems that rival those in an office or home. The addition of these new dimensions to the electronics market is being refereed to as the convergence of automotive electronics and consumer electronics. This shift is already underway with consumer products designed to withstand the operating extremes of the passenger compartment (-40 C to 85 C) so that they can be integrated into vehicles.

4) The fourth challenge and a market discriminator is related to environmental concerns. The ability to respond to environmental legislation and to customer specifications requiring the elimination of hazardous materials and recyclability will be mandatory. Most automotive electronics are exempt from the EU Directives on Waste Electrical and Electronic Equipment (WEEE) and Restriction of Hazardous Materials (RoHS). The products that are under these

12


directives are consumer electronics that are aftermarket products. Vehicle manufactures are asking that new business proposals for the 2007 and 2008 time frame be compatible with these directives.

For Automotive products, some notable initiatives that are emerging as we progress into the first decade of the new millennium:

MEMS (Micro-electromechanical Systems) are finding increased applications for automotive sensor products (accelerometers, yaw rate, pressure, etc.).

The onset of the hybrid electronic vehicle initiatives are requiring automotive grade power electronics components for high current (100A to 1000A) and high voltage (42V to 1000V) applications.

The current analog sensor and actuator architecture keeps the peripheral devices simple, but requires multiple wires to each device. The use of digital sensors would eliminate the analog signal processing components in the engine controller and would simplify the polling requirements of the microprocessor.

As the size of most electronic components shrink, the functional modules’ dimensions are now limited by the automotive connector’s size. To go to smaller electronics with high-speed data busses, new connectors will have to be developed. The factors that play a key role in the marketplace for the automotive sector are Cost, Size, and Complexity. The marketplace is motivating the drive to reduce each of these areas. The winner will be the one who masters such reduction.

Defense / Aerospace Product Sector The main factors that distinguish the Harsh Environment – Aerospace/Defense Product Emulator from the other iNEMI product emulators is the environment in which the products must perform, the volume of products demanded, and the period of time these products are expected to perform and be supported. Aerospace/defense products require flawless performance on demand, in a multitude of rugged environments and must sustain this performance over long periods of continuous service. These environments include space, air, sea, and ground; both manned and unmanned. These products demand leading edge performance that is typically only offered in the latest commercial-off-the-shelf (COTS) technologies and products.

Defense/aerospace products must perform reliably in manned and unmanned space, missiles, fighter aircraft, helicopters, ships, ground vehicles, etc. Many of the product’s attributes, such as component and PWB technology, overlap with the other emulators. Increasing density is important for the aerospace/defense applications because of the resultant size, weight, and cost reductions. Military and space products are typically in the upper cost, high reliability, high density, and low volume categories. Military electronics are only a small portion of the electronics industry in relationship to computers, communications, and consumer electronics. The challenge for the aerospace / defense emulator is to adapt other emulators’ technologies to meet it’s performance requirements but to modify (ruggedize) the designs to meet the thermal, vibration, humidity, salt fog, and other environmental and reliability requirements associated with DoD platforms - in a cost effective manner.

As we move forward, the factors that play a key role in the market place for the aerospace/ defense products are maximizing performance and reliability, while at the same time reducing size, weight, power, and cost. The winners will be the ones who master such reductions. Accordingly, these factors are the drivers behind all that is addressed in this chapter.

13


Chapter 2: Research Priorities Summarized by Research Area Over eighty topics were recommended as research needs in the 2004 iNEMI Roadmap. Each of these topics is listed in Appendix B, indexed first by research area and then by the roadmap chapter in which it appears. The Research Committee attempted to group these. The following lists are the Research Committee’s consensus of the top 5 research priorities in each research area in order of importance, while the remaining priorities are listed in no specific order.

Manufacturing Processes Top Research Priorities

1. Manufacturing processes that support rapid miniaturization • Development of automated printing, dispensing, placement, and rework equipment compatible with

the rapid miniaturization of components • A new approach to organic substrate fabrication addressing needs for increases in density, reduced

process variability, improved electrical performance, and a radical reduction in cost • An effective optical/electronic design and manufacturing concept which increases wiring density

while reducing cost 2. Significant cost reduction in manufacturing processes

• Organic substrates • Optical interconnect • Ceramic substrates. • Testing/inspection (with appropriate process feedback/ yield improvement)

3. Development of new testing technologies to improve yield, and reduce test cost. Possibilities include: • Electromagnetic Signature • Acoustic Microscopy • Photoelectric Effect • E Beam Test Systems • Thermal Imaging

4. In-circuit electrical and optical test technologies that can be incorporated into the build process 5. Improved packaging manufacturing processes for semiconductors, SiPs, RF components, and RF MEMs

• Increased pick and place speed for die attach, flip chip, and SMT • Improved accuracy and reduced cost per placement for die attach, flip chip, and SMT • Improved cutting processes for package singulation reducing cost/damage, increase speed

Additional Research Needs

• Improved final assembly processes for the electronics industry • Self/passive Opto- Electronic part alignment • Development and deployment of surface micromachining technologies for sensors

System Integration Information Integration

As the design/supply chains become more distributed in today’s outsourced environment, the ability to manage these global processes puts increased demands on targeted information flow, resource utilization/management, and decision making. At the same time, the industry excesses of the late nineties have created an environment characterized by more conservative investments, demanding better accountability for ROI in the IT area.

Top Research Priorities 1. IT infrastructure research is needed to assure fast and secure transfer of increasingly larger amounts of data

across the supply chain 2. Research is needed to develop tools for meeting environmental compliance regulations – cost effectively and

securely

14


4. PLIM security research is needed to guarantee that only the data that is needed is supplied to the appropriate supply chain nodes

3. Improved planning/PLIM methodologies that deal with the ever changing market needs over the life cycle of a product and optimize the resources of the entire system (from concept to end-of-life)

Technology Integration

As we continue to combine disparate technologies into product architectures, a number of integration challenges face the industry. These limit our ability to improve cost, size, and reliability.

Top Research Priorities 1. OE ICs, integrated electrical-optical devices with multi-wavelength sources, modulators, detectors, etc, on a

device 2. Robust and low cost out-of-plane optical coupling, e.g. surface mounted device to embedded waveguide 1. Semiconductor lasers with shorter wavelengths and optical lens configurations with higher numerical

apertures for optical storage devices 2. Development of Low Voltage MEMS: Innovative MEMs structure designs, monolithic integration with ICs 3. Low cost, high reliability, dirt tolerant, optical interconnects


• High frequency VCSELs, 20 GHz and above • Optical self test modules • High speed Analog/Digital converters with higher resolution (14 bits)

Energy and the Environment Top Research Priorities

1. Development of good scientific methodologies to assess the environmental impact of materials and potential trade-offs of alternatives

2. Development of cost-effective, energy efficient power sources 3. Development of a common, meaningful, straightforward definition of sustainability

Materials & Reliability Top Research Priorities

1. New interconnect and packaging technologies deploying nanomaterials supporting decreased pitch and increased frequencies

2. Low cost, higher thermal conductivity packaging materials, such as adhesives, thermal pastes, and thermal spreaders, need to be developed for use in products ranging from high performance computers to automotive applications

3. Improved materials for energy sources and energy storage systems 4. Next generation of lower cost solder alloys 5. Reliable materials systems for first level of interconnect between new device technology and new substrates,

including laminate materials with higher density, competitive cost, improved dimensional stability, planarity, low moisture absorption, and low warpage


• Second level under-fill technologies compatible with board assembly processes • Improved materials for embedded resistors, capacitors, and inductors • Capabilities for embedding non-discrete in substrates during build-up process

15


• Materials, processes, and design for perpendicular and patterned recording media, tunneling magneto resistive heads (TMR), and heat-assisted magnetic recording (HAMR)

• Low cost, SS OE devices, tunable • Nanostructures for OE modules • Development of “optical solders” • Combination of materials and fabrication research to support development of monolithically integrated optics

and electronics that take advantage of electronics infrastructure • Development of photonic crystals or other materials that are amenable to size scale reduction (compared to

Moore's Law for electronics) • Medium power, low loss, high selectivity filters are a cost, weight, and size bottleneck to consumer and

portable products • New materials optimized for spintronics with regard to electrical, delamination, moisture issues • Emerging technologies such as holey fibers, photonic band gap materials, and new semiconductor material

based devices and solutions, hold promise of providing alternative methods of transmitting and controlling optical signals in ways that may be commercially important in the future.

• Die attach solder for Tj >200C • Wirebond materials that enable 20 micron pitch without wire sweep and barrier metals for Cu wirebond pads

to reduce intermetallics • Aluminum capacitors and film capacitors lagged behind ceramic and tantalum capacitors in their readiness for

surface mount, and now lag in their readiness for 260°C maximum reflow • Realization of the potential for embedded sensors will require development of miniaturized sensor elements,

integrated control systems, and micro-actuators that can all be interconnected in a single package with a small form factor

• Wider selection of biocompatible materials for packaging of biosensors backed by long-term reliability and safety data

• Miniaturization is especially critical for implantable devices, and one of the keys to achieving this is the availability of ultra-small, stand-alone power sources with long life. Although continued research into miniaturized fuel cells is required, this area may benefit from breakthrough developments in nanotechnology for energy storage devices.

Design Top Research Priorities

1. Thermo-mechanical models for nanoscale level technology such as experimental tools capable of measuring electrical & thermal and mechanics phenomena/material properties at smaller scales, for which scale dependent algorithms will be needed that have the ability to shift scales

2. Co-design of mechanical, thermal, and electrical performance of the entire chip, package, and associated heat removal structures

3. Simulations (thermal and mechanical) for optimizing hybrid lamination schemes for electrical and optical needs

4. Mechanical and reliability modeling 5. Improved and integrated design tools for emerging technologies like embedded passives, nanotechnology,

and optoelectronics PWBs, including low cost, fine line substrate and PWB routing technologies enabling denser package I/O and multi-physics simulations (e.g. Joule heating in sub 50 nm interconnects, electro-chemical phenomena in bio-MEMS devices)


• Thermal and thermo-fluid simulation • Integrated mechanical, electrical and thermal package/product design • Integrated design and simulation tools needed for RF modules and devices • Modeling & simulation of spin as electrons move through different materials; spin relaxation modeling

16


Chapter 3 - Significant Issues from Roadmap

Strategic Gaps Six strategic gaps have been identified from the 2005 iNEMI gap analysis:

Active Device Technology The predicted end of traditional CMOS semiconductor scaling is generating significant reverberations in approaches to and structures of computing systems. The first consequence is the gradual but certain reduction of emphasis on the microprocessor frequency metric and the corresponding increase in importance of the system’s throughput metric.

Thermal Management Another consequence of the expected demise of the traditional scaling of semiconductors is the increased need for improved cooling and operating junction temperature reduction due to large leakage currents and increase in chip power.

Increased Communications Bandwidth The third consequence of the end of the traditional semiconductor and the gradual but certain reduction of emphasis on the frequency of microprocessor metric is an increased demand for higher bandwidth to and from the microprocessor in any system. At some point, circa 2010-2015, the limits of electrical transmission in high performance systems may be reached. Optical systems may provide part of the solution, particularly if optical ICs are developed.

Next Generation Packaging Technology Every score of years the electronics industry has introduced a new generation of packaging technology which has taken a decade to develop. It is apparent that a new generation of low-cost packaging technology will be required during the next decade. This new generation will need to interface with the interconnect structures of the new nano-devices, provide the required thermal cooling and electrical performance, and be compatible with a number of devices and applications including rf, optical, sensors, MEMs, and biological. This next generation packaging technology will require materials with enhanced properties and advanced manufacturing processes to address the rapidly decreasing dimensions and cost expectations.

Design and Simulation Tools The lack of tools for mechanical, electrical, thermal, environmental, optical co-design from the device to the system level is delaying the introduction of new technology. Significant research and development is needed both in tools and techniques for effectively partitioning the design and simulation. The new simulation areas that need research (Opto, Nano-Scale / Spintronics) will become critical after 2007. Simulations related to Nanotechnology could have a large impact on the semiconductor and the electronic materials industry. It is clear that a combined business and quantum-mechanics perspective is required for nanotechnology to be successful. Proper funding of simulation tools will be critical to nanotechnology growth and funding will be contingent upon future business conditions.

Sustainability Metrics The concept of sustainability is being widely adopted and promoted by companies throughout the electronics supply chain as a core tenet and measure of performance of their business operations, largely in response to heightened interest by stakeholders in corporate social and environmental responsibility. The electronics industry would benefit greatly from a more unified, coherent, and meaningful approach to sustainability reporting.

Evolutionary Gaps A number of gaps have been identified which will require significant R&D over the next decade to keep technology abreast of market needs. Consumers are pushing for increased functionality, smaller form factor, and less weight in their portable devices, but at reduced cost. To meet these cost constraints, manufacturers have pushed component suppliers to integrate more functionality into their parts. The net result is a significant reduction in total part count for an equivalent function. Electronic component suppliers are looking to utilize embedded passive components, systems

17


in package, system on chip, or any other means to densely pack ICs with increased functionality. There are two dominant gaps that are caused by the evolutionary development of increased functionality and density of packaging; manufacturing challenges from miniaturization and materials reliability.

Manufacturing Challenges from Miniaturization In many SiP modules today, SMT is dominated by Pb-free 0201 attachment, pushing the limits of solder, components, and process equipment. Continued improvements in Pb-free solder pastes that provide more consistent reflow and ease of cleaning are critical. For SiP applications in particular (where die attach and wire-bonding are performed after SMT mounting), a clean solder process is required to assure good adhesion and reliable bonds.

It is projected that these immediate challenges will continue throughout the decade as I/C package sizes continue to increase and SiP packages become more common, while passive packages and no-lead (QFN/MLF) packages decrease in pitch and size. This continuing challenge of cost-effective assembly miniaturization will require innovative processes, equipment and materials.

Materials Reliability The IC industry has been struggling to deal with poor mechanical properties of low K dielectrics. It has taken almost a decade to solve the low K dielectric problem though flip chip and wirebond processes. While the low K achieved so far is about 3.0, the industry has been planning to lower the K value to 2.2 -2.5. It appears that there is no way to achieve this lower K without the dielectric being porous, thus making the problem correspondingly worse than with K of 3.0. New solutions, such as sealing the ultra low K to make it more hermetic and more mechanically strong, is a starting point. More important contributions need to come from packaging and assembly technologies, with stresses on the interconnection that are lower than the fracture strength of the dielectric.

18


Chapter 4 - Options for Innovation This chapter explores newly discovered phenomena and future technologies that may assist in meeting the R&D needs defined in the iNEMI Research Priorities.

Active Device Technology The predicted end of traditional CMOS semiconductor scaling is generating significant reverberations in approaches and structures of computing systems. The first consequence is the gradual but certain reduction of emphasis on the microprocessor frequency metric, and the corresponding increase in importance of the system’s throughput metric. The emphasis has changed from “how fast can it go” to “how much can it do”.

1. Implementation of advanced, non-classical CMOS devices with enhanced drive current Given that we are reaching a transistor switching density which can require transient currents of up to 200 amps and can generate up to 200 watts per square centimeter we need to find new ways to accelerate the performance and utility of CPU systems. Although we are in a temporary respite where clock speed is no longer the benchmark by which processors are judged and parallel on-chip systems, distributed computing, liquid cooling and other solutions are effective work-arounds, we will again reach the crunch point where we are unable to increase utility without increasing power consumption and heat evolution. This is particularly marked in laptop computers where battery life is a critical selling point, and in communications equipment where rack equipment in server farms can generate kilowatts of heat per square foot of thermal load that must be managed.

Enhancing CMOS, “putting afterburners on CMOS” as it was dubbed at a 2004 NSF meeting, will require the use of new materials (such as carbon nanotubes to enhance thermal conductivity or nano-sized copper in coolants to improve boundary-layer conductivity), improved thermal modeling, architecture changes within the wafer, and computing architecture changes among other innovations. Other changes will include the integration of sensor networks with logic on the wafer using directed-assembly nanowires and other components to increase utility, following the same logic as Intel’s Centrino chipset which integrates wireless functionality for laptops. Many of these new structures will have a lower thermal resistance than conventional systems and interconnect methods such as reflow soldering will have to change to accommodate lower process temperatures. Novel conductive adhesive systems, especially UV and radiation-cured systems, will be required.

2. Identification, selection and implementation of advanced devices (beyond CMOS) and new system architectures Once we reach the nanometer scale, quantum effects become significant. The wavelength of a 7eV copper conduction electron is approximately 0.5 nm (calculation courtesy of Ron Powell at Novellus), which means that, as we approach nanometer scale features, tunneling leakage becomes a dominant problem. Other ways to create a binary state or a variable state by measuring spin or other atomic or particle states are being developed under the general heading of “Quantum Computing”. Although general principles have been described and some measurements have been made, no structure has yet been developed which would be manufacturable and competitive with conventional systems.

3. Printable active devices Printed active devices have the possibility to transform low density, low cost, high volume segments of the electronics industry. They can be produced on flexible substrates using conductive, dielectric, semiconductor, and light emissive inks.

iNEMI must, through the Research Committee and Innovation Committee, stay strongly connected with the leading edge device developers through links with NSF, Sematech, SRC, SEMI, and others who are active in these areas. This connection will allow us to identify the new devices and architectures that are starting to show promise and identify the supply chain requirements and constraints inherent in their use.

19


Thermal Management Another consequence of the expected demise of the traditional scaling of semiconductors is the increased need for improved cooling and operating junction temperature reduction due to large leakage currents and increases in chip power. The continued reduction in semiconductor features sizes and increased clock speeds have made the need for improved electronics cooling a priority in many electronic markets. Although ICs will be designed to minimize thermal dissipation, electrical design will not solve the cooling issue, which will continue to grow in importance. As a result, the cost associated with cooling systems will escalate. Active cooling solutions such as liquid and refrigeration cooling are currently being used in some consumer electronic products and the use of these technologies will continue to expand by 2015.

As the cost associated with active cooling affects more electronic products, there will be significant market for lower cost, high performance cooling technologies. Candidates for these lower cost solutions will include single/two phase liquid, air, heat pipe, direct immersion, refrigeration, and thermoelectric technologies. In some electronics applications, performance and cost issues will require chip integrated cooling technologies to reduce the number of thermal interfaces between the chip and the environment.

Increased Communications Bandwidth The third consequence of the end of the traditional semiconductor and the gradual but certain reduction of emphasis on the frequency of microprocessor metric is an increased demand for higher bandwidth to and from the microprocessor in any system. Current designs have already utilized 3-6 Gbps data rates and designs in the near future are expected to increase this bandwidth capability to 10-12 Gbps for electrical interconnects. At some point, possibly 2010-2015, the limits of electrical transmission in high performance systems may be reached. Optical systems may provide part of the solution, particularly if optical ICs are developed.

Next Generation Packaging Technology Every few decades the electronics industry has introduced a new generation of packaging technology which has taken a decade to develop. It is very apparent that a new generation of low-cost packaging technology will be required during the next decade. This new generation will need to interface with the interconnect structures of the new nano-devices, provide the required thermal cooling and electrical performance, and be compatible with a number of devices and applications including rf, optical, sensors, MEMs, and biological. This next generation packaging technology will require materials with enhanced properties and advanced manufacturing processes to address the rapidly decreasing dimensions and cost expectations.

1. Innovative packaging for giga-function system in package (SiP) Advanced packaging concepts are focused on the change from traditional interconnections of discrete components to thin film component integration. This “package integration” is now starting to emerge on CMOS ICs as an overlay, in a variety of SiP (System In Package) modules. Digital convergence that can include not only portable electronics, but also bio-electronics is driving the need for the package integration, not only at the IC and module level, but also at the system level. SIP, system-centric technology, based on embedded thin film components in organic boards or packages, together SoC (System On A Chip) devices, and battery and user interfaces, can lead to multi-functional systems in the short term and more integrated giga–function systems in the long run. This concept seeks to integrate disparate technologies to achieve multiple system functions in a single package, while providing an ultra-small form factor. 2. Materials and reliability Materials continue to pace the introduction of packaging technologies to meet the major manufacturing requirements of low cost through increased modularity, integration for smaller size, and higher bandwidth for more functionality. In addition to these product-specific attributes, there are general requirements for environmentally friendly materials systems (e.g., bio-based polymers) that use low energy processes. While traditional technologies have focused on materials systems for electronic performance, future materials requirements will need to embrace optical, mechanical, and chemical performance for electro-optical, mechanical (MEMS), and chemical sensor systems, respectively.

20


Advanced fillers for nanocomposites and nanoengineered materials with property improvements that are not possible with micron fillers, such as low CTE with high toughness, high electrical conductivity with low thermal conductivity (high ZT), and high compliance with high current carrying capability offer the promise to meet some of these enhanced performance requirements.

Several technologies may impact material packaging trends in the near and mid-term. In addition to shrinking the IC with higher density printed wiring boards (PWB), embedded passives technologies (resistor, capacitors and inductors) lead the drive towards small size. A key goal is higher dielectric constant materials to produce embedded passives for de-coupling capacitors. Solder systems have migrated to higher reflow temperature ranges due to the drive for no-lead solder systems, placing severe thermal loads on the existing material systems as they traverse the reflow zones. Underfill processes, which improve mechanical performance of parts soldered to the PWB, increase assembly time leading to increased assembly cost. To reduce the time delays associated with underfill assembly for ICs or semiconductor packages, underfill materials will be pre-bonded or be thin films applied to the PWB by pick-and-place machines. As alternatives to solder attach, conductive adhesives (liquid or thin film) will grow in prominence for low temperature processing and fine pitch assembly, assuming they can satisfy the bandwidth requirements for higher performance systems.

Printed electronics will develop in the longer term. Produced on flexible substrates and using conductive, dielectric, semiconductor, and light emissive inks, these materials have the possibility to transform segments of the electronics industry. Ink sensitivity to humidity, oxygen, and light will require modification to existing material systems, or the development of new systems, to make the inks compatible with the environmental goals.

3. Sensors The rapid acquisition, processing, and action resulting from sensor data (mechanical, acoustic, thermal, chemical, seismic, environmental, biological, etc.) will be increasingly critical to numerous populations in future years (homeland security, industrial monitoring, aging populations, quality of life, etc.). Intelligent integrated sensing and control systems are migrating from islands of automation, to interconnected solutions, and subsequently to intelligent self-managing highly scalable systems, i.e., autonomous active control and monitoring systems. Such an evolution requires coordination and leverage across multiple technologies such as sensing, monitoring, control, and communications. Sensor technologies, management tools, and gateways will play a central role in enabling the higher level of integration needed in the development of these new intelligent sensing and control systems. Beyond these Sensor Fusion elements, architectural considerations are required to coordinate this evolution to address scalability issues such as performance, global universal object identification, system management, and security. These large sensor networks will require unique solutions in the acquisition, transmission, and processing of the extensive amount of information gathered for robust networks.

The development of the sensor network involves the deployment of the sensors as well as the network elements to collect and transmit the data for analysis and action. These sensors and their local processors and communication function may take the form of a system in package (SiP). Thus a packaging base that can support mechanical, acoustic, thermal, chemical, seismic, environmental, biological, sensors, and optical and rf communications will be required. It is important to identify the correct number of nodes and sensors that are needed to get the most accurate information at the lowest possible cost. The networks are likely to have a multiplicity of sensors and the need to determine the number and placement of the sensors within the “network field” is of vital importance. Redundant sensors provide more accurate information, however, with the large amount of data that can be collected from the deployed sensors, it is important to identify how and where to fuse the data.

Secure self organizing wireless nodes will form the basis for new networks and will require some localized signal processing capability. Data from the network of sensors will be transmitted to redundant gateway devices, which will have the processing power to determine the relevance of the data based on field conditions. The gateway devices will store, analyze, and relay the data as needed. This transmitted data would be used to make appropriate decisions on what actions need to be taken. The processing power of the gateway device will be determined by the number and type of sensor nodes deployed. The sensor networks also need to be insensitive to environmental conditions in their deployment.

21


Design and Simulation Tools The lack of tools for mechanical, electrical, thermal, environmental, optical co-design from the device to the system level is delaying the introduction of new technology. Significant research and development is needed both in tools and techniques for effectively partitioning the design and simulation. The new simulation areas that need research (Opto, Nano-Scale / Spintronics) will become critical after 2007. Simulations related to nanotechnology could have a large impact on the semiconductor and the electronic materials industry. It is clear that a focus on the combination of business and quantum-mechanics perspectives is required for nanotechnology to be successful. Proper funding of simulation tools will be critical to nanotechnology growth and funding will be contingent upon future business conditions.

Sustainability Metrics The concept of sustainability is being widely adopted and promoted by companies throughout the electronics supply chain as a core tenet and measure of performance of their business operations, largely in response to heightened interest by stakeholders in corporate social and environmental responsibility. There is currently no industry-wide consensus on how to measure or report sustainability. While there are differing views on the degree to which such comparisons are practical or useful, the electronics industry would benefit greatly from a more unified, coherent, and meaningful approach to sustainability reporting.

Nanotechnology The rapidly increasing functionality and availability of microelectronics have combined over the last five years with new scientific discoveries to create public demand for systems that dramatically increase the quality of life. One manifestation of this public demand in the United States is the National Nanotechnology Initiative (NNI), the multi-agency U.S. Government R&D program focused on accelerating the "discovery, development, and deployment of nanoscale science, engineering, and technology" (The National Nanotechnology Initiative - Supplement to the President's 2006 Budget). The NNI’s definition of Nanotechnology is “working at the atomic and molecular and supramolecular levels, in the length scale of approximately 1-100 nm range, in order to create materials, devices and systems with fundamentally new properties and functions because of their small structure”. The goal of NNI is to translate novel nanoscale properties into significant economic benefits for the Nation and its citizens. The updated NNI Strategic Plan (NSTC 2004) defines major subject areas of investment that are critical to accomplishing the goals of NNI, including transferring the technology into products for economic growth, jobs, and public benefit. These areas of investment are fundamental nanoscale phenomena and processes; nanomaterials; nanoscale devices and systems; instrumentation research, metrology, and standards for nanotechnology; nanomanufacturing; major research facilities and instrumentation acquisition; and societal dimensions.

Currently, academic, government, and start-up company developments put the USA in a strong, but not necessarily dominant, technology position world-wide. With a total federal investment in nanotechnology of $4B since 2001, the estimated US Government support of $1B in 2005 represents approximately 25% of global spending in nanotechnology R&D. Most applications of nanotechnology are still in the pre-commercial stage and will need development not only of the product but also of modeling techniques and metrology. Products are being developed and commercialized in large and small companies with strong nanotechnology initiatives. The 2005 PCAST Report noted that approximately 600 companies were involved in the R&D, manufacture, sale, and use of nanotechnology in 2004. It is widely recognized that there is the potential to grow significant businesses based on nanostructure materials, devices, and systems. Some estimates predict a $1 trillion per year industry in the 2010-2015 timeframe, which is as large as the present automobile industry or the electronics industry.

Nanotechnology in its various forms has strong implications for iNEMI members and the competitiveness of the electronics industry as a whole. The combination of significant nanotechnology R&D with technology transfer obviously represents a business opportunity for our members. Nanoscale materials are already well represented in the electronics industry today, including nanocomposites and nanostructured packaging materials, sub-100 nm IC structures, and thin-film giant magnetoresistive (GMR) read heads for high density disk drives. Nanotechnology advances will contribute a number of products and processes that could be especially relevant to electronic processes and products, ranging from the extremely long range and innovative to the short range and drop-in, including:

22


• Quantum computing using “peapod structures” based on carbon nanotubes (CNT) or Bose-Einstein atomic clusters (long term)

• DNA strand self-assembly of electronic structures (long term) • Transistors based on CNT or GMR layered structures (long term) • Mixed nano and MEMS sensor technology (medium term) • Plasma display technology based on CNT emitters (medium term) • Advanced fillers for nanocomposites and nanoengineered materials with property improvements that are not

possible with micron fillers, such as low CTE with high toughness, high electrical conductivity with low thermal conductivity (high ZT), and high compliance with high current carrying capability (short term)

The commercialization of these products will depend on cost and performance when substituting for existing systems, and on the dynamics of new market creation. The iNEMI Roadmap process will continue to monitor advances in nanotechnologies and to assess how these advances meet the R&D needs of the electronics industry. Appendix C highlights some of the needs from the 2004 Roadmap that might be addressed by nanotechnology. The following list summarizes applications for which nanotechnology is expected to make significant contributions to developing markets.

• Nanoelectronics and Information Technology

• Nano-structured microprocessor devices with orders of magnitude lower energy use and cost per gate, thereby improving the efficacy of computers by a factor of millions

• Communications systems with higher transmission frequencies and more efficient use of the optical spectrum to provide at least ten times greater bandwidth

• Small mass storage devices with capacities at multi-terabit levels, 1000 times better than today • Integrated nano-sensor systems capable of collecting, processing, and communicating massive amounts

of data with minimal size and power consumption • Programmable, “smart” materials for multifunctional applications

• Programmable, “smart” materials for multifunctional applications

• Artificial, inorganic, and organic nano-scale materials cells to play roles in diagnostics (e.g., quantum dots in visualization), and also for use as active components such as systems which self-assemble on tumors and can be used to destroy them by local induction or other heating

• Effective and less expensive health care using remote and in-vivo devices, which will be required as elder care moves out of hospitals and nursing homes into a residential setting in order to minimize cost and infection risks

• Hearing and vision aids, and other sensors that compensate for impaired human signal processing, such as the retinal implants currently being pioneered at USC

• Aeronautics and Space Exploration

• Low-power, radiation-tolerant, high performance computers and nano-instrumentation for space applications

• Next generation avionics made possible by nano-structured sensors and nano-electronics • Environment and Energy

• New, “green” processing technologies that minimize the generation of undesirable by-product effluents such as CMP residues

• Efficient, small, light, cheap, portable, long-lasting batteries and fuel cells for portable and consumer electronics matched to lower energy consuming systems such as quantum dot / enhanced LEDs

• Biotechnology and Agriculture

• Nano-array based testing technologies for DNA testing • National Security

23


• Enhanced automation and robotics for all combat vehicles to offset reductions in military manpower, reduce risks to troops, and improve vehicle performance

• Higher performance (lighter weight, higher strength) military platforms with lower failure rates and reduced life-cycle costs

• Chemical/biological/nuclear sensing and improved instrumentation for casualty care • Sensor systems for nuclear non-proliferation monitoring and management

24

Appendix A: 2004 Roadmap Overview

Appendix A: 2004 iNEMI Roadmap Overview

Introduction Electronics is the single largest manufacturing employer in North America, and, based on the value of manufactured product; it is the second largest manufacturing industry. Electronics is also an essential input to the competitiveness and productivity of many other sectors of the economy. These include the automotive and aerospace industries, which are becoming increasingly reliant on electronics components for their manufacturing and operational activities.

The Electronics Sub Committee (ESC) under the Civilian Industrial Technology Committee of the National Science and Technology Council worked with the private sector to create the National Electronics Manufacturing Initiative (NEMI). NEMI was established to promote collaborative development by industry, government, and academia of technology and infrastructure required to facilitate the manufacture of new high-technology and electronic products in North America. This initiative, now industry-led and open to all in the electronics industry, seeks to meet industry goals of improved profitability and global competitiveness.

In 2004, NEMI changed its name to iNEMI (International Electronics Manufacturing Initiative) to reflect the growing global impact on members supply chains and the need for an ever increasing global view of the roadmap participants. A proactive effort was made to recruit more international participation in the roadmap effort and has resulted in participation from 11 different countries.

The structure of iNEMI borrows three successful elements from the Semiconductor Industry Association (SIA) and SEMATECH: A road mapping and technology planning process driven by major end users; a council made up of executives/senior managers of electronics manufacturers; and a focus on results, particularly on infrastructure development and deployment. Unlike SEMATECH, however, iNEMI relies on a decentralized organization to leverage existing resources within industry and other consortia. The fundamental activities of iNEMI include the creation of electronics manufacturing technology roadmaps and the implementation of projects that develop and promote the necessary supply chain infrastructure.

iNEMI Roadmap Process The 2004 International Electronics Manufacturing Technology Roadmap is driven by the seven product sectors defined in Table 1. Each product category is characterized by key cost, performance, and technology drivers which were developed by OEMs with a market vision in each sector. The leading electronic systems manufacturers are basing their strategic planning for new products on the assumption that the electronics infrastructure will develop and implement the technology to meet these key drivers. These drivers were then used to develop individual technology roadmaps that identify needed technology. The 2004 iNEMI Roadmaps were developed by nineteen Technology Working Groups (TWGs), in response to inputs from representatives of OEMs in seven Product Sectors. These twenty-six groups included more than 470 individuals recruited from over 220 private corporations, consortia, government agencies, and universities from 11 countries. The Product Sectors are defined in the following section. The nineteen TWGs can be classified into four categories:

• Business Technologies - Product Lifecycle Information Management (formerly EIT-Enterprise Information Technology, SCM-Supply Chain Management and FIS-Factory Information Systems)

• Design Technologies (Modeling, Simulation, and Design; Thermal Management; Environmentally Conscious Electronics)

• Manufacturing Technologies (Board Assembly; Test, Inspection, and Measurement; Final Assembly) • Component/Subsystem Technologies (Semiconductor Technology, Packaging, Interconnection

Substrates - Organic and Ceramic, Passive and RF Components, Optoelectronics, Displays, Mass Data Storage, Energy Storage Systems, Sensors and Connectors)

Roadmap Meetings were held in Santa Clara and Chicago, and via Web/teleconference. The TWGs presented their progress at the iNEMI 2004 Roadmap Workshop in Herndon, VA (NEMI HQ), on June 21-23, 2004. As a result of this process, participants defined industry needs and identified strategic issues.

25


Table 1. Product Sectors of the 2004 iNEMI Roadmaps Emulator Characteristics

Portable/Consumer High-volume consumer products for which cost is the primary driver, including hand-held, battery-powered products driven by size and weight reduction

Office/Large Business Systems Products that seek maximum performance, with cost as a secondary consideration

Netcom Products Products that serve the networking, datacom, and telecom markets and cover a wide range of cost and performance targets

Automotive Products Products that must operate in an automotive (i.e., harsh) environment

Aerospace / Defense Products that must operate in extreme environments

System-in-Package (SIP) Complete function provided in a single package to a system manufacturer

Medical Products Products that must operate within a high reliability environment

All sectors are looking for smaller and lighter products with increased function and performance that can reliably operate in a wide range of environments. The key variables for each of the sectors are the time to market and price that the market will bear. The following paragraphs summarize the key drivers by product sector.

Portable/Consumer Product Sector Drivers • High volume products for which cost is the primary driver, such as televisions, portable radios, CD

players, and low-end portions of the cell phone and personal computer markets • Reducing total supply chain cost is a key driver and has led to a distributed supply chain • Hand held, battery-powered products driven by size and weight reduction

• Typical products are high-end cell phones, palmtop computers, PDA’s, and wireless email or short messaging devices

• The key elements are: • Packaging, materials, and processing technology • Supply chain optimization • Design architecture

• This product sector is a growth market with a demand for increased functionality • Increased energy storage and power efficiency are gating increased functionality • The need for rapid introduction of complex, multifunctional new products has favored the

development of functional, modular components • Modular design increases the flexibility and shortens the product design cycle, placing the

technology risk and test burden on the producers of the modules Office/Large Business Product Sector Drivers

• Maximum performance within a few thousand dollar cost limit to high dollar products such as low-end servers and high-end personal computers, both desktop and laptop

• Stable market with demand for increased processing and storage performance • Products have migrated towards legacy-free products • Supply chain optimization becomes critical for cost control • A continued focus on shorter, seasonal production cycles • High-end products for which performance is the primary driver, characterized by very high performance,

high clock frequency, high reliability, and high power density, such as routers, communications switches, mainframes, scientific computers, servers and cellular base stations

26


• The market has seen significant erosion particularly in telecommunications and long haul fiber optic systems

• Packaging and design technology has changed from proprietary to commercial • Materials and component cost reduction are increasingly important • Material changes for increased performance is a gating item • Cost-effective communications bandwidth is a key driver with uncertain cost-performance trade-offs

between copper, optical, and wireless technologies

Netcom Product Sector Drivers • Networking, Datacom and Telecom products

• Small Office/Home Equipment, including low-end switches and routers, modems, WLAN cards • Enterprise Equipment, including mid-range switches and routers, PBXs, VOIP phones • Service Provider Equipment, including wireless base stations, central office switches, cable modem

termination systems, core routers • This product emulator group spans a hardware product portfolio from consumer access to long-haul

transport • Addresses system capacities ranging from kilobits per second to terabits per second • Utilizing a diverse set of technologies, including wireless, wired over copper, and wired over optical • Converging data and voice networks • Sustained need to offer more functionality, performance, and capacity, while keeping costs at bay

Automotive Product Sector Drivers • Products that must operate in automotive environments • Extreme reliability required to meet warranty requirements and mission critical applications • Sensor and actuator technology are important drivers for new applications • Product life-cycle and design cycle are significantly longer than for other sectors • Need to respond quickly to environmental legislation and customer specifications is driving a faster

design cycle • Conversion of analog sensor components to digital will require a systems architecture change to a system

that can distribute power at a higher voltage and control commands at a high frequency on a single wire • Typical products require both forward and backward traceability as well as stringent control of material

and process change

Defense/Aerospace Product Sector Drivers • High performance signal processing of sensor data • Typical products require both forward and backward traceability as well as stringent control of material

and process change • The long design cycle and product life cycle is inconsistent with the life cycle for commercial

components • Products that must operate in extreme environments • Extreme reliability required to meet mission critical applications

SIP Product Sector Drivers

• Functional modules that permit faster time to market with less risk to system integrators • Ceramic and/or organic substrates using perimeter and/or array I/O connections • The need for more compact, more highly integrated electronics • A mix of digital and analog functions on a functional module. • Specialty devices used in electronics which cannot be implemented in CMOS based technology

• The integration of optical and electronic devices in SiPs • The integration of MEMs (micro-electromechanical systems) based sensors with electronics

27


• Future applications may include the integration of biological structures, nanostructures and chemical devices

Medical Product Sector Drivers

• Products that require high reliability • Typically these products require forward and backward traceability as well as stringent control and

qualification of material and process changes • Encompasses implantable medical devices, information technology used for patients’ records, medical

diagnostic tools, and monitoring devices • Curbing the ever increasing costs of healthcare • Explosion of wireless sensing for remote healthcare monitoring • Growing need to utilize telemedicine to curb healthcare cost

28

Appendix B: Priorities by Research Area and TWG

Appendix B: Research Priorities Indexed by Research Area and TWG The following sections further elucidate specific research needed by 2015. Within each research area, the needs are further organized into Technology Working Group (TWG) lists.

Manufacturing Processes • Board Assembly:

• Development of automated printing, dispensing, placement and rework equipment capable of the pitch requirements for SiP package assembly at current process speeds

• Packaging: • Improved manufacturing processes for semiconductors, SiPs, RF components and RF MEMs:

• Increased pick and place speed for die attach, flip chip and SMT • Improved accuracy and reduced cost per placement for die attach, flip chip and

SMT • Improved cutting processes for package singulation reducing cost/damage, increase

speed • Final Assembly:

• Improved final assembly processes for the electronics industry • Interconnect Substrates – Ceramic:

• Technology by ceramic substrate interconnection industry to provide cost effective electronic system solutions

• Interconnect Substrates – Organic: • Improved fabrication addressing needs for increases in density, reduced process variability,

improved electrical performance and a radical reduction in cost • Low cost, in-circuit test technologies that can be incorporated into the build process

• Optoelectronics: • Assembly processes and equipment that support integrating electronics and optics into single

packages • Self /passive OE part alignment • Built–in self testing of OE modules

• Sensors: • Development and deployment of surface micromachining technologies for sensors

• RF Components & Subsystems: • Test and measurement

• On-wafer power measurements • Parallel device testing • RF and mmW part fixturing

• Test, Inspection & Measurement: • Several technologies exist that should be evaluated for use in other areas of test and inspection,

including the following alternative strategies: • Electromagnetic Signature – • Acoustic Microscopy - • Photoelectric Effect – • E Beam Test Systems – • Thermal Imaging – • Plasma Based Testing -

29


System Integration • Mass Data Storage:

• Semiconductor lasers with shorter wavelengths and optical lens configurations with higher numerical apertures for optical storage devices

• Optoelectronics: • Low cost, high reliability, dirt tolerant, optical interconnects • Robust and low cost out-of-plane optical coupling, e.g. surface mounted device to embedded

waveguide • OE ICs, integrated electrical-optical devices with multi-wavelength sources, modulators, detectors,

etc, on a device • High frequency VCSELs, 20 GHz and above • Optical self test modules

• Product Lifecycle Information Management: • PLIM security research is needed to guarantee just the data that is needed is supplied to the supply

chain • IT infrastructure research is needed to assure fast and secure transfer of increasingly larger data

transfers to the supply chain • Environmentally Conscious Electronics:

• Research needed to allow tools for meeting environmental compliance regulations – cost effectively and securely

• RF Components & Subsystems: • Innovative MEMs structure designs, monolithic integration with ICs • High speed Analog/Digital converters with higher resolution (14 bits)

Energy and the Environment • Environmentally Conscious Electronics (ECE):

• Development of good scientific methodologies to assess the environmental impact of materials and potential trade-offs of alternatives

• Development of cost-effective, energy efficient power sources • Development of a common, meaningful, straightforward definition of sustainability

Materials & Reliability • Board Assembly:

• Next generation of lower cost solder alloys • New interconnect and packaging technologies deploying nanomaterials supporting decreased pitch

and increased frequencies • Second level under-fill technologies – compatible with board assembly processes

• ECE: • Cost-effective, reliable, environmentally conscious alternatives to materials of concern that meet

functional and performance requirements • Energy Storage Systems:

• Improved materials for energy sources and energy storage systems • Interconnect Substrates - Organic and Packaging:

• Low loss resins for low cost, high frequency laminate applications • Improved materials for embedded resistors, capacitors, and inductors • Substrate materials with improved dimensional stability, planarity, low moisture absorption, and

warpage • Higher density, lower cost substrates • Capabilities for embedding non-discrete in substrates during build-up process

30


• Die attach solder for Tj >200C • Wirebond materials that enable 20 micron pitch without wire sweep, barrier metals for Cu wirebond

pads to reduce intermetallics • Impact of BEOL including Cu/low κ on Packaging:

• Direct wirebond and bump to Cu or improved barrier systems bondable pads • Bump and underfill technology to assure low-κ dielectric integrity including lead

free solder bump system • Improved mechanical strength of dielectrics and other interfaces • Interfacial adhesion • Reliability of first level interconnect with low • Mechanisms to measure the critical properties need to be developed • Probing over copper/low due to damage and bonding over probe mark

• Wafer Level CSP: • Low cost placement and lead free solder capability for I/O pitch between 250um

400um with less than 200 I/O • Compact ESD structures • Low cost, high reliability wafer encapsulation and bumping technology • Wafer thinning and handling technologies which support larger format 300 mm

wafers • Passive Components:

• Aluminum capacitors and film capacitors lagged behind ceramic and tantalum capacitors in their readiness for surface mount, and now they lag in their readiness for 260°C maximum reflow.

• Mass Data Storage: • Materials, processes, and design for perpendicular and patterned recording media, tunneling

magnetoresistive heads (TMR), and heat-assisted magnetic recording (HAMR) • Improved subcomponents and materials supporting higher volumes

• Optoelectronics: • Low cost, SS OE devices, tunable • Nanostructures for OE modules • Development of “optical solders” • Combination of materials and fabrication research to support development of monolithically

integrated optics and electronics that take advantage of electronics infrastructure • Development of photonic crystals or other materials that are amenable to size scale reduction

(compared to Moore's Law for electronics) • Emerging technologies such as holey fibers, photonic band gap materials, new semiconductor

material based devices and solutions, hold promise of providing alternative methods of transmitting and controlling optical signals in ways that may be commercially important in the future.

• Thermal Management: • Low cost, higher thermal conductivity packaging materials, such as: adhesives, thermal pastes, and

thermal spreaders need to be developed for use in products ranging from high performance computers to automotive applications

• RF Components & Subsystems: • Medium power, low loss, high selectivity filters are a cost, weight and size bottleneck to Consumer

and Portable products • Faster pace of R&D in materials for RF MEMs, RF components SiP and semiconductors

• Measurement, Simulation & Design Tools: • New materials optimized for spintronics – electrical, delamination, moisture

• Sensors: • The need exists for a wider selection of biocompatible materials for packaging of biosensors backed

by long-term reliability and safety data. • Realization of the potential for embedded sensors will require development of miniaturized sensor

elements, integrated control systems, and micro-actuators which can all be interconnected in a single package with a small form factor.

31


• Miniaturization is especially critical for implantable devices, and one of the keys to achieving this is the availability of ultra-small, stand-alone power sources which have long life. Although continued research is required into miniaturized fuel cells, this area may benefit from breakthrough developments in nanotechnology for energy storage devices.

Design • Board Assembly:

• Low cost, fine line substrate and PCB routing technologies enabling denser package I/O • Modeling Simulation and Design Tools :

• Mechanical and reliability modeling • Thermal and thermo-fluid simulation • Improved and integrated design tools for emerging technologies like embedded passives,

nanotechnology and optoelectronics PWBs • Co-design of mechanical, thermal and electrical performance of the entire chip, package and

associated heat removal structures • Thermo-mechanical models for nanoscale level technology, such as experimental tools capable of

measuring electrical & thermal and mechanics phenomena/material properties at smaller scales. Scale dependent algorithms with the ability to shift scales will be needed.

• Modeling & simulation of spin as electrons move through different materials, spin relaxation modeling

• Multi-physics simulations: (e.g. Joule heating in sub 50 nm interconnects, electro-chemical phenomena in bio-MEMS devices)

• Simulations (thermal and mechanical) for optimizing hybrid lamination schemes for electrical and optical needs

• Optoelectronics: • Improved and integrated design tools for emerging technologies like embedded passives and

optoelectronics PWBs • Thermal Management:

• Integrated mechanical, electrical and thermal package/product design • Thermal and thermo-fluid simulation • Co-design of mechanical, thermal and electrical performance of the entire chip, package and

associated heat removal structures • RF Components & Subsystems:

• Integrated design and simulation tools needed for RF modules and devices • Mechanical and reliability modeling

32

Appendix C: R&D Needs

Appendix C: R&D Needs Sections from 2004 iNEMI Roadmap This appendix reviews the R&D needs findings from the 2004 roadmap. The needs are described for each Technical Working Group in the following order:

• Board Assembly • Connectors • Energy Storage Systems • Environmentally Conscious Electronics • Final Assembly • Interconnect Substrates - Ceramic • Interconnect Substrates - Organic • Mass Data Storage • Modeling, Simulation and Design Tools • Optoelectronics • Packaging • Passive Components • Product Lifecycle Information Management • RF Components • Sensors • Test, Inspection, and Measurements • Thermal Management

Board Assembly For the 2004 roadmap, the board assembly TWG places the priority on the following research and development needs:

• Defining the next generation of solder materials to replace the high cost silver containing alloys • New interconnect technologies deploying nano-materials to support decreased pitch and increased

interconnect frequencies • Development of automated printing, dispensing, placement, and rework equipment capable of the pitch

requirements for SiP package assembly at current process speeds

Connectors Electronic connectors are well-positioned to fulfill global industry needs well into the future. Those areas where barriers will emerge are being addressed, either with conventional connector technology, or with alternative technologies.

Energy Storage Systems The development of high-rate batteries for HEVs is an active area for RD&E. It is this application which offers the best potential for tremendous growth in advanced battery production. This application may create new opportunities for world-wide manufacturing.

Fuel-cell systems, Li--metal/Li-SPE technology, and thin-film batteries have the potential to provide very high energy density and specific energy. These technologies are recommended for further development on an R&D level. Fuel-cell systems that operate near ambient temperature are possible power sources for portable electronic devices. Identifying the appropriate technology to store fuel for portable applications needs to be addressed, as does hybrid power system design.

33


Environmentally Conscious Electronics • Development of a common, meaningful, straightforward definition of sustainability that is relevant to the

electronic industry and its supply chain, can be applied quantitatively at the business level, can be easily communicated to stakeholders, can be used to set targets, and that encourages an integrated, lifecycle sustainability strategy

• Support of research and development to create a sustainable infrastructure and reliable recycled materials market for use in new products

• Development of cost-effective, energy efficient power supplies • Need for development and implementation of good scientific methodologies to assess true environmental

impacts of materials and potential trade-offs of alternatives • In terms of the sheer volume of environmental information that must now be maintained and managed

for product communications to users and recyclers, technologies like radio frequency identification tags could be explored for environmental data repositories on the products themselves

Final Assembly Optoelectronic component assembly today is still mostly manual, but with increasing complexity and volume, automated assembly is desirable. While there are many processes for which a 20-micron system may be adequate, active alignment techniques of lasers to connectors drive accuracy requirements below 1 micron. This is not possible with today’s equipment.

Interconnect Substrates - Ceramic The ceramic substrate interconnection technology industry must make a conscious effort to change and adopt innovative technology to provide cost effective, electronic system solutions.

Interconnect Substrates - Organic • Research Needs

• High-performance laminates that are competitively priced • Low dielectric constant • Low loss • Integral materials for resistors/capacitors/inductors • Self reinforced materials • Liquid crystal polymer that can be introduced into a PCB shop manufacturing process • New non contact testing techniques and new cost effective electrical test methods • Boards without surface finishes • Improved dimensional stability materials • Waveguide materials and manufacturing techniques

• Development Needs • Microvia technology improvement • Microvia metallization • Continuous cycle time reduction for quick turn boards • Flexible circuit quick turn facilities • Improved design tools for emerging technologies like embedded passives and optoelectronic PCB’s

• University Research • There are very few universities actively involved in interconnection substrate research, the most

notable being Georgia Technical Institute (Georgia Tech). Other universities doing research include, Auburn University, Binghamton University, Pennsylvania State University, and the University of Arkansas.

34


• The following issues need to be solved before embedded optical interconnects become commercially viable.

• Low cost/high yield transmitters and receivers at data rates of greater than 10 Gbps. • Device attachment materials and processes • Low loss wave guide materials and processes • In/out light coupling materials and structures • Low loss light turning processes and/or materials • Complete integration of the link components on the PCB substrate

Mass Data Storage Research areas needing development to support the growth of magnetic recording include perpendicular and patterned recording media, tunneling magnetoresistive heads (TMR), heat assisted magnetic recording (HAMR), recording channel and preamplifier design, dual actuator servo systems, and advanced motor and actuator design.

Some unconventional approaches to data storage are also enabled by MEMS. Magnetic data storage systems using non-rotating and non-streaming media (but not all solid state, as with MRAM) still require relative motion of a transducer over the medium. Probes and arrays of transducers have been proposed and some are under development for the reading and writing of non-rotating media. Various MEMS approaches include moving each transducer individually, and moving an entire block or array of transducers simultaneously. MEMS are also making possible other storage technologies such as holographic optical storage. Advanced micro-machined light modulators have been developed using these new materials and processing techniques.

In the current configuration of optical storage devices, areal data density is a function of wavelength of the laser and the objective lens numerical aperture. This implies that technologies needed for improvement in this area are semiconductor lasers capable of operating at shorter wavelengths, and optical lens configurations with higher numerical apertures. As classical methods reach the end of their capabilities, super resolution technology must be implemented. Design approaches that include land/groove recording, multi-layer, multi-level recording, magneto-optic "super-resolution" (such as domain wall displacement), and improved data coding and detection and error correction must be pushed to their limits. Improved areal density techniques being investigated by some companies include near-field recording and flying optical head technologies. Development of high-volume, low-cost manufacturing approaches will be the key to success for these new technologies. Unfortunately, most of the R&D for future optical storage is being done in Asia.

Increasing the per channel data transfer rate will require higher speed digital encode/decode channels, higher power semiconductor lasers, higher speed spindles, improved recording materials, and increases in track and bit density. The overall device data transfer rate can also be increased by use of multiple recording heads and/or multiple data recording surfaces; however, this is expensive to implement and has proven unreliable in practice (other trades may be necessary). The higher data rates will generate a strong need for improved circuit board design tools that incorporate the RF design factors and transmission line layout into the process. This will become increasingly important to maintain satisfactory signal quality and robust data transfer performance in future products.

Decreasing cost will require a reduction in parts count, materials cost, and more efficient assembly and test processes. This, in turn, will require simplification of the mechanical elements of the drives as well as greater electronics integration. This is a well-developed art for current and next-generation optical storage devices. The real challenge will come with future-generation optical storage products.

Decreasing device size will require minimizing dimensions of mechanical and optical elements, as well as increasing the electronic packaging density. Decreasing device size also increases the difficulty of heat removal and may necessitate reduced power dissipation for equivalent performance.

35


NAND Flash Storage (Semiconductor Storage) The technology necessary to reach 2015 with 35 nm wafer processing is bound to happen. This technology development will be funded by the semiconductor business as a whole. There is nothing that is unique to flash to keep the business from reaching this node. If we agree that flash will not endure past 35 nm, then we must look for sources of R&D funding to develop those technologies which will be used as an alternative. Small investments are currently being made in this direction by the most forward-looking firms, but such funding will need to be dramatically increased in the future for the transition to occur.

The biggest technical hurdle to overcome in extending the areal density of magnetic recording for HDD is increasing the signal to noise ratio (SNR). It takes approximately a 5-6 dB signal-to-noise ratio increase to double the density. This can come from the channel, the media, and/or the heads. Better channels, improved reader and writer, and improved media have all contributed to increasing areal density. Now the improvements are becoming increasingly difficult. There are still channel improvements but they come at the cost of increasing complexity and we are approaching the theoretical limits of data decoding.

More sensitive heads have been built and head sensitivity can still be improved but we are nearing the limits for conventional spin valves and will soon need CPP (Current Perpendicular to the Plane) and possibly TMR (Tunneling Magneto-Resistive) heads. Write head improvement in conjunction with improved media (smaller grains, higher coercive force Hc) has been responsible for significant gains in SNR. However, write heads have used higher moment materials to achieve increased write fields and are now reaching the limit of known materials; 2.4T. The higher write field is necessary due to the increased anisotropy field Hk of the grains. In addition, to increase SNR, the grain size in the media must be lowered without precipitating thermal decay. The introduction of perpendicular recording, HAMR, and patterned media offer solutions to extend density in the face of these increasing difficult constraints.

Disk drive track density is a key technology for increasing areal density, particularly since linear density may be limited by thermal sensitivity. The track pitch is defined by the width of the write (and read) heads. These dimensions are created using lithographic techniques borrowed from the semiconductor industry. As a consequence, we may expect that lithographically defined track dimensions may equal those for semiconductors by 2005. After this is achieved, the rate of track density growth will be driven by lithography development.

Some of the current technical gaps for optical storage devices and media include the following:

• Blue and UV laser diodes with sufficient output power and compatible optical recording layers that can meet or beat the dates shown in the roadmaps

• Information recording layers capable of supporting higher areal densities and data rates • Transformation of multilevel and multilayer recording techniques into practical product configurations • Spindle and motor configurations capable of rotating the disks at speeds in excess of 10,000 rpm without

excessive axial and radial runout or actual destruction of the disk • Solving materials and reliability challenges associated with the head-disk interface of near-field optical

heads combined with removable plastic substrate media, especially for WO and RW media

We recommend that the U.S. focus on the next technology after floating gate flash. Formation of a consortium to cooperate on well-funded R&D (in a single direction) would allow development of a process to replace floating gate technology ahead of other companies. This will serve not only to help the U.S. to regain an important position in semiconductor storage for the next phase of semiconductor memory, but it will also help U.S. companies develop patent portfolios that can generate impressive licensing fees and royalties for the next few decades.

Among promising emerging technologies are MRAM (Magnetic Random Access Memory), FRAM (Ferroelectric Random Access Memory), OUM (Ovonic Unified Memory), PMCm (Programmable Metallization Cell memory), and probe-based storage. Of these, FRAM has already been used in video games, and MRAM is close to commercialization with parts being sampled (2004) in the marketplace. Each of these technologies will be briefly covered in turn. Holographic memories will be briefly covered even though reliable rewritable mass storage utilizing this technology is not currently on a product horizon.

36


Modeling, Simulation, and Design Tools Tools and Methodology Research and Development Needs in the Manufacturing Systems/Supply Chain Management Areas: Table 2 illustrates the developments in both data collection and software development that would be necessary for the effective and efficient implementation of simulation for proactive planning and reactive problem solving in the electronics manufacturing arena. The needs are quantified in the areas of Data Collection, Software Generation, Inferencing Techniques, and Interfacing Ability until year 2007.

In addition, the issues regarding scalability, testability, and schema integration for simulation constructs need to be addressed. Since it is difficult to predict how systems grow and the degree of their growth, any simulation modeling technique needs to be scalable. Developing customizable XML simulation constructs specific to the electronics manufacturing domain is therefore a necessity. It should also be kept in mind that these simulation models need to be verified and validated with relatively little effort. Techniques specific to the electronics manufacturing processes must be devised to accommodate the needs of verification and validation of these models.

Emerging Areas for Electronics Packaging There are several new areas in which electronics packaging promises to play a major role. Table 3 summarizes the needs in these areas of MEMS, nano-scale, Spintronics and the Opto-PWB up to the year 2007. The key simulation issues in each of these areas and the critical year is highlighted. These areas are in the nascent research stages and there seems to be enough time for research, with the only exception being MEMS device proliferation in the automotive sector by 2005, which needs urgent attention.

37


Table 2: Projected Development and Research Needs For Manufacturing Simulation Factor 2003 2005 2007 Data Collection

Automated data collection tools are available. Significant implementation time.

Do not automatically interface with manufacturing simulation programs. Do not provide statistical information as inputs to the simulation program.

Data collection tools should take days to implement instead of weeks. Should have mechanisms to facilitate data interface with manufacturing simulation programs. Should provide some statistical processing for simulation programs.

Data collection tools need to be truly plug-and-play. Should be easy to configure (and re-configure), allowing implementation to be 'instantaneous'. Ease of re-configuration should make the system also extremely flexible.

Software

Generation

Simulation packages are available. They are not typically customized for electronics manufacturing. Require significant skill levels. Do not interface with Materials Requirement planning (MRP) systems and shop-floor data collection systems. Do have graphical user interfaces. Relatively expensive.

Software packages need to become much easier to program and use. Must facilitate 'sensitivity' analysis. Must interface in real-time with shop-floor data collection systems and with MRP systems. Must allow the user to program applications using a graphical user interface. Should be able to 'talk to' procurement & inventory functions, and the software/simulation to be able to talk to each other.

Simulation software should be easily configurable. Target should be to make them as widespread as spreadsheets today. Must be able to plug into shop-floor control systems, data collection systems, MRP systems, and B2B procurement systems.

Inferencing Techniques

Currently, it is primarily statistical. Needs a programmer with a statistical background to ensure that results are accurate. Do not assist with model validation.

Need to reduce the 'statistical skill level' needed by the user. Should automatically output parameters such as confidence levels. Should proactively assist user with ensuring validity of the model and the outputs. Need to incorporate newer data search and analysis mechanisms such as data-mining, neural-networks, heuristic optimization, genetic algorithms, and fuzzy logic.

Simulation packages need to become more 'intelligent'. They need to help the user with model validation, have the ability to use newer and more effective inference mechanisms and optimization techniques in the simulation program, and proactively select the 'best' technique.

Interfacing Ability

Are not geared towards assimilating and using data that is available in enterprise-wide databases. Also not geared towards easily utilizing data generated and available on the shop-floor.

Need to be capable of interfacing, with minor modifications and programming, with enterprise-wide databases and shop-floor data collection systems. This would help modeling (such as yield modeling) and simulation tremendously with better data input.

Should become truly plug-and-play, which would consequently allow for easier installation and use. This would remove a major hurdle in the use of simulation

38


Table 3: Projected Development and Research Needs For Simulations in Emerging Areas

Emerging Simulation Area

Needs by 2005 to 2011 Comments

Opto-PWB Robust tools to predict and optimize optical loss

Predictability of the effect of thermal, moisture, stresses on optical loss

Integration of optics and electronics in PWBs – design methodology

Simulations (thermal and mechanical) for optimizing hybrid lamination schemes for electrical, optical needs

Limited wave propagation tools available

Critical in 2005-2007

MEMS Multi-phase flows simulations, bioMEMS

Multi-physics simulations: (e.g. Joule heating in sub 50 nm interconnects, electro-chemical phenomena in bio-MEMS devices)

Multi-scale simulations from sub 50 nm to mm (similar issue to nano-scale methodology)

Methodology to predict failure of MEMS devices – e.g. delamination, cracking – surface & material science

Analog & digital design

High volume production is a challenge – many custom processes different from usual Si foundry

Lack of standards

New experimental techniques may also be needed to verify the modeling algorithms. Very accurate displacement measurements will be required.

New failure modes and mechanisms will need to be identified.

Limited commercial software packages available, criticality in 2005

Spintronics

New materials optimized for spintronics – electrical, delamination, moisture

Modeling & simulation of spin as electrons move through different materials, spin relaxation modeling

Exploit spin for storing data

10.6 B Flash memory market, 35 B RAM market

Exploit for Logic?

Wafer-package convergence

Long-term/research phase – keep a watch; critical 2005 - 2010

Nano-Scale Modeling & Simulations

Thermo-mechanical models for nano-scale

Experimental tools capable of measuring electrical & thermal and mechanics phenomena/material properties at smaller scales

Scale dependent algorithms will be needed – ability to shift scales

Drivers:

Wafer-package convergence

Device/package circuitry moving to smaller scales: < 65 nm in 2007

Advanced materials e.g. TIM (Thermal Interface Materials

Issue: How is the property and behavior different from bulk behavior/macro-scale?

Critical in 2007

39


Optoelectronics Device Technology The level 0 (Jisso 1) chapter highlighted the current state-of-the-art chip technology of photonic devices for optical communication and solid state lighting. Technology advancements in these areas are driven by performance towards higher speed, higher efficiency, and lower cost. In particular, new VCSEL developments promise cost benefits for various short distance optical interconnects including POF and PCS fiber systems. VCSELs operating at long wavelength could transfer the cost benefits of short reach fiber systems to the intermediate and long reach systems. In summary, drivers for future technology developments on the chip side are:

• Higher performance: • High speed modulators (40Gb/s, 100Gb/s) • High speed directly modulated lasers (VCSELs and quantum dots) • POF over longer distance and Gigabit/sec speed levels

• Lower packaging complexity/cost:

• 1300nm VCSEL (future 1550nm, 650nm) • Large area, high speed photo detectors for POF/PCS fiber • Integration (laser modulator EAM and MZ, multiple tunable laser arrays)

Optoelectronics Packaging (JISSO Level 2) In spite of years of research, array packages are still costly; consequently, array transceivers are still not widely implemented. Volumes are required to reduce cost, and to implement yield enhancement activities that bring cost down further. However, volumes will not increase until costs are lower.

Cost-effective packaging is required for long wavelength single mode VCSELs. The key factor limiting adoption of long wavelength VCSELs is currently chip performance, i.e. achieving robust performance over the full temperature range with good yields and demonstrated reliability. However, once this barrier has been overcome, the market will expect cost benefits of the VCSEL versus edge-emitters. his will not necessarily be immediately obvious. While the vertically emitting nature of the VCSEL creates additional packaging options, the single mode alignment problem is quite similar to that for edge emitters, but perhaps even more difficult, since one does not currently have power to throw away in the alignment as one often does with edge emitters, at least at the shorter distances. Finding a packaging cost benefit of long wavelength VCSELs remains a gap.

Optoelectronic Subsystems (JISSO Level 3&4) • Assembly

• An effective Optical electronic circuit board (OECB) technology is needed to reduce, if not eliminate fiber handling cost and provide a “standard” assembly method that will develop higher volume and thus lower costs.

• Pluggable modules using fiber connectors eliminate an important level of potential difficulty for optical modules by avoiding soldering processes. That approach does, however, introduce concerns associated with fiber connectors that are being dealt with through the iNEMI-TIA and the IPC standards efforts.

• Reduced assembly complexity at level-2 (Jisso 3&4) by elimination of fiber “spaghetti” and the associated manual processes, notably fiber splicing and handling.

• Materials and Material Based Structures

• Primary attach and rework processes are needed for Pb-free higher melting point SAC solder alloys that are compatible with OE components. These needs are similar to those for conventional Sn63 soldering in electronic assembly with the usual restrictions imposed by the relatively low temperature (<85 °C) tolerance of many OE components.

40


• Laminated and embedded wave guides may enable high-speed optical backplane and chip-to-chip interconnect. Waveguide structures, parallel optical interconnects (edge and surface), VCSEL, wave guide to receiver coupling, and small radius 90° bends are all needed.

• Improved VCSELs, for 1310 and 1550 nm transmission are needed, with sufficient reliability for thousands to be used in systems that have >15 year lifetime.

• Reliability

• The reliability of “datacom” components using lower cost innovative technology must be demonstrated to the satisfaction of users. Potential innovative examples include use of epoxy adhesives inside OE packages, non-hermetic materials, such as LCPs (liquid crystal polymers), etc.

• While testing to “telecom” standards, such as Telcordia GR1221 continues, the need is growing for “lite” standards, suitable for products with 10-15 years field life.

• Thermal Dissipation - Managing higher power dissipation, especially in lasers, to achieve lower

operating temperatures and desired wavelength stability, requires the use of: • Higher thermally conductive materials such as carbon nanotubes, carbon fiber, aluminum nitride and

even diamond. Processes to design with these (often inhomogeneous materials) and to make, form, shape, stabilize and attach them are needed

• Embedded active cooling including, potentially, heat pipes, microchannel arrays and fluid jets • Improved thermoelectric cooling, maybe incorporating materials (e.g. Peltier effect) with a higher

figure of merit

• MSA (multi-source agreement) -type OE Modules • The SFP multi-source agreement form factor is being widely adopted for almost all optical

transceiver applications below 10 Gbps. • With growing availability of chipsets that can drive 10 Gbps differential links, the XFP multi-source

agreement is expected to dominate at 10 Gbps. • The widespread adoption of these MSAs by equipment and system designers mean that new

transceiver related technological advances applied to ≤10 Gbps will struggle for acceptance unless they can be applied within the definition of the SFP and XFP MSA. For example, tunable laser technology will need to achieve this very small form factor to see widespread adoption in networks.

• Roadmap above 10 Gbps is still unclear and it appears that implementing multiple channels of 10 Gbps (e.g. using CWDM or DWDM) offers lower cost than moving to higher data rates.

• There is a 10 Gbps optical sub-assembly MSA (i.e. ROSA (receiver optical subassembly) and TOSA (transmitter optical subassembly), but no equivalent at lower speeds.

• Customers are still required to perform lengthy evaluations of each transceiver’s performance and reliability as these vary significantly between vendors and models. Customers would prefer off-the-shelf assurance of performance and reliability.

• Test and Measurement

• Further reduction in the cost of test per unit is required to match the trend lines of lower product costs. This will largely be accomplished by lower test equipment costs, faster test times, and process improvements that minimize or eliminate tests.

• As integration increases, parametric tests will be focused at the lowest level components, while module and card level tests will only apply a functional test to the product, ensuring that the product operates correctly. This will save significant test and re-work costs at the higher-level assemblies.

• With increasing integration, optical products are more likely to include protocol functionality, beginning at the card level and moving to the module level over the next five years. Often this will require the test equipment to communicate via these protocols, chiefly SONET/SDH, IP/MPLS, or FC/iSCSI, for WAN, LAN and SAN applications respectively. Some metro-edge applications may require all three, and at any combination of available rates.

• To achieve the needed cost goals and maximize capital utilization, test systems will need to be rapidly reconfigured to test the changing mix of products and protocols. This will place demand on test equipment to offer multi-protocol functionality in a modular and reconfigurable architecture.

41


• Longer Term Gaps and Needs

• Emerging technologies such as holey fibers, photonic band gap materials, and new semiconductor material based devices and solutions, hold promise of providing alternative methods of transmitting and controlling optical signals in ways that may be commercially important in the future. Optical electronics technologies are emerging continually and we need to watch for important developments and keep an open mind about them.

Packaging The most difficult challenges facing the assembly and packaging industry are presented in the following paragraphs. These challenges are intended to provide a mechanism to allow the research community to focus resources in the areas of greatest need.

Materials There is a broad base of new materials requirements for semiconductor packaging applications which are driven by the new requirements for improved reliability, performance, cost, and compatibility with new semiconductor materials. Some of the desirable attributes for materials in selected applications are listed below. These general criteria should be used as guide lines for target areas.

• Wirebond - Materials that enable 20 micron pitch without wire sweep, barrier metals for Cu wirebond pads to reduce intermetallics

• Underfills - Ability to support 100 pitch on large die, reduce stress on low-K and compatibility with lead free reflow

• Thermal Interfaces - Increased thermal conduction, improved adhesion, higher modulus for thin applications

• Materials Properties - Methodology and characterization database for freq. above 10 GHz • Molding Compound - Low modulus materials that reduce stress on low-wafer structures with low

moisture absorption for high temperature lead free applications • Lead free Solder Flip Chip Materials -- Solder and UBM the supports high current density and avoids

electromigration

Technology The key technology challenges for the packaging industry are summarized in the list below. Each of these needs can be addressed with a number of different research areas and many of them are closely intertwined. For example, in organic substrate development, the needs for higher Tg, low moisture absorption, and reduced warpage (at high temperature) are all related the materials system utilized. A research project on new organic materials, with improved properties, may be able to address all of these factors.

• Close gap between die level and package level interconnect - Improved Organic Substrates • Tg compatible with Pb free solder processing • Increased wireability at low cost Improved impedance control and lower dielectric loss to support

higher frequency applications • Improved planarity and low warpage at higher process temperatures • Low-moisture absorption • Low-cost embedded passives, R. L, C • Embedded active devices • Substrate cost is barrier to flip chip wide spread adoption today • Increased via density in substrate core • Alternative plating finish to improve reliability • Solutions for operation temp up to 200C

42


• Interconnect density scaled to silicon (silicon I/O density increasing faster than the package substrate technology

• Production techniques will require silicon-like production and process technologies after 2005

• Coordinated Design Tools and Simulators to address Chip, Package, and Substrate Co-design Mix signal co-design and simulation environment

• Integrated analysis tools for transient thermal analysis and integrated thermal mechanical analysis • Electrical (power disturbs, EMI, signal integrity associated with higher frequency/current and lower

voltage switching) • Commercial EDA supplier support • System level co-design is needed now. EDA support for “native” area array is required to meet the

Roadmap projections • Educational programs required to train engineers in these technologies/requirements

• Impact of BEOL including Cu/low κ on Packaging • Direct wirebond and bump to Cu or improved barrier systems bondable pads • Bump and underfill technology to assure low-κ dielectric integrity including lead free solder bump

system • Improved mechanical strength of dielectrics and other interfaces • Interfacial adhesion • Reliability of first level interconnect with low � • Mechanisms to measure the critical properties need to be developed • Probing over copper/low � due to damage and bonding over probe mark

• High Current Density Packages • Must be addressed through materials changes together with thermal/mechanical reliability modeling • Whisker growth • Thermal dissipation

• Flexible System Packaging • Conformal low cost organic substrates • Small and thin die assembly • Handling in low cost operation • Different carrier materials (organics, silicon, ceramics, glass, laminate core) impact • Establish infrastructure for new value chain • Establish new process flows • Reliability • Testability • Different active devices • Electrical and optical interface integration

• Wafer Level CSP

• Low cost placement and lead free solder capability for I/O pitch between 250um 400um with less than 200 I/O

• Compact ESD structures • Low cost, high reliability wafer encapsulation and bumping technology • Wafer thinning and handling technologies which support larger format 300 mm wafers

Recommendations on Potential Alternative Technologies To address the needs for packaging substrates, the development of new approaches to substrate fabrication are required. These approaches should address the needs for dramatic increases in density and reduced process variability,

43


improved electrical performance, and significant cost reduction. The development of electroformed imprint technology and phase shifted dielectrics are two possible avenues for these radical improvements.

Wafer level packaging has the potential to shift the industry to a new lower cost structure by eliminating many of the semiconductor packaging assembly operations and test processes used today.

Modeling and Simulation The ITRS Assembly and Packaging TWG has identified chip-package co-design as one of the key crosscut challenges for the 2003 ITRS. This challenge specifies the need for modeling and simulation of mechanical, thermal, and electrical performance of the entire chip, package, and associated heat removal structures as a combined system. This capability is critically needed to insure cost effective and timely assembly and packaging solutions for devices employing new generations of Cu-low k metallization and high power, low voltage (SoC) System on a Chip structures.

Research and Development Needs Key development and research needs in the modeling and simulation areas are defined here. Also, the concern of who will perform this research must be addressed simultaneously. The newly emerging areas in 2002, namely Opto-PWBs, Nano-scale modeling/Spintronics and MEMS all need research and are discussed in this section.

Reliability, Mechanical Analysis and Simulations • Interfacial Delamination: In spite of research into the areas, the ability to predict the delamination between

interfaces is not well understood. Both research and engineering work is needed to develop the ability to predict the nucleation of cracks and their subsequent propagation under static and cyclic loads. Also required for higher operating temperatures is the ability to accurately characterize and quantify the nature of interfacial strength and the interfacial fracture toughness under static (manufacturing & assembly steps) and dynamic conditions.

• Moisture Modeling: Considerable progress has been made in this area in terms of simulation methodology. It is important to have the ability to mechanistically predict the moisture performance of the package, including the diffusion of moisture, the transient thermal analyses combined with the associated stress analyses, the prediction of interfacial stress due to moisture desorption during reflow, and the effect on interfacial crack propagation. Currently, this is done mostly by build and test. Developing engineering capability to predict the moisture performance of the packages would result in cycle time reduction. Research is also needed in the relationship between moisture and delamination.

• Solder Joint Reliability Modeling: Much activity has focused on predicting the solder joint fatigue life for various package families, designs, and application environments. A look at iNEMI emulators shows that the basic predictive capabilities currently in place would be good for the future. However, the current methodologies need to be re-examined as they lack close agreement to the results of temperature cycling. Large acceleration factors in the automotive and military sectors continue to be a challenge. Also, these methodologies need to be examined closely for the lead (Pb)-free solders which are on the horizon. The latter may include establishing some of the well-known Coffin-Manson type of relationships for the Pb-free systems. The ability to predict the formation and effect of intermetallics on solder joint reliability is required both for Pb and Pb-free solder systems. With trends towards finer pitch (Hand Held & Portable systems) and smaller form factors, the current carrying capacity may be nearing the joint limits. This can cause high current densities and lead to the consideration of electromigration effects in solder joints and their under bump-metallurgies.

• Material Characterization: Another area that needs maturity in implementation is the ability to measure mechanical properties of stand alone dielectric (organic) materials below and above their glass transition temperatures. This would include time-dependent behavior such visco-elasticity and visco-plasticity. New simulation techniques such as molecular modeling could be pursued to evaluate their ability to predict global properties based on simulations at the atomic or molecular level. This could help the business issues about the laboratory infrastructure required for modeling inputs of property.

• Process Modeling: Traditional areas of process modeling such as solder joint formation during reflow (for Lead and Lead-free systems), and underfill flow with finer and higher pitches, need to be revisited. One area that has been overlooked is the ability to model wet processes, such as copper and under bump

44


metallurgy plating. This involves combining the simulation of fluid mechanics of the process and equipment with the associated electro-chemistry. The electro-chemistry, will be quite challenging and will require the collaboration of academia, National Laboratories, and industry. Correct implementation in this direction will lead to a reduced number of trials in equipment and process selection for the high-density substrates and finer pitch printed wiring boards.

Research and Development Needs in Thermal and Thermo-fluid Simulations Passive thermal management simulation challenges require faster solutions to the coupled non-linear thermal and fluid problems due to buoyancy driven flows. Though the conjugate solvers of today are capable of solving such coupled problems, algorithms are required for faster execution and convergence. Efficient and accurate thermal radiation solvers need to be incorporated into the tools since radiation is a significant effect in portable products’ passive thermal management systems. Due to duty cycle issues, transient simulations become important to evaluate the response at different times (this is a memory and graphics-intensive phase for the engineer, since it is equivalent to analyzing numerous steady state problems at once.)

Due to the drive to less expensive technology and power increases, an increased need for system level simulations will arise in the Business Systems as well as the Office Systems sectors. The needs in this area include simulation of whole system efficiently; and the questions becomes, should a brute-force modeling approach be used to simulate the system and components in one large simulation or should the problem be broken down into system level and component level with a macro-micro modeling approach? The issues in the former case are the availability of sufficient memory and processor speed, and in the latter it is how to interface the component and system levels; should there be a compact model or some other approach? At the system level, efficient turbulence models need to be developed to address both the higher Reynolds number flows and swirling flows from air-moving devices. Research in turbulence modeling is also required in the area of large-scale data centers. Significant thermal and simulation expertise may be required for turbulence modeling. Due to higher current densities, complexity increases in boards, and the need for high-density connector systems in the military sector, research needs to be done in the area of coupled electrical-thermal conduction problems, with the inherent difficulty in handling physical phenomena at a contact.

For rapid deployment of BGA technologies in both the Business Systems and Office Systems sectors, there is a need to understand how the overall heat transfer in ovens affects the melting and solidification process. Excellent simulation methods and solvers that incorporate radiation, convection, conduction, and phase change are required. This is an excellent example of how traditional areas such as thermal and mechanical analyses could be incorporated into process development.

Passive Components Manufacturing and the use of environmentally green components continues to be a driver of R&D in passive components. This was initially driven by European legislation such as RoHS, but other countries are beginning to adopt similar laws. Component suppliers have made great strides in eliminating banned substances from devices; no major gaps exist in this area. However, the use of components in Pb-free soldering processes in more problematic. As maximum temperatures of reflow soldering rise from ~240°C to ~260°C, the stress on the component greatly increases. As Table 4 shows, the gap between what is achievable now and what needs to be achieved varies by component. The most ready components are ceramic which are not affected by the maximum reflow temperature (but are sensitive to the rate of rise and fall of temperature). Components encapsulated in plastic packaging, like tantalum surface mount, are more affected by maximum reflow temperatures because of the different rates of expansion of the plastic package and the internal elements. Tantalum capacitors with the MnO2 electrode system are ready now, but tantalum capacitors with the conductive polymer electrode system still require some further development. Aluminum capacitors and film capacitors lagged behind ceramic and tantalum capacitors in their readiness for surface mount, and now they lag in their readiness for 260°C maximum reflow.

45


Table 4: Product Conformance -Being Environmentally Green and Meeting Pb-Free Soldering Profiles

• Meets Requirements?

Type Component Size 2003 2005 2007 2009 2015 Product Most Yes Yes Yes Yes

Ceramic Pb-Free Attach Yes Yes Yes Yes Yes

Product Most Yes Yes Yes Yes Tantalum/ MnO2 Pb-Free Attach Yes Yes Yes Yes Yes

Product Most Yes Yes Yes Yes Tantalum/ Polymer Pb-Free Attach Some Some Yes Yes Yes

Product Most Yes Yes Yes Yes Aluminum

Pb-Free Attach Few Some Some Yes Yes

Product Yes Yes Yes Yes Yes Film

Pb-Free Attach None Technology Show Stopper?

Product Lifecycle Information Management The following list identifies needed technology, critical gaps and showstoppers.

• Manufacturing Process

• Areas For Innovation - A set of standards is needed which can characterize a set of manufacturing assets and processes, in terms of product-centric manufacturing cost, first-pass-yield, cycle-time, according to product attributes which can be affected by design and supply chain decisions. This will enable the beginnings of a structured feedback loop from manufacturing to enable the designer to design products for lowest ultimate cost and minimum EC (engineering change) loops.

• Implementation Tools - Manufacturing processes with regard to information technology have in the large part not changed dramatically over the last several years. What has changed is the need for increased communications of the manufacturing processes between supply chain partners; additional information required to meet customer requirements, environmental reporting requirements, and quality feedback processes; better engineering change management; and transparent material, order, and production information exchange. Although most of these areas are being addressed by existing or proposed standards committees, major OEM and EMS companies, and third party implementers, the most critical gap is the lack of broad implementation and industry wide adoption down to the smaller companies in the supply chain. Existing tools for implementation are still viewed as expensive implementations with little or no return on investment by the vast majority of smaller supply chain partners. Although promising new and cheaper implementations are planned and the industry is working on these aspects, they nevertheless remain a critical gap in cross supply chain implementation.

• Cross-Cutting System Issues - MCD Compliance issues affect manufacturing process systems, supplier information exchange systems, logistics import/export systems, engineering change management systems, and enterprise resource planning systems. Although being addressed, the requirements of MCD Compliance will not achieve full integration implementation (system wide within the enterprise nor in the industry) by the implementation deadline requirements. This will result in manual and ad hoc stop gap measures being implemented until automated integration is realized.

• RFID is gaining in adoption within manufacturing processes, materials and logistics operations and in financial feedback/auditing capabilities. System Integrators, third party solution providers and

46


internal information technology departments are working toward customer demand compliance, but have not achieved critical mass

• Environment/Energy

• Areas For Innovation - Although in process, MCD compliance by adding additional data elements to existing information exchange standards has not been accomplished as previously discussed. Until standards are passed and implemented, this remains one of the largest critical gaps in environmental compliance.

• Implementation Tools - No third party solution providers have finalized their product offerings to accommodate MCD compliance. Although in development, MCD compliant automated information exchange and associated management and reporting tools have not been finalized, released or implemented.

• Cross-Cutting System Issues - Currently no industry wide database exists to retrieve or verify accuracy of internal and external MCD information collected. Although great strides are being made by several content solution providers, no solution provider or industry leader/implementer is claiming to have a 100% solution to accurate MCD information verification.

• Enterprise Systems

• Areas For Innovation - A defined way to achieve a single version of the truth about any element of Product Lifecycle Information, based on web technology, encrypt/decrypt, controlled access methods and acceptable speed and scalability. The defined method must be achievable in a series of coherent steps, which enable the participants to reach the goal at minimum cost without breaking existing mission-critical business processes.

• Data element additions to existing standards that allow supporting security policies to the data element level.

• Implementation Tools - Limited tools for data verification and data cleansing are available. These are not integrated as part of the Enterprise Systems in existence but are in most cases only utilized by the largest OEM and EMS providers as third party solutions. The lack of inexpensive and easy to use tools to audit and verify data in smaller supplier systems, results in inaccuracy throughout the information supply chain.

• Cross-Cutting System Issues - Implementation of security to the data level is increasingly required by both industry and government. Data level security has increased from “are you authorized” to “are you authorized” at the location from which you requesting this information at the time when are you requesting information” and potentially other security qualifiers. There exist gaps across Enterprise Systems for data level security and tremendous system wide changes will have to be accomplished in order to meet these new security requirements.

• Materials & Reliability

• Areas For Innovation - A standard that defines the “as built BOM” of a single instance of an electronic product, including the material composition, processes used, and who did what, when. This will provide the value-chain participants with a standard way to hand-off to each other the definition of a single-instance serialized product, and minimize the integration costs between different Factory Information Systems in the value chain. Data element additions to existing standards that allow support of financial referential links and new data element financial fields at all levels of the BOM and product/assembly enabling financial audit capabilities to the component level. This will also enable the creation of an “as built” financial record (or referential link) that can be embedded within the product’s birth certificate information (ex: RFID (radio frequency identification device) imbedding)) and support detailed cost tracking. This will also allow more accurate cost rollups enabling more accurate “as planned” and “as built” information comparison for management. Cost plus contracts with EMS vendors can be better reported and thereby controlled by the OEM, enhancing the appeal of this business model for both EMS and OEM companies.

47


• Implementation Tools - Lack of low cost tools to smaller supply chain participants and the lack of wide spread adoption to smaller suppliers are critical issues in realizing true supply chain visibility throughout the supply chain. No one vendor or solution currently has achieved critical mass. Critical mass adoption is required in order for OEM’s and EMS providers to better manage their supply chain.

• Cross-Cutting System Issues - MCD compliance issues have already been noted. Due to overall increasing globalization of manufacturing and lack of standardized import/export requirements, there is an increase in single country implementation solutions and compliance, which in turn dramatically increases the costs for system implementations and maintenance

• Design

• Areas For Innovation - A set of standards for defining electronic component content at the level usable by CAD/CAM applications, covering form (3D), fit, function is needed. This will provide a common basis for legacy data scrubbing and normalizing.

• Implementation Tools - Although slowly changing, many vendors opt for proprietary solutions to the design space, selling vertical solutions. This inefficiency of being unable to effectively exchange complete design data between different proprietary systems increases inefficiency and costs.

• Cross-Cutting System Issues - Design data must include information from other systems so that preferred vendors, quantity negotiated discounts, lead times, availability, and other material associated business processes can be included in the design process. Although this has been addressed and partial or full solutions are available by several solution providers, OEM’s, and large EMS providers, these are by far the exception.

• PLIM business process mapping

• Areas For Innovation - A set of definitions need to be created which describe at a high level the business processes of PLIM between the different players in the lifecycle of the product. The definitions should model the data flows, define the functional responsibilities, and state where the “truth” of each element of the product information should reside, according to the principles of best practice. Several models may be required to cover variable aspects, such as OEM/ODM/EMS combinations and various country, language, and political considerations. The purpose of this modeling exercise is not to define a standard of how the industry should work, but to provide a context into which the technology solutions can be mapped, such that the probability of adoption of standard methods, protocols and formats across value-chains is maximized and the benefits of adhering to the standards are clearly visible to the people who make the investments.

Rf Components & Subsystems Recommendations on Priorities and Alternative Technologies The most difficult challenges facing the RF component industry are presented in Table 5. These challenges are intended to provide a mechanism to allow the research community to focus their resources in areas of greatest need.

48


Table 5:.Technology Challenges and Scope Challenge Scope Reduce IC design cycle time

design and modeling tool sets

test and measurement infrastructure

Reduce size and parts count in consumer and portable products

alternate approach to RF channelizing filters

digital representation of RF functions

COB, FC BGA (package-less) attach (EMC issue)

More MMICs, fewer discrete devices (actives & passives)

Development of Low Voltage MEMS Innovative MEMs structure designs, monolithic integration with ICs

Low profile high efficiency antenna Critical for Consumer and portable products, Wireless LAN,

High speed Analog/Digital converters with higher resolution (14 bits)

High speed ADC critical for base station power amplifiers

Monolithic Filter integration Highly integrated transceivers

Heterogeneous network radio platform Integration

Radio Co-existence, interoperability

Multi-and antenna Technology, reconfigurable RFICs

RF component issues related to spectrum crowding

PA, passive component linearity improvements

Cell sectorization (smart antennas)

RF performance enhancement of packages/substrates

Impedance control, line loss reduction, improved thermal properties in MCM materials and lead frames

COB, FC, BGA RF parasitics and frequency limits

Underfill materials with suitable RF, thermal, mechanical properties

Test and measurement On-wafer power measurement

Parallel device testing

RF and mmW part fixturing

Sensors Realization of the potential for embedded sensors will require development of miniaturized sensor elements, integrated control systems, and micro-actuators, all of which can be interconnected in a single package with a small form factor.

Technology gaps hindering the full realization of market opportunity exist for ultra-small and implantable biosensors and self-contained sensors integrating miniaturized energy technologies. The need exists for a wider selection of biocompatible materials for packaging of biosensors backed by long-term reliability and safety data. Miniaturization is especially critical for implantable devices, and one of the keys to achieving this is the availability of ultra-small, stand-alone power sources which have long life. Although continued research is required into miniaturized fuel cells, this area may benefit from breakthrough developments in nanotechnology for energy storage devices.

We recommend more aggressive development and deployment of surface micromachining technologies for sensor components, as well as LIGA (Lithografie, Galvanoformung, Abformung (German: Lithography, Electroplating, and Molding; a micromachining technology)), a process involving forming structures in a polymer mold photo-

49


lithographically, and then electroplating into these cavities for high volume production of integrated sensor/actuator systems.

These needs especially affect the automotive industry, which currently utilizes a multi-package module for many sensor applications which require integration of sensors, microprocessors, signal conditioning, communications, power source, and memory functions. Packaging technology must evolve towards higher levels of integration using system-in-package solutions, sometimes as an intermediate step towards eventual system-on-chip implementation.

Test, Inspection, and Measurement As with the last roadmap, several technologies exist that should be evaluated for use in other areas of test and inspection, including the following alternative strategies:

• Electromagnetic Signature - An extension of near-field scanning technology for diagnostics and signature analysis has been proposed as a means for non-contact testing over a spectrum of uses, including printed circuit boards. This extension requires the integration of a number of technologies and resolution of some practical considerations to ensure ongoing competitiveness against existing AOI/AXI technology and its adoption into the production test environment. The one significant benefit it may have, particularly in the area of RF, high performance, or harsh environments, is that it potentially could offer performance as well as structural test information. Important areas for research and development include probing compliant arrays with hybrid probes using both E-field and H-field to improve resolution and reduce analytical ambiguities. Smaller probe arrays may possibly be embedded in membranes for greater resolution. New probe technologies are needed to extend the range to GHz bands. Additionally, development of at-speed methods of functionally stimulating units under test, through advanced signal injection techniques (which may, in part, rely upon industry-wide consensus on DFT) needs to be evaluated. Another need is the development of powerful image analysis software that enables an integrated system to detect, isolate, and evaluate faults rapidly. These innovations include de-blurring algorithms for improving resolution (by 3x-5x) through processing array signatures. Finally, we need to ensure that the probe arrays and software can be integrated into affordable platforms for testing a wide spectrum of electronics, ranging from chips with custom test fixtures, through single-chip and MCM (multi-chip module) packages, to bare and populated boards.

• Acoustic Microscopy - Acoustic microscopy was introduced in the late 1980s. In operation, focused high frequency acoustic waves are reflected from internal surfaces, interfaces, or voids, typically in a pulse-echo mode where the attenuated and phase-shifted signals reflected from internal interfaces are interpreted using signal processing software. As an analysis tool at the device level, acoustic microscopy has demonstrated success in areas such as cracks and lid-seal flaws in ceramic flat-packs, separated gull-wing solder fillets, popcorn cracks in plastic packages, and voids in molding compounds. Die tilt and de-bonds from plastic to die and from plastic to lead frame are examples of defects effectively revealed via acoustic microscopy imaging (AMI) techniques. While it has equally been used for finding bad solder joints and imperfect BGA solder fillets or solder bumps on flip chips, it has not been applied as a board level test technology. Although visual inspection and electrical tests can detect most defects, and X-ray and thermal imaging can provide significant detail, certain defects and internal features consistently resist discovery by these methods, prompting the potential need for research in alternative technologies.

• Photoelectric Effect - This technique uses a high powered UV laser to generate secondary emission of electrons that can be simulated and measured from a grid that is immediately above the board under test. This technique is targeted at bare boards and by applying charges to nodes and looking at stray charges and discharge rates, shorts and opens can be detected. These systems will be priced in the $500K to $1M price range.

• E Beam Test Systems - This technique is very similar to Photoelectric, with the bare board placed in a vacuum and steerable electron beams used to produce secondary electron flow. Grids can be on both sides of the board to act as source and detectors to find both shorts and opens. While lagging behind photoelectric, the technology is well known and can probe 2 mil pads and make more than 1000 tests per second.

50


• Thermal Imaging - In this case, a PWB is stimulated by elevating the temperature. Heat pattern flows of the boards, including solder connections, can be analyzed using infrared (IR) imaging. The time-temperature pattern is measured against a good joint and opens and shorts can be detected. Even marginal cases can be detected. The issue with this type of testing is test time, with 10 to 20 minutes required to heat and cool the unit under test.

• Plasma Based Testing - The plasma head for this technique needs to be mechanically staged similar to a flying probe system but the final probe is replaced by plasma generated by a high power laser. Ionized particles create an electrical pathway to the UUT where conventional stimulus and measurement can be performed. This is a non-contact testing technique that builds on flying probe technology having potentially finer resolution.

•

Thermal Management Research Needs The continued development of new and improved thermal management technology will require the combined efforts of industry based development and university based research with a focus on practical application. Research in extensive heat transfer, thermo-fluid, and thermo mechanical is needed to define new opportunities (i.e. path breaking) and to improve predictability and reliability (i.e. gap-filling). Research outcomes required to satisfy the thermal technology needs identified in the previous section are outlined below.

• Thermal Spreaders • Inexpensive, high thermal conductivity, materials (possibly composites) offering a closer expansion

match to the CTE of silicon • Algorithms for optimizing thermal/thermo mechanical design of thermal spreaders • Techniques for achieving improved thermal spreading within a chip to alleviate hot spots due to

localized high heat flux concentrations • Correlations and analytical models of dry-out and rewetting of micro-channels and micro-porous

structures to facilitate design of micro-heat pipes

• Thermal Interfaces • Thermal pastes, epoxies, and elastomers loaded with high thermal conductivity nanoparticles • New interface materials based on carbon nanotubes and other materials • Novel techniques/materials to minimize interfacial stresses • Correlations and analytic relations to predict fatigue life of bonded interfaces • Standardized method to characterize thermal performance of interface materials • Self-contained solid-to-solid phase change materials or micro-encapsulated materials as suitable

interface materials for a range of applications including harsh environment

• Heat Pipes • Flexible heat pipes • Heat pipes that handle high heat fluxes • Low cost heat pipes that can transport heat effectively over large distances (>0.5 m) • Designs to reduce the gravitational orientation impact on heat pipe efficiency, especially for avionics

applications • Heat pipe technology capable of withstanding harsh environments • Sound numerical models and optimization tools for predicting the performance and operational limits,

including dry-out, in heat pipes • Correlations and algorithm for design of thermosyphons (i.e. wickless heat pipe)

• Air Cooling • Models and correlations to predict heat transfer in transition and low Reynolds number flow over

packages and in heat sink passages

51


• Low Reynolds number turbulence models for use in CFD codes • Heat sink design and optimization procedures for the minimization of heat sink thermal resistance,

subject to mass and volume constraints • Advanced manufacturing techniques for metal and composite material heat sinks • Concepts for higher head-moderate flow, low noise, compact fans • Novel, low power consumption, low acoustic emission micro-fans for forced convection cooling in

notebook computers and handheld electronics, including low-frequency and ultrasonic piezoelectric fans

• Novel miniature fan concepts including low-frequency and ultrasonic piezoelectric fans for minimal noise emission and power consumption

• High pressure/high flow blowers with low acoustical power

• Water Cooling • Miniaturized components that have high reliability and provide enhanced performance (e.g. pumps and

heat exchangers) • MEMS and meso-scale components to create low-cost, low-noise, water-to-air heat exchangers • MEMS and meso-scale components to create low cost, package-size cold plates • Microchannel heat sinks with novel integrated micropumps to minimize package volume for high heat

flux applications • Methods to enable direct water cooling of chips or chip packages

• Direct Liquid Immersion

• Single-phase and two-phase heat transfer correlations for new families of dielectric coolants • Nanoparticles for addition to dielectric coolants to create a nanofluid with enhanced heat transfer

characteristics • Convective and phase change cooling correlations to account for highly non-uniform heat flux

boundary conditions • CHF (critical heat flux) models to account for highly non-uniform surface heat fluxes • Characterization of boiling and two-phase flow in narrow passages and 3-D structures • MEMS and mesoscale components to enhance convective, as well as pool and flow boiling, heat

transfer • Correlations and models for evaporative spray cooling heat transfer

• Sub-Ambient and Refrigeration Cooling

• Highly reliable miniaturized components such as compressors, condensers and evaporators • MEMS and mesoscale components to create low-cost, low noise refrigerators using solid-state, vapor

compression, or absorption cycles • MEMS and mesoscale components to create low-cost, package- size cold plates using solid-state, vapor

compression, or absorption cycles • New thermoelectric materials and fabrication methods that can improve the performance of

thermoelectric coolers

• Low Temperature Refrigeration • Application of Auto-refrigerating Cascade (ARC) systems to provide low temperature cooling for

electronic packages • Application of mechanically cascaded (2-stage) refrigeration systems to provide low temperature

cooling for electronic packages Recommendations The following constitutes the major cooling technology areas for development and innovation:

52


• Low cost, higher thermal conductivity, packaging materials, such as adhesives, thermal pastes, and thermal spreaders need to be developed for use in products ranging from high performance computers to automotive applications.

• Advanced cooling technology in the form of high performance heat pipe/vapor chamber cooling technology, thermoelectric cooling technology, direct liquid cooling technology, as well as high performance air-cooling and air-moving technologies, need to be investigated.

• Closed loop, liquid-cooling solutions, which are compact, cost-effective, and reliable should be developed. • High-performance cooling systems, which will minimize the impact to the environment within the customer’s

room and beyond, need to be developed. • Advanced modeling tools which integrate the electrical, thermal, and mechanical aspects of package/product

function, while providing enhanced usability and minimizing interface incompatibilities, need to be developed.

• Advanced experimental tools for flow, temperature and thermo mechanical measurements for obtaining local and in-situ measurements in micro cooling systems

• It is further recommended that action should be taken to pool resources to fund cooling technology development, promote the involvement of university/research labs and establish a closer working relationship with vendors.

53

Appendix D: Nanotechnology

Appendix D: Extract of Comments from the 2004 iNEMI Roadmap Relevant to Nanotechnology

Design Technologies • The size reduction of discrete devices has hit a wall at 0201 package size, with many facilities having

difficulty maintaining optical recognition. • The extension of Moore’s Law beyond 2005 will require new packaging technologies to reduce the cost of

packaging. • Co-design of mechanical, thermal, and electrical performance of the entire chip, package, and associated heat

removal structures is one of the key crosscut challenges. • New materials, components, and processes need to be developed, qualified, and introduced in 2004-2006 to

enhance recyclability, improve energy efficiency, and reduce ecological impact.

Manufacturing Technologies • The component underfill process has provided faster flow times, but difficulties exist with reworkable and

pre-applied underfill for standard SMT operations. • For board assembly of optoelectronics to be competitive in high cost regions, it is important to develop low

temperature soldering and automated fiber handling and assembly. • Board assembly will be challenged with providing material control and identification standards during the

transition between lead free and eutectic materials and throughout the product life cycle.

Component/Subsystem Technologies • There is a need for cost and performance models that help highlight the benefits of embedded passives. • Faster improvements in RF filter technology are needed to meet the cost/performance demands of both the

Portable and the Large Business Systems Product Sectors. • Faster improvement in antenna design and development will help meet the implementation needs of portable

units like software defined radios, dual-band, and dual-mode radios. • Assembly processes and equipment are needed to support integrating electronics and optics into single

packages. • Technologies must be developed that support effective hybrid integration of components into lower cost,

smaller, higher functioning subsystems. • The development of improved optoelectronic subcomponents and materials that allow automation will be the

key to expanding the market.

iNEMI Technical Projects • iNEMI should create a nanotechnology presence to track the progress of its practical uses due to

nanotechnology's potential for being highly disruptive to member companies. • EMS companies need to take the lead in electronics manufacturing technology development, utilizing pre-

competitive consortia like iNEMI to leverage their R&D and technology deployment budgets. Continued cooperation with government and academia should also be adopted to help address the R&D gap.

• iNEMI should establish new projects in 2004-6 to develop, qualify, and introduce new materials, components, and processes that will enhance recyclability, improve energy efficiency, and reduce ecological impact.

Materials Development • A project needs to be developed in the interconnect arena to improve the dimensional stability of organic

materials used in PWB manufacturing.

54

Appendix D: Nanotechnology

• A combination of materials and fabrication research is needed to support the development of monolithically integrated optics and electronics that take advantage of the electronics infrastructure.

• Technology development is needed in areas such as optical solders and board level wave-guides to enable more complex and higher density board level products.

Technology Development • Fuel cell systems, lithium-metal/SPE technology, and thin-film batteries have the potential of providing very

high energy density and specific energy. These technologies are recommended for further development on an R&D level. Fuel cell systems that operate near ambient temperature are a possible power source for portable electronic devices. Identifying the appropriate technology to store fuel for portable applications needs to be addressed, as does hybrid power system design.

• Disruptive technology (such as fuel cells) for overcoming substantive technical hurdles (e.g., developing batteries for electric/hybrid vehicles) offers a window of opportunity. In order to ensure success, manufacturers must be willing to invest with a long-term perspective in mind.

55

Appendix E: iNEMI Overview


iNEMI History NEMI grew out of two separate efforts to recapture American leadership in electronics manufacturing. In 1993, the American Electronics Association (AEA) conducted a study (led by Mauro Walker, Senior Vice President and Director of Manufacturing at Motorola) on the U.S. Electronics manufacturing infrastructure. Based on the results of this study, AEA recommended that the Administration create a national initiative in electronics manufacturing, focusing on strategic electronic components and electronics manufacturing systems. The second effort originated with the National Science and Technology Council’s Electronics Subcommittee (ESC), chaired by Lance Glasser, Director of the Electronics Technology Office at the Defense Advanced Research Projects Agency (DARPA).

In 1994, Dr. Walker and Dr. Glasser teamed up to initiate the creation of NEMI. Over a 10-month period, they met with representatives of government, executives from many of the industry’s largest manufacturers of electronic home and business products, and executives from the suppliers that manufacture the constituent materials, components, and subassemblies. The first NEMI Roadmap was produced in 1994 and NEMI was incorporated in January, 1996. In 2004 NEMI changed its name to iNEMI (International Electronics Manufacturing Initiative) to reflect the growing global influence of the industry’s supply chain.

iNEMI Mission iNEMI is a North American based consortium dedicated to providing leadership for the global electronics manufacturing supply chain for the benefit of both member companies and the industry as a whole.

iNEMI Organization The iNEMI Board of Directors (see fig.#1) defines the policy, strategy and direction of iNEMI; has operational responsibility for the iNEMI organization; is responsible for the iNEMI Executive Secretariat; and reviews the performance of all projects and programs.

The Executive Secretariat includes the Executive Director/CEO and a small support staff. This group provides the administrative support for the organization, including coordination of meetings, maintenance of data bases, publishing and distribution of documents, providing assistance to the Technical Committee, and performing other necessary administrative duties.

The iNEMI Technical Committee facilitates and coordinates all iNEMI technical activities and is comprised of OEM and supplier representatives as well as government, university, and industry association representatives. The Technical Committee reports to the Board of Directors and Co-chairs of the committee are ex-officio members of the Board. The Technical Committee develops and maintains the prioritized iNEMI Technical Plan; approves development and deployment projects; appoints Technology Integration Group and Project chairs; organizes, manages, and publishes the biannual iNEMI Roadmap; and recommends priorities for potential university research programs.

The Technology Working Groups (TWGs), who report to the Director of Roadmapping, are responsible for developing the technology roadmap, identifying the technology gaps, and recommending key technical areas for the TIGs to work on. TWG membership is open to the entire industry, worldwide. Currently some 19 industry groups, along with 7 Product Emulator Group Chairs assist the Director of Roadmapping in roadmapping activities.

The Technical Integration Groups (TIGs), currently six in number, report to the Director of Planning and are responsible for developing the Technical Plan, identifying technology needs, and establishing projects to address them. TIG membership is limited to iNEMI members.

56


iNEMI Organization Chart

Roadmapping

19 TWGs

Implementation

7 TIGs

Technical Committee

OEM, Supplier & Government

Representatives

Technical Committee

OEM, Supplier & Government

Representatives

Executive Secretariat

Secretary to BODCommunications

Membership Development

Technical Facilitation

Executive Secretariat

Secretary to BODCommunications

Membership Development

Technical Facilitation

iNEMI Organization

iNEMI CounciliNEMI

Council

TechnologyWorkingGroup






PLIMTIG

PLIMTIG

Optoelectronics TIG

Optoelectronics TIG

Board Assembly

TIG

Board Assembly

TIG

Environmentally

Conscious Electronics

TIG

Environmentally

Conscious Electronics

TIG

Substrate TIG

Substrate TIG

iNEMI ProjectiNEMI ProjectiNEMI Boardof Directors

Elected by iNEMI Council

Representatives

Strategic ObjectivesOperational

Responsibility

iNEMI Boardof Directors

Elected by iNEMI Council

Representatives

Strategic ObjectivesOperational

ResponsibilityiNEMI ProjectiNEMI Project

iNEMI Project iNEMI Project

iNEMI ProjectiNEMI Project


iNEMI Project iNEMI Project


SiPTIGSiPTIG iNEMI ProjectiNEMI Project

BusinessLeadership

Team

BusinessLeadership

Team

Product Emulator

Group

Product Emulator

Group

Product Emulator

Group

Product Emulator

Group

7 PEGs

Heat TransferTIG

Heat TransferTIG


iNEMI Deliverables iNEMI activities produce five key resources and benefits for electronics manufacturers:

• Roadmaps showing a projection of the needs of the electronics industry • A list of identified gaps in the manufacturing infrastructure • iNEMI stimulates R&D projects to fill these gaps • Integration projects designed to eliminate these gaps • Encouragement of standards activities to speed the introduction of new technology

Standards In the course of the project activities, iNEMI work often affects existing standards or requires new standards to be written. Each project considers standards activities required before it is formally approved by the Technical Committee. However, iNEMI does not choose to release or maintain standards. Instead, iNEMI has developed relationships with existing standards organizations, such as IPC, EIA and IEEE, to develop, release and/or influence standards of interest to iNEMI members.

57


Further Information For availability of referenced documents (such as the latest iNEMI roadmap or Annual Report) and other information on iNEMI, please contact us at:

Voice - 703-834-0330

Email – [email protected]

Web site – www.inemi.org

58

mailto:[email protected]

http://www.nemi.org/

Appendix F: Glossary

Appendix F: Glossary 3D – Three Dimensional ADC – Analog to Digital Converter AESF - American Electroplaters and Surface Finishers Society, Inc. AMI – Acoustic Microscopy Imaging AOI – Automated Optical Inspection ARC – Auto – refrigerating Cascade Atm – Atmosphere ATP – Advanced Technology Program AXI – Automated X-Ray Inspection B2B – Business to Business

BEOL- Back End of Line

BGA – Ball Grid Array CAD – Computer Aided Design CAMX – IPC Computer Aided Manufacturing Standard CAPEX – Capital Expenditures Cd – Cadmium CD – Compact Disk CHF – Critical Heat Flux

cm² - centimeter squared CMOS – Complementary Metal Oxide Semiconductor CrSi – Chromium Silicate CrVI – Chromium CSP – Chip Scale Package CTE – Coefficient of Thermal Expansion Cu – Copper DfA – Design for Assembly DfE – Design for the Environment DfM – Design for Manufacturing DfT – Design for Test DIMM – Dual In-Line Memory Module DNA - Deoxyribonucleic Acid DPMO – Defects Per Million Opportunities E Field - Electric fields, created by voltage and measured in volts per meter ECE – Environmentally Conscious Electronics EDA – Electronic Design Automation EDI – Electronic Data Interchange EIA – Electronics Industries Association EICTA – European Information & Communication Technology Industry Association EK – Electro – Kinetic EMI – Electro Magnetic Interference EMS – Electronics Manufacturing Services ENIG – Electroless Nickel Immersion Gold Plating EPA – Environmental Protection Agency ESD – Electrostatic Discharge FBGA – Flip Ball Grid Array GaAs – Gallium Arsenide Gb – Giga bit GHz – Giga Hertz H Field - Magnetic fields, induced by alternating current (AC) and measured in gauss or Tesla HAMR – Heat - Assisted Magnetic Recording HDI – High Density Interconnect

59


Hg – Mercury High k – High Dielectric Material Used in Capacitors For Higher Value Capacitors High n/sq – High ohm per Square I/O – Input / Output IC – Integrated Circuit INSIC – International Storage Industry Consortium IPC – Institute for Interconnecting and Packaging Electronic Circuits IT – Information Technology ITRS – International Technology Roadmap for Semiconductors JIETA – Japan Electronics & Information Technologies Industries Association KW/cm² - Kilo Watts per Square Centimeter LCA – Life-Cycle Analysis LGA – Land Grid Array Low k – Low Dielectric Constant Materials Used with Copper Interconnects for Lower Delays in Integrated Circuits LTCC – Low Temperature Co-Fired Ceramic MEMS – Micro – Electro – Mechanical Systems mm – Millimeter MRAM – Magnetic Random Access Memory MTBA – Mean Time Between Failures N² - Nitrogen Gas NEMI - National Electronics Manufacturing Initiative NiCr – Nickel Chromium NIST – National Institute of Standards and Technology NNI – National Nano-technology Initiative NPI – New Product Introduction OE – Optoelectronics OEM – Original Equipment Manufacturer OPEX – Optoelectronics Expenditures Pb – Lead PBB - Polybromated Biphenyl PC – Personal Computer PDA – Personal Digital Assistant PLIM – Product Life - Cycle Information Management PTF – Polymer Thick Film PWB - Printed Wiring Board R & D – Research and Development RF – Radio Frequency SECs/GEM - Semiconductor Equipment Communications Standard (SECS) and Generic Equipment Model (GEM) standard Si – Silicon SIA – Semiconductor Industry Association SiGe – Silicon Germanium SIP – System in Package SMT – Surface Mount Technology SOA – Semiconductor Optical Amplifier SoC – System on Chip SPE – Solid Phase Epitaxy SPM – Standard Process Modules TaNx – Tantalum Nitrate Tg – Glass Transition Temperature TIG – Technical Implementation Group TIM – Test, Inspection and Measurement TIM – Thermal Interface Materials TMR – Tunnel Magnetic Resistance TWG – Technical Working Group

60


UBM – Under Bump Metallurgy UF – Micro-Farad UL – Underwriters Laboratory um – Micron UUT – Unit Under Test VCSEL – Vertical Cavity Surface - Emitting Laser W/cm² - Watts per square centimeter WLP – Wafer Level Packaging XML – Extensible Markup Language

61

Appendix G: Contributors

Appendix G: Contributors Alan Rae, Nanodynamics (Chair, Research Committee)

Bill Ballard, 3M

Bob Pfahl, iNEMI (Secretary, Research Committee)

Carol Handwerker, NIST (Co-Chair, Research Committee)

Chris Matayabas, Intel

Chuck Richardson, iNEMI

D.H.R. Sarma, Delphi

Jim McElroy, iNEMI

Marc Chason, Motorola

Rao Tummala, Georgia Tech

iNEMI Technical Committee

62

research priorities - inemithor.inemi.org/webdownload/ri/inemi_2005_research...•...

Documents