report
1.8.3 ICT: General Market Description

Brief Market Description

In Europe, the value of the ICT industry is around 8% of Europe's total GDP. 6% of the European workforce is employed in ICT-related fields[1]. ENIAC reported that the global market for electronics was US$1340bn in 2005, with the value of semiconductors alone at US$265bn. This share of European employment equates to about 13 million jobs - which includes a number of activities related to the development or use of ICT (such as software development), but which are not directly linked to the manufacture of electronic devices.

The largest companies in this sector are based in the United States or Asia. Fortune magazine's Global 500 ranking of the world's largest corporations by revenue lists eight ICT companies (Samsung, Hewlett-Packard, IBM, Hitachi, LG, Matsushita, Sony) before coming to the first European representative - Nokia, in 88^th position. Other European ICT companies in this listing include Ericsson and Phillips[2]. The revenues of other European companies are too low to appear on this list, but are nonetheless substantial companies (ST Microelectronics, 2008 revenue US$ 9.84B; Infineon Technologies, €4,321M, and ASML, €2,9M).

Companies involved in this industry range from large, integrated suppliers of computing equipment and related services (such as IBM); to manufacturers of particular component types like memory or displays. In general it is a highly competitive industry, with rapid technological development and relentless pressure on price and performance.

Improvements in ICT also have societal benefits. As the recent "Strategy for ICT R&D Innovation in Europe"[3] states:

ICT is also essential to address Europe's societal challenges. It brings unique responses e.g. to the growing needs for sustainable healthcare and ageing well, for better security and privacy, for a lower carbon economy and for intelligent transport.

Nanotechnology Impact

Electronics will be one of the industries upon which nanotechnology will have the most sustained, significant impact, as so much of the technology development in this area is scaling to (or is already at) nanometre length scales.

More Moore

Moore's Law, coined by Intel's Gordon Moore, states that transistor density on a processor will double every two years. As the size of the processer itself remains fairly constant, this means that more and more transistors need to fit into ever smaller spaces. Intel's latest generation of chips, Penryn and Nehalem, have 45nm feature sizes and in a quad-core configuration have over 731 million transistors[4].

Moore's Law was intended to be a function of size and cost; the number of transistors would not only be function of the number that could physically fit on the chip, but that this would also be at the optimal price point, beyond which defect rates and other factors would increase the cost of the chip.

This scaling of features cannot continue indefinitely. Already the lithography process by which silicon chips are fabricated has had to change to reflect the fact that feature sizes are smaller than the wavelength of light - visible light has wavelengths down to 380nm, and so Extreme Ultraviolet (10nm) is increasingly being used. Eventually quantum effects will be seen, in which electrons pass through barriers of 1nm, despite not normally having sufficient energy to do so.

In terminology widely used by the electronics industry, nanotechnology which leads to continued miniaturization belongs to the More Moore technology domain. Whilst this does involve working with nanometre length scales, it is often considered that this is not true nanotechnology, as the main task is scaling existing processes. Novel processing methods, such as self-assembly, belong to the Beyond CMOS domain.

Figure 1: Moore's Law and More[5]

Beyond CMOS

The Beyond CMOS domain refers to a set of disruptive functions that, in the long term, will complement or replace conventional silicon technology. The novel devices should show significant advantage over the "ultimately scaled" CMOS transistors in power, performance, density, and/or cost to enable the semiconductor industry to extend the historical cost and performance trends for information technology.

Two main approaches are used to develop technologies that would extend the functional scaling in the 2020 or beyond timeframe. One is heterogeneous integration of new technologies with the CMOS platform (functional diversification - see next section "More than Moore"). The second is to develop fundamentally new approaches to information processing.

Nanotechnology will enable developing alternative schemes to encode and process information in order to discover a new mechanism for computing that goes beyond simply improving today's transistor. For a large part, future information processing will be done on information where the state variable is something other than electronic charge. This information includes optical images, image sequences, speech, and data sets derived from physical sensors. Alternative options for new information carriers include e.g. spin, molecular state, photons, phonons and quantum state.

Nanotechnology will also develop new methods of processor manufacture, including nanoimprint lithography, printed electronics and even self-assembly. These technologies have the potential to displace existing silicon CMOS fabrication methods, and reduce the requirement for multi-billion Euro manufacturing facilities. Before doing so, these methods will need to demonstrate the cost-effectiveness and reliability currently demonstrated by Silicon CMOS.

More than Moore

Nanotechnology is believed to also have a significant impact on the development of other functional components beyond processors and memory.

Nanophotonics, the control of light at sub-wavelength scales, is one of the most promising areas of nanotechnology development, with applications across a wide range of sectors. The technology domain includes 1 dimensional, 2D, and 3D nanostructures, quantum dots, metamaterials which demonstrate negative refraction, plasmonics, and a number of other research areas.

Applications for nanophotonics include displays, lighting, optical data communication and interconnection, as well as photovoltaics, sensing, and imaging/instrumentation. Europe has an active photonics area, including both large industrial players (such as Philips, Osram, Thales) and newer companies, often spun out from research centers.

Nanotechnology advances in displays can be divided into two broad categories; those which rely on the field emission properties of nanostructured materials (such as Field Emissive and Surface Emissive Displays) and those which involve the use of new materials and patterning technologies to allow the use of organic layers and/or flexible substrates.

New memory technologies enabling exponentially higher storage capacities will also be enabled by nanotechnology. A wide range of technologies is being considered, including MRAM, FeRAM and phase change memory.

Because of the importance of nanotechnology for electronics, almost every major manufacturer has research activities in one form or another. The largest European initiative is the Nano 2012/ Crolles3 programme, funded by the French government with approximately €457M, and with STMicroelectronics as the main corporate partner[6]. IBM and Hewlett-Packard have some of the most productive nanotechnology research, having recently been responsible for the development of racetrack memory and the memresistor, respectively.

Developments in nanotechnology for ICT will also have an effect on the countless industries in which ICT plays an important role. To take a seemingly trivial example, miniaturisation of memory enabled the development of the iPod, which in turn has had a significant impact on the music industry, changing it from being a producer of physical products (CDs) to digital goods.

Drivers and Barriers to Innovation

As has been mentioned, the electronics industry is highly competitive and demanding of continuous performance improvements. This means that the electronics industry has a high propensity for innovation. Of the 20 global firms that spend most on R&D, six are in the electronics sector (Nokia, Samsung, IBM, Intel, Matsushita and Sony). Combined, these firms spent 35bn USD on R&D in 2007.[7]

Electronics firms have also been the largest industrial funder of nanotechnology research, investing US$2.88bn in 2006, accounting for 54% of total industrial investment[8]. This is indicative of the extent to which so much electronics research and development extends into nanotechnology domains.

Drivers of Innovation

One of the major drivers of innovation is that the highly competitive electronics marketplace spurs continuous product improvements, in order to eke out competitive advantage, however temporary. Exponential development is seen in areas of electronics beyond integrated circuits, with dramatic increases in data storage density, for example.

Cost reduction is another driver, though it is worth noting that this could also be seen as a barrier. New technology approaches, unless they demonstrate dramatically enhanced performance, are unlikely to gain market share until they become cost competitive with existing approaches. Cost drivers can be seen in areas such as CNT thin film transistors, in which one of the drivers is to replace an expensive material input, Indium Tin Oxide.

Changing end-user needs provide an impetus for innovation. As the ENIAC SRA argues, ICT is being incorporated into a wider variety of domains, including healthcare, transport, and security. These new applications require different technologies.

Barriers to Innovation

A number of features of ICT drive innovation and development in other industries in turn. However, that is not to say that there are no barriers to innovation in ICT - they exist, and at both global and regional levels.

For one, the investment that has been sunk into CMOS research, process development, and fabrication facilities gives rise to path dependency - companies would rather prioritise R&D activities that prolong the life of CMOS (a rational economic strategy). This is clearly a factor in deciding where to allocate resources, and one which becomes even more acute in challenging economic times. However, it has not seemed to prevent companies in this space from considering more future orientated nanotechnology research.

The dominance of the CMOS paradigm is also indicative of the highly (and continuously) optimised state which it has reached. To supplant this for semiconductor production and alternative technology would need to demonstrate cost effectiveness, low fault tolerance, and high throughput reliability across billions of produced units.[9]

The report "Shaping the ICT Research and Innovation Agenda for the Next Decade"[10] identified several potential barriers to the development of an ICT supply industry in Europe. The three most important were found to be:

Weak public financial support for R&D
Market fragmentation
Lack of world-class research and innovation clusters

‘Weak public financial support for R&D' refers not only to direct public funding of ICT, but also the ways in which the public sector can support ICT development indirectly. One lever available to the public sector to stimulate ICT development, public procurement, is more heavily utilised in the United States. The current European level of pre-commercial ICT procurement is at €1bn, one tenth of the US figure.[11]

Market fragmentation increases the complexity of innovation and business activities, and refers specifically to there being a very large number of potential customers, operating within differing local market structures, conditions and regulations.

The point that Europe was lacking world-class research and innovation structures drew a polarized response in regard to micro-nano electronics; 23% felt that Europe was strong in this area, whilst 24% felt that Europe was weak. It is certainly the case that a number of initiatives including ENIAC, and the new European Commission ICT strategy published in March 2009[12], intend to develop European research excellence in ICT.

Relevant Sector Segmentation and Applications

This report considers five of the major applications for nanoelectronics and photonics:

Integrated circuits
Electronics manufacturing equipment
Displays
Data storage

This correlates closely to the technology area sector segmentation in the ObservatoryNano technical reports. Typically each of these applications then has a very wide range of end-user uses - in the case of integrated circuits; for example, there are a very large number of devices that utilise an IC in some way.

Possible Future Products and Time Range

This timeline summarises the product introduction timelines that are discussed within each application sector.

Application	Commercially Available	1-3 years	3-5 years	5+ years
Integrated Circuits	ICs with 45nm features (Intel)	Continued shrinking of feature sizes		Beyond CMOS ICs
Manufacturing Technologies	Atomic Layer Deposition Nano Imprint Lithography stamps
Displays	OLED displays, up to 13”	CNT thin films. SED displays Larger OLED displays	FED displays
Memory	MRAM modules FeRAM	Continued development of MRAM	Phase Change Memory Resistive Random Access Memory	Racetrack Memory