Microsystems for Bioelectronics

Subjectively, it seems reasonable to postula te that an embedded system of this size would contain

only minute and harmless amounts of materials that in larger quantities might be harmful to the body

and, furthermore, that the normal body waste disposal processes could manage the removal of

nanomorphic cells when they have reached the end of their useful lives. The nanomorphic cell would

need to employ some sort of triggering mechanism to sign al its elimination from the body. Of course,

this is all hypothetical and would need to be verified, e.g., by careful toxicology studies. The in vivo

functional nanomorphic cell is used as an example throughout the text as a vehicle to motivate the

study of the impact of extreme scaling on syst em component performance limits.

1.2 ELECTRONIC SCALING

Electronic circuits and systems are constructed from a number of components, the most basic of which

is the semiconductor transistor (see Chapter 3), that is used in digital applications as a binary switch.

Tremendous progress has been achieved in reducing the physical size of semiconductor transi stors –

within the last 40 years, the number of transistors in a ~1 cm

2

integrated circuit (IC) chip increased

from several thousand in the 1970s to several billion in 2010. The long-term trend of transistor scaling

is known as Moore’s Law: The number of tran sistors in an IC chip approximately doubles every two

years (see Fig. 1.1a). Moore’s Law has been one of the major drivers for the semiconductor industry.

The increased complexity of the ICs, accompanied by exponential decline in cost per function, h as

resulted in increased functionality and expansion of application space of semiconductor products.

The tremendous increase in the number of transistors per chip was enabled by scaling – devel-

opment of technologies to make smaller and smaller transistors, whose critical feature size decreased

from ~10

m

m in the 1970s to ~20 nm (transistor gate length, L

g

) in 2010 (for brief definitions of feature

sizes see Box 1.1). According to the International Technology Roa dmap for Semiconductors (ITRS)

[1], feature sizes may be as small as ~5 nm by 2022 (the trend of decreasing critical device size is

shown in Figure 1.1b). At this nanoscopic scale, the prope rties of matter may be different from those of

bulk materials and the physi cal effects of these nanomaterials may play a role in nanodevice operation.

Fabrication of such tiny structures requires significant technological advances, thus nanofabrication.

The entire field is often called nanotechnology and the subfield related to electronics is called

nanoelectroncs. To be sure, the chips themselves typically have a footprint on the order of 1 cm

2

;in

other words the current and most of anticipated nanotechnological solutions are implemented with

devices with nanometer features while the integrated system features are in the millimeter range.

While the scaling limits of individual devices can be estimated from physical considerations [2],

the question remains open of how small a functional system can be and still offer useful functionality.

To address this question, it is important to understand not only the device characteristics, but also how

connected systems of these devices might be used to perform complex functions. In the context of this

text, a ‘system’ performs a number of ‘functions’ (F) in response to external stimuli, which are taken to

be information flows (I). A schematic representation of a microelectronic system is shown in

Figure 1.2. It cont ains six essential units:

S – A sensing unit that receives inputs (information) from the outside world.

A – An actuat or that performs an ‘action’ on the outside world.

C – A communication unit that transmits information to the outside world.

M – A memory unit that stores instructions , algorithms, and data.

2 CHAPTER 1 The nanomorphic cell