This post continues our evaluation of the use of small digitally-controlled “fabbers” for small-scale manufacturing as a response to energy decline and economic collapse. Digital fabbers are a key part of the strategy which thinkers such as John Robb have for creating “resilient communities” that can weather the changes wrought by globalism and the hollowing-out of nation states by criminals, “terrorists,” and the extremely rich. (Mr. Robb's vision for resilient communities is driven primarily by the continued availability of cheap, advanced electronic technology. For an example of this, see Global Guerrillas: RESILIENT COMMUNITY: Technological Acceleration.)
I have only limited time for a post this week. Therefore I will give a brief summary of the things I've found relating to a vital component of digital fabbers: the microelectronics that control them. An advanced industrial society that depends on the localized, decentralized manufacture of the things it needs must be able to produce most or all of its necessary goods in this way. This applies to its electronic and microelectronic devices as well. In the last couple of posts, we have examined inorganic, silicon-based microelectronics manufactured in the ways currently employed in commercial industry.
This has revealed that the conventional manufacture of conventional silicon-based microelectronics uses large amounts of energy and costly equipment. The energy and equipment are necessary in order to produce the ultrapure silicon compounds that comprise the nanocomponents contained in most chips. According to a paper by Nestor Gonzalez Brasero of the Fundacion Telefonica, it takes 1.6 kg of fossil fuel, 72 grams of chemicals and 32 liters of water to produce a 32Mb DRAM chip whose final weight is only two grams. In 2005, 81 million desktop computers were manufactured in China, at an energy cost of 54 terawatt-hours.
It seems reasonable to assume that our expectation of the increasing availability of ever more advanced and cheap silicon-based microelectronics will one day hit a dead end. This is especially true for silicon-based components made in the standard way. Thus there has been a great deal of interest in the promise of electronics that can be made from organic compounds or other materials that don't require the high energy inputs and large-scale manufacturing plants needed to make ultrapure silicon wafers. From the turn of the 21st century various industry groups, researchers and labs have announced their interest in creating electronics, usually using inexpensive organic semiconducting compounds, that can be built by simple ink-jet printing onto a plastic substrate.
My digging around the Web has turned up the following facts:
The semiconducting property of certain organic chemicals has been known since at least the 1970's. In fact, melanin, a component of dark or tanned skin, is itself a natural organic semiconductor with some surprising switching properties.
Organic semiconductor electronics are now being used in some flat-panel displays. (The Pioneer Corporation has been producing a car stereo with an organic electronic display since 2001.)
Organic semiconductor electronics are being studied for use in printable RFID tags.
Organic semiconductor electronics are nowhere near fast enough to compete with crystalline silicon electronics at present. This is because charge transport velocities in organic semiconductors are orders of magnitude slower than in many silicon-based components. Their low charge transport velocity translates into switching speeds much lower than conventional silicon microelectronics.
Organic electronics require airtight protective coatings without which their performance degrades rapidly. In any case, their actual useful lifetime is limited to just a few years, and they will degrade whether they are used or not.
In order to create printable microelectronics with speed and performance comparable to conventional silicon-based chips, researchers are having to augment their organic electronics with metal atoms or to resort to the use of exotic materials, such as carbon nanotubes or silicon nanofibers. This detracts from the initial promise of low cost and easy fabrication of organic electronics. The making of these exotic materials still requires the same sort of high-energy, large-scale manufacturing plants that are now typical of silicon-based microelectronics. In addition, manufacturing of industrial-grade carbon nanotubes creates some serious potential pollutants.
I don't have a good handle on the energy cost involved in making organic electronics, as the technology is still in its infancy and many companies now doing research aren't interested in divulging their secrets. But I do have a couple of hunches. First, I suspect that printable and organic electronics are not yet ready to lead some next-generation “digital revolution.” Secondly, I think that large-scale, energy-intensive centralized manufacturing will still be required for many years in order to produce the fast, high-performance electronics of the sort that could power decentralized, localized manufacture of high-tech components and systems. This is a natural consequence of the need to use a lot of energy to turn high-entropy, highly disordered materials into materials with a high degree of order on a submicroscopic scale.
In short, I expect that the widespread manufacture of microelectronics will continue to remain an energy, resource and capital intensive process, although cheap, low-energy alternatives will be preferred for devices that don't have to be especially fast. It is here that printable and organic electronics can make a difference, as long as their users don't expect too much. Because of the requirement for energy and resources, I think the availability of cheap and powerful silicon-based electronics will decline as time passes. Devices based on silicon microelectronics will get more expensive. (And even if we ignore the probability of global energy decline, I think there's still "a ways to go" before you 'll be able to “print” a laptop in your garage using your own handy-dandy digital fabber ;) )
There's more to say, but I'm out of time. (Gotta get away from this computer screen!) I'm including a bunch of links below, for anyone who wants to dive in to this subject further. Enjoy.
For Further Reading