National Academies Press: OpenBook

Technology and Environment (1989)

Chapter: FRAMEWORKS FOR ANALYSIS

« Previous: Technology and Environment: An Overview
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 23
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 24
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 25
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 26
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 27
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 28
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 29
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 30
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 31
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 32
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 33
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 34
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 35
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 36
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 37
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 38
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 39
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 40
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 41
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 42
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 43
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 44
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 45
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 46
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 47
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 48
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 49
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 50
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 51
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 52
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 53
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 54
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 55
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 56
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 57
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 58
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 59
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 60
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 61
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 62
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 63
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 64
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 65
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 66
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 67
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 68
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 69
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 70
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 71
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 72
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 73
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 74
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 75
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 76
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 77
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 78
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 79
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 80
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 81
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 82
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 83
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 84
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 85
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 86
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 87
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 88
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 89
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 90
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 91
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 92
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 93
Suggested Citation:"FRAMEWORKS FOR ANALYSIS." National Academy of Engineering. 1989. Technology and Environment. Washington, DC: The National Academies Press. doi: 10.17226/1407.
×
Page 94

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Technology and Environment 1989. Pp. 23~9. Washington, DC National Academy Press. Industrial Metabolism ROBERT U. AYRES We may think of both the biosphere and the industrial economy as systems for the transformation of materials. The biosphere as it now exists is very nearly a perfect system for recycling materials. This was not the case when life on earth began. The industrial system of today resembles the ear- liest stage of biological evolution, when the most primitive living organisms obtained their energy from a stock of organic molecules accumulated during prebiotic times. It is increasingly urgent for us to learn from the biosphere and modify our industrial metabolism, the energy- and value-yielding pro- cess essential to economic development. Modifications are needed both to increase reliance on regenerative (or sustainable) processes and to increase efficiency both in production and in the use of by-products. In this chapter, mass flows for key industrial materials of environ- mental significance, and the waste emissions associated with them, are reviewed along with the environmental impact of the waste residuals; eco- nomic and technical forces driving the evolution of industrial processes; long-range tendencies in the development of the industrial metabolism; and some applications of "materials-balance" principles to provide more reliable estimates of outputs of waste residuals. The Appendix contains theoretical explorations of the biosphere and the industrial economy as materials-transformation systems and lessons that might be learned from their comparison. Before presenting the main analytic framework, it is useful to begin with some positive examples of how industrial metabolism can shift in the direction of increased efficiency in materials flows and waste streams. It 23

24 ROBERT U LYRES has been justly remarked that the history of the chemical industry is one of finding new uses for what were formerly waste products (Multhauf, 1967~. One of the most interesting early examples of such an innovation was the Leblanc process (the predecessor of the Solvay process) for manufacturing sodium carbonate circa 1800. As a raw material, it made use of sodium sulfate, an unwanted by-product of the eighteenth-centu~y process for manufacturing ammonium chloride (sal ammoniac). Sal ammoniac was used for cleaning metals and as a convenient source of ammonia, but sodium sulfate had no use. The Leblanc process reduced sodium sulfate to sodium sulfide by heating it in a furnace with charcoal. This, in turn, was heated with calcium carbonate (chalk), which induced a reaction producing the desirable sodium carbonate and a new waste product, calcium sulfide (Multhauf, 1967~. Meanwhile, the market for ammonium chloride failed, so sodium sulfate had to be produced in the same complex by reacting sulfuric acid with common salt. This yielded hydrogen chloride (hydrochloric acid) as another by-product. In this case, the hydrochloric acid quickly found a practical use in the manufacture of chloride of lime as a bleach for the rapidly growing textile industry. The other waste product, calcium sulfide, was not successfully used until the 1880s, as a source of sulfur for the manufacture of sulfuric acid. One of the most important waste products of the early nineteenth century was coal tar, which was generated in large amounts by gasworks making "town gas" for illumination. It was a systematic search for useful by- products, initiated by German chemists, that resulted in the creation of the modern organic chemical industry. The synthetic aniline dyes introduced after 1860 were all essentially derived from chemicals obtained from coal tar. Phenolic resins, aspirin, and the sulfa drugs are later examples of derivatives. Until recently, even natural gas was regarded as a by-product of petroleum production. Although it was used almost from the beginning as a fuel for refinery processes and illumination in local areas, much of it was wasted by "flaring" (indeed, this is still true in remote areas of the, world). It had no significant chemical uses until World War II, when natural gas became the feedstock for producing ethylene, and thence butadiene and styrene, the major ingredients of synthetic rubber. Chlorine is a final example. It is manufactured jointly with sodium or sodium hydroxide (lye) by electrolysis of salt or brine. When this process was introduced In the 1890s, it was the sodium hydroxide that was wanted for a variety of purposes, including petroleum refining and soap manufacturing. At the time, chlorine was a low-value by-product, which was luckily available for use in municipal water treatment The development of a number of valuable chlorine-based organic chemicals (e.g., the most

INDUSTRIAL METABOLISM TABLE 1 Mass of Active Materials Extracted Commercially in the United States, 1960-1975 (million tons) Year Material1960196519701975 Food and feed crops267295314403 (excluding hay) Meat, fish, and dairy products82858584 Cotton, wool, hides, and tobacco5544 Timber (who moisture basis)256267271249 Fuels (coal, lignite, oil, gas)9901,1761,4581,392 Ores(Fe, Cu. Pb, Zn)400435528460 Nonmetallics2ooa24Oa266255 Total2,2~a2>ooa2,9262,847 25 - aEstimated value. NOTE: Vegetable material harvested directly by animals has been omitted for lack of data, along with some obviously minor agricultural and horticultural products. Figures for metal ores exclude mine tailings and gangue removed to uncover ore bodies. Inert construction materials such as stone, sand, and gravel have also been omitted. Inert materials account for enormous tonnages, but undergo no chemical or physical change except to the extent that they are incorporated in concrete or paved surfaces. The table also excludes soil and subsoil shifted during construction projects or lost by erosion. SOURCE: U.S. Bureau of the Census (1960 1975~. common industrial solvents and refrigerants, pesticides, herbicides, and plastics such as polyvinyl chloride) actually reversed the position. By the l950s and 1960s, chlorine was the primary product and sodium hydroxide was the less valuable by-product. MASS FLOWS AND WASTE EMISSIONS Our economic system depends on the extraction of large quantities of matter from the environment. Extraction is followed by processing and conversion into various forms, culminating in products for "consumption." The U.S economy extracts more than 10 tons of "active" mass per person (excluding atmospheric oxygen and fresh water) from U.S. territory each year. Of the active mass processed each year, roughly 75 percent is mineral and "nonrenewable," and 25 percent is, in principle, from renewable (i.e., biological) sources as shown in Bible 1. Of the latter, much is ultimately discarded as waste, although much of it could (in principle) be used for energy recovery. It is difficult to estimate the fraction of the total mass of processed

26 ROBERT U LYRES active materials that is annually embodied in long-lived products and capital goods (durables). None of the food or fuel is physically embodied in durable goods. Most timber is burned as fuel or made into pulp and paper products. At least 80 percent of the mass of "ores" consists of unwanted impurities (more than 99 percent in the case of copper). Of the final products made from metals, a large fraction is converted into "consumption goods," such as bottles, cans, and chemical products, and "throwaway" products such as batteries and light bulbs. Only in the case of nonmetallic minerals (if inert materials are ignored, as before) is as much as 50 percent of the mass embodied in durable goods (mainly Portland cement used for concrete and clays used for bricks and ceramics). The annual accumulation of active materials embodied in durables, after some allowance for discard and demolition, is probably not more than 150 million tons, or 6 percent of the total. The other 94 percent is converted into waste residuals as fast as it is extracted. Combustion and carbothe~mic reduction processes are the major sources of atmospheric pollutants today but by no means the only important ones; nor is the atmosphere the only vulnerable part of the environment. From a broader environmental perspective, the production and dispersal of thousands of synthetic chemicals many new to nature, and some highly toxic, carcinogenic, or mutagenic- and the mobilization of large tonnages of toxic heavy metals are of equally great concern. The complexity of the problem Is too great to permit any kind of short summary. However, two points are worthy of emphasis. First, as noted above, most materials "pass through" the economic system rather quickly. That is to say, the transformation from raw material to waste residual takes only a few months to a few years in most cases. Long-lived structures are very much the exception, and the more biologically "potent" materials are least likely to be embodied in a long-lived form. The second point is that many uses of materials are inherently dissipa- tive (Ayres, 1978~. That is, the materials are degraded, dispersed, and lost in the course of a single normal use. In addition to food and fuels (and additives such as preservatives), other materials that fall into this category include packaging materials, lubricants, solvents, flocculants, antifreezes, detergents, soaps, bleaches and cleaning agents, dyes, paints and pigments, most paper, cosmetics, pharmaceuticals, fertilizers, pesticides, herbicides, and germicides. Many of the current consumptive uses of toxic heavy met- als such as arsenic, cadmium, chromium, copper, lead, mercury, silver, and zinc are dissipative in the above strict sense. Other uses are dissipative in practice because of the difficulty of recycling such items as batteries and electronic devices. Bibles 2 and 3, which summarize estimates of emission coefficients for various heavy metals by use category and annual emissions attributable to

INDUSTRIAL METABOLISM 27 dissipative uses, make clear the heterogeneity of the materials flows. In some cases the dissipation is slow and almost invisible. For instance, paints (often containing lead, zinc, or chromium) gradually crack, '~weather," and turn to powder, which is washed or blown away. Tires, which contain zinc (and cadmium) salts, are gradually worn away during use, leaving a residue on the roads and highways. Similarly, shoe leather, containing up to 2 percent chromium salts, is gradually worn away to powder during use. Incinerator ash contains fairly high concentrations of heavy metals from a variety of miscellaneous sources, ranging from used motor oil to plastics and pigments. On reflection, many dissipative uses (food and fuel again excepted) are generally seen to be nonessential in the sense that technologies are the- oretically available, or imaginable, that could eliminate the need for them. 1b take one example only: hydroponic agriculture in enclosed, atmospher- ically controlled greenhouses, with genetically engineered pest controls, would eliminate all losses of fertilizers and pesticides to watercourses by way of surface runoff. ENVIRONMENTAL IMPACT OF WASTE RESIDUALS Materials do not disappear after they are used up in the economic sense. They become waste residuals that can cause harm and must be disposed of. In fact, it is not difficult to show that the tonnages of waste residuals are actually greater than the tonnages of crops, timber, fuels, and minerals recorded by economic statistics. Although usually unpriced and unmeasured, both air and water are major inputs to industrial processes and they contribute mass to the residuals-especially combustion products. Residuals tend to disappear from the market domain, where everything has a price, but not from the real world in which the economic system is embedded. Many services provided by the environment derive ultimately from "common property," including the air, the oceans, the genetic pool of the biosphere, and the sun itself. Distortions in the market (i.e., prices) are unavoidably associated with the use of common property resources. Clearly, environmental resources such as air and water have been unpriced or (at best) significantly underpriced in the past. For this reason, such resources have generally been overused.2 As noted, the total mass of waste residuals produced each year by industrial metabolism far exceeds the mass of active inputs derived from economic activities. This is because nearly half of the inputs other than air and water are fossil fuels (hydrocarbons), which combine with atmospheric oxygen and form carbon dioxide and water vapor. The carbon fraction of hydrocarbons ranges from 75 percent in methane to about 90 percent

28 TABLE2 Consumption-Related Emissions Factors: Heavy Metals ROBERT U LYRES Painting Paint Electron Other Metallic and and Tubes and Electrical Metal usea Coatingb PigmentsC Batteries d Equipment Silver 0.001 0.02 0.5 0.01 0.01 Arsenic 0.001 0 05 0.01 N^. Cadmium 0.001 0.15 05 0.02 N^. Chromium 0.001 0.02 05 N^. N^. Copper 0.005 0 1 N^. 0.10 Mercury 0.050 0.05 0.8 0.20 N^. Lead 0.005 0 0.5 0.01 NA. Zinc 0.001 0.02 05 0.01 N^. N^. = Not applicable. aAlloys or amalgams (in the case of mercury) not used in plating, electrical equipment, catalysts, or dental work. Losses can be assumed to be due primarily to corrosion, except for mercury, which volatilizes. b Protective surfaces deposited by dip coating (e.g., galvanizing), electroplating vacuum deposition, or chemical bath (e.g., chromic acid). The processes in question generally resulted in significant waterborne wastes until the 1970s. Cadmium-plating processes were particularly inefficient until recently. Losses in use are mainly due to wear and abrasion (e.g., silve~plate) or to flaking (decorative chrome trim). In the case of mercu~y-tin "silver" for mirrors, the loss is largely due to volatilization. Paints and pigments are lost primarily by weathering (e.g., for metal-protecting paints), wear, or disposal of painted dyes or pigmented objects, such as magazines. (::opper- and mercury-based paints volatilize slowly over time. A factor of 0.5 is assumed arbitrarily for all other paints and pigments. dIncludes all metals and chemicals (e.g., phosphorus) in tubes and primary and secondary battenes, but excludes copper wire. Losses in manufacturing may be significant. Mercury in mercury vapor lamps can escape to the air when tubes are broken. In all other cases it is assumed that discarded equipment goes mainly to landfills. Minor amounts are volatilized in fires or incinerators or lost by corrosion. Lead-acid batteries are recycled. Includes solders, contacts, semiconductors, and other special materials (but not copper wire) used in electrical equipment control devices, instruments, etc. Losses to the environment are primarily through discard of obsolete equipment to 1andfi11s. Mercury used in instruments may be lost through breakage and volatilization or spillage. in anthracite. Petroleum is intermediate. Every ton of fuel carbon is converted into 3.67 tons of carbon dioxide emitted to the atmosphere.3 It is thought that only about half of this amount remains in the atmosphere, but the carbon dioxide level of the earth's atmosphere has risen over the past century from about 290 parts per million (ppm) to around 344 ppm at present. Although the magnitude of the climatic warming (the "greenhouse

INDUSTRIAL METABOLISM 29 Chemical Agri- Nonagri Uses Not cultural cultural Medical, Miscella Metal Embodiedf Embodiedg Uses h Uses i Dental3 neous k Silver 1 0.40 N^. N^. 0.5 0.15 Arsenic N^. 0.05 0.5 0.8 0.8 0.15 Cadmium 1 0.15 NA. N^. N^. 0.15 Chromium 1 0.05 N^. 1 0.8 0.15 Copper 1 0.05 0.05 1 NA. 0.15 Mercury 1 NA. 0.80 0.9 0.2 0.50 Lead 1 0.75 0.05 0.1 NA. 0.15 Zinc 1 0.15 0.05 0.1 0.8 0.15 fuses not embodied in final products include catalysts, solvents, reagents, bleaches, etc. In some cases a chemical is embodied but there are some losses in processing. Losses in chemical manufacturing are included here. Major examples include copper and mercury catalysts (especially in chlorine manufacturing); copper, zinc, and chromium as mordants for dyes; mercury losses in felt manufacturing; chromium losses in tanning,; lead in desulfunzation of gasoline; and zinc in rayon spinning. In some cases, annual consumption is actually loss replacement and virtually all of the material is dissipated. Detonators such as mercury fulminate and lead azide (and explosives) are included in this category. "Uses embodied in final products other than paints or batteries include fuel additives (e.g., tetraethyl lead), anticorrosion agents (e.g., zinc dithiophosphate), initiators and plasticizers for plastics (e.g., zinc oxide), wood preservatives, and chromium salts embodied in leather. Losses to the environment occur when the embodying product is used, for example, gasoline containing tetraethyl lead is burned and largely (lasso) dispersed into the atmosphere. However, copper, chromium, and arsenic are used as wood preservatives and dispersed only if the wood is later burned or incinerated. In the case Of silver (photographic film), it is assumed that 60% is later recovered. Includes agricultural pesticides, herbicides, and fungicides. Uses are dissipative, but heavy metals are largely immobilized by soil. Arsenic and mercury are exceptions because of their volatility. i Biocides used in industrial, commercial, or residential applications. Loss rates are higher in some cases. j Includes primarily pharmaceuticals (including cosmetics), germicides, etc., as well as dental filling matenal. Most uses are dissipated to the environment through wastewater. Silver and mercury dental fillings are likely to be buried with cadavers. k Miscellaneous emissions not counted elsewhere. SOURCE: Ayres et al. (1988~. effect") due to the rising level of carbon dioxide is still quantitatively uncertain, the qualitative impact is likely to be adverse. Carbon monoxide is, of course, toxic to humans and has been impli- cated in health problems among urban populations. It is less well known that carbon monoxide plays an active, and not necessarily benign, role in a number of atmospheric chemical reactions before it is ultimately oxidized to carbon diomde. Inefficient combustion processes convert up to 10 percent

30 ROBERT U LYRES TABLE 3 Emissions from Consumptive Uses: Heavy Metals, United States, 1980 (metnc tons) Metallic Uses Electrical (E!xcept Protective Coverings Other Coatings Plating Paints Batteries Electrical and and and and Uses, Instru Metal Electrical) Coating Pigments Equipment meets, etc. Silver 0.83 a o.o 0.15 0.67 Arsenic 0.04 0.0 0.0 1.97 0.0 Cadmiumb 0.04 136.1 116 7.81 0.0 Chromium 151.80 155.7 6,490 0.0 0.0 Copper c 11,074.00 0.0 0.0 0.0 0.0 Mercury 0.0 0.0 0.0 195.91 1752 Leadc 1,249.00 0.0 48,S00 8,510.00 0.0 Zinc 514.00 8,778.0 77,750 63.00 0.0 bIncluded in first column. cl979. 1977. SOURCE: Ayres et al. (1988). Of fuel carbon into carbon monoxide; carbothermic reduction of iron ore and other metals generates even higher percentages. However, the average percentage over all processes is much smaller. Actual emissions of carbon monoxide to the atmosphere in the United States were about 110 million tons in 1970 (mostly from automobiles and trucks), with a carbon content of 47 million tons. Emission controls reduced the net output of carbon monoxide in 1980 to about 85 million tons (U.S. Environmental Protection Agency, 1986~. The discovery of chlorofluorocarbons (CFCs) in the stratosphere has raised an even more frightening prospect: ozone depletion (see Glas, this volume). In apparent confirmation of this phenomenon, an "ozone hole" has recently appeared in the stratosphere over Antarctica. This "hole" has reappeared each spring for several years and seems to be growing (Clark, 1987; Miller and Mintzer, 1986~. Chlorofluorocarbons are chemically inert gases, discovered in 1928 and produced since the 1930s mainly as refrigerants and solvents and for "blowing" plastic foams (see Friedlander, this volume). The major use is for refrigeration and air conditioning. In most uses, CFCs are not released deliberately, but losses and leakage are inevitable. The problems revealed so far may be only the beginning. If (perhaps it is better to say "when") the ozone level in the stratosphere is depleted, the effect will be to let more of the sun's ultraviolet radiation through to

INDUSTRIAL METABOLISM Chemical _Biocidal Consumer Poison Uses Industrial Uses, Nonagri Catalysts, Additives, Agriculturalcultural Medical, Reagents, Extenders, Pesticides,Pesticides Dental, Explosives,Photography, Herbicides, (Except Pharma- Miscel etc.etc. Fungicides Medical) ceutical laneousTotal _ 7.648 0 0 15; 0.4560 492.00 2,950 5,901 19.70 09,364 029 0 0 0 0.98290 1,297.03,890 0 1,038 0 2,141.0011,659 4,222.00 1,560 0 0 015,452 412.10 16 236 8.36 6.96893 OO O O 0 1,329.0056,900 2,508.018,622 188 251 1,003.00 0109,670 31 the earth's surface. One likely impact on humans is a sharp increase in the incidence of skin cancer, especially among whites. The ecological impact on vulnerable plant or animal species cannot be estimated, at present, but could be severe. Methane, oxides of nitrogen (NO=), and sulfur oxides (SO=) are other residuals that have been seriously implicated in climatic or ecological effects. All three are generated by fossil fuel combustion, as well as other industrial processes. Like carbon monoxide, carbon dioxide, and the CFCs, they can be considered as metabolic products of industry. Methane is released in natural gas pipelines, petroleum drilling, coal mining, and several kinds of intensive agriculture (especially rice cultivation or cattle and sheep farming). Nitrogen oxides are also coproducts of combustion. In effect, at high temperatures, some of the atmospheric nitrogen is literally "burned." Nitrogen oxides are implicated in many atmospheric chemical reactions (including those that create smog) and eventually oxidize to nitric acid, which contributes to acid rain. In the stratosphere, where nitrogen oxides are decomposed by ultraviolet radiation, atomic nitrogen also contributes to the destruction of ozone. Finally, oxides of nitrogen are "greenhouse gases" that contribute to climatic warming (Miller and Mintzer, 1986~. Sulfur oxides are generated by the combustion of sulfur-containing fuels-especially bituminous coal-and by the smelting of sulfide ores. Most copper, lead, zinc, and nickel ores are of this kind. In principle, sulfur can be recovered for use from all these activities, and the recovery rate is rising. However, the costs of recovery, especially from coal-burning electric power plants, are still far higher than the market value of the potential products (e.g., dilute sulfuric acid). Hence, for the present,

32 ROBERT U LYRES calcium sulfites and sulfates-as well as SOc-constitute a waste residual that must be disposed of. This is also true of fly ash.4 These examples show that although today's industrialized economic system may be in rough equilibrium in terms of supply and demand rela- tionships, it is far from equilibrium in thermodynamic terms. Enormous quantities of fossil fuels and high~uality minerals are extracted each year to drive the economic engine. The economic system is stable somewhat in the way a bicycle and its rider are stable: if forward motion stops, the sys- tem will collapse. Forward motion in the economic system is technological progress. EVOLUTION OF INDUSTRIAL PROCESSES It is generally accepted by economists that the mechanism that normally drives the evolution of industrial processes is technological innovation. The primary incentive for taking risks appears to be economic. A new process that saves one link in the chain between raw materials and finished ma- terials or final goods can usually be justified through savings in materials and energr inputs or capital requirements, if not both. Moreover, process technology is inherently easier to protect from piracy than product tech- nology. As noted, most chemical products are intermediates used in the production of other chemicals. Thus, final products result from chains, or sequences, of processes with an overall energy conversion efficiency that is the arithmetic product of the conversion efficiencies at each stage.5 If the typical chain has three steps, each of which has a conversion efficiency of 0.7, the overall conversion efficiency of the chain is about 0.34. A four-step chain would have an overall efficiency of around 0.24. That is, the available energy embodied in the final product might be somewhere between 25 and 35 percent of the available energy of the original feedstocks. Because primary feedstocks are essentially indistinguishable from fuels, efficiency improvements translate directly into cost savings. Clearly, a powerful long-term strategy for improving overall effective- ness in production is the development of new processes to shorten these process chains, bypassing as many intermediates as possible. In other words, one would like to be able to produce final products such as polyethylene or synthetic rubber directly from first-tier intermediates or even from primary feedstocks such as ethane and propane.6 Biological organisms differ from industrial organizations in that they are able to build complex molecules directly from elementary building blocks with relatively few intermediates. Thus, biotechnology offers a long-term prospect of radically higher produc- tion efficiencies and correspondingly lower costs (U.S. Office of Technology Assessment, 1982~. Another long-term strategy for increasing effectiveness is better use of

INDUSTRIAL METABOLISM 33 by-products and wastes. When a process can be justified on the basis of the market for its primely product, by-product sales can be highly profitable. This positive motivation to seek new uses is compounded by the fact that, because of some of the environmental problems already noted, waste generation is being increasingly discouraged by environmental regulation. Moreover, waste disposal is increasingly expensive, and the cost is becoming more uncertain. Firms that buried toxic chemical wastes many years ago in landfills (methods that were regarded as acceptable at the time) are sometimes finding themselves saddled retroactively with heavy costs of digging up the same wastes and disposing of them again in a safer manner. The chemical waste dump used by Hooker Chemical Company at Love Canal, near Niagara Palls, is one example. Meanwhile, suitable sites for disposal of hazardous solid or liquid wastes are becoming scarce.7 Ma- terials that are recovered and reused internally or embodied in marketable products that can be readily and effectively recycled are less likely to cause this kind of problem.8 The incentives for technological innovation in the area of waste re- duction and disposal are not always operating, especially where there are massive market failures. In practice, the "unpaid" environmental damage costs have been deferred and many of them will have to be paid by later generations. But as these costs become larger and more visible, there will be growing political pressure to force the producers and users of fossil fuels (and other materials such as heavy metals) to pay the costs of abat- ing the resulting environmental damages.9 Notwithstanding resistance by energy users, it seems inevitable that in the long run these costs will be added to the prices of fuels and materials. This, in turn, will create sig- nificant economic opportunities for innovations in the area of "low-waste" technologies. EVOLUTIONARY ECONOMICS AND GAIA One view of evolution and probably the most common view among scientists-can be characterized as "the myopic drunkard's walk" The drunkard's walk is not exactly random, but it tends to follow the path of least resistance in the short run. If a mutation or an innovation offers short- term advantages, it will be adopted. In this view, there is no long-range force or tendency to approach a distant goal. I,here is no Aristotelian "final cause." Most scientists tend to regard more teleological theories, including Lovelock's Gala hypothesis ~velock 1988), with skepticism, because of their aroma of mysticism. In the case of biological evolution, indeed, it was difficult for a long time to suggest a likely mechanism whereby short-term advantages to the individual could in some cases be overridden in favor of longer-term benefits

34 ROBERT U LYRES to the community or the species. Even when such a mechanism is suggested (e.g., the kinship hypothesis), there may be difficulties in explaining how it could have become programmed into the genetic code. Yet, such behavior is a biological fact. The only real argument is whether a "strict" myopic test of the immediate utility of every mutation would have directed biological evolution along pathways significantly different from those that have actually occurred. In the case of economic evolution, there is no difficulty in identifying possible mechanisms for selecting goal-oriented pathways. The decision makers are humans with the ability to look and plan ahead. The question here is, How far ahead do industrialists plan? The answer surely varies with individuals and circumstances, but it can be decades and is seldom less than a few years. Thus, long-term goal orientation in economic life is not particularly rare. 1b relate this general point to the present context, one might postulate a long-run evolutionary imperative favoring industrial metabolic technolo- gies that result in reduced extraction of virgin materials, reduced loss of waste materials, and increased recycling of useful materials. For conve- nience, one might refer to this overall trend if it exists as an imperative to reduce materials intensiveness, or dematerialization (see Herman et al., this volume). In general terms, the same long-term tendency can be observed in the biosphere, as noted in the Appendix. It must be pointed out that short-term economic incentives do not necessarily point in the direction indicated. For example, market forces appear to favor product differentiation and specialization, but these trends increase the costs of repair and recycling. In poor countries, such as India, there is virtually no such thing as "junk" Any manufactured product, no matter how old or obsolete, is likely to be repaired or rebuilt and retained in service as long as physically possible. When it can no longer be repaired, it will be disassembled and useful parts will be separated for further use; the remainder will be hand-separated by material (stainless steel, iron, aluminum, copper, rubber, plastic, paper) and recycled in "baclyard" operations. By contrast, in advanced countries, manufactured products are becoming more and more complex and correspondingly difficult to repair. This is particularly true of electronic devices such as printed circuit boards and cathode-ray tubes. Moreover, as products are designed to be more reliable, so that repairs are no longer "normal" and expected, disassembly is becoming more difficult and in many cases is actively discouraged. In fact, for warranter reasons, critical subassemblies are often sealed and must be either returned to the factory or discarded. Finally, the complexity of products is often reflected by the increasing complexity of materials, which makes recycling inherently more difficult. ~ take one simple example, worn-out woolen

INDUSTRIAL METABOLISM 35 suits and dresses were once routinely collected and recycled, mainly into coats and blankets. The once-profitable wool-recyclir~g industry has virtually disappeared, because most new clothes are "blends" of natural and artificial fibers that cannot be reprocessed economically. The point of the last two paragraphs is that short-term economic in- centives and trends are often inconsistent with the postulated long-term imperative. One must, therefore, also postulate other counteracting in- centives and mechanisms. Without going into detail, such incentives and mechanisms (if they exist) must grow out of social and even political responses to perceived environmental problems. It is political action, ul- timately, that creates the incentive structure fiscal, monetary, and tax policy and the regulatory environment within which economic incentives drive entrepreneurial activity. Having said all this, I believe that the "de- materialization imperative" is alive and well. The question I now raise is the following: Does this (hypothetical) evolutionary imperative toward reduced material intensiveness have any additional specific implications for future industrial processes? Can it be used as a basis for forecasting? I think the answer is a qualified yes. 1b take one example, a future industrial metabolism deriving its energy ultimately from nuclear fission, nuclear fusion, or the sun itself, rather than from fossil fuels, would necessitate a completely different chemical energy carrier system. The most likely bulk energy carrier appears to be hydrogen. Unless biotechnology offers an attractive alternative, it appears likely that hydrogen would be obtained from electrolytic or thermal decomposition of water. Of the two processes, thermal decomposition is more direct and consequently, if the scale is large enough, more efficient Intermediates that have been suggested for thermal decomposition processes include a variety of compounds of iron, calcium, strontium, mercury, copper, chromium, manganese, and vanadium with bromine or chlorine. Many, if not most, of these compounds are considered hazardous. In consideration of the extremely large scale on which any such process is likely to have to be carried out, it is obviously of ~ the utmost importance that process intermediates be recycled with extremely high efficiency (of the order of 99.9 percent). This is a severe challenge for chemical engineering and one that will probably be a major concern in future decades. As another example, if we postulate much higher prices for carbon- based fuels (to reflect the environmental damage associated with their use), the problem of mobile power takes on a new dimension. Electric propul- sion based on nuclear electricity (from "inherently safe" high-temperature gas-cooled reactors) is one possibility favored by many. But practical im- plementation on a large scale requires a technological breakthrough of unlikely magnitude in the area of compact electrical energy storage sys- tems. Hydrogen has also been suggested as a possible substitute for liquid

36 ROBERT U LYRES fuel (most likely for jet aircraft), but liquid fuels are far more convenient for most purposes. The evolution of industrial metabolism can also be addressed from the perspective of eliminating dissipative uses of tone heavy metals (arsenic, cadmium, chromium, copper, lead, mercury, nickel, silver, and zinc) and of halogenated hydrocarbons. All of these materials have been implicated in environmental problems ranging from "Minimata disease" (mercury poisoning) to erosion of the ozone layer in the stratosphere. In the long run, it appears likely that all of them will have to be replaced or used only in applications permitting an extremely high degree of recycling. It is probably safe to say that the industrial metabolism of the next century will recycle many of the waste products that are produced in the largest quantities today, notably sulfur, fly ash, and lignin wastes from the paper industry. Lignin wastes may yet turn out to be a useful growth medium for single-cell organisms, providing high-protein supplements for food products. By the same token, inherently dissipative uses of biologically active materials will have to decrease as the mistakes of the past are rectified. In particular, the underpricing of environmental and exhaustible resources must be reduced or even (temporarily) reversed. APPLICATIONS OF MATERIALS-BALANCE PRINCIPLES The materials-balance principle, a straightforward application of the first law of thermodynamics (widely used in the design of chemical engi- neering systems, for example), is a potentially valuable and underutilized tool for using economic data in environmental analysis. Frequently, a com- bination of input data (obtainable from economic statistics), together with technical process data available from engineering analysis, gives a more reliable estimate of waste residual output than direct measurements alone could be expected to do. This principle is particularly true when the pol- lutant of concern is produced in relatively small quantities and is emitted together with large amounts of combustion products or process wastewater. A good example of the mass-balance principle has been taken from a study of environmental problems in the aluminum industry. One of the major environmental problems associated with aluminum smelting in the past was the emission of gaseous fluorides from the smelter. The source of fluorine is the electrolytes (molten cryolite and aluminum fluoride) used as a solvent for alumina in the electrolytic cell. An unavoidable side reaction in the cell breaks down these electrolytes and releases some of the fluorine at the anode. Exact "recipes" for the production of aluminum are known only by the aluminum companies. However, a materials-balance analysis for the year 1973 suggested that for each 100 kilograms of aluminum produced, 2.1 kilograms of c~yolite and 3 kilograms of aluminum fluoride

INDUSTRIAL METABOLISM 37 were consumed as inputs (Ayres et al., 1978~. Based on these estimates and straightforward chemistry, the aluminum industry would have accounted for 40 percent of the known production of hydrofluoric acid in that year, which is consistent with both official (U.S. Bureau of the Census, 196~1975) and unofficial estimates (Ayres et al., 1978~. In the absence of fluorine recovery facilities, all of the fluorine consumed by the industry must have been emitted eventually to the environment. In other words, the total amount of Woolite and aluminum fluoride consumed by the aluminum industry was (and is) exclusively to replace fluorine losses. It is interesting to note that the fluoride emissions calculated by using materials-balance principles were about twice as high as the Environmental Protection Agency's published estimates at the time. The latter were based on direct (but unreliable and difficult to verify) measurements taken at a few smelter sites. Assuming the production and use statistics for hydrofluoric acid were correct, one would have to believe that the indirect estimate based on materials-balance considerations was probably more reliable than the estimate based on partial and questionable direct measurements. Another application of materials-balance methodology is in the recon- struction of historical emissions data. This is a problem of some importance to basic environmental science, because the cumulative impact of air or water pollution over long periods can be evaluated only in relation to a baseline of some sort. In this context it becomes important to know more about emissions in the past, when no measurements were made. ~ be sure, sediments and ice cores offer some help, but not enough. The picture can be clarified considerably, however, with the help of synthetic models using production and consumption data (which are often imperfect, but better known than emissions) together with engineering analysis of processes. Sometimes process information is not even needed, as when emissions are linked directly to inputs. For example, fairly good historical estimates of SON emissions, required to analyze the long-term impact of acid rain, among other things, can be reconstructed easily from historical statistics on coal consumption and copper, lead, and zinc smelting (e.g., Gschwandtner et al., 1983~. This is because the sulfur content of coal and metal ores can be assumed to be the same in the past as it is today, and until recently all of that sulfur was emitted straight to the atmosphere. Reconstruction of NON emissions data is slightly more complicated, but the approach is basically similar (Gschwandtner et al., 1983~. More complex reconstructions of historical emissions data have been undertaken recently, e.g., for the Hudson-Raritan estuary (Ayres et al., 1988~. I\NO examples of material process-product flows, taken from the above study, are presented in Figures 1 (cadmium) and 2 (chromium). Another way of using the materials-balance approach is in the analysis of materials "cycles." The water cycle, the carbon cycle, and the nitrogen

38 - A)loying Metallurgical \ - - - - \ - - \/ ~ ~ /~ ¢) / ~lenide ~sulfide Misc. Uses Including Photography & Pottery 1.3 x 105Lbs Distill in Graphite Retort, Air Oxidation Manufacture of Cadmium Ball Anodes for Plating -T i; I Oxide Ball /f ~ Low-Melt Bearings, grazings, etc. 0.5 x 105 Lbs ROBERT U LYRES Secondary Recovery _ Recovery from (Batteries) Flue Dusts (Cu. Pb) _ \ 1 Recovery from Zinc Refining Metal (Ingot or _ Sponge) ', React with H2SO4 React with HCI ~; ~ ^=N Sulfate ~Chloride ~/ React with Ammonium Hydroxide ~adm~ ( Hydroxide ~ CdOH2 J Dissolve in | \\ \ / ~ \\V ~ \\)(\ ~ ~ Cadmium ~ J \\\~ r l l ~1 Pigment* | CdSe-CdS | -HgS+CdS | | Orange | Pigment* | CdSe-CdS | Yellow | Pigment. 1 ZrlS-CdS (*Pigments, Phosphors, etc. 2.5 x 105 Lbs) React with Benzoic Acid React with Stearic Acid 1 'I 1 1 1 1 ~_ | Nickel | Cadmium | Batteries 10.4 x 105Lbs Electrm plating 1 _ ~ 16x105~sl | Stabilizer I | for | 1 PVC Mastic 1 12.5 x lO5LbSI FIGURE 1 Cadmium process-product flows, circa 1969. Except for the bottom row, rectangles indicate processes; ellipses, products or reactions. SOURCE: Ayres et al. (1988~.

INDU512IAL METABOLISM 39 cycle are familiar examples. The concept is also applicable, of course, to flows that are not truly cyclic, as in the case of arsenic (Figure 3~. This concept has been widely used by geochemists, hydrologists, ecologists, and environmental scientists to organize and systematize their work. Such a presentation helps specify geographical scales of analysis. It also facilitates such comparisons as the relative importance of natural and anthropogenic sources. Finally, and potentially most important, it provides a starting point for detailed analysis of the effect of anthropogenic emissions on natural processes. CONCLUSION It appears that we have methods to describe our industrial metabolism better, both qualitatively arid quantitatively. Initial analyses reveal several important points, for example, that many materials uses are inherently dissipative and thus pose difficulties for recycling. Analysis also shows that although residuals do not disappear from the real world of human health and environmental quality, they do tend to disappear from the market domain. Thus, many environmental resources are underpriced and overused. It is also clear that where the production and use of by-products are concerned, many industrial processes involve multiple steps, resulting in a low level of system efficiency, especially in comparison with biological systems. The sum of the argument here suggests that we should not only postulate, but indeed endorse, a long-run imperative favoring an industrial metabolism that results in reduced extraction of virgin materials, reduced loss of waste materials, and increased recycling of useful ones. APPENDIX The Biosphere as a Matenals-liansformation System Three salient characteristics mark the difference-between the natu- rally evolved biosphere and its human~esigned industrial counterpart, the "synthesphere." The first is that the metabolic processes of biological or- ganisms are derived (by photosynthesis) from a renewable source: sunlight. The second characteristic is that the metabolism of living organisms (cells) is executed by multistep regenerative chemical reactions in an aqueous medium at ambient temperatures and pressures. Most process intermedi- ates are regenerated within the cell. Reaction rates are controlled entirely by catalysts (enzymes). The energy transport function is performed in all living organisms by phosphate bonds, usually in the molecule adenosine triphosphate (ATP).~3

40 Hydroxide H2SO4 | Hydroxide . . React with with Sulfurous A c i d fi n W e a k 2SO4 Ituric ~ //1 Sulfate Alum / / I OKpi0,)~ ~ i: ROBERT U AIRS Melt in I Reverberatory Fumace with Lime and Soda or _ Potash ~;~N Iron 0(e (Chromite) `F6Cr2O~ ~ \ ~1 ~ Alkaline Roast , and Leach ~~ - ~ / / Chromate ~ K2CrO4. J Norma / / ~ / ~ / 1\ / React ~ / ~ / with Electro / IN Sulfuric Iytic . / /; H2SO4 ~ Reduction \~ / ~/_ ~ ~ ,~tass~ ~\ ~ \ ~ Bichromate Bichromate i ~\ \L K2Cr2O7 Na2Cr207 Jr\ ~\\ ~1 1 ~ - Stainless | Steel and | Cloy 1 Steel | 1 1 1 Ferro- Mordants Corrosion Lead Chromate chrome (laxatives) Inhibitor Zinc Chromate Ugno- for Dyeing (Primer) sulfate FIGURE 2 Chromium process-product flows, circa 1968. Except for the bottom row, rectangles indicate processes; ellipses, products or reactions. SOURCE: Ayres et al. (1988~.

INDUSTRIAL METABOLISM 41 The third salient characteristic differentiating the biosphere from the industrial synthesphere is that, although individual organisms do generate process wastes- primarily oxygen in the case of plants and carbon dioxide and urea in the case of animals the biosphere as a whole is extremely effi- cient at recycling the elements essential to life. Specialized organisms have evolved to capture nutrients in wastes (including dead organisms) and recy- cle them. A significant exception to this rule in the present geological epoch is the deposition of skeletal remains of zooplankton as sediments on the deep ocean floor. These remains are largely, but not entirely, calciferous.~4 Over geological time periods, some of this sedimentary material is likely to be recycled as chalk, limestone, or phosphate rock The biosphere as it now exists is a nearly perfect materials-regycling system, but this was not the case when life on earth began. The first and most critical evolutionary "invention," from which all else follows, was the process for replicating complex organic molecules. In effect, the information describing the entire living structure is stored as sequences of nucleic acids in the genetic substance known as deoxyribonucleic acid (DNA). The mechanism for storing, coding, transferring, and decoding that information apparently evolved some 4 billion years ago, before species differentiation. Both the code and the mechanism are common to all known living organisms. The first cellular organisms, which appeared about 3.5 billion years ago, were prokaryotes (i.e., cells without nuclei). They obtained the energy needed to sustain the reproduction cycle from the anaerobic fermentation of organic molecules previously created by natural geophysical processes in an atmosphere containing no free oxygen. If the primitive atmosphere had contained oxygen, organic molecules could not have survived long enough to achieve the degree of complexity needed to construct self-reproducing systemS.l5 In cellular fermentation a molecule of glucose is split into two mole- cules of pyruvate. Energy is captured in the form of high-energy phosphate bonds in ADP (adenosine diphosphate) and ATP. Fermentation of a glucose molecule has a net yield of available energy in the form of two molecules of h~gh-energy ATP, converted from the low-energy ADP form (Hinkle and McCarty, 1978~. Further reactions in the cell convert the pyruvate to ethyl alcohol, lactic acid, and carbon dioxide, all of which are excreted as wastes. No oxygen is required. The fermentation-based forms of life could not have been the foundation for a sustainable ecosystem, however, because they were using up a finite stockpile of exploitable organic molecules. The next great evolutionary innovation, about 3 billion years ago, was anaerobic photosynthesis. Me first photosynthesizers were prokaryotic pho- tobacteria. These organisms began to synthesize glucose from atmospheric carbon dioxide and sunlight, thus replacing the depleted organic "soup" of

42 cn o cS 1 . ~ ~: c) o Z UJ ~n 1 o U) a, o C (D o ~ a a) 0~) o C~ '=o a' c C) o .. j ~ ~,, ~,. ,. ,. ~ = ,., A, ~iS, .,W, . ~. ~. ~, ~ ~, ~,.,iON, ti i, a, ~ ~ ~ ~ 3 ~ ·- D ·- c, a' _ C ~ ·_ ._ ~ O ~ a) c Q _ O ~ C) - ~n e~ - o - i . ~ ~ .0 ~ ~ ~ I G E =' ~ cS a, a) O == C C m ~ ~ ~S ~ .. Oo _ o, Co ". 0 - Ct - - ~C :: ._ C~ Ct P4 C) ._ Q) o o ._ Ct - CQ oo ~o C~ Ct C~t ~ ~i ~ CO

INDUSDUAL METABOLISM 43 the primitive oceans. They also produced oxygen as a waste product. At first, the free oxygen was quickly removed by stromatolites, organisms that combined oxygen with iron dissolved in the oceans and precipitated as iron- rich reefs. This seems to have been the origin of hematite deposits being exploited as iron ores today. However, when the dissolved iron was used up, about 1.8 billion years ago, the oxygen level in the atmosphere began to rise, thereby increasing the rate of dissolution of all macromolecules. Thus, the ecosystem was still unsustainable, because it could neither tolerate nor recycle its own toxic wastes. The oxygen toleration problem was partially solved by a new class of aerobic photosynthesizers, the cyanobacteria (or blue-green algae). These appeared a little more than 2 billion years ago. A third great evolutionary innovation the substitution of respiration (using oxygen) for fermentation to obtain energy from organic molecules- solved the recycling problem. The respiration process begins the same way as fermentation, with the splitting of glucose into pyruvate (called glycoly- sis). However, in respiration, glycolysis is followed by a longer sequence of reactions, known collectively as the citric acid cycle. In effect, the pyruvates are oxidized enzymatically to carbon dioxide, with the formation of many more high-energy ATP molecules. In fact, each glucose molecule, when fully oxidized, yields 36 molecules of ATP, whereas the glycolysis stage alone yields only 2. The yield of available energy for further metabolic processes is, therefore, 18 times that of the fermentation process. Because the respiration process is far more efficient than its precursor, aerobic respirators required much less organic material to sustain them. Thus, anaerobic organisms could not effectively compete with aerobic or- ganisms in the presence of oxygen. (They still fill an environmental niche in sediments and deep oceans lacking free oxygen.) With the "invention" of the citric acid circle, the biosphere became sustainable within more self-defined system boundaries. Evolutionary de- velopments since then-of which the most important were the develop- ment of the eukaryotes (cells with true nuclei) and the advent of sexual reproduction made the system more diverse and more efficient. It is in- teresting to note that, despite the radical changes in energy metabolism which occurred, the basic scheme of macromolecular reproduction seems to have remained essentially unchanged for 4 billion years. Industrial liar~sfonnation Processes In contrast to modern biological processes, industrial processes are almost exclusively energized by the combustion of fossil fuels, which (by definition) are not regenerated within the system. In this sense, the indus- tnal system of today resembles the earliest stage of biological evolution, when the most primitive living organisms obtained their energy from a

44 ROBERT U LYRES stock of organic molecules accumulated during prebiotic times. Instead of regenerative cycles powered by solar (or nuclear) energy, industrial pro- cesses are linear sequences of discrete, irreversible transformations (Figure 4~. The sequence begins with extraction of raw matenals, followed by phys- ical separation and elimination of impunties, and subsequent reduction or recombination into convenient "first-tier" intermediates. This category includes prunary metals and other elements In pure form, cellulose, sodium carbonate, ammonia, methane, ethane, propane, butane, benzene, xylene, methanol, ethanol, acetylene, ethylene, propylene, and some others. These materials, in turn, are subsequently recombined into desired chemical and physical forms. Almost all of the processes for reducing metals from ores or producing first-tier intermediates are endothermic, that is, driven by externally sup- plied heat. Processes often use catalysts, and rates and directions are fine tuned by controlled variation of temperatures, pressures, and flow rates or dwell times. There are five major categories of endothermic processes: (1) dehydration; (2) calcination; (3) "reducing" processes for splitting metal (or other) oxides into their constituents; (4) dehydrogenation processes, of which the simplest is the splitting of the water molecule;~7 and (5) processes for combining or synthesizing molecules that do not combine spontaneously at ambient temperatures or pressures. Energy is obtained initially from combustion (oxidation) and delivered either by process steam or by direct contact with the oxidation products. Comparatively few indus- trial processes are electrolytic: the production of aluminum, sodium, and chlorine and the refining of blister copper are the primary examples. The most familiar example of dehydration is the production of plaster of parts from the mineral gypsum. Brick and ceramic manufacturing are also based to some extent on this process. The major example of calcination (and the origin of the name) is the production of calcium oxide (quicklime) from calcium carbonate (limestone) by driving off carbon dioxide. This is a major element in the manufacture of Portland cement. Both dehydration and calcination are accomplished by the simple application of heat. The so-called carbothermic reduction process by which iron ore is converted to pig iron is a typical example of the third category. In this reaction, coke is partially oxidized to carbon monoxide, which in turn reacts with the ore-at appropriate temperatures to reduce the iron oxides, while converting the carbon monoxide to carbon dioxide.~9 The temperature determines which way the reaction goes. If the temperature rises too high (above 1300° C), any pure iron that is present will reoxidize. Ammonia production exemplifies both the fourth and the fifth cate- gories. Dehydrogenation (water splitting) is an example of a process by which synthesis gas for ammonia is produced. In a first stage of the reac- tion, steam reacts with natural gas, in the presence of a catalyst, to produce

INDUS17UAL METABOLISM Embodied Negentropy (Increase in Information) Separation of ore from crust Chemical processes that increase "availability" (or free energy) Matching of specifications for trace elements, crystal size, dislocations, etc. Matching of dimensions and shape specifications 45 Materials Transformation Lost Negentropy (Increase in Entropy) Extraction (Mining, Drilling, Harvesting, etc.) . mew Materia:3 Physical Separation and Refining | (Beneficiation, Smelting, etc.) termedia~ Pure ) Recombination | (Alloying, Synthesis, Heat Treatment, Hot/Cold Working, etc.) ~ Finished \~ Materials J Macro-Forming (Casting/Molding, Machining, _ Fording, etc.) (nished Pas) Work done on environment Heat loss to environment Design | Fabrication and Construction 1 Informationlossas · J finished products 1 `( Finished ~ Products J Use junk. Refuse ] Information loss with wear Disposal ~Dispersal of materials I I FIGURE 4 Representation of the economic system as a multistage system for the extraction, physical separation, recombination, formation, and consumption of materials. At the end of the sequence, materials are returned to the environment in a degraded form as waste. The process of creating a finished product (center column) concentrates information, or negentropy, in the product at the expense of increased entropy in the environment.

46 ROBERT U AYRES carbon monoxide and hydrogen. In a second stage, known as the water-gas shift reaction, carbon monoxide reacts with added steam to yield more hy- drogen and carbon dioxide, which must be removed by a scrubber such as potassium carbonate. The hydrogen is then mixed with nitrogen gas (from the air) in a 3:1 ratio.20 At very high pressures and temperatures, these in- gredients combine endothermicall~again in the presence of a catalyst to form ammonia gas, the basis of virtually all nitrogen-containing compounds used by our industrial civilization. Most of the reactions by which ammonia, chlorine, lime, sulfur, methanol, ethanol, acetylene, ethylene, propylene, or other first-tier inter- mediates are converted to other "downstream" compounds are exothermic and in effect-self-energizing. For example, many second-tier intermedi- ates are produced by controlled oxidation (e.g., sulfuric acid from sulfur, nitric acid from ammonia, acetaldehyde from ethanol or propane, acetic acid from acetaldebyde or butane, acetic anhydride from acetaldehyde, ethylene oxide from ethylene, propylene oxide from propylene, and so on). Most hydrogenation, chlorination, and hydrochlorination reactions are also exothermic, as are most reactions between strong acids and metals or hydroxides. In effect, the first-tier intermediates are the energy carriers for subse- quent reactions. They play a role somewhat analogous to that of ATP in biochemical systems. However, whereas the ATP is cyclically regenerated within the same cell, the first-tier intermediates are not regenerated but are physically embodied in downstream products. This is another funda- mental difference between industrial metabolism in its present form and its biological analogue. NOTES The three major German firms Bayer, Hoechst, and BASE were merged in the 1920s into a single giant under the name I.G. Farl~enindustrie. Farben is the German word for "color." The name reflected their common origin as synthetic dye (color) manufacturers. 2. An important corollary is that the underpricing of environmental resources corresponds to an underpricing of those exhaustible mineral resources whose subsequent disposal as waste residuals causes harm to the environment. This is because of the lack of any link between the market puce paid (for coal, oil, or whatever) and the subsequent cost of waste disposal or more important-of uncompensated environmental or health damages such as bronchitis, asthma, emphysema, cancer, soil acidification, the greenhouse effect, and so on. Here the distinction between renewable and nonrenewable resources is critical: although renewable resources can obviously create pollution problems, such as sewage, they are almost invariably localized in nature and can be abated at moderate cost. This is emphatically not the case for combustion products of fossil fuels or dispersion of toxic heavy metals. floe total world output of carbon dioxide from fuels and cement manufacturing has been estimated to be 5.1 billion metric tons for 1982, of which 26.7 percent was attributable to North America McFarland and Rotty, 1984~.

INDUSTIUAL METABOLISM 47 4. Fly ash is primarily a by-product of coal combustion. At present it is being recovered fairly efficiently, from stack gases of large utility boilers and industrial furnaces, by means of electrostatic precipitatom However, the ash itself has become a large-scale nuisance because there exists no use or market for it. Ibe amounts are large: over 50 million tons are generated annually in the United States alone. Several possible remedies exist. Fly ash is a potential "ore" for several metals, especially iron, aluminum, and silicon. These metals could probably be recovered commercially if, for example, bauxite became unavailable (Ayres, 19823. Alternatively, fly ash could be used as a substitute, or more likely as a supplement, for Portland cement in the manufacture of concrete and concretelike products. fits major disadvantage in this application is that concrete made with fly ash does not harden and set as rapidly as the commercial variety. This has obvious economic costs, but so does the disposal of fly ash in landfills.) Another use of fly ash- already demonstrated in France is as a means for the permanent disposal of toxic liquid wastes in the folm of a hard, impermeable rocklike material suitable for long-term storage. The "chain" analogy is an oversimplification, of course, because many processes yield more than one useful product (the chlor-alkali industry is an obvious example) and many products also require two or more inputs. Thus, the structure of the system as a whole is more like a "tree." 6. For example, the first process to manufacture acetylene proceeded by way of calcium carbide production (from coke and limestone) but was displaced by direct dehydro genation of hydrocarbon feedstocks. The first production of ac~y'onitrile involved a reaction between ethylene oxide (itself the third step in a chain beginning with ethane) and hydrogen cyanide (made from ammonia). This was replaced by an acetylene cyanation process (acetylene being made from methane) and finally by a propylene ammoxidation process (propylene reacting directly with ammonia). Similarly, the fimt process in the manufacture of acetaldehyde started with acetylene (from calcium carbide) or ethanol (from ethylene). A newer process made acetaldehyde directly from ethylene. ~us, in the first example the calcium carbide stage was bypassed. In the second example the oxidation of ethylene was avoided, and in the third example the conversion of ethylene to alcohol was avoided. 7. The problem has been dramatized by several recent episodes in which cities have attempted to dispose of solid wastes by using private contractors who, in turn, thought to transport them to countries where disposal rules are nonexistent or wealcly enforced. For example, the city of Philadelphia contracted with Joseph Paolino & Sons to dispose of incinerator ash. Paolino, in turn, contracted with Amalgamated Shipping, a Bahamian concern, to transport the ash to the Bahamas. However, the Bahamian government bawd the dumping, and the ship carrying the ash was subsequently turned away from ports in the Dominican Republic, Haiti (after it had dumped 2,000 tons of ash), Honduras, Costa Rica, Guinea-Bissau, and the Cape Verde Islands. The ship apparently succeeded in dumping its load of ash somewhere in the Indian Ocean, after more than two years (New York Tones, November 10, 1988~. 8. Admittedly, it can still happen. Asbestos and polychlorinated biphenyls are two examples of materials that were once thought to be safe but have subsequently come to be regarded as hazardous and for which the major producers or users have had to spend billions of dollars to collect and dispose of them safely. Nevertheless, there would seem to be an economic opportunity for a "high-tech" resource-recove~y firm to go into business reconverting fly ash and incinerator ash into its most valuable components, light metals such as aluminum, iron, potash, and titanium (Ayres, 1982~.

48 ROBERT U LYRES A mineralized glassy residue of sodium-silica and heavy metals would, of course, remain for disposal (though it might also find uses as a construction material). 9. This is a straightforward implication of the widely accepted, but sporadically enforced, "polluter must pay" principle. A number of possible processes for large-scale thermal decomposition of water have been suggested and studied in some detail, e.g., by the European Atomic Energy Commission (EURATOM). See, for example, Marchetti (19733. 11. Ammonia, hydrazine, and other compounds have also been suggested, but there is no clear leader at this time. A new study of the economic and technical feasibility of non~arbon-based liquid fuels would be helpful in clarifying the choices. 1Z It is true that some organisms have evolved to function in the deep oceans under conditions of high pressure and salinity, others have evolved to function in surface waters, saline or fresh; still others have evolved to function in environments with very little water. Nevertheless, the internal environment of every cell is aqueous and the pressure inside each cell is essentially the same as the external pressure of the environment in which the organism lives. 13. For a concise summary of the biochemistry of energy transport in cells, see Schopf (1978~. 14. The so-called manganese nodules, which are accretions of iron, manganese, copper, cobalt, and other transition elements, are evidently the result of some combination (as yet imperfectly understood) of biological, chemical, and geological processes (see, for example, Morgenstein, 1973~. 15. Even in the absence of oxygen, it would seem that dissolution of macromolecules should proceed faster than synthesis (Wald, 1954~. The exact evolutionary mechanism leading to self-reproduction is still obscure. Fixation of nitrogen must have been accomplished during this early period, because the early atmosphere was nearly transparent to ultraviolet radiation and any free ammonia in the atmosphere would have been quickly destroyed. 16. The obvious exceptions are elements occurring naturally (such as sulfur) and hy- drocarbons that can be obtained By physical separation from natural gas (methane, ethane, propane, butane) or coal tar (benzene, xylene, toluene). Cellulose occurs naturally in some very pure forms (e.g., cotton), but it is usually obtained from wood pulp lay a chemical digestion process. 17. Other major examples of dehydrogenation include the splitting of methane, ethane, propane, and butane to produce acetylene, ethylene, propylene, butylene, and butadi- ene. 18. Pig iron is a solution of iron carbide in iron, with a typical carbon content of approximately 6 percent. The conversion of pig iron to pure (wrought) iron or steel requires removing this carbon and then adding any desired alloying elements. 19. Multistage reactions such as the reduction of iron ore and the synthesis of ammonia, involving an intermediate (carbon monoxide) that is produced by the reaction and later consumed, are quite common in industry. Less common are reactions involving an intermediate that is not produced within the process but is recycled. One of the first examples of such a process was the Solvay (ammonia-soda) process for manufacturing synthetic sodium carbonate from sodium chloride and calcium carbonate. In this process, ammonium hydroxide reacts with calcium carbonate to yield ammonium carbonate and calcium hydroxide. Ammonium carbonate is converted to ammonium bicarbonate. When this reacts with sodium chloride, sodium carbonate and ammonium chloride are produced. Finally, calcium hydroxide and ammonium chloride are reacted to recover ammonia (gas) for recycling and calcium chloride. Ibe latter is a low-value by-product.

INDUSTRIAL METABOLISM 49 20. In principle, the nitrogen gas and steam could be produced together lay partial oxidation of natural gas (or any hydrocarbon) in air to yield a mixture of nitrogen, steam, and carbon monoxide. The steam in the hot combustion products could then be reacted with additional natural gas to generate hydrogen and more carbon monoxide to be used as feedstock for the shift reaction. However, the presence of nitrogen before it is wanted complicates the engineering unreasonably. REFERENCES Ayres, R. U. 1978. Resources, Environment and Economics: Applications of the Materi- als/Energy Balance Principle. New York: John Wiley & Sons. Ayres, R U. 198Z Coalplex: An integrated enerp,y/resources system concept. United Nations Environment Program Seminar on Environmental Aspects of Technology Assessment, United Nations, Geneva, November-December 1980. Ayres, R. U., J. Cummings-Saxton, and E. Weinstein. 1978. Assessment of methodolo- gies for indirect impact assessment. Anal Report (IRT~68-R/a) prepared for U.S. Environmental Protection Agency. Washington, D.C.: International Research and Technology Corporation. Ayres, R. U., L W. Ayres, J. A. Tarr, and R. C. W~dge~y. 1988. An historical reconstruction of major pollutant levels in the Hudson-Raritan Basin: 188~1980. National Oceanic and Atmospheric Administration Technical Memorandum NOS OMA 43, 3 vols. Rockville, Md. Clark, R., ed. 1987. The Ozone Layer [Series: UNEP/GEMS Environment Library], Vol. 2. United Nations Environment Program, Nairobi. Gschwandtner, G., K C. Gschwandtner, and K. Eldridge. 1983. Historic Emissions of Sulfur and Nitrogen Oxides in the U.S. 1900 1980. Report to U.S. Environmental Protection Agency. Durham, N.C.: Pacific Environmental Services, Inc. Hinkle, P. C., and R. E. McCarty. 1978. How cells make ATP: The prevailing theory is the "chemiosomotic" one. Scientific American 23~3~: 104 123. Lovelock, J. 1988. The Ages of Gala: A Biography of Our Living Earth. New York Norton. Marchetti, C. 1973. Hydrogen and energy. Chemical Economy and Engineering Review 5~1~:7-25. Marland, G., and R. M. Rotty. 1984. Carbon dioxide emissions from fossil fuels: 195~1982. Tellus 36B(4~:23~261. Miller, ~ S., and I. M. Mintzer. 1986. The Sky is the Limit: Strategies for Protecting the Ozone I>yer, Research Report (3~. Washington, D.C: World Resources Institute. Morgenstein, M., ed. 1973. Papers on the Origin and Distribution of Manganese Nodules in the Pacific and Propects for Exploration. Symposium, Hawaii Institute of Geophysics, Honolulu. Multhauf, R. P. 1967. Industrial chemistry in the nineteenth century. Pp. 4~ in Technology in Western Civilization: The Emergence of Modern Industrial Society Earliest Times to 1900, M. Kranzberg and C W. Pursell, Jr., eds. New York: Oxford University Press. New York limes. November 10, 1988. After two years at sea, ship dumps U.S. ash. Vol. 138, lO(N), c4(L3. Schopf, J. W. 1978. The evolution of the earliest cells. Scientific American ~9~3~:11~138. U.S. Bureau of the Census. 1960 1975. Statistical Abstract of the United States. Washing- ton, D.C.: U.S. Government Printing Office. U.S. Environmental Protection Agency. 1986. National Air Pollution Emission Estimates, 194~1984 (EPA~50/4~5~14~. Research Triangle Park, N.C: Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency. U.S. Office of Technology Assessment. 1983 Genetic Technology: A New Frontier. Boulder, Colo.: Westview Press. Mild, G. 1954. The origin of life. Scientific American 191(August):45-53.

Technology and Environment 1989. Pp. 5~69. Washington, DC: National Academy Press. Dematerialization ROBERT HERMAN, SIAMAK ~ ARE, JESSE H. AUSUBEL Until recently the role of consumption as a driving force for environ- mental change has not been widely explored. This may be due in part to the difficult of collecting suitable data. The present chapter approaches the consumption of materials from the perspective of the forces for materi- alization or dematerialization of industrial products beyond the underlying and obviously very powerful forces of economic and population growth. Examination can occur on both the unit and the aggregate level of materi- als consumption. Such study may make it possible to assess current streams of materials use and, based on environmental implications, may suggest directions for future materials policy. The word dematerialization is often broadly used to characterize the decline over time In weight of the materials used in industrial end prod- ucts. One may also speak of dematerialization in terms of the decline in "embedded energy" in industrial products. Colombo (1988) has speculated that dematerialization is the logical outcome of an advanced economy in which material needs are substantially satiated.) Williams et al. (1987) have explored relationships between materials use and affluence in the United States. Perhaps we should first ask the question: Is dematerialization tak- ing place? The answer depends, above all, on how dematerialization is defined. The question is particularly of interest from an environmental point of view, because the use of less material could translate into smaller quantities of waste generated at both the production and the consumption phases of the economic process. 50

DEM47;ERL4LIZAHON 51 But less is not necessarily less from an environmental point of view. If smaller and lighter products are also inferior in quality, then more units would be produced, and the net result could be a greater amount of waste generated in both production and consumption. From an environmental viewpoint, therefore, (deymaterialization should perhaps be defined as the change in the amount of waste generated per unit of industrial products. On the basis of such a definition, and taking into account overall production and consumption, we have attempted to examine the question of whether dematerialization is occurring. Our goal is not to answer definitively the question whether society is dematerializing but rather to establish a frame- work for analysis to address this overall question and to indicate some of the interesting and useful directions for study. We have examined a number of examples even though the data are not complete. Undoubtedly, many industrial products have become lighter and small- er with time. Cars, dwelling units, television sets, clothes pressing irons, and calculators are but a few examples. Were is, of course, usually a lower bound regarding how small objects such as appliances can be made and still be compatible with the physical dimensions and limitations of human beings (who are themselves becoming larger), as well as with the tasks to be performed.2 Apart from such boundary conditions on size and possibly weight of many industrial product units, dematerialization of units of products is perceived to be occurring. An important question is how far one could drive dematerialization. For example, for the automobile, how is real world safen,r related to its mass? In a recent study, Evans (1985) found that, given a single-car crash, the unbelted driver of a car weighing about 2,000 pounds is about 2.6 times as likely to be killed as is the unbelted driver of an approximately 4,000 pound car. The relative disadvantage of the smaller car is essentially the same when the corresponding comparison is made for belted drivers. For two-car crashes it was found that the driver of a 2,000-pound car crashing into another 2,000-pound car is about 2.0 times as likely to be injured seriously or fatally as is the driver of a 4,00~pound car crashing into another 4,000-pound car. These results suggest one of the reasons that dematerialization by itself will not be a sufficient criterion for social choice about product design. If the product cannot be practically or safely reduced beyond a certain point, can the service provided by the product be provided in a way that demands less material? Ib return to the case of transportation, substituting telecommunications for transportation might be a dematerializer, but we have no data on the relative materials demand for the communications infrastructure versus the transportation infrastructure to meet a given need. In any case, demands for communication and transportation appear to increase in tandem, as complementary goods rather than as substitutes for one another.

52 ROBERT HERMAN, S1AMAK A. ARDEKANI, ID JESSE H. AUSUBEL It is interesting to inquire into dematerialization in the world of minia- turization, not only the world of large objects. In the computer industry, for example, silicon wafers are increasing in size to reduce material losses in cutting. This is understandable if one considers that approximately 400 acres of silicon wafer material are used per year by IBM Corporation at a cost of about $100 million per acre. A processed wafer costs approximately $800, and the increase in total wafer area per year is about 1~15 percent. Although silicon waters do not present a waste disposal problem from the point of view of volume, they are environmentally important because their manufacture involves the handling of hazardous chemicals. They are also interesting as an example of how the production volume of an aggressive new technology tends to grow because of popularity in the market. More- over, marry rather large plastic and metal boxes are required to enclose and keep cool the microchips made with the wafers, even as the world's entire annual chip production might compactly fit inside one 747 jumbo jet. Thus, such new industries may tend to be simultaneously both friends and foes of dematerialization. The production of smaller and lighter toasters, irons, television sets, and other devices in some instances may result in lower-quality products and an increased consumer attitude to "replace rather than repair." In these instances, the number of units produced may have increased. Although dematerialization may be the case on a per-unit basis, the increasing number of units produced can cause an overall trend toward materialization with time. As an example, the apparent consumption of shoes, which seem increasingly difficult to repair, has risen markedly in the United States since the 1970s, with about 1.1 billion pairs of nonrubber shoes purchased In 1985, compared with 730 million pairs as recently as 1981 (Table 1~. In contrast, improvements in quality generally result in dematerial- ization, as has been the case for tires. The total tire production in the United States has risen over time (Figure 1), following from general in- creases in both the number of registered vehicles and the total miles of travel. However, the number of tires per million vehicle miles of travel has declined (Figure 2~. Such a decline in tire wear can be attributed to improved tire quality, which results directly in a decrease in the quantity of solid waste due to discarded tires. For example, a tire designed to have a service life of 100,000 miles could reduce solid waste from tires by 6~75 percent (Westerman, 1978~. Other effective tire waste reduction strategies include tire retreading and recycling, as well as the use of discarded tires as vulcanized rubber particles in roadway asphalt mixes. Dematerialization of unit products affects, and is influenced by, a number of factors besides product qualitr. These include ease of man- ufacturing, production cost, size and completer of the product, whether the product is to be repaired or replaced, and the amount of waste to

DEMATERIALIZATION 53 TABLE 1 Apparent Consumption of Nonrubber Shoes in the United States Total Consumption Shoes per Capita Year (million pairs) per Year 1970 802 3.9 1975 728 3.4 1976 786 3.6 1977 746 3.4 1978 788 3.6 1979 833 3.7 1980 745 3.3 1981 730 3.2 1982 830 3.6 1983 912 3.9 1984 1,018 4.3 1985 1,097 4.6 SOURCE: U.S. Bureau of the Census (1975-1985~. be generated and processed. These factors influence one another as well (Figure 3~. For example, the ease of manufacture of a particular product in smaller and lighter units may result in lower production cost and cheaper products of lower quality, which will be replaced rather than repaired on breaking down. Although a smaller amount of waste will be generated on a per-unit basis, more units will be produced and disposed of, and there may be an overall increase in waste generation at both the production and the consumption ends. Another factor of interest on the production end is scale. One would expect so-called economies of scale in production to lead to a set of facilities that embody less material for a given output. Does having fewer, larger plants in fact involve significantly less use of material (or space) than having more, smaller ones? At the level of the individual product, the shift from mainframe computers to personal computers, driven by desires for local independence and convenience, may also be in the direction of matenalization. Among socioeconomic factors influencing societr's demand for mate- rials are the nature of various activities, composition of the work force, and income levels. For example, as a predominantly agricultural society evolves toward industrialization, demand for materials increases, whereas the transition from an industrial to a service society might bring about a decline in the use of materials. Within a given culture, to what extent are materials use and waste generation increasing functions of income?

54 ROBERT HERMAN, StAMAK A. ARDEKANI, AND JESSE H. AUSUBEL 210 200 180 '0 1 60 o ._ , _ - o ._ o 140 120 100 80 60 40 20 o - - Automobiles Trucks and Buses - 1955 1960 1965 1970 1975 Year l 1 1980 1 985 FIGURE 1 Production of automobile, truck, and bus tires in the United Stated SOURCE: U.S. Bureau of the Census (197~1985~. NOITE: Lines connecting data points are for clarity only. The spatial dispersion of population is a potential materializer. Migra- tion from urban to suburban areas, often driven by affluence, requires more roads, more single-unit dwellings, and more automobiles with a consequent significant expansion in the use of materials. The movement from large, extended families sharing one dwelling to smaller, nuclear families may be regarded as a materializer if every household unit occupies a separate dwelling. Factors such as photocopying, photography, advertising, poor quality, high cost of repair, and wealth generally force materialization. Technological innovation, especially product innovation, may also tend to force materialization, at least in the short rum For example, microwave ovens, which are smaller than old-fashioned ovens, have now been ac- quired by most American households. However, they have come largely as an addition to, not a substitute for, previous cooking appliances. In the long term, if microwave ovens truly replace older ovens, this innovation may come to be regarded as a dematerializer. National security and war, styles and fashions, and fads may also function as materializers by accel- crating production and consumption. Demand for health and fitness, local mobility, and travel may spur materialization in other ways.

DEM41ERL4L=;AnON 210 200 > 180 u, a) - 1 60 a) ._ > 1 40 a ._ ._ ~ 120 a) Q oh ,= 100 1~ 80 ~ 1955 1960 1965 1970 Year 55 Automobiles Trucks and Buses ~ - ~ I ~I 1 1 1 1975 1980 1985 FIGURE 2 Consumption of automobile, truck, and bus tires in the United States per million vehicle miles driven. SOURCE: U.S. Bureau of the Census (1975-1985~. NOTE: Lines connecting data points are for clank only. The societal driving forces behind dematerialization are, at best, di- verse and contradictory. However, the result may indeed be a clear trend in materialization or dematerialization. This could be determined only through collection and analysis of data on the use of basic materials with time, particularly for industry and especially for products with the greatest materials demand. Basic materials such as metals and alloys (e.g., steel, copper, aluminum), cement, sand, gravel, wood, paper, glass, ceramics, and rubber are among the materials that should be considered. The major products and associated industries that would be interesting to study could well include roads, buildings, automobiles, appliances, pipes (metal, clay, plastic), wires, clothing, newsprint and books, packaging materials, pottery, canned food, and bottled or canned dnnks. Hibbard (1986) reported without much detail that annual per capita intensity of materials use in the United States remained nearly constant between 1974 and 1985 at about 20,000 pounds. It would be useful to confirm this finding and extend the data to explore the extent to which such a fact might result from changes in gross national product (GNP), materials substitutions, market saturation, or other factors of the kind mentioned above. About two-thirds of Hibbard's estimate comes from

56 ROBERT HERAfAN, StAMAK A. ARDEKANI, ID JESSE H. AUSUBEL Quality )~( ~ _ ~ Size and \~ Complexity Waste ~ Generations ~ . ~ Dematerialization (\Neight or Energy) Ease of ~ remanufacturing J I've ' If/ ~ Cost ~ FIGURE 3 Factors affecting, and affected By, the dematenalization process. Economic and population growth, of course, also strongly interact with many of the factors. stone, and from sand and gravel for concrete. These materials may be less important for environmental quality than others that are more active in our "industrial metabolism" (see Ayres? this volume). Further thought should be given to defining baskets of materials whose use over time might form the most meaningful indicators with respect to environment from the point of view of processing and disposal. Bible 2 shows the consumption of carbon steel as a function of time across various major end products. As can be seen, the use of steel in two major industrial activities, namely, construction and automobile manufacture, clearly has been in decline. This significant dematerialization trend has come about by virtue of the use of lightweight, high-strength alloys, and synthetics as substitutes for steel and cast iron. The trend is especially evident in the automobile industry where large weight and size reductions were achieved by materials substitutions in the 1970s in order to conserve energy. Able 3, the estimated pounds of materials used in a typical U.S.-manufactured car, shows that the use of plain carbon steel declined by 475 pounds per car in the 10 years examined, 197~1988. On the other hand, the use of high-strength steel, plastic composites, and aluminum increased by 99, 43, and 36.5 pounds, respectively, in the same period. The result is a total reduction in weight of a typical U.S. car of about 400 pounds from 1978 to 1988. In the construction industry, however, caution must be exercised in associating the decline in steel use with dematerialization, because such a decline could be indicative of the

DEMATERLALIZATION 57 increased popularity of concrete over steel as the basic construction material for aesthetic, technical, or cost reasons. Growth in the use of advanced materials is expected to continue. For example, it is anticipated that by 1997 the world market for fabricated ad- vanced polymer composites will be almost triple its 1987 level (Miller, 1988, pp. 57-59~. These changes will significantly affect the industries producing conventional materials because the automotive industry has traditionally been a major consumer of these materials. In 1978, for example, the automotive industry used 22 percent of the total U.S. steel consumption and 17 percent of aluminum consumption (Motor Vehicle Manufacturers Association, 1982, p. 6~. The significant decline in the use of steel in the automobile industry provides strong evidence in support of dematerialization at the production end. An examination of energy consumption in selected national economies between 1973 and 1985 further underscores an industrial trend in efficiency and dematerialization (Table 4~. Although total energy consumption in most countries increased considerably during this period, the energy consumed per 1980 constant GNP dollar declined in 9 out of 10 nations examined. This result may be explained in part by energy efficiency in production or by an increasing GNP associated with the services sector. Whether increasing energy efficiency is a net dematerializer is not clear. Often, increasing energy efficiency involves substituting durable capital goods in the form of better or larger amounts of building materials such as insulation. However, further evidence of dematerialization at the production end is provided by data on industrial solid waste generation, which show a significant decline from 1979 to 1982 (Figure 4~. The generation of municipal solid waste, also shown in Figure 4, has been on the increase. Examination of trends in municipal solid waste gen- eration (total and by each component) provides insight into materialization and dematerialization at the consumption end. The data on municipal solid waste generation suggest a trend toward materialization at the consumption end. Able 5, for example, shows the total amount of paper waste in the municipal solid waste in the United States. As can be seen, from 1960 to 1982 there was an approximately 75 percent increase in the total paper waste generated, as well as an approximately 35 percent increase in the paper disposed per capita. Such increases are to be viewed in light of predictions that the advent of computers would reduce the use and wastage of paper. A possible contribution to this rise in paper waste is the increase in circulation of daily newspapers from a total of 53.8 million in 1950 to 62.8 million in 1985 (U.S. Bureau of the Census, 1975-1985~. However, other factors must be taken into account, such as changes in the average size and number of pages of newspapers over the country, as well as the amount of wastepaper that is used by the industry for printing

58 - - a> s - o o - ._ Go on 4) Al ID 4, - Ct U) - s _. ._ - :' o" - 4, - ~: o Cal o o ._ £ 3 S:: Ct At; _ as ~ 0~3 .= be Ct ~ o ._ - Ct o U] _ I: U} - So o PA ~ Cal ,D t: ~ o !1 o ._ CJ I: CO 4) ~ - o .= o As Cal JO t> on .. ooze Do o ~ - ~ ~ ~ So ~ ~ ~ ~ ~ . . . . . . . . omen oomo~ooooo`= oo--oxc~coooor~cr`+o~- CO 0 0 ~ so ~o 0 ~ ~ ~ 0 oo . . . . . . . . . . . . . ~ ~ _. ~ ~ ~ ~ ~ ~ ~ ~ U) 0 ~ 0 ~ ~o o. o. o. o~ o~ ~ oo o~ ~ ~ oo oo ~ oo ~ ~ V~ ~ ~ ~ ~ oo r~ ~ ~o ~ ~ ~ ~ O . . . . . . . . . . . . . r~ ~ ~ ~ ~ ~ <~ so ~ ~ ~ ~o oo ~ ~ ~ ~ ~ ~ ~ d. ~ O ~ ~ ~ ~ ~ ~ 00 oo ~ oo 0 . . . . . . . . . . . . . U' so 0` ~ ~ d. ~ ~ O ~ ~ O Oi Oi ~o ~ oo as 0 ~ ~ c~ o~ ~ _. mooo~ o moo m~ ou~ oo 0` ~ d. 0\ ~ 0\ ~ oQ oo o~ 0 ~ ~ CO oo oo o~ os o~ ~4 ._ - CtS o C~ 4, .c ~4 - C' _# ._ - U) - 4, - V' s: - as C~ .,_ ~ o U] _ :s .o~ g ~ _ ~o~ Ct ~> ~ ~ fL) _' Ls "0 D t; ,: ~ O ~ ~:^ g ~ ~ - C~ ~ ~C~ 4) ~ ~C~ 'CtS- ~O ~ s" · s~ - O - ., ~ . . o

DEA~lTERL4LIZAHON TABLE 3 Estimated Material in a Typical U.S. Car (pounds) Material 1978 1984 1986 1988 59 Plain carbon steel1,915.0 1,526.01,470.01,440.0 High-strength steel133.0 210.02235232.0 Stainless steel26.0 28530.531.0 Other steels55.0 54.055545.0 Iron512.0 481.0465.5457.0 Plastics/plastic composites180.0 204.0216.0223.0 Fluids and lubricants198.0 189.0181.0178.0 Rubber1465 138.01345134.0 Aluminum1125 13651395149.0 Glass865 85585585.0 Copper fabrications and electrical components37.0 43546.049.0 Zinc die castings31.0 17518.0195 Other materials137.0 1185105.01245 Total3,5695 3,232.03,17053,167.0 NOTE: Estimates are based on U.S. models only, including family vans and wagons. SOURCE: Stark (1988, pp. 33-37~. TABLE 4 Energy Intensity of Selected National Economies, 197~1985 Energy (megajoules) Change, per 1980 Dollar of GNP 197~1985 - Country 1973 1979 1983 1985 (percent) Australia 21.6 23.0 22.1 20.3 -6 Canada 38.3 38.8 365 36.0 Federal Republic of Germany 17.1 16.2 14.0 14.0 -18 q Greece" 17.1 185 18.9 19.8 + 16 Italy 185 17.1 15.3 14.9 -19 Japan 18.9 16.7 135 13.1 -31 Netherlands 19.8 18.9 15.8 16.2 -18 Turkey 28.4 24.2 25.7 25.2 -11 United Kingdom 19.8 18.0 15.8 15.8 -20 United States 35.6 32.9 28.8 275 -23 aEnergy intensity increased as a result of a move toward energy-intensive industries such as metal processing. SOURCE International Energy Agency (1987).

60 , . \\\\\ ~ ...... ..................................... ..... -.- . ~ ::::::::-::~::~:::::::::::::::::-:~:::::::::::-:~:~:::::-:-:-:-:-:~:-:-:~:- -:~:~:~:~:~:-:~:-~-:-:-: ::: :-: :.:.:.: :.:.: :...:.:.:.: :.~.:: . . .. .... . I < . ~ . ::.::::::: ::: :: -~::::::::::~: e: ::::: :::::::: :::: :::: ::: ::: ::-.: :::::: - ~e ~e ~e ~ ·~e ~ o~e ~ ~ ~ a ~e -e -e ~: e ~e ~e e ~ ~ ~ e ~e ~ee ~e ~ee I ~W i\\\\ Ail ~ \ t ~ en eta ~ eee ~ ~ e e ~ e i- ~ e en o~ eee e ~ e ~ e ·~ oeee~ of e ~ ~ ee~ ~ lea ~ em r.. :. ~ ::: ~ .: ~ ::. :~.::::::::~.::::::::::~.:::::: .: . . ~:.~. ~:~.~.~.~. ~. _ ~·~ ~ ~e~ lee eel e ~;~ en e e ~ e e he ~eel ~ ~ ee ~ oe ~ ~e ~e ~a ~ore ~ ~ o~e ~ ~en e~ ~ en see ~ he e see 1 1, I t_ o o o o ~o N _ ~ (quo} U°!ll!~) 6IS~M P!1°S o o C~ oo o) oo a) - U] o CO s a, a) Q. - ._ t: d] - Ct 3 - Ct .r; . ~ _ - .~: ~ C: Ct _ C~ ·s ._, ~ 1: ~ o o ." m ~ ~. . . 0 C~

DEAL4TERIALIZATION TABLE 5 Amount of Paper in the Total Municipal Solid Waste Generated in the United States Paper Wasted Paper Wasted (pounds per capita Paper Recycled Year (million tons) per day) (million tons) 1960 29.9 0.91 5.4 196S 37.9 1.07 5.7 1970 40.4 1.08 6.8 1975 42.7 1.08 8.2 1976 49.1 1.24 9.7 1977 50.8 1.27 105 1978 53.4 1.32 10.6 1979 555 1.35 11.6 1980 54.1 1.30 11.8 1981 555 1.32 11.4 1982 52.2 1.23 10.8 SOURCE: U.S. Bureau of the Census (1975-1985~. 61 new newspapers. Wastepaper consumption in the newspaper industry rose from about Z6 million short tons in 1977 to 3.6 million short tons in 1987 (Institute of Scrap Recycling Industries, 1988, p. 22~. Recently, there has been a great deal of interest in the paradox associated with the proliferation of paper in our sociotechnical culture. The following discussion on this point is based on a recent article by Tenner (1988~. We were all encouraged in the past to believe that information technology, as a by-product, was going to reduce the consumption of paper significantly. As we all now know, the reverse has transpired, with paper prices rising and trees in jeopardy. Consumption in the United States of writing and printing paper increased in 1959-1986 from about 7 to 22 million tons, and in the short period 1981-1984 the use of paper by U.S. businesses rose from 850 billion to 1.4 trillion pages. It is estimated that between 1986 and 1990, printed material may increase from about 2.S to 4 trillion pages. In 1988 newsprint production was approaching capacity at about 12 million metric tons, and in the Pan Am Building in New York Cibr a newsstand is reported to carry more than 2,000 magazines! Banks have rid us of the savings account passbook, but in its place there is a spate of paper. Consumers have resisted reliance on home computer on-line services. Moreover, attempts by banks not to provide customers with canceled checks have failed; in 1985 U.S. banks processed some 45 billion checks. Plastic credit cards generate considerable amounts of paper, as do automated teller machines. The Rush Medical Library, Chicago, used about 188 Linear miles of paper in its photocopy machines

62 ROBERT HElUfAN, SLAMAK A. ARDEKANI, ID JESSE H. AUSUBEL in the year 1982-1983, and the Princeton University computer center used close to 6 million pages of letter-sized laser paper in 1986, plus about 4,500 cartons of impact printout paper. Harvard's computer printers use more than 22 million pages a year, not including personal and faculty computers. The question has been asked, What was wrong with the assumption that electronics would substitute for paper? Apparently nobody anticipated that the microchip would catalyze the burgeoning of paper to such an enormous extent It would appear that the information age technicians did not understand that the amount of information was not fixed and that electronic information was not simply a substitute for paper. Computers are storing greater quantities of more kinds of information than ever before in extremely compact form, but people prefer reading from the printed page rather than the average computer screen, which in order to have excellent resolution must be improved by a factor of about 10. In addition, there is an increase in office workers compared to those in manufacturing jobs, and this shift leads to an increase in precisely the kind of people who generate paper. Note also that it is easy to produce photocopies compared to the old days, when making carbon copies was indeed a great burden. In 1959 when Xerox introduced its dry copier, a consulting company estimated that no more than 5,000 such copiers would be required in the entire United States. The huge mailings today from businesses and various organizations would not be feasible without the backup of the copier and the computer. In 1986 businesses in the United States bought 200,000 photocopiers, and this market is expected to increase for some years to come. It is difficult to comprehend that in 1986 about 45 billion pieces of bunk mail alone were handled by the U.S. Post Office. Notwithstanding the popularity of electronic mail, facsimile machines are materializing by the millions and spewing forth even more paper. One factor that further encourages the storage of data on paper is that it is unsafe to assume that electronically stored records will be readable for even a small fraction of the 200- or 300-year lifetime of acid-free paper (National Research Council, 1986~. Even if the data are imprinted on poor paper, it is always possible to photocopy it and obtain a better copy than the original, before the sulfite sheet crumbles into its acid grave. Evidence of our insecurity about electronic memory is that, although 90 percent of securities trades now take place through electronic means, they are, as one can surmise, backed up by mountains of paper. So perhaps it is not surprising that in the information era, the trees of the world are at risk Moreover, the equivalent of about 1,500 pounds of petroleum is required to make a ton of paper (Tenner, 1988~. One wonders which will last longer-energy or the trees. Imagine the implications for the environment if a cost-effective, but nonbiodegradable, plastic substitute were found for paper! Parenthetically, we might add that biotechnology,

DEMATERLALIZATION 63 operating at the genetic level, might be expected to bring about demate- rialization to an extent even beyond that anticipated for the information technologies. However, if the end result is not only a new gene but also an enormous "supercow," then the effect again may well be materialization The increase in paper waste is related closely to the broad arena of efficiency in use as well as recycling. Examination of municipal and industrial waste (solid and liquid) shows that the annual generation rate per capita in the United States was estimated in the mid-1970s at approximately 3,600 pounds (Tthobanoglous et al., 1977~. Japan was closest to the United States with an estimated average of 800, followed by the Netherlands at 680, and the Federal Republic of Germany at 500. The reliability and comparability of the estimates are uncertain because Cointreau (1982), for example, shows only a factor of two difference in daily per capita waste generation between New York and cities such as Hamburg and Hong Kong. Moreover, comparable estimates of which we are aware do not include emissions of environmentally important substances such as gaseous air pollutants or carbon dioxide. The human race now discharges to the atmosphere more than 5 billion tons of carbon dioxide annually, or 1 ton per person. The considerably smaller rates of waste generation in other industrial countries are often attributed to either a lower consumption rate of goods or a more serious effort to recover and reuse the wastes (Ithobanoglous et al., 1977~. In this connection, it would be instructive to examine questions such as how much paper is sold per capita in the United States, what fraction of a newspaper is recovered, whether more envelopes can be designed for reuse, what fraction of paper wasted is still usable, and what fraction of paper available for recycling is actually recycled. According to one estimate (Hagerty et al., 1973), only 24 percent of the 47 million short tons of recyclable paper in U.S. solid waste was recovered in the early 1970s. Although paper makes up the greatest fraction of solid waste (3W 35 percent), it has one of the lowest recovery rates, following textiles (17 percent) and zinc (14 percent). These low recovery rates are more than likely due to economic reasons. A recent Wall Street Journal article (Paul, 1989) stated, "The bottom, has fallen out of the market for recycled newspapers, exacerbating the nation's already critical garbage problems." It is reported that just a few months ago municipalities were receiving as much as $25 per ton for their newspaper waste, whereas they must now pay about $5 to $25 per ton to have old newspapers hauled away. This situation is counter to the myth that recycling should always make money. In this volume, Ayres, Ausubel, and Lee each argue that perceived scarcity of physical resources usually leads to technological substitutions. If substitution is not possible, then recycling is considered. From a purely

64 ROBERT HERMAN, StAMAK A. ARDERANI, ID JESSE H. AUSUBEL TABLE 6 Scrap Use in the United States Total Consumption (million short tons) Material 1977 1982 1987 Percentage of Total Consumption in Recycled Scrap 1977 1982 1987 Aluminum6.49 5.94 6.90 24.1 33.3 29.6 Copper2.95 2.64 3.15 39.2 48.0 39.9 Lead158 1.22 1.27 44.4 47.0 54.6 Nickel0.7S 0.89 1.42 55.9 4S.4 45.4 Steel/iron142.40 84.00 9950 29.4 33.4 465 Zinc1.10 0.78 1.05 20.9 24.1 17.7 Paper60.00 61.00 76.20 24.3 245 25.8 SOURCE: Institute of Scrap Recycling Industries (1988~. economic standpoint, high-grade resources are exploited before lower grade resources and recycling are considered economically viable. An overall view of scrap usage in the United States during 1977-1987 is shown in Table 6, where data on total consumption and percentage of total consumption in recycled material for a number of metals, as well as paper, are presented. Among the metals, there was an increase in total consumption of aluminum, lead, and nickel over the 10 years examined, whereas there was a decrease in steel and iron, as mentioned earlier. During this same period the percentage of total consumption in scrap increased for aluminum, lead, and steel. Zinc and paper have the lowest percentage of total consumption in recycled scrap, namely, 17.7 and 25.8 percent, respectively, in 1987. It is difficult to see exactly what correlations may exist or the underlying reasons for the observed variations. The availability of scrap might be expected to depend on total consumption, but it is also a function of usage, costs, and other factors. Another question to be raised in connection with the economics of consumption and disposal is what the "true" cost of consumption and processing of the generated waste is to society. What is the true cost of burning fossil fuel for transportation when, for example, the finiteness of resources and consequent long-term damage to the environment are con- sidered? Should high-grade resources be made available at much higher cost so that profits may be reinvested toward development of the capital and the knowledge to permit the use of lower-grade resources and the development of technological substitutes? What is the actual disposal cost of municipal and industrial wastes? To what extent is the cost of waste col- lection subsidized by different societies and different segments of a society?

DEAN 4TERlALIZATlON 65 Would a higher cost for garbage collection effectively encourage recycling, sorting recyclable materials at the generation source, and dematerializa- tion? Would it encourage more illegal dumping? Can societr truly afford to continue functioning in its present "throwaway" mode of products such as food, clothing, diapers, and shoes, as well as watches, radios, flashlights, light bulbs, cameras, calculators, pens and pencils, razors, knives, spoons, and forks? A practice potentially very risly to society is the emission of chlo- rofluorocarbons (CFCs) to the atmosphere (see Glas and Friedlander, this volume). Projected depletion of the ozone layer, attributed to the envi- ronmental release of CFCs, resulted in the U.S. ban of nonessential CFC aerosol propellants in the mid-1970s. Combined release of CFC-ll and CFC-12 in the United States traditionally accounted for about one-third of the total worldwide release of these substances. The aerosol ban, however, resulted in only a gradual and temporary decline of production and emis- sions levels (see Glas, this volume, Figure 5), because CFC-ll and CFC-12 have had other, growing, nonaerosol industrial applications such as in re- frigeration, air-conditioning, cleaning electronic and computer equipment, and foam manufacturing (Warhit, 1980~. According to a 1987 international treaty the industrial countries agreed to cut CFC production in half by the year 2000. In March 1989 there was a conference in London, attended by over 100 nations, at which a proposal for the total elimination of CFCs by the year 2000 was entertained. Although it certainly seems prudent to reduce or eliminate CFC use, one wonders whether their elimination may yet result In further materialization, for example, through a need to have bulkier refrigerators again. Lead is another example of a substance whose wide use presents a cleanup problem. Lead-containing aerosols, paint, and vehicular exhaust are among major sources of lead in the environment. It has been estimated that an effective program to reduce exposure to lead paint from the interiors of the nation's housing stock would cost between $28 billion and $35 billion (Chapman and Kowalski, 1979~. Although ingestion of lead-based paint chips is regarded as the major cause of lead poisoning in children, lead exposure results from a combination of sources, including automotive lead emissions. It is estimated that 70 percent of the lead in gasoline is emitted into the atmosphere and that this accounts for about 90 percent of airborne lead emissions (Boggess and Wixson, 1977~. In the 1970s the U.S. Environmental Protection Agency (EPA) enacted a phased reduction schedule for the lead content of gasoline that has resulted in installation of lead-intolerant catalytic converters in virtually all cars produced in the United States. The national average lead content of all grades of gasoline declined from about 2.5 grams per gallon in 1968 to less than 0.1 gram per gallon in 1988, and sales of unleaded gasoline

66 ROBERT HERMAN, SIA~iK A. ARDEKANI, AD JESSE H. AUSUBEL have increased consistently. Lead was introduced as an antiknock additive to gasoline in the 1930s to increase the efficiency of automobile engines. As such, lead may have contributed to the dematerialization of cars in terms of either weight or energy. But we did not foresee sufficiently that the increasing quantity of lead in our environment would itself become a serious problem. In a recent study the EPA (1988) identified some 30 broad categories of environmental problems (see Frosch et al., this volume, Able 1) and ranked the seriousness of these problems according to the risk they posed to the population in terms of total incidence of disease and other factors. The risks considered included cancer risk noncancer health risks, eco- logical effects, and welfare ejects such as materials damage to industrial, agricultural, commercial, and residential properties, among others. Lead and CFCs along with, for example, sulfur dioxide, suspended Articulates, carbon monoxide, and nitrogen oxides were included in three air pollutant categories regarded as having relatively high risks. Industrial dematerial- ization would have a significant impact on reduction of the various risks associated with these air pollutants. Other problems evaluated in the EPA report in which materialization is a central factor include nonhazardous municipal and industrial waste, as well as mine waste. Discharges of direct and indirect effluents and municipal sludge into surface waters and wetlands are also among the high-risk problems that might be associated with materialization. In this connection, discharges of sludge and medical waste into the oceans are pressing problems with high news visibility. During the past few years, the Atlantic Ocean has been regurgitating progressively more garbage and waste onto the beaches of the northeastern United States, especially around New York and New Jersey. Included in the dumping that causes this shocking situation are some 500,000 pounds of medical waste per week from New York City alone. Examples of materialization resulting from medical technology are the plastic throwaway hypodermic syrunge and throwaway needles. In the old days, glass syringes and }~igh-quality surgical steel needles were sterilized and used many times over. At present, syringes, for many good reasons, are used once and thrown away, as is much other medical material. With the burgeoning of hazardous medical waste, the disposal task es- pecially at hospitals, becomes complex and expensive. This unquestionably leads to illegal dumping to cut costs and avoid demanding procedures. It is difficult to believe that clinic and hospital authorities are not aware of the dangers associated with illegal disposal. The midnight dumping of medical wastes raises the question of the role of the entire spectrum of "criminal" activity in our society tenth regard to transport and disposal of materials. Attempts are being made to determine at what point in the disposal chain

DEMATERLALIZATION 67 the system breaks down. The solution to this type of complex problem must, of necessity, have an ethical component, with better values placed over and above such considerations as cost-effectiveness. Although no recycling process is 100 percent efficient, recycling is a promising means of dematerialization. The construction industry is one of the major generators of solid waste. What fraction of construction waste is reusable? 1b what extent are brick, wood, steel, and asphalt reused? In general, a more thorough examination of practices in the construction industry regarding waste generation and processing is warranted in studies of dematerialization. How much waste is generated in construction activities such as paving roads and building houses? What happens to the waste from building construction or from demolished buildings? What determines whether a building-should be demolished or renovated? What fraction of buildings is demolished as a result of safety considerations or to be replaced by a larger structure for economic reasons? What is the potential for recycling materials resulting from demolition operations, as well as various construction activities? 1b what extent do construction and demolition "activate" environmentally significant materials? The embalming of no- longer-usable nuclear power plants is an interesting case of permanent structural materialization. In a recent essay, Marland and Weinberg (1988) make a powerful case for a life-cycle approach to infrastructure systems, exploring connections between quality of service provided and aging of facilities. They ask three fundamental questions about a variety of infrastructure systems: What actually is the characteristic longevity of a given infrastructure? How long could it last? How long should it last? This first attempt at a demography of infrastructure needs to be pursued in many areas in connection with materialization. From an environmental perspective, what could and should be the design life of everything we create? In the area of nuclear materials we are accustomed to asking long-range questions about how materials will be transported, stored, and disposed of. Such a life cycle perspective might be applied usefully to other materials as we contemplate transforming them for human purposes, and thus provide guidance about instances in which dematerialization rather than materialization should be the eventual objective. More generally, it might be useful to undertake materialization impact assessments for selected new products and activities. Furthermore, the interplay between dematerialization and transportation costs in terms of weight and bunk should be examined. The questions raised and discussions set forth in this chapter pO=t tO a number of overall objectives, namely, to single out the important driving forces behind trends in materialization and dematerialization, to determine whether on a collective basis such forces drive society toward materialization or dematerialization, and to assess the environmental implications of these

68 ROBERT HERMAN, SLAMAK A. ARDEKANI, ID JESSE H. AUSUBEL long-term trends. Many questions remain to be answered quantitatively; for example, how much basic material and how many of each major product are used per capita over time and what is the lifetime of venous manufactured products? If we consider that for every person in the United States we mobilize 10 tons of materials and create a few tons of waste per year, it is clearly important to gain a better understanding of the potential forces for dematerialization. Such understanding is essential for devising strategies to maintain and enhance environmental quality, especially in a nation and a world where population and the desire for economic growth are ever increasing. \ ACKNOWLEDGMENT The authors gratefully acknowledge assistance and comments from Walter Albers, Robert Ayres, Gerald Culldn, Denos Gazis, Shekhar Govind, Ruth Reck, Richard Rothery, and Hedy Sladovich. NOTES 2. 1 In an essay published in the proceedings of the Sixth Convocation of the Council of Academies of Engineering and Technological Sciences, Colombo (1988, pp. 26 27) makes the following observation: [EJach successive increment in per capita income is linked to an ever-smaller rise in quantities of raw materials and energy used. According to estimates by the International Monetary Fund, the amount of industrial raw materials needed for one unit of industrial production is now no more than two-fifths of what it was in 1900, and this decline is accelerating. Thus, Japan, for example, in 1984 consumed only 60 percent of the raw materials required for the same volume of industrial output in 1973. The reason for this phenomenon is basically twofold. Increases in consumption tend to be concentrated on goods that have a high degree of value added, goods that contain a great deal of technology and design rather than raw materials, and nonmaterial goods such as tourism, leisure activities, and financial senaces. In addition, today's technology is developing products whose performance in fulfilling desired functions is reaching unprecedented levels.... One kilogram of uranium can produce the same amount of energy as 13 U.S. tons of oil or 19 U.S. tons of coal, and in telecommunications 1 ton of copper wire can now be replaced by a mere 25 or so kilograms of fiberglass cable, which can be produced with only 5 percent of the energy needed to produce the copper wire it replaces. It would be interesting to venture calculations about the significance for materialization of the increasing average height and weight of humans, even though this effect is small compared with that of present population growth. The increase director expands needs for textiles and food, as well as creating pressure for larger vehicles and dwellings.

DEMATERIALIZATION 69 REFERENCES Boggess, W. R., and 13. G. W~xson. 1977. Lead in the Environment. Repon NSF/RA-770214. Washington, D.C.: National Science Foundation. Chapman, E. R., and J. G. Kowalski. 1979. Lead Paint Abatement Costs: Some Technical and Theoretical Considerations. Washington, D.C.: U.S. Department of Commerce. Cointreau, S. J. 1982.- Environmental Management of Urban Solid Wastes in Developing Countnes. Washington, D.C: World Bank. Colombo, U. 1988. The technology revolution and the restructuring of the global economy. Pp. 2~31 in Globalization of Technology: International Perspectives, J. H. Muroyama and H. G. Stever, eds. Washington, D.G: National Academy Press. Evans, L 1985. Car size and safety Results from analyzing U.S. accident data. Pp. 548 555 in Proceedings of the Tenth International Conference on Experimental Safety Vehicles, Oxford, U.K., July 1-5, 1985. Washington, D.C.: U.S. Government Printing Office. Hagerty, D. J., J. L. Pavoni, and J. E. Heer. 1973. Solid Waste Management. Environmental Engineering Series. New York: Van Nostrand Reinhold. Hibbard, W. R. 1986. Metals demand in the United States: An overview. Materials and Society 10(3~:251-258. Institute of Scrap Recycling Industries (ISRI). 1988. Facts 1987 Yearbook. Washington, D.C. International Energy Agency (IEA). 1987. Energy Conservation in IEA Countries. Pans: Organization for Economic Cooperation and Development and IEA. Marland, G., and A. M. Weinberg. 198~8. Longevity of infrastructure. Pp. 312-332 in Cities and Their Vital Systems, J. H. Ausubel and R. Herman, eds. Washington, D.C.: National Academy Press. Miller, E., ed. 1988. Ward's Auto World. Detroit, Mich.: Wards Communications. Motor Vehicle Manufacturers Association. 1982. Information on the Use of Various Materials in the Automotive Industry. Detroit, Mich.: Policy Analysis Department. National Academy of Engineering. 1985. The Competitive Status of the U.S. Steel Industry. Steel Panel Committee on Technology and International Economic and Made Issues. Washington, D.C.: National Academy Press. National Research Council. 1986. Presentation of Historical Records. Commission on Engineenng and Technical Systems. Washington, D.G: National Academy Press. Paul, B. January 25, 1989. Market for recycled newspapem in U.S. collapses, adding to solid waste woes. Wall Street Journal B4(E). Stark, H. As, ed. 1988. Ward's Automotive Yearbook. Detroit, Mich.: Wards Communica- tions. Tchobanoglous, G., G. H. Theisen, and R. E. Eliassen. 1977. Solid Wastes Engineering Principles and Management Issues. New York: McGraw-Hill. Tenner, E. March 9, 1988. The paradoxical proliferation of paper. Princeton Alumni Weekly. U.S. Bureau of the Census. 1975-1985. Statistical Abstract of the United States. Washing- ton, D.C.: U.S. Government Printing Office. U.S. Environmental Protection Agency. 1988. Unfinished Business: A Comparative Assess- ment of Environmental Problems. Springfield, Va.: National Technical Information Service. Warhit, E. 1980. Regulating chlorofluorocarbon emissions: Effects on chemical produc- tion. Report EPA-560/12~0 0016. Washington, D.C: U.S. Environmental Protection Agency. Westerman, R. R. 1978. Tires: Decreasing solid wastes and manufacturing throughput. Report EPA 600/5-7~009. Cincinnati, Ohio: U.S. Environmental Protection Agency. Williams, R. H., E. D. Larson, and M. H. Ross. 1987. Materials, affluence, and industrial use. Annual Review of Energy 12:99 144.

Technology and Environment 1989. Pp. 7~91. Washington, DC: National Academy Press. Regularities in Technological Development: An Environmental View JESSE H. AUSUBEL Forward, forward let us range; Let the great world spin forever down the ringing grooves of change. Tennyson, "Locksl~y Hall," 1842 Accept for the moment that there are long-term regularities in techno- log~cal development. Suppose that the evolution and use of both individual technologies and entire technological systems are sometimes tightly con- sistent and predictable over decades and generations. Then, we can know with confidence some important sources of future stress on the environ- ment and, equally, what technologically based stresses may fade, largely through natural advancement of the industrial economy. The thesis of this chapter is that, in fact, there are such long-term regularities in technolog- ical development and that these deserve more attention for the important implications they have for environmental concerns. Let me draw you back a century to a forgotten episode of environmen- tal history. The photographs in Figure 1 show the key material in terms of bunk in the massive expansion of the railroads in the nineteenth century. It is not widely remembered that railroads, usually associated in our minds with coal and iron, were largely wooden systems in their early development. The "iron horse" was something of a misnomer. Fuel for locomotives was 70

luedmo~ penally u~a~uON uOl~u!llna nor sue pus 'sosaoq 'salvos 1oqel angel 'poor Jo somnlo~ oSml ~ulsn ~ Ins Alan Xep cod sollm 8 Jo SplOOO~ laurel ~DEJ1 JO dn-=olD :htal7q 'WOW eueluo~ lseA All shoe L~1 U! P!el SU!^ shell :~°q~ I annoy IL ~Wdol~a 7~OID070NHO~ Or sw~7nOw

72 JESSE H. AUSUBEL wood, cars were wood, some of the rails were wood, trestles were wood and most important, crossties were wood. About the turn of the century President Theodore Roosevelt spoke as follows: Unless the vast forests of the United States can be made ready to meet the vast demands which this [economic] growth will inevitably bong, commercial disaster, that means disaster to the whole country, is inevitable. The railroads must have ties.... If the present rate of forest destruction is allowed to continue, with nothing to offset it, a timber famine in the future is inevitable. Speech to the American Forest Congress, 1905 (quoted in Olson, 1971, p. 1) An industry leader in 1906 described the railroads as the "insatiable juggernaut of the vegetable world" (Olson, 1971~. Such images were echoed in Argentina, India, the Middle East, and parts of Europe as railway networks were extended at the expense of local forests. In the United States, prevention of destruction of forests was proposed through a range of both supply and demand strategies. It was proposed to cover Kansas with a catalpa forest dedicated to supplying crossties. Railroad companies were asked to plant trees along the rail right-of-way to have a renewable stock of timber for ties. Better management of remaining forests was seen as urgent; in fact, the Forest Service was in large part built under Gifford Pinchot in this era in response to the railroad-induced crisis. What eventually contributed most to averting the forecast crisis were, initially, creosote and other technologies for preserving crossties and, later, especially in Europe, replacement of wood by concrete ties. As is evident from Figure 2, around the time of the peak of the perceived crisis, a technological solution was already penetrating the market for crossties. Preservation technologies tripled the life of ties, and within a couple of decades, the juggernaut of the vegetable world was satiated. In fact, in the 1920s the railroad network itself reached saturation (see Figure 5), so that demand for both new and replacement ties decreased. Railroads today are almost always described as environmentally benign. So, in the railroad timber story, new technologies are both cause and cure of environmental problems. The new transportation system placed intense demand on natural resources, and innovations in turn alleviated the demand to the extent that today the issue is obscure or forgotten. At this point it is necessary to make a brief methodological comment. A premise of this chapter is that, as suggested by Figures 3 and 4, sociotech- nical systems, like biological systems, often grow according to basic patterns well-described by S-shaped curves, in particular, logistic functions (Hamblin et al., 1973; Lotka, 1956; Montroll and Goel, 1971; Volterra, 1927~. In the simplest case, technologies, like biological organisms in constrained envi- ronments, proceed through a life cycle of early development through rapid growth and expansion to saturation or senescence. Often two technologies are in competition for an "econiche," that is, the market; then a logistic

REGULARITIES IN TECHNOLOGICAL DEVELOPMENT 100 90 a, 80 a) a) ~ 60 a) ._ ~ 50 o a) 4o 30 20 1890 1900 1910 1920 Year 73 If/ - I ~I I 1930 1940 1950 1960 FIGURE 2 Percentage of crossties manufactured in the United States treated with chemical preservatives. It is interesting to note that the innovation penetrated the market in a characteristic S-shaped cunre, disturbed only temporarily by world war and depression. SOURCE: After Olson (l97l). substitution model applies where a new technology replaces the old and the status of the system is described by the changing fraction or share of the market held by the technologies (Fisher and Pry, 1971~. When more than two technologies are competing for a market, a generalized version of the logistic substitution model can be used (Marchetti and Nakicenovic, 1979; Nakicenovic, 1988, pp. 212-220. Logistic functions and logistic sub- stitution models are a compact way of presenting data on the history of technology and are used frequently in the following sections of this chapter. However, numerous methods exist to explore quantitatively the existence of patterns in sociotechnical phenomena (Montroll and Badger, 1974), and the method used most frequently here should be taken simply as indicative of the value of extending the search for regularities by using a variety of methods. Some examples make the case for long-term regularities and also point out hazards in identifying them. Figure 5 shows the remarkably stable and parallel growth of three major systems of transport infrastructure in the United States: canals, railroads, and paved roads. For each of these trans- port infrastructures it would apparently have been possible relatively early in the life history of the system to make quite an accurate prediction about its eventual size and scope. Such vision in turn may be translated into conjectures about environmental problems and technological opportuni- ties, indeed about technological necessity. For example, it could have been

74 JESSE H. AUSUBEL CD ~ 300 a) E - c a) _ A, 200 ._ a) I 100 J _ ! I I I I I l I 0 7 1 4 21 28 35 42 49 56 63 70 77 84 Days 2 1o1 10 ·1 10-2 fit = 7.5 weeks I ~I 0 5 10 15 Weeks 0.90 - 0.50 ,, 0.10 FIGURE 3 Abe upper panel shows the growth of a sunflower, measured in height, precisely charting a logistic curve (Reed and Holland, 1919~. The lower panel shows the same data in linear transform, which is sometimes easier to employ for visual inspection and emphasizes the predictability of the process once established. For example, the ultimate height of about 260 centimeters could be estimated quickly with the linear transform. Ibe "/\ t" refers to the time for the process to go from 10 to 90 percent completion, in this case 7.5 weeks (see Lotka, 1956; Marchetti, 1983~.

REGULARTrlES 12V 1:E:CHNOL~OGICAL DEVELOPMENT 1o1 10° 10-' 10-2 / I 1870 ., r fit - 54 yr f 1900 1930 1950 Year 75 FIGURE 4 Growth of the length of wire for the U.S. telegraph system. Notwithstanding the battles involving Western Union and its predecessors and competitors, and all the associated economic and regulatory issues, the telegraph system spread its branches just as a sunflower plant grows. It is also interesting to note that the time the system required to reach its full extent (/\ t) was slightly more than 50 yea rid SOURCE: Marchetti (l988). 100 80 60 a) a' CL 40 20 o 1 780 1800 1850 1900 Canals J o J Railways as yr of -/ 1946 I 'fit, 1~1 1 1 ~ I I 1 1 1 1 1 1 1 - - Year 1 950 2000 FIGURE 5 Growth of major transport infrastructures in the United States in terms of percentage of length of final saturation level. Both actual data and best-fit logistic curve are shown. The midpoint of the growth process is also shown. SOURCE: Grubler (l988~.

76 JESSE H. AUSUBEL TABLE 1 Vehicular Pollution Means of Emissions Transport Pollutant (grams per mile) Horses Waste, solidab 640 Waste, liquid 300 AutomobilesC Hydrocarbons CO NOx 0.25 4.7 0.4 aCalculation based on an average production of 16 kg of solid waste per day and a range of 25 miles per day. b Calculation based on an average production of 75 kg of liquid waste per day and a range of 25 miles per day. C1980 U.S. piston engine standards. clear early on that a rail system of predictable dimensions would be unsus- tainable as a predominantly wooden technology and required innovations in materials and other areas to reach forecast dimensions. Agendas for research and for entrepreneurship might have stemmed from this analysis. A similar argument can be made about the system of paved roads. This system was initially designed for horses and horse-drawn vehicles, preceding the widespread use of the automobile. From an environmental perspective, a road system of the dimensions that began to be built could have been catastrophic if the traffic were horses. Able 1, based on calculations made by Montroll and Badger (1974), shows that, from an environmental point of view, cars were a marvelous technological innovation, at least when they were not much more numerous than horses. Figure 6, showing the substitution of cars for horses, emphasizes the continuity of the demand for personal transportation service and the fact that technologies or modes compete to meet such demand. When considering the intensity of problems of urban air pollution in places such as Denver, Los Angeles, and Mexico Cigar, the time may be at hand when an improvement almost as radical as that of substituting cars for horses is needed to accommodate growth in transportation demand. It is sometimes suggested that methanol fuel or electric cars will do the trick, but methanol has few obvious advantages over gasoline used in conjunction with a catalytic converter, and a versatile and wide-ranging electric car may not be available for decades. Methane and hydrogen cars are already technologically feasible and could meet stringent new environmental constraints, but they demand

REGULARITIES IN TECHNOLOGICAL DEVELOPMENT fog 101 - O _ 1o-l ~ -2 ~/ 1 1 1 900 / 1910 Ho: At/ Year 1 920 77 / 0.99 0.90 Cal c to 0.70 <5 0.50 LL _ 0.10 0.01 1 930 FIGURE 6 Replacement of homes by automobiles in the United States. Irregular lines are historical data; smooth lines are best fit and extrapolation. SOURCE: Nakicenov~c (1988~. emergence of a substantial infrastructure of supporting service that so far is not evident. Will a breakthrough come and, if so, when? The abrupt replacement of horses by cars shows one of the short- comings of the type of framework presented here, namely, the difficulty of anticipating system bifurcations and fluctuations. Although the growth of overall demand for transportation as represented by horses or cars between 1900 and 1930 appears consistent in Figure 6, within 2~30 years a radical change occurred in the way that demand was met. Diesel technologies conquered steam with equal rapidity, and jets replaced propeller aircraft in about the same time. Could the timing of the introduction of such new technologies have been foreseen on the basis of a sound and transferable logic? How many in policy positions in government or industry would have believed that transformations of the transport system could occur so rapidly? Many might have recognized in 1900 that the horse-powered system was environmentally unsustainable and foreseen the concomitant need for technological solutions. I suspect that these solutions would more commonly have been believed to be incremental, for example, the breeding of horses that would be more powerful for their size (more fuel efficient) or somehow generate less waste. It is also interesting to note long-term regularities within the auto- mobile system, where technologies specifically employed for environmental improvement have followed Apical patterns of substitution and diffusion. Figure 7 shows the adoption of emission-reducing technologies and then

78 JESSE H. AUSUBEL 2 _ ~ 0.99 Emission Catalyst / 1o1 10° 10- ~ _ 1 o-2 ,, 1 1 1 1 L/ 1 1 - ~1 1 1 1960 1 970 -1 No Con 0.90 - 0.50 _ IL _ _ 0.10 0.01 1 990 1 980 Year FIGURE 7 Substitution of emission controls in the U.S. vehicle fleet. The category "emis- sion" refers to crankcase, exhaust, and fuel evaporation controls SOURCE: Nakicenovic (1985~. catalytic converters. Identification of historically characteristic rates of such substitutions might help in setting feasible targets for future fleet improve- ments. More examples of the implications of long-term regularities in tech- nology for environment are found in examination of the transport system in its entirety (see Figure 8~. If the road system is considered, it seems clear from Figure 8 (and Figure 5) that the challenge over the next many decades is maintenance and repair of a large, mature system. The road system is in fact fully grown and decreasing as a proportion of the length of the total transport system. However, we just seem to be coming to grips with environmentally sound operation and maintenance of the system that has been built. For example, with current practices and technology, the amounts of salts (close to 400 pounds per capita in the United States in 1980; Hibbard, 1986) and other chemicals that might be used for the next 50 or 100 years to keep the system ice-free are staggering. Their accumula- tions almost certainly pose worrisome problems for soils and water. Under the auspices of the Strategic Highway Research Program of the National Research Council (1988), technological alternatives are beginning to be explored. Accumulations of chemicals connected either with fuels that will wash off the roads or with the wearing out of tires (see Ayres, this volume) might be another issue that is now being underestimated. Conjectures can also be offered about pressures on environment from the air transport sector. Since concerns faded in the early 1970s about

REGUIARITIES TV TECHNOLOGICAL DEVELOPMENT 1o2 1ol 10° 10-1 \ ~ Canals \ __, 10-2 1 800 1850 1900 1950 Year \ /Railways\ /~ / / At, on fairways 2000 2050 79 0.99 0.90 0.70 0.50 ° 0.30 0.10 0.01 FIGURE 8 Shares of total operated intercity route mileage of competing transport infrastructures. SOURCE: Nakicenovic (1988~. stratospheric effects of fleets of supersonic transport planes (SSI§), little attention has been paid to environmental aspects of aviation. Noting the tremendous growth projected for the air transport system, one may wonder if concerns lie ahead, either in the stratosphere with a large fleet of second-generation SSTs or perhaps in a more straightforward manner in the troposphere. Could tropospheric ozone be significantly enhanced if growing emissions of nitrogen oxides (NO=) by aircraft are considered? Changes might be looked for in the main travel altitude region near 10 kilometers, especially in the northern hemisphere where most air traffic occurs (Bruehl and Crutzen, 1988~. The long-term regularities identifiable in adoption of transportation technologies are paralleled in the closely related energy sector. 1b a considerable extent, the history of environmental and safety issues is simply the underside of the history of energy development (and agriculture). On an urban scale, 700 years of this history are recounted in The Big Smoke (Brimblecombe, 1987), which chronicles London air pollution since the Middle Ages and describes how improvements in technologies for burning wood and coal and for ventilation helped population density to increase and morbidity to decline. In energy, as in transport, what is most striking is the overall consistency and stability of the evolution of the technologies favored, as illustrated in Figure 9, which shows consumption of hydrocarbon fuels wood, coal, oil,

1 OI' ASCOTS 1l DIn8.l~ SU ~UGg1 CS~O18 U98Olp/g pus UOql~ JO SOlDlMIU SU Endures poUIUI~X~ 11e one sea pug 61!° 618°o CP°°hs JI sleet OSI 1sed oIll con MINUS lSOlU OlUlOUO~ O1 posn ua~q sol 1~ dnos lmlmoq~ oq1 m uoqlm 10] U08Olp/Q JO UO~nl!ISqtlS O~ ~ Bole PIOUS O~ Ul SpUDI1 pOgS!IqUlSO -ls~q pun SunSololu! lsotu alp JO ouo 61~ol uellsq~ oIOlU ~ I seI l~lnlcu 10] 9101 O8181 ~o Cpoollpaq~ oq1 poopUI 'M~qlssod oq1 [q pologO fullenb IclUoUIuonAUo ~0 1UorUo^OldUII o~ 10; soplunlloddo 1noq~ ~u~1 o1 po~u o~ 'Su~ods [IPB°IS (8861 ' 18 19 1oqnSn~) 110^ se 139~o osnolluoo~ oq1 ~noq~ SUI~ -UOD 'D18~tU]D lOU lllq 'UGSSO1 4lE.lluB1sqtls plilOM 1I ·SUUOIqOld UlUl plOU ~UIlUIAOIIB [llu=g~lu8ls '9~oodslod IUlU9tUUOMAU3 UE =01] 1U0tUdOIDA3p 8m -lsololu~ lsotu ~ ~q Plr~°~ sc' 1~luleu uo ~ueqol lulscolouI ·~IoUo ~ld ~o ~loos ~ulpco1 o~ souIoo~q uoos s~8 l~luleu 'loporu S~1 UI ·(01 DItl8I~ O8lO~D SOIEgS 10418W OlNln] }0 SUOllOO[Old 'lOglUD pOUOllUD~ pU~ Og] ~o lapotu uo~lnl~lsqns ops~8o1 ~ Ul pololdruo o1e ulep l=~lols~ll o~ ~I o~qotuolne oq1 ~0 osn oq1 pue po Illm 1led oil~l u~ pole!oosse SUO!SSIUIo ~ON ~o 3su aq1 pue 'clo 1~°o o~ ~o xode oIll le SOZ61 o~p, U! ~od ~ pollouol llo~^ 'suolsslwo In~lnS ~o oS!l o~ 01E 1.mlld~ OS~ ·pOO~ J0 ~ 91=S-~8lUl g1~ OW83 18~ UOIlUlSOIO]Op Og1 'DIdW=D 10] '~OlSIg IBlUDWUO ,~UD ]0 18oP 18oll ~ 11 ~w So1~o mou~d o~ 0981 01 4OBq 8~08 SOIEgS lO4lEW ]0 S~lO1 ~ SOlUlS pOlIUn Ogl 103- SE' pUP (9861) L=UnoD M33~0S011 leUoll8N :~02IflOS 0861~S81 'od~ lan~ [q uolloe~; 'salelS pal!un alll ul uolldmnsuoo lan] uoqlmolp`H 6 ~L~ ~ea,< OL6 ~ 096 l 0£6 ~ 0 L6 ~ 068 ~ OL8 ~098 t::::::::::::i:::::::::::i:::::::::::::~::::::::::::i:::::::::::~:::::::::::~:::::::::::i::::::::::::i::::::::::::~::::::::::::::i::: .::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: I::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: T::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: - - .1....................................... L:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ :::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :'::: : ~ ~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::: : :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~:::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~:::::: ~: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~:::::::::::: t:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::::::':::::': L:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~:::::::::::::: ': :':':':':':':':': :':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':' ' ' ' :':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':'~:': :': :'::: :'::: :'::: :':':':: :': :'::: . ::::::::::::::::::::::::::::::::::::::::::::::~7 L~ ::~:::::::::::::::::::::::::::::::::::::::~::::::::::::::::::::::::: ':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':~:~:~:~:':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':'~::: :': :': :':: :': :':::::::: :':'::: :'::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 0 :::::::::::::::::::::: ~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::::::::::::::::::::::: =:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::x:::::::::::::::::::::::: p~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::::::::::::::::::::::::: ~i ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::::::::::::::::::::::::::: ~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::` :::::::::::::::::::::::::::: X~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::::::::::::::::::::::::::::: ~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~::::::::::::::::::::::::::::::: L::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~:::::::::::::::::::::::::::::::: ~:':':': :':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':'~:':: :':':: :':': :': :':': :':':: :':': :'::: :': :':: :':':: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::0:::::::::::::::::::::::::::::::::::::::::::::: i:::'::':':':':':':':':':':':':::::::::::::::::::::::::::::::::::::::::::::::::::~:::::::::::::::::::::::::::::::::::::::::::::: ~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: - ::::::::::::::::::::::::::::::::::::::::::::::: ': ::::::::::::::::: :J~::::::::::::::::::::::::::::::::::::::::::::::: t:::::::::::::::::::::::::::::::::::::::::::::::::::::::::APIl:::::::::::::::::::::::::::::::::::::::::::::::: :,:,:,:::,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':~#':: :':'::: :':':':':': :':':: :':: :':'::: :':':':': :'::::: :': :': :':':': :'::: `:::::::::::::::::::::::::::::::::::::::::::::::::::::ArO:::::::::::::::::::::::::::::::::::::::::::::::::: ::':':.:':':':':':':':':':':':':':':':':':::::::::::::::::::::::::::::::~:::::::::::::::::::::::::::::::'::::::::::::::: '''''''''''''''''''''''''''''''''' ::::::::::: :Ar*:.:.:.: , ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ~, `,:,:':,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:':':':':':':':':':':':':':':':':':':':':':':~':::: :': :': :':': :':'::: :'::: :':: :': :'::: :'::: :':':'::: :'::::::: :':':': r::::::::::::::::::::::::::::::::::::::::::::::~:~::::::::::::::::::::::::::::::::::::::::::::::::::::: .::::::: ::::: :::::: ::.: :::::::::::::::: ::::::::::: ::::: ::~,.:~:::::::::::::::::::::::::::::::::::::::::::::':::::::': .:::::::::::::::::::::::::::::::::::::::::~::~.::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::^,:~:::::::::::::::::::::::::::::::::':':':':':':':':':':':':':':':':':':'::: ~:::::::::::::::::::::::::::::::::::Acr:r.::::::::::::::::::~: ~:: ::::::::::::::::::::::: ::::::::::::::::::::::::::::::::~:::::::::::::::::::~:::::::::::::::::::::: .:::::::::::::::::::::::::::::A~:::::::::::::::::::::97:: ~:~:::::::::::::::::::::: ~:::::::::::::::::::::::::::~.::~:::.:.:::::::::::::::::::::::::::::::::::::::::::::::::::::: :~::::::::::::::::::::::::::A~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: a:::::::::::::::::::::::~.:.:~:.::.:.:.:.::.:.::::.::::::::::::::::::::::::::::::::::::::::::::::: ~t::::::::::::::::::::~O ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~,:,:,: :':,:,:,:,:,:,:,:,:,:,:,:,~,: :,:,:,:,:.:,:.:,:.:,:,:,:,:,:.:.:,:,:,:,:,:,:,:.: :':':':': :':':: :':':': :':': :':':':': :':::: :':':::::::::: .2..2.2.22 2..2.2. 2 2"""""""""""'""""""""""""""""""""'""'""'"'" ...... ' """""""""""''''''''''''' ~////~//~2 ' ' ''" ' ' ' "''"" ''' ' '' ' ' "' ""' '""""""" ~:: :~::: :,:.:.:.:.:.:.:.:.:,:.:.:.:,:.:,:,:,:,:.:.:,:.:.:,:,:.:.:,:,:.:,:,:.:.:.:,:.:.:.:,:,:.:.: :':':: :':': :':':':':: :':'::::: :':::::: ':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':':::::::':':':':':':':':':':': ~ :::::::::::::::::::::::::::::::::::::::::::::::::::':':::~:::::~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::: ::::::::::::: ::::::::: :::~:::::~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::: ::::::::~:~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::: : :::::: ::::::::::::::::::::: ::::: ::::::::::::: :::: :::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 7~ansn~ ·H ~SS~! Z-0 b O _. o 90 80 O l 08

REGULARITIES TV TECHNOLOGICAL DEVELOPMENT 1o2 101 10° 10 lo-2 1 ~1 1 1 ~ 1800 1 850 1 ~.~ ~ Wood to 81 0.99 0.90 ~ Coal Oil Natural G:s ~ X ~ _ 1 1 ~ 0.01 1900 1950 2000 - 0 50 -° 0.10 Year FIGURE 10 Primary energy substitution in the United States, expressed as "market share" according to the logistic substitution model, 182~2000. The natural gas cume refers to gas resources not associated with oil exploration. SOURCE: Grubler and Nakicenovic (1988~. society has been moving steadily toward an economy running on natural gas and eventually on hydrogen. As discussed by Lee (this volume), the op- portuIlities to evolve in the next decades beyond hydrocarbon fuels appear timely. So far no reference has been made to "long waves," the cycles of about 50 years that seem to have characterized the world economy for the past 200 years (Freeman et al., 1982; Grubler, 1988; Nakicenovic, 1984; Schumpeter, 1939; Van Duijn, 1983~. There is much disagreement about the strength of the signal that emerges in analyses of long time series of data of technological and economic phenomena that may be indicative of long waves. Figure 5 does show that a sequence of major transport infrastructures emerged at roughly 50-year internals. Figure 9 shows that the characteristic time required for a major energy technology to capture or lose a leading role in the energy marketplace is also about 50 years. Synchronization of the diffusion of several major technologies would logically lead to periods of especially aggressive transformation of the environment and equally to "seasons of saturation" (Grubler, 1988~, when environmental management might revolve more around accommodating mature systems (such as the interstate highway system). Analysis of the evolution of energy demand shows two pulses of growth, each lasting 40 years or more (Ausubel et aL, 1988; Stewart, 1988~. Figure 12, which shows these pulses, may be seen as a pair of logistic curves, the second surmounting the first From an environmental point of view, a

82 JESSE H. AUSUBEL 1o2 1o1 loo 10-1 10-2 0.99 . ~ _' H/C = 4 / ~l Nonfossil Of Wood H/C = 0.1 Coal H/C = 1 Oii H/C = 2 Gas H/C = 4 0.10 1700 1800 1900 2000 2100 Year 0.90 I + - I 0.50 11 o ._ co IL FIGURE It Evolution of the ratio of hydrogen ~ to carbon (C) in the world fuel mix. Ibe figure for wood refers to dry wood suitable for energy production. If the progression is to continue beyond methane, production of large amounts of hydrogen fuel without fossil energy is required (see Marchetti, l985~. several conjectures are worthwhile. One is that it may be possible to match each pulse with a dominant energy supply technology, coal in the first case and oil in the second. During each pulse of growth, this form of energy supply may reach environmental constraints (and other constraints as well) that limit the overall growth of the energy system. In other words, a characteristic density may be all that is achievable or socially tolerable for each form of energy within the context of a larger industrial paradigm in which that form of energy dominates. ~ accommodate a further increase in per capita energy consumption, a society must shift each time to a form of primary energy that is not only economically sound, but also cleaner and in some ways more efficient, especially in terms of transport and storage. At a high hierarchical level, the pycle-adjusted view suggests that there are periods when the main orientation of the system is not so much growth as consolidation, with strong emphasis on squeezing more efficiency out of the system (a collection of technologies). At other times, the system seelo; to expand rapidly and relies on introduction and diffusion of new technologies that may be "inefficient" when introduced.

REGUL4R=IES IN TECHNOLOGICAL DEVELOPMENT 1o2 10 10° 10-1 _ 10-2 _ 1850 1875 1900 1925 1950 Coal Pulse 0.3~1.0 tce/cap Jim it /~1R99 Pulse / 0.8~2.3 tce/cap .; fit ~ 1965 ~·/ /Qt = 44yr /^t= 40yr , _ /1 1 / Gas Pulse 2.0~6.0 ,' tce/cap ,' ,,'1 ,'~ 2030 ,'^ t = 45 yr ~l I - - 1975 2000 2025 2050 Year 83 0.99 0 90 ~ o ._ i_ 0.50 LL 0.10 FIGURE 12 Growth pulses in world per capita energy consumption measured in tons of coal equivalent (tee). If historical discontinuities in per capita energy consumption persist, a new pulse of growth in world energy use would be expected to take off about the year 2000, which would triple per capita energy consumption from today's average world level of about 2 lee to about 6 lee (roughly half the current U.S. level). SOURCE: Ausubel et al. (1988~. It appears that we are nearing the trough of a demand cycle now. If strong demand for energy growth does not resume for another 7-10 years, as implied by the long-wave perspective, then improved energy efficiency looks like the most important near-term energy strategy, along with preparing the way for natural gas to accommodate another growth pulse (see Lee. this volume). This perspective also implies that the United States and other industrialized countnes, almost all of which have sufficient capacibr for electricity generation and other energy carriers in the near term, will face before the turn of the century a potential leap in energy consumption, not the steady state or low-growth world that many environmental advocates would like to see persist. ~ meet renewed rapid growth in demand in an environmentally sound way, gas must almost inevitably take the leading role, probably supported in particular niches by nuclear power. It is useful to ask whether energy efficiency is always consonant with environmental improvement. At the level of particular functions such as lighting or refrigeration, it is clear that many engineering systems, indeed probably many biological systems, tend to follow steady trajectories over long periods of time toward higher efficiency (Figure 13~. In most cases it may be supposed that increasing energy efficiency will also be environ- mentally beneficial A counterexample is electricity. Its use is less efficient than more direct use of alternatives such as natural gas, oil, and even coal and yet is often environmentally preferred. Another counterexample, the lean-burn (Otto-pycle) engine, produces less carbon monoxide but much more NOR than a less efficient engine with a catalytic converter, which

84 JESSE H. AUSUBEL 10° 10-1 Watt INewcomen , Saverv I 10-2 10-3 ~ 1700 1800 1900 2000 Triple Parsons Expansion 1 :.] Fluorescent li a' /~ HT Steam ~Turbine Cornish Tungsten ~` / filament / / ,~~ ~ Prime Movers Cellulose ~ /. t 1 - 50% = 300 yr Filament:/ Edison's ,7 First Lamp/ /:- Paraffin / ~ t 1 - 50% = 80 yr Candle / Mercury Sodium// / Lamp Lamp Lamp / /~ Ammonia Production | At 1 - 50% = 70 yr - Lamps Year 50 to ce 1 .0 0.1 FIGURE 13 Examples of increasing energy efficiency. Prime movers, lamps, and ammonia production are measured as machines or processes for energy transformation according to the second law of thermodynamics. Onginal analyses are by 1~ M. Slesser, University of Strathclyde, Scotland. SOURCE: Marchetti (1983~. is currently more environmentally attractive at the cost of efficiency. A1- though the overall long-term evolution of energy systems appears to be in the direction of both efficiency and environmental compatibility, at vari- ous levels and times the system may not be optimizing for both of these objectives or they may be in conflict. From transportation and energy, let us turn to materials, which figure prominently in the chapters by Herman et al. and by Ayres in this volume. Simple extrapolations of the kind used above have often been troublesome and unsuccessful as aids in projecting consumption of materials. As shown in Figure 14, past projections of demand for certain key materials remind us why studies such as those of Meadows et aL (1972) in the early 1970s foresaw extremely severe problems of both exhaustion of mineral resources and pollution associated with mineral use. What happened to create the gap between the extrapolated trends and realibr? Systems prove to be bounded in a varieW of ways, so that exponential growth does not persist indefinitely. In the case of materials, several factors have been at work, including economic growth rates, shifts in the composition of economies from manufacturing to services, and resource- saving technologies. But most important may be smart engineering that made feasible the substitution of plastics, composites, ceramics, and optical

REGULARITIES TV TECHNOLOGICAL DEVELOPMENT 40C 30 20 10 0~ 1950 1960 800 600 Year Steel Aluminum Projected ,, I i 1970 1 980 Projected' ~~M o 1955 19601970 1980 Year 10 Zinc Projected Hi' 1960 1970 1980 Year 85 0.8 10 8 6 4 2 o 1950 1 960 .2 Copper Projected '' Actual 1970 1 980 Year Nickel Projected '' 0.0 _ 1950 1960 1970 1980 Year FIGURE 14 Actual materials consumption for five metals, 195~1985, and projections made in 1970 for 197~1985. SOURCE: Tilton (1987~. materials for metals, that is, the continuing replacement of metals with nonmetals and the associated overall decrease in metal needs. The telecommunications sector provides a vivid example (Figure 15~. In 1955, telecommunications cables were made almost entirely of copper, steel, and lead. By 1984, close to 40 percent of the materials used were plastics. If substitution of lead by polyethylene for cable sheathing had not taken place, consumption of lead by AT&T alone might have reached a billion pounds per year, an amount to create considerable anxiety from the point of view of environment, given the toxic properties of lead. Herman et al. (this volume) have examined the possible "dematerialization" of the .,

86 JESSE H. AUSUBEL automobile. In considerable part, the phenomenon again has to do with the substitution of plastics for metals, as implied by Figure 16. Overall, there appears to be a decreasing dependence on common metals, perhaps combined with greater need for less common metals (Hib- bard, 1986~. There is also growing use of metals in the form of composites, coatings, films, and artificial structures. As Ayres (this volume) suggests, use of metals in such areas as electronics may dissipate more broadly and rapidly because many of the uses are highly dispersed and thus also entail greater complexity in recycling. The difficulty is that good data are not readily available, and may not exist, to back up such generalizations firmly. In 1976 Goeller and Weinberg sought to develop baseline information for what they termed the "age of substitutability." One of the notions they introduced was that of "demandite," the average nonrenewable resource used by human society. They defined demandite by taking the total extraction in moles of elements such as copper and iron and selected compounds (e.g., hydrocarbons) and computing the average hypothetical chemical composition of one demandite molecule (or average mole percent composition). Goeller and Weinberg excluded renewable resources, such as agricultural products, wood, and water, from demandite but looked at them in another portion of their study. Able 2 shows the result for the United States and for the world, for 1968, the most current year for which Goeller and Weinberg were able to perform the calculation in the mid-1970s. The dominance of hydrocarbon is spiking. It is interesting that in 1968 the United States had a more favorable hydrogen-to-carbon ratio than the world as a whole, partly offsetting from an environmental perspective the fact that U.S. energy consumption is so high. Broadly speaking, the need is apparent for developing and applying concepts like "demandite" on a regular basis. With steady monitoring, such approaches might serve as indicators that would alert us to substitution processes, improving projections and reducing the likelihood of the kind of erroneous projections shown in Figure 14. At a specific level, it is evident that, just as some environmental concerns about metals use may be decreasing, more attention must be given to plastics and paper, as also argued by Herman et al. (this volume) and Ayres (this volume). Although according to one estimate per capita use of materials in the United States remained constant between 1974 and 1985 at about 20,000 pounds per year, use of paper increased by about 25 percent to about 650 pounds, and use of plastics increased by about 40 percent to 180 pounds (Hibbard, 1986~. The latter figure is an obvious and essential part of the explanation for the recent widely reported concerns about the deterioration of environmental quality at beaches in the United States and Europe (see Lynn, this volume).

(9861) ll~eqellloS pue day Chinos spunod Jo suo!~llm U! 't861 pue SS61 'sa!Soloulloa~L .L~.w ~ salqe~ suo!l~lun~mo~al~1 lo a~nl~e~nue~ Dog sleualem ~0 In sl ~u cunu!cunl~ - Lunu!u~nl~ L8 ooz laces b86 1 ~N~dO7~a ~IDO]ONHO~ ~ sinew

88 JESSE H. AUSUBEL 10 - ~8 a) a) 6 - ._ - a) a) 4 2 o O Ward's Data (U.S. Fleet) , ' - · Ford Fleet f it, o ~ , O O I 1 1 1 Projected 1960 1970 1980 1990 Model Year FIGURE 16 mends in plastics content of U.S. passenger cars. Included is increased use of plastics in bumped, fuel tanks, air cleaners, and wheel covers, but not in body panels, for which construction in plastic is also increasingly feasible. SOURCE: Gjostein (1986~. In the search for advanced materials, we may be creating materials that are virtually immortal. One wonders, for example, whether the new marvelously strong materials increasingly popular for heavy-duty envelopes are as readily recycled or biodegradable as old-fashioned paper. From an environmental point of view, electronic memory would indeed be a sought-after substitution for paper as a medium for storing information, if it could be made long-lived and reliably reproducible. Also attractive is the notion of replacing packaging itself; food irradiation, for example, may be environmentally desirable if it can significantly reduce the required volume of packaging materials. 1b summarize, there is intriguing evidence of long-term regularities in the evolution, diffusion, and substitution of technologies. Understanding these regularities is of value for both environmental research and man- agement. From numerous illustrations available in transport, energy, and materials it is evident that there is need to increase scrutiny of environ- mental problems and opportunities associated with growth of air transport; increasing reliance on natural gas; and disposal of plastics. Clear possibili- ties exist for the development of illuminating indicators, such as trends in the hydrogen-to carbon ratio and the composition of demandite, connected to technologies and resources that would be valuable in our diagnoses

REGUI~UTIES IN TECHNOLOGICAL DEVELOPMENT TABLE 2 Average Nonrenewable Resources Used by Man in 1968, ~Demandite~ Atomic Percent Resource United States World CH 2.14 80.22 - CH 1.71 ~ 66.60 SiO 2 11.15 21.17 CaCO3 453 8.1S Fe 1.10 1.45 N 0.76 0.68 O 053 0.4S Na 053 0.45 C1 053 0.45 S 0.23 0.23 P 0.08 0.07 K 0.07 0.07 A 0.11 0.07 Cu. Zn,Pb 0.04 0.04 Mg 0.04 0.04 X 0.08 0.08 NOTE: Here, X represents all other chemical elements: highest in order of demand are Mn, Ba, Cr. F. Ti, Ni, Ar, Sn, B. Br, Zr, others account for less than 100,000 tons per year worldwide or less than 30,000 tons per year in the United States. The term CH refers to the combination of coal, oil, and natural gas, which are all made up of carbon and hydrogen in different ratios. The subscript refers to the average hydrogen-to carbon ratio. SOURCE: After Goeller and Weinberg (1976~. 89 and prognoses of environmental quality. We should not underestimate our technological ingenuity with respect to the environment nor the enormous dimensions of the systems requiring successful application of that ingenuity. ACKNOWLEDGMENTS I would like to thank William Clark Robert Frosch, Arnulf Grubler, Nebojsa Nakicenovic, and Stephen Schneider for sharing many ideas that led to this paper and Hedy Sladovich for research assistance.

9o JESSE H. AUSUBEL NOTE 1. Mathematically, a logistic function may be denoted by x/(tc -x) = exp(<xt + pa), where t is the independent variable usually representing some unit of time; ct is a constant representing rate of growth; ,B is a constant for the location parameter (it shifts the function in time, but does not affect the function's shape); tc is the asymptote that bounds the function and, therefore, specifies the level at which the growth process saturates; x is the actual level of growth achieved; and ('c -x) is the amount of growth still to be achieved before the saturation level is reached. Substituting f = x/,c in the equation expresses the growth process in terms of fractional share f of the asymptotic level tic reached; that is, the equation becomes f/~1-f) = exp(at + ,0), the Fisher and Pry (1971) model. Taking logarithms of both sides of the equation results in the left-hand side being expressed as a linear function of time, so that, when plotted, the secular trend of a logistic growth process appears as a straight line (sometimes with perturbations). The terminology employed here is used in the figures in this chapter. REFERENCES Ausubel, J. H., A. Grubler, and N. Nakicenovic. 1988. Carbon dioxide emissions in a methane economy. Climatic Change 12:245-263. Brimblecombe, P. 1987. The Big Smoke: A History of Air Pollution in London Since Medieval Times. London and New York: Methuen. Bruehl, C, and P. J. Crutzen. 1988. Scenarios of possible changes in atmospheric temperatures and ozone concentrations due to man's activities, estimated untie a one- dimensional coupled photochemical climate model. Climate Dynamics 2~3~:173-203. Fisher ~ ~ anti R ~ Pit' 1971 A simple model of technalc~Qical change. Technological Forecasting and Social Change 3:75 88. Freeman, C, J. Clark, and L Soete. 1982. Unemployment and Technical Innovation: A Study of Long Waves and Economic Development. Westport, Conn.: Greenwood. Gjostein, N. A. 1986. Automotive materials usage trends. Materials and Society 10~3~:369- 404. Goeller, H. E., and ~ M. Weinberg. 1976. The Age of Substitutability or What Do We Do When the Mercury Runs Out? Report 76-1, Institute for Energy Analysis, Oak Ridge, Tenn. Grubler, ~ 1988. The Rise and Fall of Infrastructures. Dissertation, Technical University of Vienna. Grubler, A., and N. Nakioenovic. 1988. Ibe dynamic evolution of methane technologies. Pp. 13-14 in The Methane Age, T. H. Lee, H. R. Linden, D. ~ Dreyfus, and T. Vasko, eds. Boston: Kluwer Academic Publishem. Hamblin, R. L., R B. Jacobsen, and J. L. Miller. 1973. A Mathematical Theory of Social Change. New York: John Wiley .£ Sons. Hibbard, W. R. 1986. Metals demand in the United States: An overview. Materials and Society 10(33:251-258. Key, P. L., and 1: D. Schlabach. 1986. Metals demand in telecommunications. Materials and Society 10~3~:433-451. Lotka, A. J. 1956. Elements of Mathematical Biology. New York: Dover. Marchetti, C 1983. On the role of science in the postindustrial society "Logos," the empire builder. Technological Forecasting and Social Change 24:197-206. Marchetti, C. 1985. Nuclear plants and nuclear niched Nuclear Science and Engineering 90:521-526. Marchetti, C. 1988. Infrastructures for movement: Past and future. Pp. 146-174 in Cities and Their Vital Systems: Infrastructure Past, Presents and Future, J. H. Ausubel and R. Herman, eds. Washington, D.C.: National Academy Press. sL A-lA-Al By- ~9 ~ as. ^~. ~ ^~. ^~ ~ ^. ~ ~ -COMA ~

REGUI~IES IN 17EC}:INOLOGICAL DEVELOPMENT 91 Marchetti, C, and N. Nakicenovic. 1979. The Dynamics of Energy Systems and the Logistic Substitution Model. RR-79-13. I>xenburg, Austria: International Institute for Applied Systems Analysis. Meadows, D. H., D. Lo Meadows, J. Randets, and W. W. Behrens III. 197Z The Limits to Growth. New York: Universe Books. Montroll, E. W., and W. W. Badger. 1974. Introduction to Quantitative Aspects of Social Phenomena. New York: Gordon and Breach. Montroll, E. W., and N. S. Gael. 1971. On the Volterra and other nonlinear models of interacting populations. Reviews of Modern Physics 43~2~:231. Nakicenovic, N. 1984. Growth to Limits: Long Waves and the Dynamics of Technology. I~enburg, Austria: International Institute for Applied Systems Analysis. Nakicenovic, N. 1985. The automotive road to technological change: Diffusion of the automobile as a process of technological substitution. Technological Forecasting and Social Change 29.309~340. Nakicenovic, N. 1988. Dynamics and replacement of U.S. transport infrastructures. Pp. 175-221 in Cities and Their Vital Systems: Infrastructure Past, Present, and Future. J. H. Ausubel and R. Herman, eds. Washington, D.C.: National Academy Press. National Research Council. 1986. Acid Deposition: Long-Term [Lends. Washington, D.C.: National Academy Press. National Research Council. 1988. Annual Report of the Strategic Highway Research Program (October 31, 1988~. Washington, D.C Olson, S. H. 1971. The Depletion Myth: A History of Railroad Use of Timber. Cambridge, Mass.: Harvard University Press. Reed, H. S., and R. H. Holland. 1919. Ihe growth rate of an annual plant helianthus. Proceedings of the National Academy of Sciences 5:135-144. Schumpeter, J. A. 1939. Business Cycles: A Theoretical, Historical, and Statistical Analysis of the Capitalist Process, Vols. I and II. New York: McGraw-Hill. Stewart, H. B. 1988. Recollecting the Future. Homewood, Ill.: Dow Jones-I'win. Tilton, J. E. 1987. Long-run growth in world metal demand: An interim report. Mineral Economics and Policy Program. Colorado School of Mines. Van Duijn, J. J. 1983. Ihe Long Wave in Economic Life. London: Allen and Unwin. Volterra, V. 1927. Variations and fluctuations in the number of coexisting species. Pp. 6~236 in Ihe Golden Age of Theoretical Ecology 1923 1940, F. M. Scudo and J. R. Ziegler, eds. New York: Springer, 1978.

The Promise of Technological Solutions

Next: THE PROMISE OF TECHNOLOGICAL SOLUTIONS »
Technology and Environment Get This Book
×
Buy Paperback | $55.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Technology and Environment is one of a series of publications designed to bring national attention to issues of the greatest importance in engineering and technology during the 25th year of the National Academy of Engineering.

A "paradox of technology" is that it can be both the source of environmental damage and our best hope for repairing such damage today and avoiding it in the future. Technology and Environment addresses this paradox and the blind spot it creates in our understanding of environmental crises. The book considers the proximate causes of environmental damage—machines, factories, cities, and so on—in a larger societal context, from which the will to devise and implement solutions must arise. It helps explain the depth and difficulty of such issues as global warming and hazardous wastes but also demonstrates the potential of technological innovation to have a constructive impact on the planet. With a range of data and examples, the authors cover such topics as the "industrial metabolism" of production and consumption, the environmental consequences of the information era, and design of environmentally compatible technologies.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!