1장
The Energy Mirage: Separating Reality from Wishful Thinking
Imagine a world where electric cars dominate our roads, nuclear power is so cheap it barely needs metering, and small-scale renewable energy powers our civilization. This utopian energy future has been promised repeatedly over the past century, yet consistently fails to materialize. In "Energy Myths and Realities," Vaclav Smil-described by Bill Gates as someone who "thinks about energy systems analytically without bias"-methodically dismantles our most cherished beliefs about energy transitions. The book has become required reading in energy policy circles, with executives from major energy companies and policymakers alike citing its influence on their thinking. Beyond technical analysis, Smil's work exposes how human psychology, political expediency, and wishful thinking repeatedly cloud our judgment about one of civilization's most fundamental needs.
2장
The Electric Car's Century-Long Promise
The electric vehicle revolution seems perpetually just around the corner. During the automobile industry's earliest days in the late 1890s, electric cars actually outnumbered gasoline-powered vehicles in America. They were cleaner, quieter, and didn't require dangerous hand-cranking. Thomas Edison himself dismissed Henry Ford's work on gasoline engines as not "really useful," investing nearly a decade developing his nickel-iron-alkaline battery to power what he believed would be the dominant transportation technology.
History proved Edison wrong. By the 1930s, commercial electric car manufacturing had vanished, defeated by Ford's assembly-line manufacturing, Kettering's electric starter (eliminating the need for cranking), and Midgley's tetraethyl lead solution to engine knock. The pattern of electric vehicle enthusiasm followed by disappointment repeated throughout the 20th century. The 1973 oil crisis sparked renewed interest, while California's 1990 zero-emission mandate required 2% of vehicles be electric by 1998 and 10% by 2003-goals that were eventually abandoned when reality intervened.
Today's electric vehicle enthusiasm mirrors these historical patterns. The Tesla Roadster, essentially a British Lotus Elise loaded with 6,831 lithium-ion batteries, sold about 900 units by late 2009 at a price of $101,500. Despite media hype comparing electric cars to personal computers or mobile phones, realistic forecasts put electric vehicles at just 0.6-2% of total vehicles by 2020.
The infrastructure challenges remain formidable. Converting America's entire fleet would require approximately 980 TWh of electricity annually-25% of 2008 U.S. generation-which historically took utilities 15 years to add. Even with perfect coordination of charging schedules, America's current electricity generation mix offers electric vehicles no primary energy savings or carbon emissions advantage compared to efficient gasoline or hybrid vehicles.
Meanwhile, gasoline engines continue advancing rapidly. Daimler's innovative DiesOtto design combines spark ignition at full load with diesel-like self-ignition during partial load, potentially achieving over 60 mpg. These developments suggest the internal combustion engine will remain dominant for decades to come, with electric vehicles gradually gaining market share rather than explosively disrupting transportation.
3장
Nuclear Power's Unfulfilled Promise
"Too cheap to meter." This infamous phrase, attributed to Lewis L. Strauss in 1954, has haunted nuclear power for decades. Though Strauss never explicitly linked this claim to nuclear fission, the phrase became emblematic of technological hubris and has been used by critics to question nuclear power's credibility ever since.
Ironically, many power engineers and utility economists harbored serious doubts about nuclear development from the beginning. David E. Lilienthal, the first chairman of the U.S. Atomic Energy Commission, privately described nuclear advocates as "fanatics or zealots" driven by "salesmanship, propaganda, and overzealousness." Nevertheless, nuclear power development proceeded rapidly after 1950, driven not by economic demand but by political factors-America's desire to demonstrate peaceful applications of nuclear technology and strategic considerations about maintaining technological leadership.
The rush to deploy submarine reactor designs for civilian use established pressurized water reactors (PWRs) as the dominant global technology, creating a technical lock-in with far-reaching consequences. Nuclear generation took off in the late 1960s, with 83 orders between 1965-1969, and by 1970, the United States had 107 units online, under construction, or purchased.
Then came the 1973-74 oil crisis, which paradoxically damaged nuclear power's prospects. Higher oil prices led to inflation, lower economic growth, and electricity conservation that dramatically reduced demand growth from 7% annually to just 2-3%. Simultaneously, regulatory requirements exploded from about 100 codes in 1971 to over 1,600 by 1975. Construction times ballooned from 50 months to 130 months, and costs skyrocketed. Some projects saw catastrophic overruns: California's Diablo Canyon jumped from $450 million to $4.4 billion, while New York's Shoreham went from $241 million to $6 billion.
Even as conventional nuclear power retreated, experts pinned their hopes on liquid metal fast breeder reactors (LMFBRs). These reactors could theoretically produce 20% more fuel than consumed by converting abundant uranium-238 into fissile plutonium-239. General Electric predicted commercial breeders by 1982, dominating American electricity by 1992. Yet the American demonstration breeder, initially scheduled for 1975 at $100 million, was abandoned in 1983 after costs ballooned to $675 million. France's Superphenix operated at full power for less than ten months over eleven years before shutting down permanently.
Today, nuclear power represents what Smil calls a "successful failure"-no energy technology commercialized as rapidly, yet none fell so dramatically short of initial expectations. Despite receiving 96% of all U.S. energy research funding between 1947-1998 (about $145 billion), nuclear power faces persistent economic disputes and waste disposal challenges. No country-not even nuclear-dependent France-has devised an acceptable permanent storage solution for radioactive waste requiring millennia of isolation.
4장
The Soft Energy Illusion
In his influential 1976 Foreign Affairs essay, Amory Lovins proposed a revolutionary "soft energy" path emphasizing decentralized renewable energy and efficiency as an alternative to centralized power generation. His vision promised to eliminate nuclear proliferation risks while providing energy matched to end-use needs through naturally distributed renewable flows.
Lovins made sweeping claims about renewable energy's advantages: that natural energy flows "are always there whether we use them or not" and perfectly match geographic distribution of needs. He insisted decentralized generation would eliminate distribution costs, with savings far outweighing maintenance of dispersed systems. Beyond mere practicality, Lovins claimed soft energy would be "elegant," benefit the poor, spread democracy "from the ground up," foster community resilience, and provide "ecologically inoffensive" alternatives.
Reality proved far less accommodating. At the turn of the 21st century, no major economy relied on soft energy for more than a negligible fraction of its supply. While Lovins correctly predicted the U.S. would consume about 100 EJ by 2000, his composition forecast proved spectacularly wrong. Rather than his projected 33% from soft energy sources, renewable energies excluding large hydropower provided just 4% of U.S. primary energy-with 90% of that coming from industrial-scale operations like logging residue burning, large ethanol facilities, and commercial wind farms. True small-scale, decentralized energy contributed less than 0.5% of U.S. energy-missing Lovins' target by 98.5%.
When the soft-energy vision faded in the 1980s, Lovins pivoted to promoting the Hypercar-an aerodynamic, carbon-fiber vehicle promising to be quiet, safe, 95% less polluting, affordable to lease, and achieving up to 200 mpg fuel efficiency. Despite establishing the Hypercar Center in 1994 and launching Hypercar Inc. in 1999, no such revolutionary vehicle materialized. By 2004, the company had abandoned the concept and rebranded as Fiberforge to focus on composite materials manufacturing.
Lovins wasn't alone in his overoptimistic renewable energy projections. The InterTechnology Corporation predicted solar would supply 36% of America's industrial heat by 2000. A Harvard Business School study claimed solar could "reasonably" meet 20% of total U.S. energy needs by 2000 using existing technologies. All these predictions missed reality by orders of magnitude.
Even Maoist China embraced small-scale energy approaches, including small coal mines, small hydrostations, and family biogas digesters. Despite initial enthusiasm with over 400,000 biogas digesters by 1975 and ambitious goals of 70 million units by 1985, most installations failed due to technical challenges. By the 1990s, China had abandoned these small-scale approaches in favor of energy megaprojects.
The failure of the soft-energy vision isn't surprising when understood as a product of the countercultural revolt of the late 1960s and early 1970s-it was intended as a profound social transformation, not just an energy solution. Like other grand schemes meant to reform society, from American Technocracy to Maoism, it ultimately failed.
5장
The Peak Oil Panic
Peak oil theory has generated apocalyptic forecasts of civilization's collapse, with some adherents like Richard Duncan predicting industrial civilization's demise by 2025 and humanity's return to primitive conditions. Duncan's "Olduvai Gorge theory" rests on the incorrect claim that global per-capita energy consumption peaked in 1978, when in reality it was 10% higher in 2008.
M. King Hubbert's forecasting record shows significant flaws. While he correctly predicted the 1970 U.S. oil peak, his estimate was 18% below actual production. More importantly, his forecast assumed ultimate recovery of 200 billion barrels, yet by 2005 the U.S. had already produced 192 billion barrels while still ranking as the world's third-largest producer with 30 billion barrels in reserves.
Hubbert's global oil forecasts proved even less accurate. His 1969 predictions projected peaks either in 1990 or 2000, but neither materialized. Other peak oil predictions have similarly failed, including those by the Workshop on Alternative Energy Strategies (1990s peak), Andrew Flower (pre-2000 peak), the CIA (1980s peak), and Colin Campbell (1989 peak). Perhaps most absurd was Kenneth Deffeyes' prediction of a Thanksgiving Day 2005 global peak.
The U.S. Geological Survey's latest assessment puts global ultimate oil recovery at about 3.02 trillion barrels, including 690 billion barrels from appreciation of known fields and 730 billion barrels yet to be discovered. Major conventional oil discoveries are expected in the Mesopotamian Foredeep Basin, West Siberian Basin, Zagros Fold Belt, Niger Delta, Rub al Khali Basin, and East Greenland Rift Basin.
No sharp line divides conventional from nonconventional oil resources. The hydrocarbon continuum ranges from mobile medium-heavy oils to immobile tar sands and bitumen to oil shales. Most of the world's 4-5 trillion barrels of heavy oils are concentrated in Venezuela (1.2+ trillion barrels) and Alberta's oil sands (2.5 trillion barrels of bitumen).
Market forces profoundly influence oil production curves, contradicting peak oil theories that ignore demand factors. After OPEC's 1973 price quintupling, global consumption initially fell just 1.5% before rebounding. However, when prices nearly tripled again between 1978-1981, production dropped 15% by 1983-clear evidence of market response, not resource depletion.
While conventional oil extraction will eventually peak, possibly within two decades, this doesn't spell civilization's collapse. The combination of larger-than-estimated conventional reserves, abundant nonconventional resources, alternatives like natural gas, and our capacity to reduce demand through efficiency means oil will remain significant through mid-century.
6장
The Carbon Sequestration Challenge
With global CO2 emissions set to rise substantially in coming decades despite sustainability rhetoric, carbon sequestration has become a focal point for stabilization efforts. By 2006, China had surpassed the United States as the world's largest CO2 emitter, with emissions 2.15 times higher than 1990 levels. India's emissions doubled between 1990-2005.
Earth's biosphere functions as a massive carbon sequestration system, with photosynthesis annually moving over 120 billion tons of carbon from atmosphere to plants. Scientific consensus remains elusive regarding net carbon sequestration. Higher atmospheric CO2 has increased plant carbon storage by 1.2-2.6 billion tons annually compared to preindustrial times, but the terrestrial biosphere fluctuates between being a carbon source and sink.
New tree plantations face two fundamental limitations: sequestration durability and required scale. Fast-growing species reach maturity in 10-15 years, after which they provide negligible additional carbon storage. The scale required is staggering-sequestering just 10% of 2005's carbon emissions would require boreal plantings nearly equal to the combined forest area of North America and Russia.
Soils already store more than twice the carbon in the atmosphere and nearly four times that in terrestrial plants. Conservation tillage, cover crops, and crop rotation could enhance storage while improving soil productivity. However, complications abound: increasing tropospheric ozone can reduce plant productivity, global warming accelerates soil decomposition releasing more CO2, and rising CO2 levels reduce root-derived carbon sequestration in many European trees.
Biochar, inspired by Amazonian terra preta soils, can hold 2-2.5 times more carbon than regular soils and provide exceptionally long-lived storage-potentially 1,000 years. Despite enthusiastic claims, practical limitations abound: crop residues are needed to prevent soil erosion, the same biomass is targeted for cellulosic ethanol production, and incorporation requires tillage that contradicts no-till farming practices.
Ocean iron fertilization exploits the fact that iron controls productivity in nutrient-poor waters. However, effectiveness varies dramatically by region-while equatorial experiments showed promise, colder waters showed no carbon pump activation. The largest experiment saw iron-stimulated phytoplankton quickly consumed by amphipods, preventing further growth.
With biological sequestration methods proving unreliable or impractical at scale, technical solutions offer more controlled but often limited or expensive alternatives. Mineral carbonation offers advantages but requires at least three tons of serpentinite ore for every ton of CO2 sequestered. Controlling just coal combustion emissions (12 billion tons in 2005) would require mining over 33 billion tons of ore annually-nearly three times the mass of all fossil fuels extracted that year.
Carbon capture and sequestration (CCS) from concentrated combustion sources has received the most attention. The scale challenge is immense-sequestering just 15% of global CO2 emissions would require handling 4.8 billion tons annually, creating an industry 1.3-2.2 times the volume of the global oil industry. At $30-75 per ton for capture and compression, plus transport and storage costs, sequestering just 15% of emissions would cost approximately $300 billion annually.
Capturing and storing carbon dioxide also imposes substantial energy penalties on power generation. Coal-fired plants would increase their internal electricity consumption by 30-40% to power carbon capture systems. Post-sequestration concerns include potential leakage through unknown faults, fractures, or earthquake damage. Even a tiny annual leakage rate of 0.1% could eventually release significant carbon back to the atmosphere.
7장
The Biofuel Mirage
Liquid biofuels for transportation aren't new-Henry Ford promoted ethanol and his Model T could run on gasoline, ethanol, or a mixture. While Brazil successfully implemented a sugar cane-based National Alcohol Program starting in 1975, U.S. ethanol production remained marginal until recent acceleration, reaching 35 billion liters by 2008.
The fundamental problem remains scale-replacing global transportation fuels with biofuels would require more land than is currently cultivated in all tropical regions. In 2005, worldwide demand for liquid transportation fuels equaled about 2 billion metric tons of crude oil. Even using Brazilian sugar cane ethanol's relatively high power density of 0.45 W/m2, replacing transportation fuels would require about 600 million hectares-more than all currently cultivated tropical land and nearly 40% of the world's entire cultivated area.
America's enormous gasoline consumption combined with ethanol's low power density (only 0.25 W/m2) means corn-derived ethanol can never supply more than a small fraction of U.S. fuel needs. Converting America's entire corn harvest would yield just 13% of gasoline consumption. Despite these limitations, powerful agribusiness lobbies have secured massive subsidies worth $5.5-7.3 billion annually in 2006.
More concerning are environmental impacts: corn cultivation is America's largest source of agricultural soil loss, requires heavy nitrogen fertilizer application causing Gulf of Mexico dead zones, demands extensive irrigation depleting aquifers, and produces significant wastewater. Rather than helping climate change, intensive farming for biofuels actually increases greenhouse gas emissions through higher nitrous oxide releases.
Sugar cane offers clear advantages over corn for ethanol production, yielding about 65 tons/hectare with 12% sucrose content. Even with high conversion efficiency of 82L/ton, the 19 million hectares currently growing sugar cane worldwide would produce ethanol equivalent to less than 6% of global gasoline consumption.
Cellulosic ethanol-alcohol fermented from sugars derived from cellulose-is promoted as the perfect answer to food-versus-fuel concerns. Despite significant investment from the Department of Energy and Silicon Valley entrepreneurs like Vinod Khosla, who claimed cellulosic ethanol would be cost-competitive by 2009, reality falls far short of these dreams.
Fundamental challenges persist: crop residues aren't valueless waste but provide essential agroecosystem services by returning nutrients to soil, replenishing organic matter, retaining moisture, and preventing erosion. Even America's abundant corn stover (200 million tons annually) can't be harvested indiscriminately-conservative estimates suggest only 40% could be sustainably removed, equivalent to just 3% of U.S. gasoline consumption.
8장
Wind Power: Promise and Limitations
Wind power's global potential has been quantified by Stanford researchers Archer and Jacobson, who found that about 13% of all measured locations have winds strong enough for low-cost commercial electricity generation. Their calculations suggest global wind power potential of 630 PWh (72 TW)-far exceeding current global energy consumption.
The 21st century has seen wind elevated to a prime candidate for post-fossil fuel electricity, with impressive advances in turbine design and generation capacity. Unit capacity has risen from under 50 kW in the early 1980s to over 2 MW today, with prototype machines reaching 6 MW. Modern wind farms like Texas's Horse Hollow (735 MW) equal midsize coal-fired stations in capacity.
Global installed wind capacity rose from 4.8 GW in 1995 to 120.8 GW by 2008. Europe leads with 66 GW (55% of global total), while U.S. capacity grew from just 10 MW in 1981 to 25.1 GW by 2008-a 2,500-fold increase in 27 years.
Beyond fundamental resource limitations, wind power faces significant constraints. Wind's extraordinarily low power density presents a major challenge. Even in windy regions, spacing requirements between turbines reduce power density to just 2 W/m2 on average. Large-scale extraction would further reduce this to around 1 W/m2. Supplying half of today's electricity would require 2-4 million km2 of wind farms-areas comparable to Mexico or half of Brazil.
Wind intermittency creates integration challenges beyond about 10-20% of system capacity, requiring extensive interconnections. Transmission infrastructure presents another major obstacle. The Dakotas cannot supply California or the Northeast without massive new high-voltage lines.
Wind power's future growth trajectory remains uncertain. In 2007, wind turbines produced just 1.25% of global electricity, with Denmark leading at 21%, followed by Spain (12%), Portugal (9%), Ireland (8%), and Germany (7%), while the U.S. remained below 2%.
Technical realities make certain limits clear: isolated grids could draw most electricity from wind only with massive storage that doesn't yet exist; nations with abundant wind and strong grid interconnections like Denmark might reach 40-50%; regions with high-capacity grids could eventually derive 30% from wind; globally, 15% by 2030-2040 is aspirational, while 30% is unrealistic and 50% impossible.
9장
The Pace of Energy Transitions
The summer of 2008 brought both record oil prices and ambitious energy transition proposals. Craig Venter announced progress toward synthetic bacteria for biofuel production. T. Boone Pickens promoted his plan to replace natural gas-generated electricity with Great Plains wind power. Al Gore called for 100% renewable electricity within just 10 years, claiming the goal was "achievable, affordable and transformative."
These proposals share a fatal flaw with previous failed energy forecasts: they ignore the inherently slow pace of energy transitions. Previous energy predictions have consistently failed. Despite forecasts that coal would be obsolete by 2000, it still produces almost 50% of U.S. electricity, with nuclear at 20%. New renewables provide less than 2.5% of U.S. energy.
Historical data reveals the first major transition-from biomass to coal-occurred in the U.S. only in the 1880s, globally in the late 1890s, in Russia during the late 1920s, and in China during the 1960s. Many African countries still rely primarily on traditional biomass.
Coal's global dominance receded from 95% in 1900 to 60% by 1950, being surpassed by oil only in 1965. By 2000, coal had declined to 24% of energy supply, but then began regaining importance, reaching 29% by 2008-higher than during the 1973 energy crisis.
The pace of these transitions is revealing: oil took 50 years from its 1860s commercial introduction to capture 10% of the global energy market, and another 30 years to reach 25%. Natural gas required 70 years (1900-1970) to rise from 1% to 20% of global energy.
Energy transitions proceed gradually for various fundamental reasons. For primary energy supply, the lengthy timeline stems from the massive financial investments required to develop and perfect extensive infrastructure. The global oil industry alone handles 30 billion barrels annually through a complex system worth over $5 trillion that took more than a century to build.
The scale of our current energy transition challenge is unprecedented. While nonconventional fossil fuels and renewable energy sources receive significant attention, they represent minuscule portions of global energy supply-nonconventional oil provides only about 1% of primary energy, while all renewable conversions combined supply merely 0.5% of commercial energy globally.
Gore's claim that renewable energy costs will continuously fall like computer chips fundamentally misunderstands technological progress. His comparison to silicon chip prices suggests photovoltaic electricity costs could halve every eighteen months. This analogy fails because Moore's Law works primarily through increasingly dense transistor packing, not cheaper silicon. While PV cells have indeed become cheaper-from $20/watt in 1980 to around $4.50/watt in 2009-their efficiency improvements follow much slower trajectories.
10장
Beyond the Myths: Toward Rational Energy Policy
My systematic deconstruction of prominent energy myths offers lessons ranging from obvious to subtle. Some represent cautionary tales of misplaced zeal, while others demonstrate the folly of uncritical advocacy and hasty implementation.
The most important lesson from electric vehicle plans is that such a seemingly revolutionary shift wouldn't significantly reduce primary energy consumption. While widespread EV adoption would reduce crude oil demand, it wouldn't meaningfully decrease fossil fuel dependence unless the additional electricity came from highly efficient combined-cycle generation or renewables.
Nuclear power offers a crucial lesson about unrealistic expectations for emerging energy technologies. It represents a "successful failure"-a technology that captured substantial market share and proved reliable and economical, yet fell far short of initial expectations and faces unresolved challenges.
The failure of soft-energy conversions to capture significant market share teaches two critical lessons: energy goals shouldn't be driven by ideology, and no single approach should be elevated to an all-encompassing solution. Energy transformations will have socioeconomic consequences, but energy systems shouldn't primarily serve as tools for social transformation.
The obsession with precisely timing peak oil perfectly illustrates the futility of assigning dates to predicted future events. Such forecasts are invariably wrong and mislead those who take them seriously. Rather than obsessing over the exact timing, we should gradually shift to alternative energy sources so that peak oil production, whenever it occurs, becomes as unremarkable as past peaks in fuelwood or coal production.
The most crucial lesson from examining carbon sequestration plans is that avoiding environmental impacts is far superior to neutralizing them after they occur. Affluent nations could moderate energy use to intensify decarbonization and achieve absolute emission reductions. Modernizing countries could achieve economic gains without replicating the West's undesirable choices in building and transportation design.
Rapid expansion of crop-based ethanol exemplifies energy policy driven by overemphasis on a few positive aspects while neglecting negative consequences. The costs vastly outweigh benefits in economic, social, and environmental terms. America's strategic position would benefit more from sound fiscal policies and vehicle efficiency improvements than from spending billions converting Midwestern corn to ethanol.
Exaggerated expectations for wind power stem from selective evidence interpretation. While the resource is substantial, economically harvestable power is considerably smaller-likely less than 10% of theoretical capacity. Well-sited, well-engineered wind farms within interconnected grids have an important role in energy transition, but cannot fulfill unrealistic expectations.
To avoid energy myths, we should distrust strong claims about future adoption of new energy sources, recognize the persistence of established resources and technologies, avoid uncritically embracing unproven energies because they fit ideological preferences, remember the extensive infrastructure required before new energy systems can be widely adopted, and acknowledge that energy transitions are prolonged affairs lasting decades, not subject to Moore's law of computing.
As Terence wrote over twenty-one centuries ago: "Men believe what they want to"-hardly the best foundation for rational energy policy.
