Monday, October 1, 2018

"The Program," or, a strategy for universal prosperity in the twenty first century

Energy! What is it, how do you get it? 
Energy has an esoteric physics definition, but for the purposes of this blog, energy is the ability to do useful work. Work energy per unit time is power, while force multiplied by distance is work.
Throughout human history, if a human wanted something done, they had to use muscle power. Muscles, powered by food grown using sunlight, water, and ambient carbon dioxide, was the limiting factor in the deployment of work for useful things. There were a handful of exceptions in the form of water or windmills, of course. 
That all changed in the 18th century when engineers in Britain devised the first practical steam engines. For the first time, mechanical power could be harnessed to operate pumps, mills, and vehicles, that wasn't derived from mammalian metabolism. 
This first industrial revolution was followed by the widespread development of factories and assembly lines in the 19th century. 
At around this time the gasoline internal combustion engine was developed, such that by the 1930s rubber wheels on bitumen were overtaking trains as a preferred land surface transportation mechanism.
The Second World War saw the rapid maturation of aviation and oil extraction technology, and from 1948 until 1973, world per capita energy consumption grew at 7% per year. Physicist Gerard O'Neill predicted that by the year 2000, humanity would need space solar power to continue its incredible growth. 
What does this story mean? Gasoline is a fuel, or a mechanism for the storage and convenient dispatch of mechanical energy. In this, it is hard to beat, as it is pourable, readily available, and incredibly energy dense. While highly flammable, it can be made safe to use. 
7% annual energy growth encapsulates billions of humans being lifted from the poverty of preindustrial subsistence agriculture. With gasoline powered machines, an individual human can be so much more productive, enabling great increases in quality of life. 
Of course, this trend has not continued to the present. Instead, we had a series of oil shocks and a plateauing of wealth in developed countries. Indeed, recent trends seem to point toward a zero-sum struggle for control of limited resources, rather than confidence in limitless growth, at least to the point of post-scarcity. Indeed, energy has only gotten more expensive. 
The figure below shows 5-year smoothed data for US per capita energy consumption and the price of crude oil. The period from 1939 until 1973 was marked by robust growth and steady, slightly declining oil prices. The period from 1973 until present has endured great uncertainty and change in the price of oil (up to 20% per year for several years!), and a commensurate loss of steady growth in the exploitation of energy. 

I believe that worldwide economic stagnation in the 1980s and 1990s was headed off only by the coincidental development of consumer-accessible computers.
Computers are a very unusual case, as they have gotten twice as good every 18 months for many decades, at least until recently. To provide an example of how unusual this is, the Curiosity Mars rover uses a particular kind of space-grade chip, which is necessarily of rather poor performance compared to the state of the art. The Mars Helicopter doesn't have enough power to run this chip, so the engineers had to select an option that was smaller and less power hungry. In any other field, this would imply even poorer performance, but because this is a computer, the smaller chip is 60 times as powerful, which is a great boon to the software developers! 
Moore's Law has tapered off in recent years due to fundamental physics, but I think that we're a long way from fully exploiting the potential of current computer hardware. 
While our civilisation has failed to obtain an infinite supply of ever cheaper energy, the growth of wildly cheap computing capacity has allowed us to subvert that constraint to some extent. Although computers don't perform mechanical work in any macro sense, their prodigious abilities with logic and calculation have extended the cognitive capacity of humanity in the same way that engines have extended our muscles. 
Of course, this has further exacerbated a chronic oversupply of human labor present since the end of WWII, since all human needs could now be met in developed countries with a ten hour work week. Today, we are seeing the computerized automation of middle management, such that employees of organizations like Uber are actually dispatched by an algorithm. And when one considers the affordability of real estate it's clear that in this period of macroscale industrial stagnation much value has been stealthily inflated away.
After this necessarily imprecise historical and economic review, it's time for me to pull out my crystal ball and talk about The Program.
I was discussing an application of cheap solar energy with my cousin J when my reference to The Program elicited a blank stare. This blog is my attempt to structure on paper my thoughts about energy policy over the next 30 years.
When we think of 2050, we think of supersonic passenger jets, flying cars, good food, free education, housing, healthcare, and other trappings of a wealthy, prosperous society. Yet even in 2018, nothing can happen without oil. Oil is a dirty, messy business that destroys the environment, corrupts governments, ruins our health, and poisons the atmosphere. But for all that, we can't live without it because, as explained earlier, it can give any human on Earth the superhero capacity to dispatch, as a rough average, 100 times more work than their muscles could achieve on their own. 
Given that oil is not only poisonous, it's also finite, a replacement will have to be found and deployed, and in our lifetime. 
While it's possible to chemically synthesize fuels at huge expense, I'm going to focus this blog on the technology that will prevail, namely electricity. 
Electricity and magnetism are magic. Like gasoline, electricity is a form of energy that can flow down conductive wires and, with motors, perform mechanical work. Its technological maturation has progressed alongside fuels and is so successful that everyone can get it from outlets in their walls.
Whereas a 19th century factory may have transmitted energy from a central steam engine using belts or shafts, these days nearly everything is run using electric motors. My robot vacuum has 12 motors in it! 
But while electricity has been the go to for appliances for decades, and not just because no-one wants to run a dishwasher with a motorcycle engine, gasoline has been the traditional energy source in mobile applications. 
This reminds me of the old joke about electric helicopters, which is that they needed a very long extension cord! In all seriousness, boats, cars, planes, and rockets needed fuels they could carry with them, so electricity was not an option. Recall why gasoline is such a compelling fuel! Cheap, pourable, energy dense, safe enough. 
When my father built the house in which I grew up in 1986, he used power tools powered by the mains, and drove a petrol powered car. When I attempt to construct things, I use battery powered tools and drive an electric, battery-powered car. No-one has a gas-powered mobile phone. World battery production is growing as batteries power ever-larger vehicles. The Tesla Gigafactory will reach 35GWh production later this year, two years ahead of schedule. There are serious proposals to build battery ships to transport electricity from solar farms in North Africa to Northern Europe!
In 1973, battery technology was not very advanced. In 2018, there are battery-powered planes that have crossed the English Channel. In short, the geological and geopolitical shortage of oil that halted the mechanical progress of humans is finally ready to be circumvented. 
Lithium batteries, like all products, require the mining of certain materials which have environmental effects. Needless to say, for equivalent work, lithium batteries are much less harmful than oil! They're also highly recyclable.
To get technical for a moment, the energy density of gasoline is about 46MJ/kg. The energy density of the best batteries is about 1MJ/kg, which improves by about 5% per year. The saving grace is that electric motors can be 95% efficient, while a car engine would be lucky to reach 15% efficiency, with the rest being wasted as heat. That means that on a per kg basis, batteries+motors are only 7 times less mass efficient than gasoline+engines. Further, electric motors have much higher power density, so can be lighter. Finally, vehicles such as cars spend most of their mass budget on things other than fuel and power train, so extra batteries can be added with only marginal increases in weight. This is notably not the case with long haul jets (50% fuel mass fraction) or rockets (95% fuel mass fraction) which both consume an inordinate amount of fuel.
To consider my personal carbon footprint, I use low wattage LED bulbs, but every time I fly to Australia my share of the plane's fuel is about 500kg, or 180 gallons. Each way. Driving a Hummer wouldn't make much difference!
A frequent though inaccurate criticism of electric cars is that the electricity they use, generated by coal, just moves the pollution elsewhere. This is imprecise, as electric cars are about 5 times as efficient at using energy due to regenerative braking and better motor efficiency, while electricity power plants are about twice as efficient as car engines, primarily because they operate at steady state and have better heat disposal mechanisms available.
Incidentally, while batteries are not yet good enough to operate long haul flights, there are now hundreds of companies developing short range electric commuter planes. Further, the power density and mechanical simplicity of electric motors allows for vertical take and landing, like a helicopter. Finally, I think that electric power has unique advantages which point the way to cost effective consumer supersonic flight, which is definitely part of my awesome vision for the future.
Nevertheless, The Program does not stop once it has succeeded in supplanting gasoline as the fuel of choice for all vehicles except orbital rockets. Indeed, as far as global warming goes, humanity could burn all the oil and all the gas and do relatively little harm compared to reserves of coal. Coal, a black, carbon rich fossil of trees, stores concentrated ancient sunlight and is extremely popular as a source of electricity.
While wind power has matured in recent years, the most applicable renewable natural resource is solar power. The sun is about 110 times wider than the Earth and will burn for another five billion years. During the day, every square meter of the Earth's surface receives about a kilowatt of power. It just rains down from space for free. It powers trees, so anywhere there's green, there's solar power.
But recall that preindustrial societies are solar powered! Horses and cows eat solar powered grass and humans eat wheat, corn, cows, and so on. No-one is deriving nutritional value from coal. How can solar power produce enough energy for our civilization?
The answer rests in efficiency. A modern commercial solar panel is about 20% efficient. Combined with a 80% efficient power transmission system, a 90% efficient battery charger and a 75% efficient cordless drill, about 10% of that kW of solar power, or 100W, makes it to the work piece.
Contrast this with agriculture, which also requires arable land, fertilizer, pesticide, and irrigation. Plants spend most of their energy transpiring water to keep cool in the sun, and are less than 0.1% efficient at converting solar energy into digestible starch. Cows or yeasts are about the same. Then the human metabolism is about 5% efficient at converting consumed energy to mechanical work.
So while a solar powered electric system is 10% efficient, an agricultural system is 0.000005% efficient. This is the main reason that farms are really big. To create value, they have to capture a lot of sunlight, which is rather dispersed. This is also the reason why biofuels can never scale to completely replace gasoline. Their end-to-end efficiency is thousands of times lower than solar and batteries, and there isn't enough arable land to produce enough ethanol. Not even close.
To dwell on power transmission for a moment, let's consider trees. Trees are self-powered, but they're not generally considered to be capable of locomotion. Animals that move need to eat a lot of plants to concentrate the stored solar energy. This is why very few animals bother with photosynthesis. Likewise, some electrical applications require so little energy that they can be powered directly by solar panels. But most human machines are too energy intensive to be powered in this way.
For instance, at highway speeds my car consumes about 15kW. If this were provided by a solar panel, in addition to being unable to drive at night, the car would need to be 50m long to fit in a standard lane. In the US this may be permitted on certain roads at certain times with escort vehicles; the SpaceX Falcon 9 rocket first stage is of a similar size.
Clearly, solar panels on houses or in dedicated farms are needed to concentrate the sun's power. A transmission grid continues to dispatch the supply to the end user.
How many solar panels are needed? Lots and lots! For a rough estimate, recall that solar is about 2000 times as efficient as non-meat food production, but that per capita energy consumption is only 1% food in industrial societies. Therefore, about 5% of the area devoted to agriculture is necessary to meet foreseeable electricity needs. About 11% of Earth's land surface is used for crops, so we're talking about 0.5%.
As an example, using only desert military bases in California and Nevada, which receive a lot of sun, would produce enough power to supply the entirety of North America. In practice, a mix of rooftop solar and utility farms in sunny areas is the most robust approach.
It's instructive to consider areal land uses for other forms of energy production. In Australia's picturesque Hunter Valley, there are several enormous open cut coal mines. I computed that a solar farm operating for 20 years will produce energy equivalent in value to a coal seam 3m thick. That is, even if the coal is at the surface, and it's not, if it's less than 3m thick it's better to use as a foundation for solar panels than to dig it up and burn it. Further, there aren't that many coal seams that thick anymore! A similar argument about nuclear power, given a 20km exclusion zone, shows that more energy rains down from the sky in that area than can be produced by fission.
Coal - leave it in the ground.
Like wind, solar is not a continually available resource. It varies daily and seasonally. For this reason, a smarter grid with responsive demand, intersticial storage, and a rational pricing strategy is a worthy goal. In practice, this means that in the future power will be very cheap at noon and during summer, so all sorts of new applications are possible. My favorite examples, though far from exhaustive, are aluminium production and mass desalination for agriculture, both of which require power below the price of 1c/kWh.
Indeed, in 2017 solar power supply bids reached 2.7c/kWh in Mexico, indicative of a long awaited reversal of the decades-long trend of gradually increasing energy costs. Extrapolating is risky business, but at present rates solar power will reduce in price by a factor of 10 every 17 years. Continuing that trend, in 2050 power will be cheap enough to artificially refill rivers parched by global warming with pumps and desalination, for example.
This trend need not stop in 2050. It is my belief that by the time I die, historians will see the period between 1973 and 2013 as an anomaly, a blip, a detour into computing on an otherwise unbroken industrial trend towards ever greater deployment of useful energy for peaceful humanitarian purposes.
What is The Program? The Program is a vision for the mass deployment of solar power, smart grid technology, grid storage, and electric vehicles. It is now economically viable to talk about public funding of generational infrastructure on the supply side, while a clear roadmap will drive private innovation in vehicle design. 

1 comment:

If you want to comment without a google or open-ID account, sign enough of your name that at least I know who you are, or leave an email address or bank account details or something.