[ Seldom have I seen such a vitriolic fight among scientists. As you’ll see below, I think all of them miss the main problems with trying to create a 100% renewable energy system, though I can only summarize and leave a great deal out, though you can see the details in my book “When Trucks Stop Running: Energy and the Future of Transportation”, 2015, Springer.
Many authors have been writing for years about why Jacobson and Delucchi’s (J & D) plans for a 100% low-cost renewable energy is a cloud cuckoo-land fantasy (references below). But never so many, so loudly, and in such a prestigious journal (Clack 2017).
The 21 authors of the PNAS article felt compelled to write this because J & D’s irresponsible fairy tales are starting to influence actual policy and waste money. If cities and states set renewable goals of 100% and try to achieve them with the J & D plan, their spending will be wasted because the J & D plan leaves out biofuels, grid-scale battery storage, nuclear, and coal energy with CCS.
The most important problems with achieving a 100% renewable system are not even mentioned (Friedemann 2015c).
Renewable contraptions cannot outlast finite fossil fuels, because they are utterly dependent on fossil fuels from birth to death to mine, crush, and smelt the ore, deliver the ore to a blast furnace, fabricate 8,000 wind turbine parts at hundreds of manufacturing plants all over the world, and deliver the parts to the assembly plant. For each turbine, dozens of trucks are needed to prepare the wind turbine site so that dozens of cement trucks can pour tons of concrete and steel rebar for the platform, deliver pieces of the huge parts of the turbine, and diesel powered cranes to lift the parts hundreds of feet into the air.
In their 2011 paper, the J & D 100% renewable system would be accomplished with 3.8 million 5-MW wind turbines (50% of power), 49,000 solar thermal plants (20%), 40,000 solar PV plants (14%), 1.7 billion rooftop PV systems (6%), 5350 geothermal plants (4%), 900 hydroelectric power plants (4%), and marine hydrokinetic devices (2%). Their 2015 paper has somewhat different but equally unrealistic numbers.
It is questionable whether there’s enough material on earth to build all these contraptions and continue to do so every 20 years (wind) to 30 years (solar). Fossil fuels will grow more and more scarce, which means cement, steel, rare (earth) metals, and so on will decline as well. Keep in mind that a 2 MW turbine uses 1,671 tons of material: 1300 tons concrete, 295 tons steel, 48 tons iron, 24 tons fiberglass, 4 tons copper, .4 tons neodymium, .065 tons dysprosium (Guezuraga 2012, USGS 2011). The enormous demand for materials would likely drive prices up, and the use of recycled metals cannot be assumed, since downcycling degrades steel, perhaps to less strength than required.
The PNAS authors propose grid-scale batteries, but the only kind of battery for which there are enough materials on earth are Sodium-sulfur NaS batteries (Barnhart 2013). To store just one day of U.S. electricity generation (and at least 6 to 8 weeks would be needed to cope with the seasonal nature of wind and solar), you would need a 923 square mile, 450 million ton, $40.77 trillion dollar NaS battery that needs replacement every 15 years (DOE/EPRI 2013). Lead-acid: $8.3 trillion, 271.5 square miles, 15.8 million tons. Li-ion $11.9 trillion, 345 square miles, 74 million tons.
There are dozens of reasons why wind power will not outlast fossil fuels (Friedemann 2015b), including the scale required, the need to increase installation rates 37-fold in 13 years (Radford 2016), population increasing faster than wind turbines to provide for their needs can be built, wind is seasonal – very little in the entire U.S. in the summer, no commercial wind year round in the South East, a national grid, no commercial energy storage at utility scale in sight, plus a financial crisis or war will likely break the supply chains as companies go out of business.
Since a major factor in the Jacobson and Delucchi paper is that we can use Underground Thermal Energy Storage (UTES), here is my critique of that in my book “When Trucks Stop running”:
Concentrated Solar Power (CSP) with Underground Thermal Energy Storage (UTES)
You won’t find a CSP plant on your neighbor’s roof. CSP is a large power plant requiring considerable acreage—5.5 square miles for California’s Ivanpah—where mirrors focus bounties of sunlight to boil water for steam generation. CSP contributes only 0.06 % of U.S. electricity. The United States has 1861 MW of CSP operating or under construction, mainly in California (64 %) and Arizona (24 %) because extremely dry areas with no humidity, haze, or pollutants are required. Of the 1861 MW, only about one-quarter can also store electricity using thermal energy storage. Energy is stored as heat, usually in molten salt, with total CSP storage rated at 510 MW. CSP is more capital expensive than any other power generation plant except nuclear. Eight plants cost a total of $9 billion (Solana, Genesis, Mojave, Ivanpah, Rice, Martin, Nevada solar 1, Crescent Dunes (NREL 2013). Almost all CSP plants also have fossil backup to diminish night thermal losses, prevent molten salt from freezing, supplement low solar irradiance in the winter, and for fast starts in the morning. You can’t hurry a sunrise. CSP electricity generation in winter is significantly less than other seasons, even in the best range of latitudes between 15° and 35° (Fig. 17.4). To provide seasonal storage, CSP plants would need to use stone, which is much cheaper than molten salt. A 100 MW facility would need 5.1 million tons of rock taking up 2 million cubic meters (Welle 2010). Since stone is a poor heat conductor, the thick insulating walls required might make this unaffordable (IEA 2011b). Nevada’s 110 MW Crescent Dunes opened in 2015 with 10 hours of storage and is expected to provide an average of 0.001329 Twh a day. Multiply that by 8265 more Crescent Dune scale plants and presto, we’ll have one day of U.S. electrical storage (11.12/0.001329 TWh). Without storage, solar CSP and solar PV do nothing to keep the grid stable or meet the peak morning and late afternoon demand.
Okay, drum roll. The biggest problem is that electricity does not matter. This is a liquid transportation fuels crisis. Trucks can’t run on electricity ( http://energyskeptic.com/category/fastcrash/electric-trucks-impossible/ ).
The Achilles heel of civilization is our dependency on trucks that run on diesel because it is so energy dense. This is why diesel engines are far more powerful than steam, gasoline, electric, battery-driven or any other motive power on earth (Smil 2010). Billions of trucks and equipment worth trillions of dollars are required to keep the supply chains going over tens of millions of miles of roads, rail, and waterways that every person and business on earth depends on. Equally if not more important are off-road mining, agriculture, construction, logging, and other trucks. They not only need to travel on rough ground, but meanwhile push, lift, dig and perform other tasks far from the electric grid or non-oil distribution system.
Trucks must eventually be electrified, because biomass doesn’t scale up and has negative or break-even energy return, coal and natural gas are finite, and hydrogen /hydrogen fuel cells are dependent on a non-existent distribution system and far from commercial. In my book, I show why trucks can’t run on electricity, as well as why a 100% renewable grid is impossible.
The authors briefly point out that one way to counter wind and solar intermittency is an energy source that can be dispatched when needed. But they neglected to mention that natural gas plays most of this role now. But natural gas is finite, and has equally important uses of making fertilizer, feedstock and energy source to make hundreds of millions of chemicals, heating homes and buildings, and so on. All of these roles will have to be taken on by biomass after fossils are gone, yet another reason why biomass doesn’t scale up.
J & D propose a month of hydrogen storage to power transportation. But hydrogen boils off within a week since it is the smallest element and can escape through atomic scale imperfections. It is not an energy source, it’s an energy sink from start to finish. First it takes a tremendous amount of energy to split hydrogen from oxygen. That’s why 96% of hydrogen comes from finite natural gas. And a tremendous amount more energy to compress or liquefy it to -423 F and keep it chilled. It is so destructive of metal that expensive alloys are needed for the steel pipelines and storage containers, making a distribution system too expensive. A $1.3 million dollar hydrogen fuel cell truck would require a very heavy and inefficient fuel cell with an overall efficiency of just 24.7%: 84% NG upstream and liquefaction * 67% H2 on-board reforming * 54% fuel cell efficiency * 84% electric motor and drivetrain efficiency * 97% aero & rolling resistance efficiency, and even less than that without an expensive 25 kWh li-ion battery to capture regenerative braking (DOE 2011, Friedemann 2016). And far less than 24.7% efficient if the hydrogen were made from water with electrolysis.
J & D propose thermal energy storage in the ground. The only renewable that has storage are concentrated solar plants, but CSP plants provide just 0.06% of U.S. energy because each plant costs about a billion dollars each (and less than a quarter of them have storge). Scaled up, CSP would need to use stone, which is much cheaper than molten salt. A 100 MW facility would need 5.1 million tons of rock taking up 2 million cubic meters (Welle 2010). Since stone is a poor heat conductor, the thick insulating walls required might make this unaffordable (IEA 2011b). J & D never mention insulating walls, let alone the energy and cost of building them. The PNAS paper also says that phase-change material energy storage is far from commercial and still has serious problems to solve such as poor thermal conductivity, corrosion, material degradation, thermal stress durability, and cost-effective mass production methods.
The PNAS authors suggest bioenergy, but this is not feasible. The billions of diesel engines in trucks and equipment can’t burn ethanol, diesohol, or even gasoline. Most engine warranties don’t allow biodiesel, or up to 20% at most. Biofuels (and industrial agriculture) destroy topsoil, which in the past was the main or a major reason why all past civilizations failed. Industrial farming also depletes aquifers that won’t be recharged until after the next ice age. As I mentioned earlier, biomass simply doesn’t scale up. Burning it is far more energy efficient than the dozens of steps needed to make biofuels, each step of which takes energy (into a negative energy return if the boundaries are wide). Yet even if we burned every plant plus and their roots in America, the energy produced would be less than the fossil fuel energy consumed that year, and we’d all have to pretend we liked living on Mars for many years after our little experiment. Friedemann (2015a) has many other examples of the scaling up issues, ecological, energy, and other issues with biofuels.
Nuclear is not an option due to peak uranium, and the findings of the National Academy of Sciences about lessons learned from Fukushima. It’s also too expensive, with 37 plants likely to shut down (Cooper 2013). And leaving thousands of sites with nuclear waste lasting hundreds of thousands of years for our descendants to deal with after fossil fuels are gone in an industrially poisoned world is simply the most evil of all the horrible things we’re doing to the planet (Alley 2013). And since trucks can’t run on electricity, there’s no point in building them.
The book “Our renewable future” (Heinberg & Fridley 2016) was written to show those who believe in Jacobson and Delucchi’s fairy tales how difficult, if not impossible it would be to make this happen. Though I fear many of their major points were probably ignored or forgotten, with readers deciding that 100% renewables were possible, even if difficult, since the book was too gentle and abstract. For example, they mention that there are no ways to make cement and steel with electricity, because these industries depend on huge blast furnaces that run for 4 to 10 years non-stop because any interruption would cause the brick lining to cool down and damage it. It is not likely a 100% wind and solar electricity system to be up 24 x 7 x 365. That’s a real showstopper. But the average person believes in infinite human ingenuity assumes that an electric solution can be found, even if it has to overcome the laws of physics…
J & D include wave and tidal devices, but these are far from being commercial and unlikely to ever be due to salt corrosion, storm waves, and dozens of other problems (NRC 2013).
I’m not as concerned about the incorrect J & D calculations for GHG emissions, because we are at or near peak oil and coal, and natural gas. Many scientists have published peer-reviewed papers that based on realistic reserves of fossil fuels, rather than the unlimited amounts of fossils the IPCC assumes, and there is a consensus that the worst case scenario likely to be reached is RPC 4.5 (Brecha 2008, Capellan-Perez 2016, Chiari 2011, Dale 2012, Doose 2004, Hook 2010, Hook 2013, and 10+ more).
The PNAS authors mention of coal with carbon capture and storage (CCS) won’t work. First, coal is finite (and probably peaking globally now or soon), and carbon capture and storage technology so far from being commercial, and uses up 30 to 40% of the energy contained in the coal, that it’s unlikely to be used when blackouts start to happen more and more often (http://energyskeptic.com/category/energy/coal/carbonstorage/).
We’re running out of time. Conventional oil peaked in 2005. That’s where 90% of our oil comes from at a Niagra Falls rate. Tar sands and other non-conventional oil simply can’t be produced at such a high rate. So it doesn’t matter how much there is, Niagra Falls will slow to a trickle, far less than what we use today. And since energy is the basis of growth, not money, it is questionable if our credit/debit system can survive, since once peak oil is acknowledged, creditors will know they can’t be repaid.
Also, oil is the master resource that makes all other resources available. We don’t have enough time to replace billions of diesel engines with something else. There is nothing else. And 12 years after peak the public is still buying gas guzzlers.
The main PNAS criticisms of J & D are:
- J & D used invalid modeling tools, had modeling errors, inappropriate methods, and implausible and inadequate assumptions.
- Claimed that a 100% renewable system would be low cost and exceed current electric-utility reliability standards
- Their portfolio of options of wind, water, and solar, with no coal, natural gas, bioenergy, or nuclear power is not broad enough.
- Wind and solar are variable, so energy storage is essential. But “there are no electric storage systems available today that can affordably and dependably store the vast amounts of energy needed over weeks to reliably satisfy demand using expanded wind and solar power generation alone”.
- Parts of the economy are difficult to electrify: airplanes, cement manufacture, etc.
- Their solutions include technologies that have not been commercially proven at scale, can provide adequate and reliable energy; be built rapidly enough, and not violate environmental regulations.
- Their papers include innovations that don’t exist: hydrogen-powered airplanes and steel, multi-week energy storage systems with a capacity twice the U.S. generating and storage capacity today, underground thermal energy storage (UTES) systems in nearly every community to provide services for every home, business, office building, hospital, school, and factory, yet doesn’t account for the pipes and distribution lines.
- They vastly underestimate the cost and environmental impact of expanding hydroelectric dams, scaling up hydrogen production, or a national grid.
- J & D assume we can store 1 month of U.S. electricity in hydrogen by using twice as much energy-generating capacity as we have now.
- J&D assume that 63% of industries are flexible and can reschedule all energy needs with an 8 hour window of time. My comment: That’s simply not true, many industries can’t be rapidly curtailed, and there are many products made with continuous processes around the clock (refineries, chemical plants, blast furnaces, etc).
- J & D assume the capital cost of building all of this renewable energy at 30 to 50% of what most other studies assume.
- Hydropower is not a dependable source for always available (dispatchable) power, due to droughts and maintaining a large reservoir to provide water to cities, farms, and fish. Plus there are very few places left to put (pumped) hydropower dams physically and won’t cause ecological harm.
- Underground Thermal Energy Storage (UTES) is a central requirement of their vision because energy storage is essential once natural gas is gone. But UTES is far from commercial with just two small-scale demonstration projects for about 300 homes. But only to heat them, yet J & D propose to provide most air-conditioning and half of refrigeration with this (and ice-based systems). On top of that UTES requires energy for heat pumps but they don’t model this energy requirement. They don’t provide any reliable figures for how much this would cost, but estimate $37 to $900 billion. Yet the Drake Landing system costs if scaled up would cost at least $1.8 trillion dollars and leaves out the heating and cooling systems of homes and businesses in new homes. Retrofits are very costly. Finally the performance and cost depends on the thermal properties of the soil and total absence of groundwater (which removes the stored heat).
Other reasons listed in the PNAS paper:
- Their proposal would require 6% of the continental U.S. for wind turbines, and 100,000 square kilometers to install large-scale centralized solar PV and CSP system (an area the size of Kentucky).
- 150,000 5 MW turbines would be built offshore
- lack of electric power system modeling of transmission, reserve margins, and frequency response
- the climate/weather model used for estimates of wind and solar energy production has not shown the ability to accurately simulate wind speeds or solar insolation at the scales needed to assure the technical reliability of an energy system relying so heavily on intermittent energy sources
- their numbers in the supporting information of Jacobson (2015a) imply that maximum output from hydroelectric facilities cannot exceed 145.26 GW, which is 50% more than exists in the United States today, but figure 4B of shows hydroelectric output exceeding 1,300 GW
- There are conflicting and exaggerated figures for the amount of flexible load
- They don’t explain how we can provide an extreme excess of high power for short periods of time to industrial, commercial, and residences
- Germany is the most committed of any nation to achieving an 80% renewable energy system and made a huge effort from 2007-2014, which J & D propose we can do 14 times faster than the U.S. average for 55 years and 6 times the peak rate
- The need of AC grids to cope with power flows and need for a constant frequency. It’s questionable whether the radical changes to grid architecture required by J & D is feasible, and they don’t even attempt to model or analyze transmission capacity , power flow, transmission constraints, operating reserves, logistics, frequency regulation, or operation reliability.
- They assume perfect predictions of electricity demand and variable wind and solar so they don’t bother to calculate what the reserve requirements would be if their crystal ball isn’t always right, let alone how a drastically expanded grid with a new architecture could cope
The most devastating part of the 21-scientist PNAS critique is in the technical supplement that finds errors line by line (here).
Other articles that critique Jacobson and Delucchi:
- Smil, Vaclav. 2010. Energy Myths and realities. Bringing science to the energy policy debate. AEI Press.
- Dodge, E. November 17, 2013. Critique of the 100 Percent Renewable Energy for New York Plan. 100% Renewable Energy for New York State from Wind, Water and Solar. Really? Energy Collective.
- Trainer, T. 2013. A Critique of Jacobson and Delucchi’s Proposals for a World Renewable Energy Supply. org
- Friedemann, A. 2015. Not enough wind, solar, geothermal to replace fossil and nuclear power in the 11 western states of the WECC. California, Oregon, Utah, and Washington have already developed most of their prime-quality in-state resources. energyskeptic.com
- Lopez, A., et al. 2012. S. Renewable Energy Technical Potentials: A GIS-Based Analysis. National Renewable Energy Laboratory.
- Revkin, A. C. June 18, 2013. A Reality Check on a Plan for a Swift Post-Fossil Path for New York. New York Times
Best articles about the PNAS paper
- Bailey, R. 2017. Powering U.S. Using 100 Percent Renewable Energy Is a Total Fantasy. reason.com
- Fairley, P. 2017. Can the U.S. Grid Work With 100% Renewables? There’s a Scientific Fight Brewing. IEEE Spectrum.
- Porter, E. 2017. Fisticuffs over the route to a Clean-energy future. New York Times.
- Alley, W.M. 2013. Too Hot to Touch: The Problem of High-Level Nuclear Waste. Cambridge University Press.
- Barnhart, C., et al. 2013. On the importance of reducing the energetic and material demands of electrical energy storage. Energy Environment Science 2013(6): 1083–1092.
- Brecha, R. J. 2008. Emission scenarios in the face of fossil-fuel peaking, Energy Policy, 36: 3492–3504.
- Capellan-Perez, I., et al. 2016. Likelihood of climate change pathways under uncertainty on fossil fuel resources availability. Energy Environ. Sci 9: 2482-2496.
- Chiari, L., et al. 2011. Constraints of fossil fuels depletion on global warming projections. Energy Policy 39: 5026-5034.
- Clack, C. T. M., et al. 2017. Evaluation of a proposal for reliable low-cost grid power with 100% wind, water, and solar. Proceedings of the National Academy of Sciences.
- Cooper, M. 2013. Renaissance in reverse: competition pushes aging U.S. nuclear reactors to the brink of economic abandonment. Institute for Energy and the environment, Vermont Law School.
- Dale, M. 2012. Meta-analysis of non-renewable energy resource estimates. Energy Policy 43: 102-122.
- DOE. 2011. Advanced technologies for high efficiency clean vehicles. Vehicle Technologies Program. Washington DC: United States Department of Energy.
- DOE/EPRI. 2013. Electricity storage handbook in collaboration with NRECA. USA: Sandia National Laboratories and Electric Power Research Institute.
- Doose, P. 2004. Projections of fossil fuel use and future atmospheric CO2 concentrations, The Geochemical Society Special Publications 9: 187–195.
- Friedemann, A. 2015a. Peak soil: Why biofuels destroy ecosystems and civilizations. energyskeptic.com
- Friedemann, A. 2015b. Dozens of reasons why wind power will not outlast fossil fuels. energyskeptic.com
- Friedemann, A. 2015c. When Trucks Stop Running: Energy and the future of Transportation. Springer.
- Friedemann, A. 2016. Heavy-duty hydrogen fuel cell trucks a waste of energy and money. energyskeptic.com
- Guezuraga, B. 2012. Life cycle assessment of two different 2 MW class wind turbines. Renewable Energy 37:37-44.
- Heinberg, R. and Fridley, D. 2016. Our Renewable Future: Laying the Path for One Hundred Percent Clean Energy. Island Press.
- Hirsch, R. L., et al. February 2005. Peaking of World Oil Production: Impacts, mitigation, & risk management. Department of Energy.
- Hook, M., et al. 2010. Validity of the Fossil Fuel Production Outlooks in the IPCC Emission Scenarios, Natural Resources Res., 19: 63–81.
- Hook, M., et al 2013. Depletion of fossil fuels and anthropogenic climate change – a review. Energy Policy 52: 797-809.
- Jacobson MZ et al. 2015a. Low-cost solution to the grid reliability problem with 100% penetration of intermittent wind, water, and solar for all purposes. Proc Natl Acad Sci USA 112:15060–15065.
- Jacobson MZ, et al. 2015b. 100% clean and renewable wind, water, and sunlight (WWS) all-sector energy roadmaps for the 50 United States. Energy Environ Sci 8:2093–2117.
- NRC. 2013. An Evaluation of the U.S. Department of Energy’s Marine and Hydrokinetic Resource Assessments. National Research Council.
- Smil, Vaclav. 2010. Prime Movers of Globalization: The History and Impact of Diesel Engines and Gas Turbines. MIT Press.
- USGS. 2011. Wind Energy in the United States and Materials Required for the Land-Based Wind Turbine Industry From 2010 Through 2030. U.S. Geological Survey.