More doom-saying.

That article (The one I’m responding to, see link above) seems to demonstrate a clear failure to understand the exponential nature of technology growth. Granted, such growth can’t go on forever, but both solar power and battery storage are on a fast growth curve, doubling world deployment roughly every 2–3 years.

Of course, China is pretty far back on the growth curve. According to Wiki:

Of the 6142.5 TWh electricity produced in China in 2016 [11], only 66 TWh was generated by solar power, which is 1.07 % of total electricity production. [12]

Solar PV deployment in China. Source Wiki (above).

Now, there’s an interesting relationship between the amount of deployed technology, and the cost of deploying more. It’s called “Wright’s Law”, and states that, usually (i.e. absent other factors), the cost of a technology will decline by a particular fraction when its deployment increases by different fraction. For instance, when the deployment doubles, cost will typically drop to 75–90% of your starting point. (The exact number depends on the technology.)

My go-to paper on this subject is usually Statistical Basis for Predicting Technological Progress by Nagy, Farmer, Bui, and Trancik, (February 28, 2013 https://doi.org/10.1371/journal.pone.0052669). However, a 2015 report from the Fraunhofer-Institute for Solar Energy Systems offers some hard historical numbers, specifically that a doubling of deployment will see costs, on average, reduce to 79% of the starting point.

Now, Wright’s “Law” should probably be applied to global deployment, but I’m going to use the China numbers, mostly because those of the US and Europe have been distorted by punitive tariffs. Using the chart above, if we go back to 2008 (8 years), we get an annual increase of about 2.2 (220%) in deployed PV. If we go back 5 years, it’s 1.8 (180%).

I’m going to average those, and assume that deployment in China will roughly double every year going forward. Let’s see what that gives us for 2025:

66 TWh doubled nine times (2016–2025) gives us 33,792 TWh, 5.5 times the total for 2016. Before we consider how that energy will be used, let’s look at the cost: 79% × 79% ×79% ×79% ×79% ×79% ×79% ×79% ×79% = about 12% of current costs.

Before you reject this as “impossible”, let me show you a chart from that same 2015 report linked above:

From here. See source for more detail.

Note that they provide a “critical cost range where material costs clearly dominate”. The primary material for standard crystalline silicon PV is ultra-pure silicon.

Note that this is cost, not amount. There’s a common myth that “Moore’s Law” “can’t work for PV because there’s no way to miniaturize transistors.” This is certainly how the cost of information processing has been brought down over the last few decades, but Nagy et al. (linked above) shows that the same rule applies to many technologies that don’t require miniaturization.

There are a number of new technologies with potential to reduce the cost of ultra-pure, mono-crystalline silicon, such as float zone, one based on ball milling, etc.

There are also technologies to reduce the thickness of the silicon and eliminate waste from sawing, such as the new process developed by researchers at the Fraunhofer Institute for Solar Energy Systems in Freiburg, which uses vapor deposition, and the one I’ve been watching for a couple years, Rayton Solar. (I’m still not absolutely sure it’s for real, so do your due diligence before investing. I’ve seen nothing about having solved the SF6 issue I noticed in 2015.)

Putting all these together, I see no reason why the cost of 1-junction monocrystalline PV shouldn’t keep dropping until it’s in the same range with sheet plastic.

I’m not going to discuss batteries, I’m much less sure how long Lithium-ion will continue its price drop (which could be impacted by availability issues), and similar is true for vanadium-based flow batteries.

Instead, I’m going to look at power-to-gas (P2G), specifically methane. Solar power, which is very intermittent, can be converted to hydrogen (H2) by electrolysis. This H2 can be combined with CO2 extracted from the air or sea surface (they’re roughly equivalent for environmental purposes)to produce methane, using the Sabatier reaction. This methane can, in turn, be fed into existing natural gas infrastructure, including transport, storage, and electricity generation.

One of the most important advantages of this is that the entire gas infrastructure is no longer stranded assets. (Well, I guess fracking would be.) Which means that they can begin massive investment without worrying about losing it after only a decade or two.

Current estimates of the energy efficiency of this reaction (per Wiki) are 30%-38% from PV bus to CCGT bus. (CCGT stands for Combined Cycle Gas Turbine, the most efficient mature type of gas generation.) Given that the estimates given by Wiki aren’t stated to include costs of extracting ambient CO2, I’m going to assume the lower value, 30%, for long-term, and bring it down to 25% for pre-2030 purposes.

But the cost per watt of solar power, which is roughly equivalent to gas today, will be down to 25% of its current cost in only 6 years (79% × 79% ×79% ×79% ×79% ×79% = 24.308745552%). Which means by 2022 it might actually be cheaper to install solar panels and P2G equipment to create gas for power than to take it out of the ground.

Of course, this would depend on the relative costs of the P2G equipment vs. other ways to get gas. And at least part of it would have to be overcapitalized so it would be able to use the solar energy available only during the day. (My guess is that the H2 will be stored locally, allowing the electrolysis and downstream processes to run full time. But that’s just a guess.)

So here’s how China could deal with their fossil fuel issues:

  1. Begin an immediate conversion from oil to natural (or bio-) gas. This could extend to all fixed power (CCGT can be made fuel-flexible at a small overcharge, so the shift from oil could be leisurely) as well as larger vehicles such as trucks, railroad, and cargo vessels. This deals with the short-term “peak oil”.
  2. Develop ways to force-feed the market for ambient CO2 extraction, such as the out-of-the-box idea proposed by Professor Myles Allen, which I’ve discussed here. This is hardly the only approach, but it has priority.
  3. Develop ways to force-feed the market for “green” gas, such as requiring a fraction of all gas burned to be “green”. This could start at a small number, such as 0.1% (1/10th of one %) and double every three years until, after 30 years, it reaches 100%. This could drive the P2G market, while declining solar energy costs drive the supply.
  4. Develop a plan to transfer off of coal as the exponentially growing volume of solar-based power comes to replace it.

All this, then, could help to solve their oil problem by 2021 or so, and their gas problem by 2030. These exponential growth curves will make planning much harder, but more likely to work out than what the US and EU are doing by ignoring them.

Of course, there’s also the issue of what they’re going to do with all that solar energy. As they install it, the amount of power it generates will become a steadily larger fraction of what they need, until at some point they won’t need all of it to meet demand.

At first, redundant solar and CCGT will work fine, because the capital cost of CCGT is low, and most of the cost of electricity comes from fuel (and variable plant costs). CCGT plants can pay for themselves by producing power when solar can’t, and solar can pay for itself by reduced fuel costs.

This works until solar capacity is roughly equal to CCGT. After that, pumped hydro storage will be able to help:

With the required policy frameworks now in place, the new plan focuses on increasing pumped storage capacity, with its total volume representing just 1.5 per cent of China’s installed electricity capacity at the beginning of 2016. In order to address this shortage, the country aims to reach 40 GW total pumped storage capacity by 2020.

Implementation is well under way. In 2016, China commissioned three pumped storage projects totalling 3.66 GW — Xianju (1,500 MW), Hongping (1,200 MW) and Qingyuan (960 MW). In addition, the first batch of units at the 1,500 MW Liyang project came online in August 2016, and the project in on schedule for completion in April 2017. Furthermore, over 30 GW of pumped storage capacity was under development in China at the end of 2016.

Much of the lake area behind these dams can probably be used for floating solar PV, such as the facility located in the city of Huainan, in China’s eastern Anhui province that they just switched on.

Floating PV runs cooler and cheaper than ground-mount. In principle, the support structure can be little more than the plastic foam shipping mount inside the box of your new PC. It also reduces the need for new transmission, as unneeded solar power can simply pump water back uphill for later use.

In addition, solar power can be used to extract ambient CO2 (for sequestration), and produce H2 using electrolysis. By tuning the required fraction of “green gas” and CO2 sequestration, their government can probably keep the cost of energy low while growing the market for solar energy to drive price reductions.

There are other uses for solar power, including pumping irrigation water and desalination. Both of these applications could probably be set up so they don’t even need inverters, using DC pumps and pressurizers.

Using solar pumping (and, where necessary, desalination) water can be made available for areas where “high” average exposure to water stress over its shale oil and gas area might impact gas production.

It will also allow huge areas to be brought under cultivation, breaking the dependence between farming and oil/gas extraction.

So will China follow a plan similar to this? Probably. It’s clear to me they’ve understood the basics of Wright’s “Law” since at least 2012, perhaps as far back as 2008. Their current plans seem built on exponential assumptions.

And what about the US and EU? They’ve responded to low Chinese PV prices with tariffs and other protectionism. This won’t help, it’ll just keep essentially non-competitive companies in business. If the US (and EU) wants to catch up with China, first they’ve got to admit how far behind they are.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store