Icarus’ Legacy — Episode 4 — Solar and Wind Energy
(Google Doc version more suitable for PC and Tablet is here)
In the previous episode
Hydropower and geothermal energy were the earliest forms of renewable sources used for electricity generation, and we started to harness them over a century ago.
With hydro much more impactful than geothermal, we see them evolving and continuing to support the energy transition, but we don’t expect them to increase their weight in the overall energy mix.
This is because they are only available at scale, as a source of electricity generation, in specific regions: hydropower requires major waterways, and geothermal energy requires significant volcanic activity. The ongoing research could make these sources applicable and cost effective in an extended range of cases and regions.
Pumped storage, a variation on hydropower which requires large water basins at different elevations, but no major waterways, is used as a massive energy storage, and with the introduction of intermittent energy sources like solar and wind, we will see an increase in its importance to balance the power grid load.
Episode 4 — Solar and Wind Energy
“We cannot direct the wind, but we can adjust the sails” Dolly Parton
In this episode
We explore Solar and Wind energy, the most important of the so-called modern renewable sources, which over the past decade have contributed to the growth of renewable electricity in pretty much every part of the globe. The massive improvements in technology and the large reduction in production costs have made them competitive vs. fossil fuels in electricity generation. So much that 90% of all new installed power generation capacity of 2020 was either solar or wind energy.
Can Solar and Wind energy be the ultimate solutions for our energy transformation? Unfortunately not. They can play an important role, and they are among the best tools we currently have, but they are far from definitive, so we will need more solutions, innovations, breakthroughs.
And unfortunately, they are also far from perfect. As with all complex problems in our world, there’s never just a straight answer and we must get used to nuanced arguments. Solar and wind need to resolve a number of issues in their production, that otherwise risk creating damage to some of the issues they are trying to resolve.
Let’s see where we stand.
Modern Renewable Energy Sources
From promise to reality, in 10 years
In the early 2000s many governments around the world started to fund schemes that would get modern renewables from concept to large scale implementations.
At that time, the cost of energy from solar and wind was prohibitively expensive, so it was only through generous subsidies (and a lot of scepticism) that these new industries were effectively kick-started.
The drivers were both ecological as well as strategic: reducing pollution and emissions by producing green electricity, and reducing dependencies from fossil fuel producers towards energy security and independence.
And it worked! Government subsidies led private companies and individuals to good investment decisions, which in turn led to large scale deployments, technology improvements and reduction in prices (solar -82%, onshore wind -40% in cost of electricity generation in 10 years), in a virtuous cycle that is still ongoing, and keeps delivering better value and increased renewable energy generation.
The time of subsidies is now practically over for wind and solar, which regularly come first in auctions for new power generation. In 2020 solar and wind combined contributed to 9% of global electricity production. In 2010 this was only 1.8%. With the nuclear share slightly reducing, and hydro remaining stable over this period of time, the growth in solar and wind pushed low-carbon sources, for the first time in history, to overtake coal in global electricity generation (40% vs. 34%).
At the end of 2019, China and the US owned nearly 50% of the global installed solar and wind capacity, with over 580 GW.
But other countries in Europe (Germany, UK, Spain and Italy) and in Asia (India and Japan) invested significantly and boasted over 30 GW of combined solar and wind power each. Only 47 countries in total had an installed capacity of over 1 GW, which shows that the big transformation hasn’t started everywhere yet.
The predictions of growth for solar and wind are extraordinary: by 2050 we expect them to represent over 60% of electricity generation (up from current 9%), with electricity at the same time growing to represent over 50% of the overall energy consumption (from current 20%). The overall energy consumption is also expected to grow by 50% as the world population grows and develops.
The 2050 prediction implies a 23-fold increase of solar and wind energy production!
Note that we would expect electricity to be fully low-carbon (renewables and nuclear) while non-electrical access to energy would leverage renewables sources such as biomass and hydrogen and, to a minimal extent, what remains of fossil fuels usage.
From weather forecast to power forecast
Once a location is chosen for a solar plant or a wind farm, their energy production will remain highly dependent on seasonal and weather factors.
Solar energy, over long periods of time (months/years) tends to be relatively predictable and will generate the expected results, but over shorter periods of days/weeks, the variability is still large. Wind is even less predictable. Long studies are carried out to find the most promising locations, but then even over relatively long periods of time, performance may still vary considerably.
This unpredictability plays an important role in two ways: it reduces the capacity factor of plants, and it makes their output wildly variable through the day, like night and day, cloudy and sunny, quiet and blustery, from zero to their maximum capacity.
The following chart shows an example of how the availability changes during the day and over seasons.
The capacity factor (CF) of a plant can be defined as the power that is generated over a period of time, relative to its maximum production potential (the installed capacity).
As an example, a power plant with an installed capacity of 500 MW has a 50% CF on a given day, when it generates 6 GWh of energy (500 MW x 24 hours x 50%).
Wind and Solar tend to be at the bottom of the energy sources for capacity factors.
Energy sources like nuclear, most fossil fuels and even geothermal can get very close to 100% CF for long periods, as they only need to be stopped for brief maintenance activities.
Hydropower CF vary depending on water availability and season, but on a global scale and over full years, they tend to be in the 40–50% range.
Wind farms, typically fall in the 20–40% bracket, over full years, with offshore farms, normally exploiting stronger ocean winds, falling on the higher end of this range.
Solar plants are dependent on the location and typical weather (latitude and insolation) and their CF vary from as little as 10% (e.g. in Northern Europe) to around 20% (e.g. Southern Europe, Southern US States) to as much as 30% (e.g. desertic regions in Northern Africa).
If we look at the actual CF of energy sources in a given country (see US in 2019 in the chart below) we can observe that fossil fuel power plants have a much lower CF than their potential CF, as the grid operators will prioritize nuclear and renewable sources, and only use fossil fuels when required by the grid demand.
The low CF for solar and wind impacts the economics of power plants, as we will need to install larger plants (more solar panels, more wind turbines) to achieve a desired energy output over a period of time.
As an example, if you install solar panels for residential use in a location where the expected CF is 10%, and you want to generate an average of 1 KW of power, you will require 10 KW worth of solar panels. In a year you will generate around 365 x 24h x 1 KW = 8,760 KWh, which is very close to the average consumption of a US household (Oh… and more than twice that of a UK household, and more than three times that of an Italian household — but this is a different issue — we’ll discuss later).
What is remarkable is that, given the very low prices, even regions with around 10% CF like Scotland or part of Canada, find it convenient to install solar energy, as the installation cost payback time is still fairly short.
Power grids don’t like intermittent energy
When the sun doesn’t shine or the wind doesn’t blow, there’s no output from these plants. That’s pretty obvious. But from the perspective of a power grid, that needs to respond to the demand of all sectors of society, this is a massive problem. No one expects power cuts when the sun goes down.
Solar and wind are called intermittent sources, as they can disappear and reappear at variable levels, in no time. If they constitute a small percentage of the grid capacity, it is fairly easy to adapt: the grid will pretty much take whatever it gets from these sources, and use other “more dependable” sources to reach the required load.
In a fairly simplistic way, a power grid has to react to variable demand, typically with peaks in the morning and in the evening, high consumption through the day, lower during the night, and generally low demand during weekends.
Electricity generation (in GW) on a sunny winter day in Italy. We can observe:
- Higher load during the day, with morning and evening peaks
- Large solar and hydro output
- Limited wind and geothermal generation
- Base load provided by thermal energy (primarily natural gas and some coal)
- Thermal and hydro called to step up to cover both morning and evening peaks, and throttle down as soon as the solar increased its output
Planning production and balancing the actual load during the day, normally relies on three types of plants:
- Base load plants: can operate predictably at very high capacity factors, pretty much uninterrupted, and they cannot very easily be throttled to produce more or less. Nuclear and coal plants fall into this category. They can provide a large base for the required electricity of a region or a country, but they are difficult to stop and restart, and a full cycle might take hours or even days, so they cannot be used to follow the grid demand
- Dispatchable plants: these are plants that in addition to possibly operating at full capacity for at least a few hours, they can also operate at reduced capacity (curtailment), or even stop, and change output almost in real time. Hydroelectric and natural gas plants fall into this category. A hydro plant (including pumped hydro) can respond within seconds to a grid “command”
- Intermittent sources: these are like solar and wind, where a base prediction can be made, but then the actual output will vary during the day, so other sources will have to accommodate for the shortfall or the overproduction coming from these plants
The ideal grid would operate with dispatchable plants only, as they can be used for base load, as well as flexible load balancers — their production can be programmed in advance, and can be adjusted during the day. This is reflected in the higher prices paid for dispatchable energy by national grids.
A grid cannot operate efficiently with too many intermittent sources, as it would need to keep many other dispatchable plants active but on standby, such as “gas-peakers”, increasing overall system operational costs, and increasing the probability to waste energy.
Other ways to deal with intermittent non-dispatchable sources are:
- Introducing utility-level energy storage, such as pumped hydro plants. When we produce excess energy we load them, so they can be quickly activated when production is lower than required
- Introducing excess intermittent capacity, so the probability of shortfall is much lower, and excess production can be exported to neighbouring regions/countries
Both methods increase grid complexity and costs.
As an example, the chart below shows the production of wind and solar energy in Italy over a period of 6 years, with monthly observations. Installed capacity only changed marginally over this period (sadly). We can observe the expected summer/winter pattern of solar, while the wind contribution is much more variable.
The intermittent nature of wind and solar plants impacts the economics of these sources: they command lower prices from the grid as they require other plants to be ready to compensate for the periods of low production.
As we will see, there are ways to make solar and wind plants dispatchable, so in the near future this problem could be largely resolved.
Digging For Renewables
Steel, Cement and Plastic, build and recycle
A solar plant or a wind farm will eventually generate zero carbon energy, but this is not the beginning, and it’s not the end of their stories.
Producing these plants requires material that must be manufactured, transported and installed, including steel, cement and plastic (all very carbon intensive to produce, today). This is true, in different measure, for all types of power plants, including nuclear, hydro and fossil fuels.
At the end of their useful life, all these materials should be reused and recycled as much as possible, to avoid further pollution, but this is far from reality today. It is a one-off production that will deliver much larger advantages (in reduced pollution and emissions) typically for 30 years or more, but still, their impact must be accounted for, and as such, reduced to the minimum, if we want to achieve net zero.
A recent study found that it can take 6 months of energy production for an onshore wind turbine to pay back the energy expended in manufacturing, transportation and installation, and a significant amount of the materials used would eventually end in landfills. A solar plant will also require significant construction materials and the energy payback time is also typically 6 months to 1 year.
Not so rare but very dirty earths
Even more importantly, both technologies, require a large amount of so called “rare earths”. These metals are in fact quite common all over the world, but they are present in very limited concentrations, which makes mining them expensive. Rare earths are required in a number of modern applications, from smartphones to electric batteries and, indeed, renewable energy tech.
For instance, neodymium, praseodymium and dysprosium are required in large quantities (tons) to produce the large permanent magnets, which are key to the workings of wind turbines.
At the moment their extraction creates significant concerns and issues:
- Geopolitical: the near-monopoly of China, currently extracting 85% of global rare earths, within its territory or abroad. Significant deposits are present in many other countries, but they have been undercut by China’s low mining and processing costs. Recent “trade wars” have made the supply of rare earths critical and at risk, so consumer countries are revisiting their supply chains
- Environmental: the pollution of soil and water that is generated on-site, due to the complex extraction processes, requiring acids and other chemicals, and the not very stringent environmental standards employed in China and some African countries
- Social: impact on local communities access to basic resources such as water, and the conditions of miners employed for the extraction
For all the materials involved, solar and wind apparatus designs will have to be revisited to improve circularity (reuse, recycle) and reduce the requirement for rare earths. While new designs are ongoing, it is important that end user countries and companies apply strict environmental, social and governance regulations to the full supply chains, and the full production processes, or these issues will water down the advantages of the renewable energy transformation.
Solar photovoltaic (PV) cells similar to the ones we know today, made from silicon, for electricity production, were invented in 1954 at Bell Labs. It took decades to improve efficiency and reduce costs to a level where widespread commercial use was deemed possible.
In the 1970s, as an example, a solar panel could cost around $100 per Watt, while today this price is under $0.50. It wasn’t until the early 2000s that we saw a large deployment of residential installations, and due to high costs, these had to benefit from generous subsidies.
But over the past two decades prices have dropped massively, and today the photovoltaic solar energy is among the cheapest and most available sources at most inhabited latitudes.
Large scale deployments, mass production and improved efficiencies are still contributing to decreasing prices of installation and production.
Solar power is predicted to be the largest source of energy production by 2050, which shouldn’t surprise us, as it is also the most available in our planet — to a massive scale: the sunlight that hits the Earth in one single hour is equivalent to all the energy used by human processes in the world for an entire year (roughly 430 Exajoules or 120,000 TWh).
Most countries and energy suppliers have schemes that allow residential solar power owners to sell excess electricity back to the grid. Also, as the cost of batteries is dropping significantly, residential use is sometimes coupled with batteries for electricity storage, so that the electricity generated during the sunny hours can be used when the sun is not shining.
The power of bottled sunshine
Solar power is also used to generate heat, in concentrated solar power (CSP) plants, that make use of mirrors or lenses (instead of PV panels) to concentrate the solar beams on a specific point, and then operate like a conventional thermal power plant with steam and turbines.
These plants also have the advantage that they can incorporate energy storage in their design, in the form of molten salt, which is heated when the sun is shining, is held in insulated containers, and can release energy when required, even days or weeks later.
This is an example of dispatchable solar energy plant, which is much more valuable for a grid operator, than a simple “intermittent energy” solar plant, as the output can be planned in advance and controlled in real time to meet the load requirements, thanks to its on-site molten salt storage capabilities.
We can achieve the same result by coupling other types of storage (e.g. electric batteries) to a solar photovoltaic plant, and although battery costs are still fairly high, in recent times we have seen the first installations of this type. We will cover energy storage in more detail in a separate episode.
Solar Thermal energy is viable also in the small scale, for domestic heating purposes, by using thermal solar panels (as opposed to photovoltaic ones) to heat water.
In recent times special dual-purpose solar panels have been produced that offer both photovoltaic (electricity production) and thermal (heating) functionality.
When coupled with a heat pump for all-seasons heating and cooling, and electric batteries, to store and balance the output over one or more days, these can basically cover all the domestic energy uses, and reduce residential energy-related emissions to zero.
As someone said, wind power has been used as long as humans have put sails into the wind.
Hero of Alexandria, e Greco-Egyptian mathematician and engineer, is credited to have designed the first earliest machine powered by a windwheel — the organ on the picture.
Wind power is one of the fastest growing source of energy. Wind turbines and electric generators were both known and experimented already in the late 19th century, but it took until the 1990s to see their usage scale to utility level.
Farms can be installed onshore and offshore, with projects that can take a number of years, given their complexity. While the installation alone might only take a few months, these projects spend years in feasibility studies, planning, consultations, applications and authorizations, before they can order the turbines, prepare the site and begin installation.
In comparison, it’s simpler to execute solar farm projects as they have fewer uncertainties: solar power can be predicted with good accuracy, given the location identified, while winds are much less predictable. There’s also no impact on aviation and less impact on the landscape. Both energies are improving their density in MW/km2, as technologies advance.
Over the past few years, the development of offshore wind farms accelerated, due to the potential of remote ocean areas (with strong, predictable winds) and to the reduced disruption on the environment. Large manufacturers have also started to deploy floating turbines, which allow the building of wind farms on deep ocean areas (normal offshore turbines cannot be anchored in depths of over 50 meters) and also it will allow for their displacement, in case the originally selected positioning reveals suboptimal.
The output of a wind turbine depends on the size of the turbine and the length of its blades, and it is proportional to the cube of the wind speed, so doubling the wind speed will generate roughly eight times the energy. Scaling to utility level required scaling individual turbines: the typical capacity of a large turbine in the 1980s was 50 KW, while today we achieve over 10 MW (200 times larger) with turbines using blades more than 200 meters long. A single rotation can power an average US home for a whole year.
Development and scaling has been so rapid that today there’s a growing business of repowering existing wind farms, by replacing turbines with much larger, powerful and efficient ones.
Onshore wind farms have sometimes received opposition from local communities due to the visual impact on the landscape, but they are becoming more and more common. They are generally cheaper than the offshore versions, due to simpler installation and operation, but finding onshore areas with good winds and reduced landscape impact is generally problematic.
As for solar plants, a key improvement will be the ability to couple wind farms with energy storage, so they can be treated as dispatchable energy plants, and provide a much better service to the grid: continue to generate the expected energy without waste, even when there’s excess or insufficient production.
As winds tend to be less predictable than solar energy, the size of the on-site energy storage would be even more critical, and potentially increase wind farm costs significantly. Electric batteries, compressed air or hydrogen are among the possible energy storage solutions that we will discuss in a separate episode.
Wind and solar energy are key to the energy transformation towards clean and renewable sources. They have shown, over the past 10 years, that they can reliably deliver at competitive costs in large scale deployments, and they are accessible to most countries. They are the best tools we currently have to deliver increased electricity output, as the transport, residential and industrial sectors reduce their use of fossil fuels in favour of electric vehicles, heat pumps and other electric appliances and processes.
Most of the current investments in renewable energy go towards new wind farms and solar plants, as most countries implement their plans towards “net zero”.
Solar and wind are also to a certain extent complementary, in terms of seasonality, timing of energy output, and geographical availability. However, while over long periods of time the output and returns from these plants are fairly predictable, their short term predictability is limited, so to become predominant in future electricity grids, they will also require suitable utility-scale or on-site storage capability.
A number of improvements will need to be introduced in the design and processes for the sourcing of materials, manufacturing and installation of these plants, to reduce impact on the environment and society in general.
While we continue to deploy solar and wind technologies, we will also need to look at the other tools at our disposal, which are not necessarily receiving the necessary attention, or that are not ready for prime time. In the following episodes we will look at modern biomass, nuclear energy and hydrogen as other important sources for the energy transition.