ANU team cracks solar thermal efficiency of 97% – a world record

ANU team cracks solar thermal efficiency of 97% – a world record

Print Friendly, PDF & Email

ANU scientists achieve record solar thermal efficiency of 97% – a breakthrough that could cut CST electricity costs by 10%.

Print Friendly, PDF & Email

A team of Australian National University scientists has brought economically competitive, grid-scale solar thermal energy generation closer to reality, after achieving a new record in efficiency for the technology that could compete with the cost of electricity from fossil fuels.

The ANU solar thermal dish. Image: Stuart Hay, ANU

The ANU team, whose CST technology harnesses the power of the sun using a 500 square meter solar concentrator dish, made the breakthrough by redesigning the system’s receiver in a way that halved its convection losses and boosted its conversion of sunlight into steam from 93 per cent to 97 per cent.

According to the ANU’s Dr John Pye, the new design could result in a 10 per cent reduction in the cost of solar thermal electricity.

“Ultimately the work in this project is all about reducing the cost of concentrating solar thermal energy,” he said. “Our aim is to get costs down to 12 cents per kilowatt-hour of electricity, so that this technology will be competitive.

“I’m optimistic that our technology can play a role in the grid, by helping to provide power at night without fossil fuel power stations running.”

The team, from the ANU Research School of Engineering is part of a broader group of scientists working in the area, with funding from the Australian Renewable Energy Agency, to generate cheaper base-load electricity from renewable energy sources.

“When our computer model told us the efficiency that our design was going to achieve, we thought it was alarmingly high. But when we built it and tested it, sure enough, the performance was amazing.”

Screen Shot 2016-08-22 at 12.54.19 PM

At 500 square metres, the ANU solar concentrator is the largest of its kind in the world, and works by  focussing the power of 2,100 suns onto the receiver, through which water is pumped and heated to 500°C. The energy is stored in molten salt.

The team’s new receiver design is described as “a cavity that resembles a top hat with narrow opening and a wide brim.

When sunlight is focused onto the pipes, it heats the water as it enters at the brim and spirals up into the cavity, reaching peak temperature deep inside the cavity, thus minimising heat loss.

Heat that does leak out of the cavity can be absorbed by the cooler water around the hat’s brim, the team said.

“The overall efficiency of this receiver, with the measured as-built dish optics and at the reference solar elevation angle of 30° …was 98.7%,” the report said.

“This accounts for reflection, emission and convection losses, and is a ratio of the heat absorbed by the working fluid to the total reflected solar irradiance incident upon the receiver surface.”

The ANU team says it has already had commercial interest in the solar thermal system.

The receiver design was presented at the SolarPACES 2015 conference, a full paper is available here.

Print Friendly, PDF & Email

  1. Diego Matter 4 years ago

    Is the goal of 12 cents per kilowatt-hour including molten salt storage costs?

    • Brunel 4 years ago

      Good question. They should say what is the cost of storage or the cost of electrons at 5am.

  2. Brunel 4 years ago


    Well solar PV is cheaper at US 3-6c/kWh. And I am not sure how well solar thermal would work in Germany or on cloudy days.

    • David Boxall 4 years ago

      To quote from the article:
      “I’m optimistic that our technology can play a role in the grid, by
      helping to provide power at night …”
      Unless something’s changed recently, solar PV plus storage costs a bit more than 6c/kWh.

      • Brunel 4 years ago

        Of course molten salt can provide electrons at night. So can batteries.

        The question is, at what price.

        And batteries can be installed at the point of use to avoid transmission losses – molten salt tanks will be large and located at power stations.

        • Rhys Lloyd 4 years ago

          Why not both? The ultimate solution will be a big mix of all these technologies.

          • Brunel 4 years ago

            Because CSP is more costly.

        • David Boxall 4 years ago

          Exactly, but you were comparing PV without storage to stored CST. To quote again from the article: “The energy is stored in molten salt.” Apples and oranges.

          • Brunel 4 years ago

            You make no sense. First you say that solar PV + batteries have to be 6c/kWh or less.

            Then you say 12c/kWh is cheap enough?

          • David Boxall 4 years ago

            You misrepresent, both what I said and what you aserted. You compared PV without storage to stored CST, asserting that the former is 6c/kWh. I just pointed out that the two are not comparable.

            Before you edited it, your response read:
            “So you mean 12c/kWh will be the price of electrons at 5am?

            There is no reason why batteries will not be able to store electrons for 5c/kWh in the near future.”

            Ignoring the first line, which is a red herring, you conceded that the price of storage will soon be 5c/kWh. Add that to the cost of PV power, which you’ve already put at about 6c/kWh and we’re in the same ballpark.

          • Brunel 4 years ago

            You could have said “the price of solar PV + batteries is more than 12c/kWh.”

            But you said “the price of solar PV + batteries is more than 6c/kWh”.

          • David Boxall 4 years ago

            I could have said any number of things, but I referred to your comment, which compared CST with molten salt storage to PV without storage.

          • Brunel 4 years ago

            Solar PV in Dubai and Chile is US 3c not 6c.

            Boffins outside AUS are working on cutting the cost of storage to 5c/kWh.

            On a cloudy day, solar PV still makes electrons.

            I doubt CSP can compete any more.

          • David Boxall 4 years ago

            I doubt that either you or I know enough to judge. Different solutions will work best, depending on circumstances.

            A monoculture is rarely best. A variety of technologies will probably emerge. The right mix of technologies will vary with conditions.

          • Mark Yates 4 years ago

            it’s different currenciens anyway. When people talk about cents on the dollar they normally are quoting US $. But this is an article written in Australia using the Australian $. And there’s about 2 of those to the US $. Hence why $12c/kWh seems expensive compared to the cheapest solar.

    • Mark Yates 4 years ago

      $12c/kWh is an Australian $ price. Isn’t there about 2 of those to the $US – hence the difference.

      • Brunel 4 years ago

        Obviously the pricing should be in £ or €. Not “$”.

        Anyone can convert £ into AUD.

  3. Chris B 4 years ago

    You’d get another 35% replacing all of that wasted latent heat of vapourization with sCO2.

  4. calmly_observing 4 years ago

    Aren’t stand alone CSP design like this only effective (most efficient) with highly precise tracking systems? There is no mention here. But I thought the axial alignment is typically very delicate in order to get maximum production for a given amount of sun. In other words, reflected and scatter sunlight is of no value.

    • Rhys Lloyd 4 years ago

      Can’t imagine it would be all that difficult. It’s not like the sun moves at random.

      • calmly_observing 4 years ago

        If you’ve ever tried to position a mirror to maximize the reflection to, for example heat or light some flint, you know precise alignment with the sun is important.

        • Rhys Lloyd 4 years ago

          I don’t doubt that. What I’m saying is it wouldn’t be all that difficult to program where the sun will be at any given time (this data is already available), then add a layer of dynamic optimisation to ensure it’s running at maximum efficiency at all times.

          • calmly_observing 4 years ago

            I am simply pointing out that while the efficiency peak is significant and laudable, the true output is perhaps even more reliant on a well designed tracking mechanism to realize the gains.

            Yes, heliostats and even 2-axis PV trackers are an established technology. PV panels do not require such precision. But productivity improvements can only put more of a burden on CSP tracking. There is a reason why the Brightsource project has kept their tracking algorithms and methods as trade secrets (proprietary), for example. Perhaps I should have posed the question differently. Such as, “Will the added efficiency require a more precise tracking capacity than what is currently used?”

          • Rhys Lloyd 4 years ago

            > “Will the added efficiency require a more precise tracking capacity than what is currently used?”

            Fair question.

        • Tisham Dhar 4 years ago

          Of course not done by hand. Heliostats are a proven technology, it is hard for me to solder 0201 SMD components by hand but machines do it all the time.

Comments are closed.

Get up to 3 quotes from pre-vetted solar (and battery) installers.