Santiago Arias, Technical Director at Gemasolar, supplying 24 hour solar power

Santiago Arias is Technical Director of Operation and Maintenance at Torresol Energy's Gemasolar concentrated solar thermal power (CSP) plant in Seville, Spain. The 20MW Gemasolar plant with molten salt storage has been in commercial operation since May 2011 and supplies 24 hour solar power to about 27,500 Spanish households every year.

Gemasolar is the first commercial concentrated solar power plant with central tower receiver and molten salt storage technology. The plant has produced over 100 million kWh in over one year of operation. Its record is 404MWh in a 24 hour period.

Santiago explains how Gemasolar has improved over the year and measures its contribution to the Spanish electricity grid. With a 15 hour buffer in the molten salt storage technology, Gemasolar provides predictable, reliable power generation. As the operation of the steam turbine to generate the electricity is independent of weather conditions, CSP with storage has advantages over other renewable technologies such as wind and solar PV (which are continuing to realise rapid cost reductions).

Gemasolar was recenlty visited by the South African government:

In August 2011, Santiago hosted Australian climate change advisor Ross Garnaut, federal independent MP Tony Windsor and BZE's Matthew Wright at the Gemasolar plant, to show what could be possible for Australia.

Listen to more from Santiago:

Santiago Arias - Interview

You are missing some Flash content that should appear here! Perhaps your browser cannot display it, or maybe it did not initialize correctly.



 Matthew Wright: On the theme of zero emissions, today we are very delighted to be speaking with Santiago Arias, from Torresol Energy, a project developer of Sener’s Solar Thermal Plants.  Santiago is the Chief Operating Officer of the plant.  He started out in the 1970s as a nuclear engineer with Sener.  Hello Santiago.

Santiago Arias: Hello.

Matthew Wright: We want to find about what I think is the most exciting renewable energy project in the world, Gemasolar.  Could you please share with us how Gemasolar is and how it operates?

Santiago Arias: Torresol is a special branch - a concentrated solar thermal plant - no PV, just mirrors: 2,650 heliostats.  These are structures that move mirrors to reflect and concentrate the sun’s light to 1000 times onto a small spot.  The spot is located on top of a tower.  On this spot we get a huge amount of energy: we get 120MW of solar energy.  This spot gets very hot. To dissipate this heat we use molten salt, so we are pumping molten salts through the receiver, which is like a radiator, and the solar energy is transferred to the salt, getting very high temperatures in the salt.  At the end of the solar process we have molten salt at 565 degrees, and this is stored in the hot molten salt tank.  This is the end of the solar process.  The rest of the plant is just taking that thermal energy, and converting that thermal energy into electricity in exactly the same way as any other thermal plant does, for example, a coal-fired plant - but  instead of having coal, we have molten salt at very high temperature. 

From the molten salt we produce steam, which is delivered to a high pressure turbine, and then the turbine rotates to produce electricity in an alternator.  Exactly the same process as any coal plant.  The advantages when compared to other renewables is that the production of electricity is not linked to the environmental conditions.  It’s not linked to the sun’s radiation, it’s linked to the level of the hot molten salt tank.  We have an over-sized solar field, which means during the day we are not just generating enough energy to deliver to the grid, but part of the energy is stored, because the level in the hot molten salt tank rises.  We take that energy that has been captured during the day to produce electricity during the night.  So in summer time we work continuously, 24hrs at full capacity.  We have already achieved that last summer time, many weeks we were running 20MW day and night, in the same pattern as say a nuclear plant.  In winter time, days in Spain are shorter, so at the end of the afternoon we don’t have enough energy accumulated to run the turbine for the whole night, so we run the turbine till 10:30pm until the consumption of electricity has dropped and then we drop the power of the turbine from 15MW to 14 MW in order to ensure that tomorrow morning at 9am there’s still molten salts in the tank.  Then we don’t start and stop the turbine every day, it’s just like increasing, then decreasing the power during the night when the demand is lower and the price of the electricity is lower.  So we have a plan that in summertime it works as a baseload system, and in wintertime it’s working even better, adjusting power to the demand.  And that’s Gemasolar.

Matthew Wright: That’s fantastic.  Gemasolar has been operating for over a year, can you tell us how long it’s been operating, and how the experience has improved over that time?

Santiago Arias:  Gemasolar was the first plant of its kind, therefore the start-up was not that easy.  The first summer we were learning how to do it, the plant was performing better and better every day. Starting up in the morning was taking 2 hours, and we were concerned with problems with draining the receiver in the afternoons, so were very conservative in stopping the production of hot molten salts 1-2 hours before the sunset.  This is not the case anymore.  We are learning how to do this better and better.  This winter we have already achieved 404MWh in 24 hrs.  That was supposed to be the maximum because we considered that the turbine was not supposed to produce more than 20MW, and 10% was to be absorbed by the mechanism, so reaching 404MWh in 24hrs was a record.  I’m very proud of that.  My team, the day when they achieved that, they took a picture of the screen showing that, and delivered that to me.  In summertime then we reached 428MWh in a single day.  The plan is no longer a technological dream, it has already produced more than 100 million kWh in a little over 1 year – I say a little over, as we were working for 6 months in the early stages of a learning curve.  Right now we are reaching very high performance, very close in a monthly basis to the targeted budget amount in normal production.  We have already produced a lot of energy, and are very happy with performance. 

Matthew Wright: You mentioned that in relation to the MWh, what does that correlate to in the number of houses that you are supplying electricity to?

Santiago Arias: We can fulfil 20,000 houses, based on average consumption: close to 100,000 people of domestic production. 

Matthew Wright: Your plants have storage.  Given that rooftop PV is so much cheaper for daytime production, is it true that solar thermal plants have to be produced with storage?

Santiago Arias: In the long term PV will be cheaper and cheaper.  Just like computers, with performance and cost.  My impression is that PV prices will continue dropping, so we cannot compete against PV. My vision is we can enlarge the amount of PV installed in a certain country because we produce electricity not just during the day.  We may store and manage the amount of electricity we produce.  For example, last month we passed the manageability test, in which we say what will be the hourly production tomorrow, very accurately.  Last month, during the whole month, we were able to predict the hourly production tomorrow with a deviation of lower than 4.29%, which means we predicted an accuracy of greater than 95% of the amount of MW that we’ll produce between 3:00 and 4:00, 4:00 and 5:00, and so on.  So our plant is like a swiss clock – very predictable.  It’s not predictable because equipment is different to other solar plants, rather the turbine operation is not related to the environmental conditions – we have a buffer of 15 hours.  This allows us to be very precise in how much electricity we are going to deliver tomorrow.  Having this type of plants will help compensate the little problems that are created by using wind or PV energy.  Most countries prefer to use electricity during the early part of the night – in the case of Spain, the two hours just after sunset are when the demand of electricity is highest, and the cost of the electricity is highest.  If your plant isn’t producing at this time, your plant isn’t adding any power in the system, it’s just reducing the hours of operation of someone else that has to be there as a backup.  We don’t need that, we produce in a very continuous way.  Besides, we have another opportunity.  Since we have a thermal step in the process, storing thermal energy is very easy, and combine the solar part of the plant with any other way of getting heat (burning biogas, and so on), for the bad days in the year.  So going to sunny places and having the possibility to burn coal or wood to produce heat, you can produce electricity how you want when you want.

Matthew Wright: We’re very interested in the reliability issues, how reliable solar thermal technology is.  Can you give an insight to the cost reduction curves as the technology develops over time?  Where are the big gains in solar thermal, as the technology progresses?

Santiago Arias: CSP has made a tremendous effort.  In Spain we have installed 64 plants of this kind (CSP).  There’s a clear percentage reduction  in cost - the manufacturers are talking 20%, the plants are also more effective.  We have the experience of Valle 1 and Valle 2.  They are in theory identical to Andasol 1 and Andersol 2 that were made 5 years ago.  The cost was cheaper and the performance is getting better because some of the components are getting better, and so there are less losses.  We are reducing the cost, particularly trough technology, which is improving day by day.  The company I used to work developed the SENERtrough ® and they have to be very good, as they produce more than 2,000 kms of the trough.  To be sold not only in the plants they are the major contractor, but to plants in which the major contractor are competitors of Sener.  This trough is lightweight, simple to assemble on site, and really reduces costs of installations.  This is just the first step – improving components we are always evolving the technology. For example, with the storage system, we now we have two tanks – one for cold molten salt tank, and one for hot molten salt.  Sener has made a prototype for 10,000 tonnes of molten salt in a single tank, containing hot molten salt in the top, and the cold molten salt in the bottom.  By doing that, we are practically reducing the cost by half, in tanks, pumps and so on.  So every day they are inventing new techniques that allow us to reduce the costs.

 Regarding the tower technology, you have to consider that Gemasolar is very small – only a 20MW plant.  Right now, Sener is considering delivering a project for South Africa, the same as Gemasolar, but with 100MW power.  The power is multiplied by 5, but the cost is multiplied by around 2.5.  So we have a relevant reduction in cost because the size is going to be bigger.  Deloitte did an analysis of possibilities of cost reduction by not only increasing the size of the plant, but by locating groups of plants (as we do in Spain for most cases , for example, Valle 1 and Valle 2 are operated with a single team, single control room) there is a clear reduction of cost.  Just imagine that in the whole world there is only 2GW of power installed in CSP, and wind has 140GW installed capacity.  So, wind has been evolving and reducing price just because of the amounts of plants that have been installed is larger.  So it’s producing high opportunities for cost reduction.  So what we are demanding is the same, we are at the beginning of this technology, we want it to be expanded to have thousands of plants in order to reduce the cost.

Matthew Wright: With the South African proposal, is that 17 hours storage, or will that storage ratio be adjusted?

Santiago Arias: I believe it will be less, but at this stage it is just a proposal.  In South Africa they aren’t interested in 15 hours storage, rather the clients are requesting around 7-10 hours storage, as they don’t want flat production.  They are looking for the winter model, for example that produces less at night, to keep coal plants producing at night.  This is very important: the same plant with the same field can be modified according to the request of the owner, or according to the demand.  In Spain we prefer to run flat for 24hrs, as that minimises the cost per kWh, but in other places producing during the night (for example 2-3am) is not interested, as there is excess capacity.  They need capacity during the day and during the first 2-3 hours of the night, as such 3-4 hours of storage is enough.  In South Africa, where they asked for 7-8 hours, that’s more than enough to operate the way we operate in winter time.

Matthew Wright: Recently in Spain, 60% of the nation’s power in a day was from renewable sources.  I assume Gemasolar contributed to that, can you tell us about those occasions?

Santiago Arias: Gemasolar contributes very little to that.  Spain has hydro, wind, PV, CSP.  CSP contributes on the best days 2% of the total energy consumption.  For CSP, there are 64 plants in operation next year (20-30 plants bigger than Gemasolar), so Gemasolar is producing a small proportion of the renewable package that Spain’s able to produce.  People from South Africa visit because they see this as a good example to combine wind, hydro and solar with and without storage to achieve 40% in many months.

Matthew Wright: It’s a fantastic achievement.  Do you see Spain heading towards 100% renewables?

Santiago Arias: I don’t think so.  There are plenty of new installed gas turbine plants with high efficiency, but these people also need to live.  I think that the trend will be a slow one.  There is another opportunity which is much more realistic: Spain, Italy and Portugal (the ones creating the problems to the Euro) are the sunny countries, could be internally producing desert-tech, and producing electricity for central European countries.  Germany has promised to stop nuclear plants so this demand for electricity needs to be fulfilled with something else.  We volunteer to install 200 Gemasolars in the local area, where there’s plenty of empty space for these plants, to be able to generate for Spain, and most European countries.

Matthew Wright: We’re not short of space here either.

Santiago Arias: Yes, there are many places in the world where space is not an issue, and these could be used to create electricity without CO2.

Matthew Wright: Thanks Santiago for joining us.

 Transcript by Dale