Two-tank molten salt storage is currently the CSP industry standard, but it is extremely costly. Could single-tank storage be the answer?

By Toby Price in Barcelona

To date, the two-tank thermal storage system has worked well enough to have been commercially applied in several plants in the US and Spain. It will also be used in Abengoa Solar’s 280 MW Solana plant in Arizona, which, once completed will be the world’s largest CSP plant.

As the name implies, two-tank thermal storage uses two tanks; one containing the high-temperature fluid and the other holding the cooler fluid, so the two cannot mix.

This arrangement makes operation relatively simple and results in high performance because both fluids are always delivered at their design temperatures.

However, two-tank storage has a significant a drawback – namely, cost. Firstly, the capital cost of a CSP plant’s storage system is effectively doubled because two tanks and duplicate piping are required.

Secondly, two-tank storage uses a fluid – usually molten salt - exclusively as the storage medium. Molten salt is a nitrate salt, which is mined and, as a commodity product, is therefore subject to price pressures, which can negatively impact a CSP project’s bottom line.

Consequently, a number of researchers have been looking at possible alternatives, one of which is single-tank or thermocline storage.

This approach uses a single vessel containing a fluid with a thermal gradient running vertically through the tank, where hotter fluid (lower density) is at the top of the tank and colder fluid is at the bottom. The thermal gradient separates the two temperature potentials.

“With thermocline storage, you only have one tank instead of two, which reduces cost,” explains Greg Glatzmaier, a senior engineer at the National Renewable Energy Laboratory (NREL).

Single-tank storage also reduces the molten salt requirement to as little as 30 percent of the storage medium, explains Dr. Thomas Mancini, CSP Program Manager at Sandia National Laboratories (SNL). The remaining 70 percent comprises low-cost filler such as sand or rocks.

As well as providing the bulk of the thermal capacitance of the system (its ability to store heat) and preventing convective mixing, these fillers decrease the amount of expensive storage fluid required, leading to further costs reductions.

Mancini also notes that it may be possible to install a coil-like exchanger inside the tank and use a different, cheaper working fluid, thereby reducing the molten salt inventory still further to achieve even greater savings.

As a result of this streamlining Glatzmaier calculates that thermocline storage probably offers a reduction in cost of about 25 percent to 40 percent, which he estimates roughly equals a drop in the cost of electricity of about US$1 cent per kWh.

NREL and SNL are both conducting performance modelling and cost analysis of the thermocline system. Meanwhile, SNL has successfully demonstrated a 2.5 MWh prototype using molten salt as the storage fluid and quartzite rock and sand for the filler.

Streamlined, but more complex

Despite clear cost advantages, single-tank storage is, more complicated to operate and performance can suffer.

The main drawbacks of the single-tank system, according to Mancini, are the difficulty of adding and removing heat at the same time, and maintaining the thermal stratification.

It is essential that the thermocline zone be maintained in the tank, so that it does not expand to occupy the entire vessel. Over time, heat transfer between the high and low-temperature regions causes the thermocline to spread, creating an intermediate region.

Much of the thermal energy in this region is unusable because it is too cold to be discharged as high-temp fluid and too hot to be discharged as low-temp fluid.

“Thermal energy at this intermediate temperature is usually discarded. This loss results in lower performance relative to the two-tank system,” explains Glatzmaier .

In a recent study, SNL scientist Doug Brosseau and colleagues  also found that significant salt/metal interactions can arise in single-tank storage, accelerating deterioration of the stainless steel storage vessels.

These interactions may not be of significant concern as long as the molten salt is maintained at or below 450°C, according to the report. However, further research in this area is required if thermocline storage is to be used at higher operating temperatures.

Commercially viable?

“Commercial application [of single-tank storage] could occur in a few years if one of the solar industry members decides to pursue it,” concedes Glatzmaie. However, he adds:“ At this time, there are still too many unknowns in terms of performance for someone to initiate that investment”.

Jose C. Martin, CEO of Sener Engineering and Systems, similarly warns: “There is a long way ahead before we see the commercial use of these concepts”.

While this may be the case, SNL engineer, James Pacheco, argues thermocline storage is a feasible option for parabolic trough plants. He says a number of companies are analysing whether or not it is a suitable alternative at commercial scale.

Sener itself has already designed a small-scale prototype of a single-tank concept that could take advantage of some of the thermocline model while avoiding some of the disadvantages. Martín says this will soon be tested in one of Sener’s Spanish projects.

NREL is also contemplating a pilot project next year, although Glatzmaier says: “That decision has yet to be made and will depend on the results of the analysis and on industry interest.”

Either way, it is hoped that the analysis being performed at NREL, SNL and elsewhere will quickly help the industry decide if single-tank storage is the way forward.

To respond to this article, write to the Editor:

Rikki Stancich: rstancich@gmail.com