3.3.2. Moving the Storage Capacity outside of UPHES, toward Thermal Stores
In response to the medium-high Capex for pure UPHES, a further innovative leap in technology and cost structure is needed as the third improvement of UPHES to achieve a Capex compatible with long-duration storage (5–500 h), one order of magnitude lower than the previously computed value of 200 USD/kWh for storages of short and medium duration.
Since the required volume of such pressurized UPHES would prove too large and therefore too costly for long duration storage, the storage of energy must be diverted elsewhere. We opt for large, low-cost, insulated, atmospheric, thermal pits or thermal tanks as described in
Figure 5. One pit, called Hot Store contains hot water at 95 °C, similar to the 40,000–200,000 m
3 pits used in Denmark for seasonal thermal storage, losing only 0.1 K per day [
23]. A second pit, called Cold Store contains a salted ice slurry at −20 °C.
The exergy from a thermal energy source is defined by the amount of mechanical work recoverable by a Carnot machine. Very interestingly, one shall realize that exergy of hot water is very high, much higher than energy of water subject to the sole earth gravity. For instance, exergy of water at 100 °C related to ambient temperature is equal to 12 kWh/m3. This is one order of magnitude higher than the mechanical energy of the same water in pumped-hydro, equal to 1.2 kWh/ m3 for a hydro plant of 360 m head.
Figure 6 illustrates the principles of energy storage for a PTES using a UPHES for electricity conversion, while
Figure 7 describes its main components. The radical innovation is that those two stores are thermally charged and recharged by the heat supplied and extracted during the time hydroelectricity is charging the UPHES, the slow moving CO
2-water interface acting as a novel type of heat-pump.
An alternative configuration is shown in
Figure 8 below, devoid of the CO
2 smaller pressurized cavern and where the necessary heat exchanges for CO
2 happen instead by using an ancillary multiphasic CO
2 blower/pump and external CO
2 tubing leading to external heat exchanger(s). A bypass circulation pipe, fitted with a valve, is allowing the CO
2 heat exchanges during ied times when the flexible membrane gets fully collapsed.
Figure 9 describes the system operation on the Pressure-Enthalpy diagram related to CO
2. It is specified that placements of cycles vary with the level of completion/depletion of thermal stores, in particular with temperatures available from the Hot Store, and that optimal cycles for campaign of energy charge can be significantly different from optimal cycles for campaign of energy discharge.
For the Hot Store to work in the domain of 95 °C temperature, it is pointed-out that the pressure and temperature of pure CO2 in the cavern needs to exceed pressure and temperature of CO2 at the critical point (7.39 MPa, 31.1 °C). Such charging and discharging operations in trans-critical phases do not create specific irreversibilities of their own, except for the added complexity that overall heat exchanges for the expansion and contraction of supercritical CO2 at the upper constant-pressure ceiling needs to be performed by a glide of several heat exchanges at increasing and decreasing temperatures. It is specified than despite incursions of the working fluid in supercritical phase, the hydro-pump and the hydro-turbine remain the only two machineries needed by the novel storage.
A few companies are presently assisting to determine whether it could be valuable, instead of using pure CO2 at every occasion, to consider using binary/ternary mixtures of CO2 with selected molecules in low molar fraction, in order to displace the critical point of the CO2 mixture. If results of simulations and experiments were to prove positive, the novel storage could benefit of several types of CO2 mixtures, stored in several smaller dedicated caverns, to fulfil different situations (production of hydro mechanical energy only, co-generation of cooling, co-generation of heating, operation as hybrid UPHES-PTES).
It is pointed out as well that, compared to most other PTES, thermal stores temperatures selected by this novel storage are classified as low. Several other PTES are using temperatures of 500–750 °C for the Hot Store in pursuit of a higher storage density, at the cost of higher Capex of energy and of sophisticated operating procedures. Other PTES using CO
2 as working fluid for energy conversion, albeit in continuous high-speed circuits of vapour compression differing greatly from the discontinuous expansion/contraction of CO
2 in closed caverns, use temperatures of 120–140 °C for the Hot Store in order to gain in energy density [
24]. Our analysis is that the challenge of lowering CO
2 emission worldwide can only be achieved with the simultaneous higher penetration of low-cost and low-emission district heating and district cooling to residential buildings, office buildings and to some industrial developments, such as pulp and paper, pharmaceuticals, agro-food industry, textile and so forth. In particular, our vision objects to the growing installation of stand-alone electric heaters and stand-alone air-conditioners because in most countries those low-efficiency devices will remain powered by electricity of high carbon intensity, often from coal power plants, for many decades to come. Therefore, we foresee that the low temperatures selected for our thermal stores shall allow utility operators to readily integrate those stores, as well as the UPHES conversion caverns, for the production of low-cost and low-emission district heating and district cooling.
From a single transcritical cycle as drawn in
Figure 9, one can also exploit the alternative of operating two adjacent cycles by drawing a third horizontal segment placed between the low-pressure and the high-pressure horizontal line, in such way that this segment is positioned at the outright saturation vapour pressure of the CO
2 fluid for the ambient temperature (i.e., an horizontal segment, from 100% liquid phase till 100% gas phase, at a pressure of 58 bar when ambient temperature is 20 °C). By doing so, the storage operator gets, in discharge mode, the flexibility of prioritizing the thermal energy of the Hot Store towards heating networks or of prioritizing the thermal energy of the Cold Store towards cooling networks, while still being able to generate electricity from one of the two adjacent cycles, by using the environment to replace the missing store. Similarly, in charge mode, the operator enjoys the flexibility of using the environment to restore whatever imbalance by storing electrical energy in one of the thermal store whenever the second thermal store gets close to maximal capacity. This split in two adjacent cycles with the use of the environment could also prove being handy to initialize the storage system by charging independently the Cold Store and the Hot Store for a first time with the highest flexibility.
An additional benefit for this UPHES-PTES hybrid storage to work in the “low temperatures” domain is that this feature provides to the system a larger occurrence to increase its RTE, when very-low-grade geothermal energy or very-low-grade industrial heat gets available from the nearby environment (very-low-grade being hereby defined from being in the 50–65 °C temperature range). Because very-low-grade heat is considered “end-energy,” it comes “for free” and goes unaccounted-for in the definition of system efficiency. Consequently, RTE of the novel storage could exceed 100% in conditions where very-low-grade heat is available at sufficient power and temperature.
When an external source of low-grade energy is freely available (low-grade being hereby defined from being in the 65–95 °C temperature range), the same “low temperatures” feature of the working fluid shall allow the UPHES to work as an autonomous thermal engine, producing net-work without depending on thermal stores nor on external electricity source. Examples of such net production of electricity are the availability of low-grade geothermal energy or low-grade industrial heat. When the low-grade source of energy is of suitable temperature however of low power (i.e., of weak flow), the procedure is to include this inexpensive energy in the heating process of the CO2 mass or to store it in water tanks to operate only a few hours per day in net-work mode. Operational modes taking benefit of very-low-grade or low-grade energy usually require the use of the ambient environment as a secondary thermal source.
With or without the availability of external thermal sources supplementing the two thermal stores, in discharge mode the effect of applying a hot source and subsequently a cold source, to a stand-alone UPHES cavern creates a near-reversible thermal engine, slowly expanding/contracting CO2, pushing-out hydro-water to the hydroelectric turbine and subsequently sucking-in hydro-water back. The favourite configuration to maintain steady power is to rig two caverns A and B asynchronously in parallel to a single turbine set and to use two receiving ponds or one larger receiving pond, located at the ground surface. While cavern A pushes-out hydro-water at high pressure (9.5–15 MPa) thanks to the expansion of CO2 heated by 95 °C water at its evaporator heat exchanger, cavern B gets its condenser heat exchanger connected to a −20 °C coolant, reducing gradually its inner pressure down to 2 MPa and allowing hydro-water to be sucked-in from the surface with close to no pumping work. Before cavern A gets entirely emptied, cold CO2 of cavern B had been pre-heated or had been swapped with an external mass of pre-heated CO2 and get ready to take-over the pushing-out of hydro-water to the same turbine set thanks to the expansion of CO2 heated by 95 °C water, while cavern A gets prepared to suck-in the fresh water needed for the continuous operation of the turbine set.
In charge mode, for transferring heat from Cold Store toward Hot Store in a steady way for the power network while the storage is in heat-pump mode, the favourite configuration is also to rig two UPHES caverns asynchronously in parallel to a single pump set and to use one or two receiving ponds located at ground surface.
Whenever the Cold Store, due to design or to temporary dysfunction, is of temperature significantly higher than −20 °C or when the Cold Store is just replaced by the ambient environment as it often happens in PTES, the favourite operative configuration would be to rig two caverns A and B in opposition, communicating by the mean of hydro-water and by the mean of either one single pump set when charging in heat-pump mode or of one single turbine set when discharging in thermal engine mode. Thus, the operation of heat-pump and of thermal engine by two caverns rigged in opposition avoids the detrimental back-work ratio of having to repeatedly spend mechanical work to reintroduce hydro-water at mid pressure inside the colder cavern A or B due to the insufficiently low temperature of the Cold Store. In such a configuration of caverns in opposition, illustrated in
Figure 10, receiving ponds are left unused/idle.
As one can figure-out from the Pressure-Enthalpy diagram in
Figure 9, there is a much heavier exchange of heat happening both ways than the 95 °C heat required from the Hot Store or than the −20 °C heat extracted by the Cold Store. Therefore, as with every other PTES, efficient regeneration of heat is one necessary component of high-level RTE. In heat-pump mode as well as in thermal engine mode, most of this regeneration can be achieved through synchronous exchange of heat between cavern A and cavern B (through CO
2-coolant exchangers or through CO
2-CO
2 exchangers where possible). High-grade heat which cannot be exchanged synchronously needs to be stored in buffer water tanks in order to be used during the next operation.
As summary, in this third level of improvement dedicated to long duration storage, UPHES is not the storage per-se but the necessary mechanism of a two-way thermo-electrical energy conversion, slow and as near-reversible as possible, providing hopes that PTES storages could fulfil their long-time promise. As seen above, upon conditions of operations, caverns can be set-up as stand-alone or in parallel (synchronous or asynchronous) or rigged opposed back-to-back.
Figure 11 provides some illustrations of scale and some density of energy for the UPHES-PTES storage.
3.3.3. Economic Targets of the UPHES-PTES Hybrid Long-Duration Storage
Target for RTE of our novel storage, at 60%–70% may seem an ambitious figure compared with results currently achieved for pilot plants of other PTES. Besides similar thermal irreversibilities due to imperfect heat exchanges, other existing PTES use high-speed circuits of a working fluid (steam or refrigerant) and need 4 complex turbine or piston machineries working only in gaseous and supercritical phases (1 compressor and 1 expander for the heat-pump mode, 1 compressor and 1 expander for the thermal machine mode). Kim et al. show that, for a traditional vapor compression process bearing a typical back-work-ratio of 36.5% and assuming realistic grid-scale efficiencies of 80% for each machinery, the best RTE which can be attained in theory by high-speed trans-critical circuit PTES is 45% [
24]. Dietrich et al. realized in 2016 a model using butane as the working fluid and computed a maximum RTE of 27.3% only [
8]. While the Siemens Gamesa company expects a full-scale deployed system to reach in the future round-trip efficiencies of 50%, a RTE of only 25% has been so-far obtained in June 2019 in Hamburg, Germany, after several years of development on a pilot plant of 130 MWh per week [
9]. Although there is no reason why large efficient gas compressors and gas turbines could not be developed in a later future, the present lack of high-range RTE in pilot and demonstrator systems is probably the cause impeding the worldwide unfolding of grid-scale PTES. In the favourite configuration of the novel hybrid UPHES-PTES, only 1 operation of a hydro pump and 1 operation of a hydro turbine are needed for a conversion round trip. Thermal irreversibilities of the hybrid UPHES-PTES differs slightly from those of the simpler UPHES as described in paragraph 3.2.4. by the minor fact that hot water and cold water get stored and loose some energy to the ambient environment during storage time (i.e., 3% loss per month). To ascertain the value of RTE in the hybrid mode relatively to the 60%–70% target, is planned for 2020 the construction of an aerial prototype made out of steel vessels, of 35 mm metal thickness, for a total pressurized volume of a few cubic meters.
Thanks to the low cost of atmospheric insulated thermal storages, assessed at 40 USD/ m
3 for the Hot Store pit [
21] and evaluated at 100 USD/ m
3 for the Cold Store ice slurry pit, the expected Capex of a hybrid UPHES-PTES assisted by two large thermal stores can be evaluated, using the same other cost components than in paragraph 3.2.4. above, to a Capex of electricity equal to 2160 USD/kW + 10 USD/kWh (meaning a compounded Capex of 20 USD/kWh only, for a 216 h storage).
Calculations can be counterintuitive because in this configuration the cavern and its receiving pond are not any longer components of energy capacity Capex but, as loci of energy conversion, components of power Capex. Based on the same required power of 100 MW but with an energy capacity increased to 100 h and to 10 GWh, based on two caverns in asynchronous parallel mode achieving a complete cycle in an arbitrary time of 2.5 h, the calculation results in 200,000 m3 of caverns, 200,000 m3 of water receiving pond, 987 000 m3 of Hot Store and 525,000 m3 of Cold Store to achieve the required power and energy capacity.
The power component of Capex is therefore increased, still for 100 MW, of an extra cost of 200,000 m3 X (300 USD + 30 USD), thus 660 USD/kW, raising this power Capex to 2160 USD/kW.
The energy component of Capex needs to be calculated in a table like
Table 2 below.
Using the same components of Capex, as a crude estimate of the Capex for the pure supply of heating and cooling, we can simply divide the Capex for electricity by 500%, which bring a Capex of heat and cooling equal to roughly 432 USD/kW Thermal + 2 USD/kWh Thermal (meaning a compounded 4 USD/kWh only, for a 216 h storage).