The potential of metal hydrides paired with compressed hydrogen as thermal energy storage for concentrating solar power plants

Highlights

• Metal hydrides with H₂ gas storage have potential for low-cost heat storage.

• Lined rock caverns have practical advantages over lower-cost salt caverns.

• Parasitic compressor power depends on metal hydride and H₂ storage method.

• The sensible heat of H₂ gas has significant effect on system characteristics.

Abstract

Concentrating solar power (CSP) plants require thermal energy storage (TES) systems to produce electricity during the night and periods of cloud cover. The high energy density of high-temperature metal hydrides (HTMHs) compared to state-of-the-art two-tank molten salt systems has recently promoted their investigation as TES systems. A common challenge associated with high-temperature metal hydride thermal energy storage systems (HTMH TES systems) is storing the hydrogen gas until it is required by the HTMH to generate heat. Low-temperature metal hydrides can be used to store the hydrogen but can comprise a significant proportion of the overall system cost and they also require thermal management, which increases the engineering complexity. In this work, the potential of using a hydrogen compressor and large-scale underground hydrogen gas storage using either salt caverns or lined rock caverns has been assessed for a number of magnesium- and sodium-based hydrides: MgH₂, Mg₂FeH₆, NaMgH₃, NaMgH₂F and NaH. Previous work has assumed that the sensible heat of the hydrogen released from the HTMH would be stored in a small, inexpensive regenerative material system. However, we show that storing the sensible heat of the hydrogen released would add between US$3.6 and US$7.5/kWh_th to the total system cost for HTMHs operating at 565 °C. If the sensible heat of released hydrogen is instead exploited to perform work then there is a flow-on cost reduction for each component of the system. The HTMHs combined with underground hydrogen storage all have specific installed costs that range between US$13.7 and US$26.7/kWh_th which is less than that for current state-of-the-art molten salt heat storage. Systems based on the HTMHs Mg₂FeH₆ or NaH have the most near term and long term potential to meet SunShot cost targets for CSP thermal energy storage. Increasing the operating temperature and hydrogen equilibrium pressure of the HTMH is the most effective means to reduce costs further.

Introduction

A viability assessment was performed on high-temperature metal hydrides (HTMHs) that use compressed gas hydrogen storage as thermochemical heat storage in concentrating solar power (CSP) plants. The intermittent nature of solar energy is driving research for different energy storage options. HTMHs show potential as thermal energy storage materials for CSP plants due to their high gravimetric and volumetric energy density compared to state-of-the-art molten salt systems. However, storing the hydrogen gas released from the HTMH until it is needed for power generation is expensive and, when stored in a low-temperature metal hydride (LTMH), introduces additional complexity associated with thermal management. The development of a generalised approach for exploring the potential of compressed gas hydrogen storage paired with HTMHs for thermal energy storage can help guide research directions that will lead to cost reduction.

Concentrating solar power plants with thermal energy storage represent an attractive alternative to conventional fossil fuels for base-load power generation. However, though CSP has low running costs they have substantial upfront capital costs [1] and their widespread implementation will only be possible with further reductions in their installation costs. A number of technology areas have been identified for cost reduction including: the heliostat field; receiver technology and heat transfer fluid [1]. Another step towards reducing the overall cost of CSP, and the focus of this work, is a means of thermal energy storage with reduced cost. CSP plants with thermal storage currently utilise the heat capacity associated with a temperature change in a molten nitrate salt mixture (NaNO₃/KNO₃) [1]. The thermal energy storage system density can be increased by the use of phase change materials or reversible thermochemical reactions and reviews of the current state-of-the-art in thermal storage materials are available [2], [3], [4], [5], [6].

Metal hydrides are an alternative class of thermochemical candidates for heat storage applications and a number of different metal hydride systems have been explored for this purpose [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22]. The use of metal hydrides for concentrating solar thermal energy storage is typically comprised of paired metal hydrides [9]: the high-temperature metal hydride (HTMH) as the thermal energy storage medium and a low-temperature metal hydride (LTMH) for hydrogen storage. During periods of sunlight, incoming solar radiation is focused by mirrors to generate heat. A portion of this heat is transferred to a heat engine to generate electricity and a portion of the heat is directed to the HTMH to release H₂ in an endothermic reaction. The released H₂ is temporarily stored in the LTMH, Fig. 1a. During night time or periods of cloud cover, the reactor temperature of the HTMH begins to fall; causing the system gas pressure to drop and H₂ is consequently released from the LTMH and absorbed by the HTMH in a self-regulating cycle. This absorption by the HTMH hydride is an exothermic reaction, where the heat released is used to drive a heat engine and generate electricity, Fig. 1b [9].

Heat storage using metal hydrides has been explored since the mid-1970s [24] but was generally applied to temperatures below 200 °C due to the nature of hydrides known at that time. In the early 1990s, the development of low-cost magnesium hydride (MgH₂) with rapid hydrogen absorption/desorption kinetics led to its research as a heat storage medium for solar thermal energy storage up to temperatures of 420 °C [25], [26], [27], [28], [29]. More recently, SunShot cost targets for thermal energy storage as part of CSP [30] have driven a renewed interest in low-cost, high-temperature metal hydrides for thermal energy storage. Metal hydrides based on Mg, Na and Ca appear to be the most promising high-temperature metal hydrides to meet these targets [12], [13], [20], [31], [32]. The early breakthrough of low-cost MgH₂ was hampered by the fact that the only suitable hydrogen storage option was using low-temperature metal hydrides that have limited hydrogen capacity, typically 1 to 2 wt% H₂, and were based on expensive transition or rare-earth metals. The cost of hydrogen storage in a LTMH for CSP applications has improved with the advent of reversible hydrogen uptake in NaAlH₄ and Na₃AlH₆ [33], [34], but the issue of thermal management of the LTMH remains. The LTMH itself produces an appreciable quantity of heat during hydrogen absorption and this same amount of heat needs to be supplied to release the hydrogen again. For example, early work using MgH₂ as the HTMH used a TiMn₂-based alloy, HWT 5800, as the LTMH hydrogen store where the heat generated by the LTMH during hydrogen absorption was ∼30% of the thermal energy stored by the MgH₂ [25]. The enthalpy of hydrogen absorption in HWT 5800 is approximately −23 to −25 kJ/mol·H₂ [35], [36]and so the issue of thermal management becomes even more problematic for NaAlH₄ and Na₃AlH₆ where the hydrogen absorption enthalpies are −38 and −47 kJ/mol·H₂, respectively [33].

In order to circumvent the thermal management issues associated with hydrogen storage in a LTMH, the hydrogen could instead be stored as compressed gas [12]. If the LTMH is replaced with compressed gas hydrogen storage, there are two basic ways in which the HTMH TES system can operate at constant temperature (see Fig. 2). In operation mode one (OM1), Fig. 2a, the daytime minimum pressure, P_min, of the compressed gas storage vessel is equal to the equilibrium pressure of the HTMH, P_eq, and the compressor must do all of the work to boost the hydrogen up to the maximum storage pressure, P_max. During night time operation, the hydrogen pressure in the storage vessel is above P_eq of the HTMH and the absorption of hydrogen from the compressed gas hydrogen storage will proceed automatically.

In operation mode two (OM2), Fig. 2b, the HTMH also operates at constant temperature during both the daytime and night time cycles. In this case, the compressor operates during part of both the daytime and night time cycle. The latter operating mode is preferable in the context of CSP as, for the same amount of compression work, the pressure ratio P_max/P_min is substantially higher (see supporting information Eq S(1) and associated discussion). All of the results presented in this work refer to OM2.

For compressed hydrogen gas storage as part of a HTMH TES system for CSP, the most feasible options from a cost and practicality perspective, are underground storage in salt caverns or lined rock caverns (LRCs). The underground storage of hydrogen gas has been receiving increased attention in recent years as it has the potential to store renewable energy in the form of hydrogen produced via electrolysis in times of excess electricity production from intermittent sources such as wind power and photovoltaics [37], [38], [39], [40], [41]. Hydrogen storage in underground caverns has the potential to play a role in buffering seasonal energy demands, to provide continuity in instances of supply chain disruption and to reduce the storage costs of hydrogen delivered to consumers for use in fuel-cell electric vehicles [42]. The feasibility of large-scale underground hydrogen storage for this purpose is now actively being explored in the USA [42], France [37], Poland [43], Sweden [44], Germany [39], Spain [45] and others [41].

Salt caverns are formed by solution mining of geological suitable salt deposits and have a long history of being used for storage of compressed natural gas (CNG) [37], [42], [46], [47], [48] and for compressed air energy storage (CAES) [49]. For example, in the USA there were 39 salt dome facilities being used to store more than 13 billion cubic metres of natural gas at standard temperature and pressure as of the end of 2014 [47]. There are currently three locations worldwide that store hydrogen underground in salt caverns. Two of these are in the USA and both have volumes of ∼580,000 m³ while the facility in the UK is comprised of three caverns of ∼70,000 m³ each [41]. Salt caverns are particularly attractive for large-scale storage of hydrogen as the cost is ∼2 orders of magnitude cheaper than the storage of hydrogen using above ground pressure vessels [46]. Salt caverns have the advantage of being naturally self-sealing due to the plastic properties of salt, and the size and shape of the cavern can often be customized. Challenges associated with the use of salt caverns for compressed gas hydrogen storage as part of a HTMH TES system include:

• The uneven geographic distribution of suitable salt deposits, and that they do not necessarily coincide with areas of high solar irradiance and demand.

• That salt caverns require a relatively large quantity of cushion gas, on the order of 30% of the total cavern capacity, in order to maintain their structural integrity [41], [42].

• That, other than maximum and minimum working pressures, the salt cavern operations are limited by the maximum allowable rate of pressure change within them [41].

• The need to purify hydrogen after storage in the salt cavern due to the presence of water vapour that evaporates from brine solution left over from the construction process [39], [41].

A lined rock cavern (LRCs) is composed of mined rock cavern with an impermeable liner for gas storage and represents a more recent development than salt caverns. The general principles related to hydrogen storage in lined rock caverns (LRCs) are covered by Refs. [44], [50] and the most advanced LRC technology for underground gas storage is comprised of an impermeable inner steel liner, a sliding layer and a reinforced concrete layer for transferring load from the steel liner to the rock face [44], [50], [51]. They are higher in cost than salt caverns but have been extensively developed in Sweden, where there is a lack of suitable geological formations available for constructing salt caverns. Since 2002, a LRC demonstration plant for storage of compressed natural gas (∼40,000 m³, 200 bar pressure), has been operating in Skallen, Sweden [41], [52]. This technology has been considered as part of the HyUnder project [37], [41], [45] and is now being applied to the storage of hydrogen gas as part of the HYBRIT project for the production of fossil-free steel [53], which recently began construction on a pilot plant [54]. While more expensive to construct, LRCs have a number of advantages over salt caverns including:

• A wider range of suitable geological locations.

• That their fully engineered nature means that hydrogen purification should not be required [52].

• Higher injection/withdrawal rates for hydrogen.

• Only ∼10% cushion gas is needed [44].

The objectives of this paper are to explore the potential of HTMHs paired with compressed hydrogen storage as thermal energy storage for CSP, and to compare these results with state-of-art molten salt heat storage and SunShot targets for next-generation CSP heat storage materials. The operating temperature and thermodynamics of the HTMH in conjunction with the charging time of the thermal energy storage system are used to derive all other factors such as hydrogen flow rates during charging, compressor performance and characteristics, the range of operating pressures in the compressed gas hydrogen storage system and the quantities of hydrogen gas required. The sensible heat contribution of hydrogen released from the HTMH and its impacts were also considered.