Elsevier

Energy

Volume 107, 15 July 2016, Pages 542-549
Energy

Ultra high temperature latent heat energy storage and thermophotovoltaic energy conversion

https://doi.org/10.1016/j.energy.2016.04.048Get rights and content

Highlights

  • We model a novel conceptual system for ultra high temperature energy storage.

  • Operation temperature exceed 1400 °C, which is the silicon melting point.

  • Extremely high thermal energy densities of 1 MWh/m3 are attainable.

  • Electric energy densities in the range of 200–450 kWh/m3 are attainable.

  • The system can be used for both solar and electric energy storage.

Abstract

A conceptual energy storage system design that utilizes ultra high temperature phase change materials is presented. In this system, the energy is stored in the form of latent heat and converted to electricity upon demand by TPV (thermophotovoltaic) cells. Silicon is considered in this study as PCM (phase change material) due to its extremely high latent heat (1800 J/g or 500 Wh/kg), melting point (1410 °C), thermal conductivity (∼25 W/mK), low cost (less than $2/kg or $4/kWh) and abundance on earth. The proposed system enables an enormous thermal energy storage density of ∼1 MWh/m3, which is 10–20 times higher than that of lead-acid batteries, 2–6 times than that of Li-ion batteries and 5–10 times than that of the current state of the art LHTES systems utilized in CSP (concentrated solar power) applications. The discharge efficiency of the system is ultimately determined by the TPV converter, which theoretically can exceed 50%. However, realistic discharge efficiencies utilizing single junction TPV cells are in the range of 20–45%, depending on the semiconductor bandgap and quality, and the photon recycling efficiency. This concept has the potential to achieve output electric energy densities in the range of 200–450 kWhe/m3, which is comparable to the best performing state of the art Lithium-ion batteries.

Introduction

LHTES (Latent heat thermal energy storage) employs energy to cause the phase change transition in a material that subsequently stores energy in the form of latent heat. That material is referred to as PCM (phase change material) and is the key element determining the overall performance of the storage system. PCMs promises one of the highest energy densities and lowest costs of existing LHTES materials [1], [2]. However, current LHTES solutions are subject to a very low heat extraction rate from the storage medium, which is attributed to the low solid-phase thermal conductivity and moderate latent heat of PCMs. Current research efforts focus on developing relatively sophisticated PCM encapsulation so that thermal conductivity is notably enhanced [1], [3], [4], [5], [6]. However, these strategies inherently have a lower energy density potential, since part of the volume is dedicated to the PCM host.

An alternative solution consists of directly using PCMs with higher thermal conductivity and latent heat. As a general rule, the heat of fusion of materials increases with melting temperature [1], [7]; thus, there is an interest on moving towards higher melting point PCMs. However, in LHTES for power generation there is a maximum temperature imposed by the HTF (heat transfer fluid) that is used to carry the heat from the PCM to the heat engine, which degrades at high temperatures. Maximum temperatures are typically below 500 °C [8]. Other technological options exist though; such as TPV (thermophotovoltaics) [9], [10], thermionic [11] or hybrid thermionic-photovoltaic [12] devices that do not require HTFs and consequently have the potential to operate at extraordinary high temperatures.

Previous works have proposed conceptual system designs for solar thermal energy storage based on very high melting point PCMs, such as pure silicon and boron (melting points of 1410 °C and 2076 °C, respectively) and TPV converters [13], [14], [15], [16], [17], [18], [19]. The first experiments on molten silicon for LHTES applications have been recently carried out at the University of South California with the aim of developing a solar thermal propulsion system for microsatellites [20]. The container damage due to freezing expansion of pure silicon was the most relevant engineering concern. Besides, a highly asymmetric freezing profile was observed due to the use of non-adiabatic container walls, which produced regions of molten silicon encased in solid silicon that ultimately resulted in high stress and container damage. It is worth noting that these issues could be solved in future designs by several means, such as using quasi-adiabatic container walls, i.e. improved container thermal insulation, or reducing the container fill factor [20]. In the opinion of the authors, an especially interesting solution consists on using silicon alloys instead of pure silicon, in order to reduce the freezing expansion coefficient of the PCM. We believe that, among all the possibilities, the silicon-boron system is particularly interesting due to the extremely high latent heat of boron (4650 J/g) and the moderately low melting temperature (1385 °C) for the eutectic Si0.92B0.08 [21], [22]. Besides, the silicon lattice parameter contracts upon alloying with boron [21], which suggests that freezing expansion issues could be eliminated. Other practical concern is the thermo-chemical compatibility between the container and the PCM at those high temperatures. A variety of refractory materials have been extensively used for casting solidification of all kind of metals at high temperatures, including silicon and boron. Some examples are BN (used in Ref. [20]) SiC and Si3N4. In these cases, important selection criteria are the wetability, solubility and reactivity of the container with the PCM, along with evaporation and oxidation of the PCM in oxygen-rich atmospheres. Thus, it is still needed an extensive research on the compatibility of these materials for the particular application of LHTES, especially concerning cycling and long term reliability.

In this work we present a conceptual LHTES system design for both S2H2P (solar-to-heat-to-power), commonly referred to as CSP (concentrated solar power), and P2H2P (power-to-heat-to-power) applications. Notice that other P2H2P concepts have been previously proposed to store the excess of electricity in the grid and co-generate heat and electricity [23]. The concept presented in this paper is based on the same operation principles than previously proposed systems in Refs. [13], [14], [15], [16], [17], [18], [19], i.e. high temperature PCM and TPV energy conversion. The proposed system differentiates from the previous designs in its geometrical configuration, which facilitates the thermal insulation of the PCM and the integration of an independent (mobile) TPV generator, providing a tunable power discharge rate (from zero to full-discharge mode). We assess this concept theoretically to predict its performance under several assumptions, which range from idealistic (to provide the upper bounds of this concept) to more realistic, to provide its actual short-term potential.

Section snippets

System description

Fig. 1 shows two possible configurations of the LHTES system presented in this paper for P2H2P [24] (left-hand side) and S2H2P or CSP (right-hand side) applications. In the P2H2P case, an electric heating system is used for melting the PCM. Among all the possible options, an inductive electric heater could be used if the PCM is magnetic or electrically conductive (e.g. iron or metallurgical silicon). Other options include resistive or microwave heating. In any case, electrical energy is stored

System model

In order to describe the transient performance of the LHTES system, we assume a quasi-1D analytical model in which the solid–liquid interface is a moving cylinder at a distance rm(t) from the axial center of the system (Fig. 3). To solve the problem we follow the quasi-stationary approach used in Ref. [15] assuming an adiabatic (loss-less) container and neglecting natural convection in the liquid. Natural convection in the liquid can be disregarded, as we will see later, due to the very low

System performance with ideal TPV cells

In this section, the TPV cells are assumed to be ideal (i.e. ηint = 1) with a bandgap of 0.5 eV, which could be manufactured for instance using InGaAsSb alloys on GaSb substrates. Under this assumption, Equations (2), (3), (4), (5), (7) are solved for the silicon parameters listed in Table 1 and Table 2

Table 3 shows the model results for the discharge of the LHTES system with different geometries and for two values of ρBR (the ideal case of ρBR = 1 and a more realistic one of ρBR = 0.8).

System performance with realistic TPV cells

Realistic TPV cells are modeled in this work by introducing the internal and external photoluminescent efficiency, ηint and ηext respectively, which account for non-radiative recombination. The best performing III-V semiconductor based PV cells have demonstrated ηint and ηext values above 95% and 35%, respectively [31]. However, semiconductors with non-direct transitions between valence band and conduction band, such as silicon, have much lower efficiencies, in the order of ηext ∼0.1–1% [32].

Conclusions

A conceptual LHTES system utilizing high temperature silicon PCM and thermophotovoltaic cells has been presented. The proposed LHTES system is fully scalable in terms of power (from kW to MW), energy (from tens of kWh to tens of MWh) and discharge time (hours to days) and enables an ultra high thermal energy storage density of up to ∼ 1 MWh/m3. The attractiveness of this concept, besides the extreme energy density, is the possibility of using silicon as PCM, the second most abundant element on

Acknowledgment

Authors acknowledge the financial support of the Comunidad de Madrid through the Programme MADRID-PV (Grant number S2013/MAE-2780) and from the Spanish Ministerio de Economía y Competititvidad through the Project PROMESA (grant N. ENE2012-37804-C02-01).

References (33)

  • A. Datas

    Optimum semiconductor bandgaps in single junction and multijunction thermophotovoltaic converters

    Sol Energy Mater Sol Cells

    (2015)
  • H. Chen et al.

    Progress in electrical energy storage system: a critical review

    Prog Nat Sci

    (Mar. 2009)
  • X. Luo et al.

    Overview of current development in electrical energy storage technologies and the application potential in power system operation

    Appl Energy

    (Jan. 2015)
  • T. Bauer, W. D. Steinmann, D. Laing, and R. Tamme, “Thermal energy storage materials and systems,” in Annual review of...
  • J.W. Raade et al.

    Development of molten salt heat transfer fluid with low melting point and high thermal stability

    J Sol Energy Eng

    (Jul. 2011)
  • T. Bauer

    Thermophotovoltaics: basic principles and critical aspects of system design

    (2011)
  • Cited by (113)

    View all citing articles on Scopus
    View full text