Assessing the use of copper slags as thermal energy storage material for packed-bed systems

Highlights

• Assessment of the use of copper slags as filler material in packed-bed systems.

• Thermal characterization of the copper slags from samples from local foundry.

• Copper slag shows better thermal capacitance than other industrial byproducts.

• Development of a 1D model for assessing the packed-bed thermal performance.

• Study of the impact of the aspect ratio and tank’s volume on energy efficiency.

Abstract

Thermocline tanks using packed-bed of rocks have become feasible candidates for improving the performance of Concentrated Solar Power plants, enabling high operating temperatures and reduced capital costs when industrial byproducts are employed as filler materials and low-cost working fluids, being competitive against molten salts thermal storage systems. The present work assesses the potential of using copper slags in packed-bed systems as filler material. Through a thermal characterization, it is demonstrated that copper slags show similar properties to other slags proposed in the literature for thermal storage medium and better thermal capacity (1.4–1.5 J/(gK)). A heat transfer model was developed to predict the cyclic behavior of a packed-bed storage using copper slags and employed in a parametric analysis to assess the impact of storage dimensions on 1^st and 2^nd law efficiencies for different storage materials, allowing to identify several design considerations depending on tank’s volume. The main findings indicate that the high thermal capacity of copper slags favors the development of a steeper thermocline, keeping a low rate of exergy loss at storage’s outlet, and also higher energy density stored of 138 kWh/m³ against 129 kWh/m³ of other byproducts under similar storage dimensions.

Introduction

The current trends in energy supply and use are highly unsustainable socially, economically, and environmentally. The need for a substantial change on the development path has encouraged the scientific community to study environmentally friendly energy systems as a measure to achieve a substantial reduction of greenhouse gas emissions. In that context, one of the main difficulties for increasing the share of renewable energy in the energy matrix is the availability of reliable and affordable energy storage solutions. Energy storage technologies can contribute to a better integration of electricity and heat network systems, playing a crucial role in improving energy systems’ efficiency and enabling the introduction of large share sustainable sources [1]. For instance, Concentrated Solar Power (CSP) plants are equipped with thermal energy storage (TES) systems, which provide heat in a dispatchable way to a power block, and give operational flexibility even during periods of low solar radiation.

CSP plants commonly use molten salt mixtures as heat transfer fluid (HTF), which is also employed as sensible heat material for thermal storage due to its high thermal capacity. The most common mixture of salts in CSP plants is 60% KNO₃ and 40% NaNO₃ (weight) [2], which is commonly called “Solar Salt.” However, the use of molten salts has several techno-economical disadvantages, such as a higher operational temperature of around 600 °C [3] due to their chemical stability limit. That constraint also represents a thermodynamic limit for achieving higher conversion efficiencies through the implementation of advanced and efficient power cycles. In addition to that, the salt must not operate at temperatures lower than the freezing point, which is 228 °C for solar salt [4].

The CSP industry and the scientific community have devoted significant efforts during the last years to develop new concepts in central receiver systems and thus achieve higher energy conversion efficiencies. In that context, the utilization of compressible gases cycles has received considerable attention in recent years. Among the cycles analyzed, the use of supercritical CO₂ and atmospheric air have been pointed out as interesting candidates for the future developments, since these operate in a wider temperature range and achieve higher conversion efficiencies [5,6].

Atmospheric air has been analyzed since the early 90s as a potential working fluid in CSP plants [7]. The use of atmospheric air as a working fluid has several advantages: it is nontoxic, low-cost, readily available, environmentally friendly, and chemically stable at high temperatures [8]. This last feature is crucial for achieving a higher conversion efficiency [9]. In that regard, several prototypes of volumetric receivers have been tested [10], which have led to the construction of two demonstrative plants using air as a working fluid: Jülich (Germany - 1.5 MW) and Daegu (Korea - 200 kW). Although those plants were able to validate the concept, the technology still has to overcome several challenges to achieve its commercial maturity. One of those critical issues is the integration with thermal storage systems [11].

One of the strengths of using compressible gases as a working fluid is the possibility of exploring thermal storage options with higher energy density and lower costs than the molten salts’ technology (as shown in Table 1). Among the storage technologies that have been analyzed for coupling to CSP systems using air as a working fluid, single-tank packed-bed storage has been pointed out by several authors as the most suitable alternative [12,13].

A packed-bed thermal storage system consists of using solids as a heat storage medium and a heat transfer fluid (HTF) in direct contact with solids to convey heat [18]. Fig. 1 shows a generic scheme for a central receiver plant coupled to a packed-bed storage where the solar radiation is absorbed in the receiver of the solar tower, and the heat transfer fluid is later conveyed to the power block. In cases when a surplus of energy exists, it is conveyed to the thermal storage system. Later, during low radiation periods, the thermal storage is discharged by delivering heat to the power block, allowing it to continue its operation. Comparing packed-bed systems with the currently dominating storage technology in CSP plants (molten salts), the main advantages of packed-bed systems are the following: the operating temperature constraints due to chemical instability of the HTF or the rocks are eliminated; the operating pressure can be close to ambient, avoiding the need for complex sealings; and thermal storage can be incorporated directly after the receiver, eliminating the need for heat exchangers between the HTF and the thermal storage medium [12].

Some of the disadvantages of packed-bed systems using air as HTF are associated to the large mass flow rate and the surface area needed, due to air’s low volumetric heat capacity and thermal conductivity [12]. These drawbacks imply higher pressure drops and energy losses [13]. That issue is explained because the absolute and relative dimensions of the tank as well as the solid elements of the bed influence flow distribution and velocity profiles, significantly affecting the heat transfer phenomena and the thermal stratification in the storage [24]. Research has demonstrated that the relative influence of the walls over velocity profiles is higher in small tanks than in large tanks due to edge effects [25]. Flow channeling near the walls is also influenced by the tank-to-particle diameter ratio, which must be carefully addressed for storage system’s yield assessment [24].

In terms of the design variables of the tank that affect the thermal behavior of the storage system, the cross section area is fundamental. Although most of the systems reported a cylindrical shape [12,26,27], recent works have opted for a truncated conical tank [25,28], aiming to reduce the thermomechanical stresses observed in cylindrical tanks (even though it reduces its thermal performance and increases the pressure drop). Consequently, the design of packed-bed storage systems results in a trade-off evaluation among mechanical performance, flow distribution, pressure drop, thermal losses, and stratification easiness. The packed-bed systems reported in the literature commonly present height and diameter of the same order of magnitude [24], which increases compactness of the storage and hence reduces thermal losses.

Achieving high efficiency for energy storage in packed-beds is closely related to keeping a high degree in thermal stratification during cycling operation. Research has shown that high energy recovery is achieved when a small amount of mixing occurs between the hot and cold zones in the storage tank. Thus, it is crucial for the control system and the shape of the thermal front. Fluid flow conditions also need to be considered (e.g. the minimization of the pressure drop allows the extraction of more useful energy from the system [27]).

Many studies have proposed assessing the performance of packed-beds through the first law of thermodynamics, or what is called “round-trip efficiency.” The latter is defined as the ratio between the useful energy released from the storage system during the discharging process and the energy delivered to the system during the charging process [29,30]. Li et al. [31,32] carried out a thorough analysis that considered multiple dimensionless numbers. The authors stated that the first law’s efficiency (η) for these systems is primarily a function of the following: the ratio between the dimensionless charge and discharge time, a dimensionless quantity that relates the mass flow rate, the heat capacity of the fluid, the height of the packed-bed, and the surface area of the filler material per unit length; and the ratio between the fluid heat capacitance and the filler heat capacitance.

Other authors have indicated that thermal storage systems used for power generation should be designed and evaluated by methods considering the 2^nd law of thermodynamics [33] (exergy). Exergy analysis assesses the potential power losses generated from operational conditions, such as pressure drop and its effect on the energy required for pumping [34]. Torab and Beasley [35] showed the existence of a tipping point where the total exergy available increases with decreasing particle diameter and enlarging the height of the storage tank, resulting in pumping energy requirements. Moreover, regarding the mass flow rate, exergy efficiency is higher for smaller flows because the thermal diffusion and mixing effects produce low exergy destruction [36]. McTigue et al. [37] went one step forward in the assessment, establishing a balance between the entropy rejected at the exit of the storage and the internal entropy generated by irreversibilities. That balance allowed them to determine a trade-off between round-trip efficiency and energy density.

As reported in the literature, the pilots of packed-bed storage systems implemented are mainly at the laboratory and prototype scales, while very few at the commercial scale [24]. Several filler materials have been tested (e.g. natural rocks, asbestos, concretes, and industrial byproducts) and assessed in terms of cost reduction potential. One of the potential options is using industrial byproducts that have an extremely low cost and favorable thermal properties [38,39].

Filler materials in packed-bed systems should be able to operate in a wide range of temperatures (from 50 °C to 1100 °C) and have a stable performance for a large number of thermal cycles, as reported in several studies [38,[40], [41], [42]]. The use of recycled materials particularly from metallic industrial processing, such as the steel making process and its variations (electric arc furnace, induction furnace), has been assessed due to high availability, the large amount of material produced, and the thermal performance of the slag. Additionally, due to the absence of a market for industrial slag, its cost is practically null as compared to other sensible heat storage materials presented in Table 1.

Ortega-Fernández et al. [22] concluded that steel slag from the electric arc furnace (EAF) process is thermally stable up to 1100 °C; moreover, the structure of the slag influences in the thermal properties as it increases its temperature: higher crystallinity results in a decrease of the specific heat and an increase on thermal conductivity (See Table 3). Wang et al. [43] analyzed the thermophysical properties of EAF slags as well. However, they considered samples from two different countries and included a wear behaviour analysis of the slag rocks, concluding their suitability for TES in a packed-bed configuration. As for other types of industrial byproducts, copper slag from the pyrometallurgical processing of copper ore has been brought to discussion in the last decade, as reported by Navarro et al. [39], where the thermophysical properties of copper slag as an aggregate for mortar mixtures and other types of recycled materials were characterized and compared to molten salts based on their techno-economical advantages. This case study determined that these materials performed better in terms of energy density at a lower cost than molten salts.

A parametric study developed by Ortega-Fernández et al. introduced guidelines for the optimal design of a packed-bed using industrial byproducts. The authors analyzed the behavior of storage capacity and round-trip efficiency by varying the aspect ratio of the storage, mass flow rate, and particle diameter. Hence, for a fixed mass flow rate, aspect ratios close to 2 showed promising results in terms of thermal performance. These results also suggest the importance of maintaining a certain balance between mass flow rate and particle diameter to compensate for heat transfer mechanisms in storage and prevent the spread of the thermocline [44].

The first attempt at using copper slag as sensible heat material was proposed in 1980 by Curto [45], due to the advantages offered by its thermal properties, especially considering that this byproduct has a high percentage of ferrous compounds. In addition to that, up to date copper slag has no subsequent use, and therefore, it has a negligible value. It is a waste material with high availability, especially in Chile.

As stated by Curto [45], the application of copper slags as a filler material in packed-bed storage systems has significant advantages for being used in gas power cycles and industrial processes. It could be used in a relatively high operation temperature range: from 650 °C to 1200 °C, remaining chemically, thermally, and mechanically stable. The thermophysical properties reported initially ranged between 4.3 and 4.4, for the specific gravity, while the heat capacity ranges between 0.670 kJ/(kg K) to 1.004 kJ/(kg K), within a temperature range between 20 °C and 1000 °C. Curto [45] assessed the use of copper slags in a packed-bed tank of 21,000 m³ using air as HTF and showed outstanding performance. Despite the high potential and cost effectiveness exhibited by copper slags, this performance has not been sufficiently studied in the literature or in engineering applications.

The present work describes the assessment of the potential use of copper slags as filler material in packed-bed storage systems. One of the contributions of the present study is extending the characterization of copper slags by including data of slag from two Chilean foundries, considering the high heterogeneity expected in the thermophysical properties of copper slags. Such heterogeneity is due to their origin and solidification rates after the smelting process. Thus, the variations in the thermophysical properties impact significantly on their performance as a storage material. Another achievement of the present work is a comparison of the thermal performance of copper slags against other industrial byproducts commonly used, and considering several storage configurations.

The article describes the processes for the thermal characterization of the material (section 2). Considering the thermal features of the slag, a numerical 1D model is developed and described in section 3, which has been validated against data from the literature (subsection 3.2). The analysis considers 1^st and 2^nd law efficiencies as performance indicators. A parametric analysis is conducted for assessing the performance of packed-bed storage systems using copper slags, configured in applications of constant Reynolds number. The proposed analysis aims to identify the best application scenarios for the slags and to extend the methodology initially proposed in Ref. [44], allowing a direct comparison between the systems (section 4). Finally, a detailed analysis of the results is presented in terms of the round-trip and exergy efficiencies (section 5).