Development of a medium-high temperature waste heat recovery hybrid thermal energy storage layout, based on red mud, a disregarded and potentially hazardous solid waste of the aluminium industry.

Project Information

REDTHERM

Grant agreement ID: 101068507

DOI

10.3030/101068507

Project closed

EC signature date 30 September 2022

Start date 1 January 2023

End date 28 February 2025

Funded under

• Marie Skłodowska-Curie Actions (MSCA)

Total cost

No data

EU contribution

€ 155 994,72

Coordinated by

ARISTOTELIO PANEPISTIMIO THESSALONIKIS
Greece

Periodic Reporting for period 1 - REDTHERM (Development of a medium-high temperature waste heat recovery hybrid thermal energy storage layout, based on red mud, a disregarded and potentially hazardous solid waste of the aluminium industry.)

Reporting period: 2023-01-01 to 2025-02-28

Summary of the context and overall objectives of the project

As the necessity toward climate neutrality becomes of paramount importance, increasing energy efficiency and minimizing industrial waste are critical priorities. A significant portion of industrial energy—particularly in high-temperature processes such as ceramics, metallurgy, and aluminium production—is still lost as waste heat. Capturing and reusing this heat represents an enormous opportunity to reduce greenhouse gas emissions, enhance resource efficiency, and lower operational costs across a wide range of sectors.

The REDTHERM project was developed in response to this challenge, aiming to demonstrate an innovative approach to thermal energy storage (TES) using red mud (RM)—a common, energy-intensive industrial by-product of the aluminium industry. REDTHERM sought to combine sustainable material valorization with advanced system engineering to deliver a new generation of hybrid waste heat recovery (WHR) units. The ambition was to transform red mud from a costly waste stream into a core component of a scalable energy storage system.

The core objective of REDTHERM was to develop and validate, at laboratory scale, a real Waste Heat Recovery-Thermal Energy Storage unit based on Red Mud-derived phase change materials (PCM). This system was designed to work under industrially relevant conditions, bridging the gap between early research and full-scale deployment. Beyond material development, the project emphasized intelligent design, modeling, and optimization—leveraging artificial intelligence (AI) and computational fluid dynamics (CFD) to simulate performance, improve efficiency, and guide industrial scale-up.

A central innovation of the project was its dual-tank hybrid TES configuration, capable of operating in both parallel and serial modes, offering unprecedented flexibility in capturing and releasing thermal energy. Furthermore, the system used a stratified cascade of PCMs with different melting points, maximizing thermal storage density across a broad temperature range.

REDTHERM’s approach reflects European strategies in energy transition, digital transformation, circular economy, and industrial competitiveness. Its outputs—spanning advanced materials, prototype hardware, and intelligent design tools—are intended to contribute directly to EU goals related to decarbonization, resource efficiency, and sustainable reindustrialization, with specific alignment to the European Green Deal, Industry 5.0 and Circular Economy Action Plan.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The REDTHERM project followed a structured research and development pathway to achieve its objective of designing, building, and optimizing a hybrid thermal energy storage (TES) unit using red mud-based composite phase change materials (CPCMs).

1. Prototype Development and Experimental Evaluation:
The core technical achievement of REDTHERM was the construction and successful testing of a novel dual-tank TES prototype capable of both serial and parallel operation—a unique feature in these type of systems. The pilot was tailored for medium-high temperature applications and operated under realistic laboratory conditions, with air as the heat transfer fluid at temperatures up to 500 °C and flow rates of 900 L/min. It was designed to replicate industrial WHR settings while enabling high experimental control. Using a cascade of four distinct CPCMs with melting points between 150 °C and 310 °C, the system achieved a total storage capacity of ~30.5 kWh. Charging and discharging cycles were conducted to assess performance, with a charging efficiency of 82%, discharging efficiency of 76%, and total round-trip efficiency of 62% under serial operation.

2. Material Scale-Up and Characterization:
The red mud CPCM materials were successfully scaled up from lab to pre-industrial levels in collaboration with manufacturing partners. These CPCMs were cast into large composite ceramic plates (30x27x3.5cm) and evaluated for thermal stability, mechanical stability, latent heat, and thermal conductivity.

3. Advanced Modeling and AI-Based Optimization:
To support system design and improve predictability, a suite of Computational Fluid Dynamics (CFD) models were developed using COMSOL and ANSYS platforms. These models were validated against literature and internal data to ensure accuracy. Parallel to this, a deep learning model was trained to predict storage performance, achieving an R² of 0.975. In addition, this was couple with metaheuristic optimization algorithms to generate optimized high-performing TES configurations. This tool supports both performance optimization and early-stage techno-economic evaluation. To enhance understanding and guide design choices for the tool, Explainable AI (XAI) methods such as SHAP and LIME were applied, identifying key performance drivers. The project also developed new multimodal packing strategies using various particle sizes and particle size distributions, which improved energy density by up to 15% and reduced melting times by 8.6%, as confirmed by a validated numerical study.

4. System-Level Scale-Up and Techno-Economic Evaluation: A scale-up case study was conducted based on real industrial data from a Greek ceramic factory and energy demand from an office building in the Central Macedonia region. The system was optimially scaled to meet this demand using the AI tool, and a techno-economic analysis was performed to assess cost-effectiveness, payback periods, and energy savings. The analysis used very conservative performance assumptions and gas/electircity pricing demonstrating that the system could be economically viable even under real-world conditions with a payback period of 5.3 years.

Collectively, REDTHERM’s scientific achievements demonstrate the feasibility of combining waste valorization, thermal storage innovation, and AI-driven optimization to address the challenges of energy efficiency in industry. The work lays a solid foundation for future upscaling and deployment of WHR-TES systems based on circular materials like red mud.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The REDTHERM project has produced several key advancements that push the boundaries of current capabilities in thermal energy storage (TES) and waste heat recovery (WHR):

1. Dual-Mode Hybrid TES System Design:
A novel two-tank hybrid TES layout was designed and experimentally validated, enabling operation in both parallel and serial configurations. This modular and switchable flow design allows tailoring charge/discharge strategies to different heat stream profiles—addressing a limitation in existing packed bed latent heat systems, which are typically locked into a single flow logic.

2. Scale-Up of Red Mud-Based CPCMs via Industrial Manufacturing:
REDTHERM successfully demonstrated the manufacturing-scale production of composite phase change materials (CPCMs) based on red mud (RM), an industrial waste material. Unlike prior lab-scale work, the CPCMs were fabricated using real-world industrial equipment in collaboration with Roka Refractories—bridging the gap between academic material development and market-ready implementation.

3. Integration of Deep Learning for TES System Prediction and Design:
The project introduced deep neural network models for accurate prediction of CPCM based TES system performance—both charging and discharging—based on multiple design and operational parameters. These models reached R² values exceeding 0.97 and were able to susbstantially accelerate and improve existing experimental setups.

These outcomes demonstrate REDTHERM’s core impact: the creation of scalable, data-driven, and sustainable TES systems that directly target the operational complexity and economic barriers of industrial WHR integration.

Lab-scale prototype developed under REDTHERM featuring red mud–based materials