Review of the Properties and Degradation Mechanisms of Refractories in Aluminum Reduction Cells

Abstract

This review examines the degradation of refractory materials in aluminum reduction cells, focusing specifically on contamination caused by the cryolite-based bath. Aluminosilicate refractories, particularly Ordinary Refractory Bricks, play a vital role in maintaining the structural integrity and thermal balance of these cells under demanding operational conditions. The interaction between the molten bath and refractory linings leads to chemical reactions and mineralogical changes that modify the mechanical and thermal properties of the material over time. The study integrates findings from industrial autopsies, laboratory experiments, and a comprehensive review of the existing literature to identify and analyze the mechanisms of degradation. By analyzing the findings obtained from these methodologies, this review explores how cryolitic infiltration triggers transformations that compromise performance and reduce the lifespan of refractory linings. Covering a broad temperature range (665–960 °C), the study addresses key challenges in understanding bath-induced contamination and provides insights into how to improve the durability and efficiency of refractory materials in aluminum production.

1. Introduction

The performance and lifespan of aluminum reduction cells depend heavily on the quality and durability of their refractory linings, particularly aluminosilicate-based Ordinary Refractory Bricks (ORBs) [1]. These materials serve as essential barriers, protecting insulation layers from extreme temperatures and aggressive chemical environments. With the development of high-amperage technologies such as AP40, the degradation mechanisms affecting ORBs have evolved. Higher operational temperatures, especially along the sidewalls, now subject the refractories to more severe conditions than those encountered in earlier cell designs [2,3].

Over the years, extensive research has been conducted to understand the degradation of aluminosilicate refractories in aluminum reduction cells. Foundational studies, such as those by Siljan et al. [4] and Tschöpe et al. [5], explored the chemical and mineralogical transformations caused by interactions with sodium vapor and cryolitic bath components. These works highlighted key degradation mechanisms, including the formation of nepheline (NaAlSiO₄) and albite (NaAlSi₃O₈), which significantly weaken the mechanical and thermal properties of refractories. Similarly, studies like those by Proshkin et al. [6] underscored the challenges posed by bath infiltration, emphasizing the importance of enhancing refractory resistance to chemical attack. However, much of the existing literature focuses on conditions near or above the sodium vaporization point (883 °C), where sodium vapor plays a dominant role in degradation [7]. This perspective overlooks critical temperature ranges, particularly those below 883 °C, where sodium vapor activity diminishes, and the cryolitic bath becomes the primary reactive agent [8].

Recent observations from industrial autopsies and laboratory studies reveal that the bath begins to react with ORBs at temperatures as low as 700 °C. This highlights a significant gap in the understanding of bath-induced degradation, which is particularly relevant for modern high-amperage cells [2,8].

To address this gap, experiments were conducted to replicate real-world conditions by exposing ORBs to cryolitic bath contamination across a temperature range of 700–960 °C. These tests revealed a progression of mineralogical transformations, including the formation of nepheline and albite, consistent with those observed in industrial autopsies [8,9]. The results confirm that at temperatures below 883 °C, the bath is the primary driver of degradation [8,10]. These findings align with the industrial data, showing that cryolitic bath interactions are especially significant in the sidewall regions of modern cells, where temperatures frequently fall within this range [11].

This review synthesizes existing research with these recent findings to provide a comprehensive understanding of ORB degradation. By focusing on bath-refractory interactions across a broader temperature range, it highlights the evolving challenges posed by high-amperage technologies and offers insights for developing more resilient refractory materials.

The review is intended for researchers, material scientists, and industry professionals working in aluminum production. It explores key aspects of refractory degradation and highlights strategies for optimizing their performance in demanding operational environments. By connecting academic research with real-world applications, this work aims to be both scientifically rigorous and practically valuable, offering meaningful contributions to the advancement of the field.

2. Production of Primary Aluminum

2.1. The Hall Héroult Process

Aluminum, like most metals, does not exist in its pure state in nature but rather in the form of oxides and silicates [12]. The industrial technique currently used to produce primary aluminum was independently developed by Louis-Paul Toussaint Héroult and Charles Martin Hall in 1886 [13]. In the Hall–Héroult process, aluminum is obtained in its liquid phase through the reduction in alumina (Al₂O₃, extracted from bauxite) dissolved in an electrolytic bath. The overall electrochemical reaction in the reduction cell [14] can be written as follows:

2 Al₂O_3(dissolved) + 3 C_(anodes) → 4 Al_(liq) + 3 CO₂

(1)

At the anode, the following reaction [14] leads to the formation of CO₂:

O²⁻ + 1/2 C(anodes) → 1/2 CO₂(g) + 2e⁻

(2)

Due to its lower density, the electrolytic bath floats above the aluminum. The main functions of the bath are to dissolve alumina, conduct current between the anode and cathode, and generate the heat required for reduction through the Joule effect [15,16]. The electrolyte also acts as a separator between the two main products of reduction: aluminum produced at the cathode and CO₂ produced at the anode.

The cathodic reaction leading to the reduction in aluminum [15] can be simplified as follows:

Al³⁺ + 3 e⁻ → Al_(liquid)

(3)

Since molten aluminum has very high electrical conductivity, the interface between the two fluids (bath/metal interface) serves, from an electrochemical perspective, as the cathode [17]. Thus, aluminum is produced at this level.

As depicted in Figure 1, a cell comprises a multitude of specific components to maintain good operating conditions regarding the mechanical, electrical, thermal, and chemical issues [18]. The reduction process operates in a specialized steel shell inside which all the other components are installed to ensure the efficient mechanical confinement of these components as well as the thermal balance required by the process [3]. The lining of the sidewall and the bottom of the cell includes several insulating materials, such as refractory bricks. In recent high-amperage cell technologies, the sidewalls also include graphitized carbon-based blocks and silicon carbide blocks bonded by silicon nitride [19]. Due to the thermal gradient over the sidewall, a ledge of solidified bath forms on the sidewall, limiting the degradation of the cell due to the progression of the bath toward the refractory bricks [19]. The anodes are primarily prepared by baking a mixture of crushed petroleum coke and coal tar pitch at 1100 °C, mixed at 160 °C, and shaped by pressing or vibro-compaction [20]. These anodes supply electrical current to the reduction zone and provide the necessary carbon for reaction. Each anode has a lifespan of approximately 24 days [20]. The cathodes are placed on the refractories at the bottom of the cell and consist of carbon blocks in which steel collector bars are sealed to collect electrical current. They also contain ramming paste, which ensures the cathodic plane is sealed against liquid infiltration [14]. The molten aluminum is a highly homogeneous, incompressible, viscous fluid with a density of approximately 2400 kg/m³, higher than that of the bath [21]. The heavier metal formed settles at the bottom of the reduction cell. The injection of alumina into the cells is automated and nearly continuous. Two essential operations are still performed manually: the recovery of molten aluminum by suction and the regular replacement of consumed anodes [20]. The molten electrolyte bath predominantly contains cryolite (Na₃AlF₆), a solvent for alumina. The alumina content is approximately 5% the mass of the bath. Cryolite melts at 1010 °C, but the addition of additives such as AlF₃ and CaF₂ lowers the melting point to around 960 °C [22].

Generally, two criteria are used to measure the efficiency of this manufacturing process: Faraday efficiency and energy efficiency. Faraday efficiency is the ratio between the quantity of aluminum produced and the theoretical quantity predicted by Faraday’s law, considering the supplied electric current [23]. It reaches 95% in the best cells, with the average being around 90% [23]. The values are calculated using Faraday’s equation as follows [23]:

$$m={\displaystyle \frac{It}{nF}}$$

(4)

where $m$ is the quantity of material produced (or consumed) in moles, $I$, the current (A), $t$, the time (s), $F$, the Faraday’s constant ($F=\mathrm{96,485}$ C/mol), and $n$ is the number of electrons exchanged during the electrochemical reaction [23]. It is noticeable, among other things, that the productivity of an electrolytic cell, meaning the quantity of aluminum produced over time, depends only on the electric current imposed in the cell.

Various factors contribute to this efficiency. Part of the produced aluminum tends to recombine with carbon dioxide according to the reverse reaction [14]:

2 Al_(l) + 3 CO_2(g) → Al₂O_3(dissolved) + 3 CO_(g)

(5)

This is the essential cause of the decrease in Faraday efficiency. To limit this reaction, efforts are made to increase the bath’s density and interfacial tension (to prevent molten aluminum intrusion into the bath). This can be achieved by decreasing the electrolyte’s temperature and modifying its chemical composition; for example, by adding AlF₃ [21]. Other factors, such as agitation of the bath/metal interface, gas evolution at the anode, or too low an alumina concentration, also contribute to reductions in Faraday efficiency.

Energy efficiency is a more significant yet challenging quantity to define. It refers to the ratio between the thermodynamic energy required for the oxidation-reduction reaction and the electrical energy supplied to the cell. This ratio particularly considers Faraday efficiency [23].

Pelletier, Allaire et al. [24] elaborate extensively on the significance of Faraday efficiency and cell voltage concerning energy consumption. Their work underscores the importance of maintaining the lowest possible voltage during reduction.

2.2. The Electrolytic Bath

Various authors have studied the phase diagram of the NaF-AlF₃ system [23,25,26]. This diagram is characterized by specific compositions [27], which are summarized in

Table 1. Key compositions and characteristics of the NaF-AlF₃ phase diagram.

Composition	Description	Melting Point (°C)
Na3AlF6 (Cryolite)	Congruent melting compound at 25 mol% AlF3	1010
Na5Al3F14 (Chiolite)	Incongruent melting compound at 37 mol% AlF3	734
Eutectic Point 1	Located at 13.8 mol% AlF3	888
Eutectic Point 2	Located at 45.2 mol% AlF3	695
Peritectic Point	Results from the incongruent melting of chiolite, positioned between 39.4 and 40.8 mol% AlF3	734

.

The composition of the electrolytic bath used in the aluminum reduction process is defined by the “Cryolitic Ratio: CR,” which corresponds to the ratio of the molar fractions of NaF and AlF₃. A ratio of 3 represents pure cryolite. In most modern reduction cells, “acidic” ratios are used, indicating an excess of AlF₃ compared to cryolite, with the cryolitic ratio often set at 2.2 [26,28].

The ionic structure of the electrolytic bath is complex and remains a topic of discussion in the literature. Numerous models have been proposed, generally agreeing that only sodium is present in cationic form (Na⁺), while other species form anionic complexes [29]. In 1924, a scheme illustrating how cryolite dissociates during fusion and specifying the chemical species present in cryolitic baths was proposed [30]. A total dissociation of cryolite into Na⁺ and AlF₆³⁻ occurs according to the following reaction:

Na₃AlF₆ → 3 Na⁺ + AlF₆³⁻

(6)

Subsequent dissociation of the AlF₆³⁻ ion occurs as follows:

AlF₆³⁻ → AlF₄⁻ + 2 F⁻

(7)

Initially, as the only positive ions in the bath, Na⁺ ions accumulate at the cathode surface due to the electric field. Aluminum, present in the anionic complexes of fluorine and oxygen, is discharged at the cathode, while NaF moves away [29]. From an ionic standpoint, aluminum is not bound to sodium, so only F⁻ ions and aluminum complexes have a net diffusion movement, making F⁻ the only current carrier in the boundary layer (since it must move away from the cathode when aluminum is discharged to maintain the electroneutrality of the mixture) [29].

It is also observed that diffusion processes lead to an increase in the net concentration of NaF at the surface of aluminum. This contact between the molten electrolyte (NaF) and aluminum causes several interactions between the metal and the bath [29]. Among these interactions, the formation of sodium at the interface occurs according to the following reaction:

3 NaF + Al → 3 Na + AlF₃

(8)

3. Generalities on Refractories

3.1. Definition

Refractory materials are non-metallic, non-alloyed materials designed to withstand high temperatures (≥1500 °C) without deformation or failure, as defined by standards such as AFNOR NF B 40-001 and ISO/R36 [31]. They are critical in high-temperature industries, including aluminum reduction cells, where they serve as linings to ensure thermal stability and protect against chemical corrosion.

In aluminum reduction cells, high-silica bricks are commonly employed beneath the carbon cathode to maintain thermal equilibrium and safeguard insulating layers from the extreme operational conditions. Aluminosilicate refractories, the most prevalent type, are classified by their alumina content [1,32], as detailed in

Table 2. Classification and characteristics of aluminosilicate refractories.

Category	Description
Chamotte Bricks (High-Silica Bricks)	Alumina content between 10 and 45 wt%; balanced in cost, thermal stability, and resistance to stress.
High-Alumina Bricks	Alumina content exceeding 45 wt%; offers enhanced performance in high-temperature environments.
Shaped Refractories	Delivered in their final usable form, such as bricks.
Unshaped Refractories	Supplied as loose materials requiring shaping, such as concretes and ramming pastes.

.

Table 2 concisely summarizes the classifications and characteristics of aluminosilicate refractories, providing a clear reference for understanding their applications in aluminum reduction cells.

3.2. Impact of Alumina/Silica Ratio on Refractory Performance

The alumina-to-silica ratio is a critical factor influencing the performance of refractories used in aluminum reduction cells. A high alumina content (>45 wt%) provides enhanced resistance to high temperatures (>1500 °C) and limits the formation of liquid phases, favoring solid phases such as nepheline (NaAlSiO₄) [1]. These solid phases reduce bath infiltration and improve the chemical stability of refractories. However, prolonged interaction with sodium can cause volumetric expansion, increasing the risk of cracking. Conversely, a higher silica content (>53 wt%) promotes the formation of albite (NaAlSi₃O₈), a liquid phase that acts as a viscous protective barrier against corrosion [1]. This property limits the progression of corrosive infiltration, although excess silica may decrease the mechanical strength of refractories at high temperatures. Therefore, achieving a balance between alumina and silica is essential for optimizing refractory performance.

To limit sodium expansion in the upper layers beneath the cathode, it is preferable to use chamotte bricks. These bricks allow for corrosive agents such as NaF to react with the SiO₂, forming a protective vitreous phase. This layer effectively prevents further penetration into the lower refractory structure. High-alumina bricks are strategically placed once it is ensured that the vitreous phase has been created. Their excellent mechanical strength and thermal resistance provide additional structural support to the base of the cell, reinforcing its durability under severe operational conditions.

While bricks within the same category, such as ORB HQB or high-alumina bricks, share similar classifications, their properties, such as their alumina and silica content, porosity, and thermal conductivity, can differ between manufacturers. These variations are typically adjusted to meet the specific operational requirements of different industries while maintaining a consistent performance in aluminum reduction cells.

3.3. Thermal Conductivity and Density

Thermal conductivity and density are crucial properties that define the performance of refractory materials in high-temperature applications. These properties dictate how well a material can manage heat transfer while maintaining mechanical stability under intense thermal and chemical aggressions. Porosity plays a central role in this balance, as it directly influences both thermal conductivity and density [32].

Dense refractories, with low porosity, tend to have higher thermal conductivity and are ideal for load-bearing applications where strength is critical [33]. Conversely, insulating refractories prioritize higher porosity to enhance their thermal insulation properties, making them suitable for environments where minimizing heat transfer is essential. Chamotte bricks, for example, strike a balance between these attributes, making them particularly effective in aluminum reduction cells [1,32]. As illustrated in Figure 2, porosity can be categorized as either open or closed, with each having distinct implications for thermal and chemical performance.

The relationship between porosity, density, and thermal conductivity is illustrated in the literature. As shown in Figure 2, open porosity allows for the infiltration of aggressive agents like the cryolitic bath, which can accelerate degradation. In contrast, closed porosity enhances resistance to chemical attack and improves the material’s durability under operational conditions. For instance, Equation (9) demonstrates that as porosity increases, the thermal conductivity of lightweight refractory materials, such as fireclay bricks, decreases, as presented in Figure 3 [32]. This behavior is critical for applications in aluminum reduction cells, where effective thermal management is necessary to prevent overheating or excessive heat loss.

The dependency of the thermal conductivity on the porosity is described by the following equation [34]:

$$\lambda ={\lambda }_0\left(1-b{P}_t\right)$$

(9)

where ${\lambda }_0$ is the heat conductivity for the material with zero porosity $P_t$ is the total porosity, and the values for the coefficient $b$ are given in Table 2. The total porosity is a sum of the open and closed porosity such that

$$P_t=P_0+P_c$$

(10)

The specific definitions of these parameters are provided in

Table 3. Definitions of porosity parameters.

Symbol	Parameter	Definition
$P_t$	Total Porosity	Ratio of the combined volume of open and closed pores to the total material volume, including solids.
$P_0$	Open Porosity	Ratio of the volume of open pores to the total material volume.
$P_c$	Closed Porosity	Ratio of the volume of closed pores to the total material volume.

, summarizing their significance in the context of refractory materials [1].

Additionally,

Table 4. Thermal conductivity values of common refractory materials (e.g., fireclay and alumina bricks) at different operating temperatures [1]. Reproduced with permission from Springer Nature.

Material	Thermal Conductivity at °C, W/m·
	20	200	400	500	600	800	1000	1200
Fireclay	1.16	-	-	1.34	-	1.47	1.51	1.55
Silica	1.16	-	-	1.4	-	1.5	1.63	-
High alumina (85% Al2O3)	-	2.33	-	2.2	2.1	2.1	2.1	2.1

provides values of thermal conductivity for various refractory materials, including fireclay and alumina bricks, at different temperatures [34,35]. These data underscores how material selection can be tailored to specific industrial requirements.

The quantitative relationships between porosity and thermal conductivity are further outlined in

Table 5. Quantitative relationship between porosity and thermal conductivity for different refractory types [36].

Porosity, %	<10	10–15	15–20	20–25
$b$	1.5	2.0	2.4	2.6

, which highlights how these properties evolve for different refractory compositions.

Understanding these relationships is key to optimizing the design and selection of refractories for aluminum cells. By controlling porosity and selecting materials with an appropriate density and thermal conductivity, it is possible to achieve a balance between thermal insulation, structural stability, and resistance to chemical attack.

3.3.1. Measurements of Porosity and Density

To calculate porosity and density of refractory bricks, ASTM standards C20 and C134 offer clear guidelines. ASTM C20-00 (2022) [37] involves measuring three key weights: the dry weight (D), the saturated weight (W), and the suspended weight (S). From these, porosity is calculated by determining the ratio of the volume of pores (open and closed) to the total material volume. The bulk density is found by dividing the dry weight by the total volume, which is calculated as the difference between the saturated and suspended weights. Meanwhile, ASTM C134-95 (2023) [38] offers a simpler approach for determining bulk density. This method uses the physical dimensions of the brick, its length, width, and thickness, alongside its dry weight. These calculations are vital for understanding the material’s properties and ensuring its quality for industrial applications.

3.3.2. Measurements of Thermal Conductivity

Thermal conductivity is measured using two main approaches: stationary and transient (dynamic) heat flow methods, each with its own advantages and limitations [1]. Stationary methods, outlined in standards like ASTM C182-88 (1998) [39] and ASTM C417-93 (1998) [40], are particularly suited for materials with low to medium thermal conductivity. They provide accurate and reliable results under conditions that closely mimic actual refractory use, though they can be less practical for high-temperature applications. On the other hand, dynamic methods, such as the hot wire technique (ASTM C1113-99 (2019) [41], ISO 8894-1:2010 [42]) and the laser flash method (ASTM E1461-13 [43]), excel in measuring materials with high thermal conductivity. These methods are faster and effective even at elevated temperatures, though they may produce higher conductivity values compared to stationary techniques and could face challenges in sample preparation for certain materials. Both methods play a crucial role in understanding and optimizing material properties for industrial applications.

3.4. Thermomechanical Characteristics

The mechanical properties of refractory materials are critical when selecting them for specific applications and designing industrial equipment [1]. Key properties include compressive strength (CCS), flexural strength (modulus of rupture, MOR), and the elastic modulus (Young’s modulus). These properties contribute to the material’s ability to withstand mechanical loads, resist abrasion and wear, and maintain performance under harsh conditions. Compressive strength is often used to assess the overall robustness of refractories, while flexural strength provides an indirect method to approximate tensile strength, particularly for brittle materials like refractories. Direct tensile testing is often challenging for these materials due to their inherent brittleness and susceptibility to stress concentrations, which can lead to premature failure. Flexural tests, by applying bending loads, induce tensile stresses at the surface opposite to the applied load, offering valuable insight into the material’s tensile behavior. Understanding this relationship is crucial, as tensile stresses play a significant role in crack initiation and propagation under operational conditions. These characteristics are influenced by factors such as porosity, microstructure, and the degree of sintering. Additionally, the elastic modulus, measured using static or dynamic methods, is closely linked to the stiffness and deformation resistance of the material. Significant variations in mechanical properties can indicate microcracks or structural inconsistencies, which may impact the lifespan and overall performance of refractories in demanding industrial environments [1].

3.4.1. Thermal Expansion

Thermal expansion is a fundamental property of refractory and heat insulation materials that reflects their ability to withstand dimensional changes when exposed to varying temperatures. This behavior is crucial for materials used in high-temperature environments, as excessive thermal expansion or contraction can lead to cracks, warping, or mechanical failure, compromising the structural integrity of the furnace or equipment [1].

Refractory materials generally expand linearly with increasing temperature, with their thermal expansion coefficients influenced by the mineralogical composition and structure of the material. Understanding these coefficients helps engineers design linings and structures that can accommodate the stresses caused by temperature changes, thereby improving durability and performance [1].

Table 6. Thermal linear coefficient of expansion for refractory and heat insulation materials (0–1000) K [1].

Material	×106 K−1
Fireclay (15% Al2O3)	7–9
Fireclay (30% Al2O3)	4, 5–6

summarizes the thermal expansion coefficients of common refractory materials, highlighting their behavior under different temperature ranges and providing valuable insight into their suitability for various industrial applications. Materials with lower alumina content, such as fireclay bricks with 15% alumina, exhibit higher thermal expansion coefficients (7–9 × 10⁻⁶ K⁻¹), while those with higher alumina content (30%) display lower coefficients (4.5–6 × 10⁻⁶ K⁻¹). This trend reflects the stabilizing effect of alumina on thermal expansion. Lower thermal expansion coefficients are particularly desirable in industrial settings like aluminum reduction cells, as they minimize the risk of cracking and mechanical failure during thermal cycling, making high-alumina bricks more suitable for demanding applications.

3.4.2. Thermal/Chemical/Mechanical Interactions on Refractories

Refractories are subjected to various strains mechanisms during operation, causing stresses in the material that can significantly impact their performance and lifespan. These strain mechanisms can be categorized into three main types, as detailed in

Table 7. Types of strain mechanisms on refractory materials [44].

Type of Strains	Description
Mechanical	Instantaneous elastic (reversible), plastic (irreversible), and time-dependent viscoelastic (reversible) and viscoplastic strains due to compression, tension, and bending, which can cause structural damage.
Thermal	Caused by thermal shock and differential thermal expansion due to fluctuating temperatures (reversible).
Chemical	Corrosion resulting from interactions with the cryolitic bath, sodium vapor, and other aggressive agents (irreversible).

.

These phenomena often interact. For example, chemical infiltration by the cryolitic bath can alter the mineralogical structure, weakening the material’s mechanical and thermal properties [45]. Key interactions and their resulting effects are summarized in Figure 4.

3.5. Degradation of Refractories

Refractory degradation in aluminum reduction cells is a multifaceted process driven by the combined effects of chemical, thermal, and mechanical factors. A key contributor to this degradation is the interaction between the cryolitic bath and aluminosilicate refractories at temperatures above 700 °C. The process involves the formation of secondary phases, as detailed in

Table 8. Secondary phases formed during refractory degradation [46].

Phase	Formation Mechanism	Impact on Refractory Material
Nepheline	Reaction of cryolitic bath with alumina and silica.	Weakens structural integrity.
Albite	Interaction with bath components, especially in high-SiO2 conditions.	Reduces thermal stability and increases porosity.

.

The formation of these phases is influenced by several factors, as summarized in

Table 9. Factors influencing refractory degradation.

Factor	Description
Bath Composition	Relative proportions of Na2O, SiO2, and Al2O3 in the cryolitic bath.
Refractory Properties	Includes porosity, silica-to-alumina ratio, and mineralogical structure of the material.
Environmental Conditions	Temperature gradients, surrounding gas composition (e.g., sodium vapor, CO2), and vapor pressure of the bath.

:

The Na₂O-SiO₂-Al₂O₃ phase diagram (see Figure 5), calculated using FactSage™ software (version 8.3), provides critical insights into these interactions. It maps the stability regions for nepheline and albite, illustrating how variations in bath composition, temperature, and environmental factors like vapor pressure drive the degradation process [24].

3.5.1. Degradation of Ordinary Refractory Bricks

• Previous Works

The main degradation agents in aluminosilicate refractories are liquid sodium fluoride and metallic sodium, and a combination of both was studied by Pelletier and Allaire [24,47]. The molten electrolyte (represented by molten NaF) or the percolating bath reacts and erodes the lining material immediately after the cell starts operating, continuing relatively intensively during the initial operating phase [48]. As the depth of penetration increases, the rate of infiltration slows down but continues throughout the cell’s lifespan as the electrolyte continues to percolate through the cathode. This attack leads to significant mineralogical transformations in all lining materials in contact with molten fluorides [49]. Over time, this impacts the material’s performance and, consequently, the cell’s lifespan or, in the worst-case scenario, results in complete pot failure. The current understanding of the degradation process suggests that, in addition to the well-known attack by the molten fluoride mass [50], metallic sodium vapor is the primary attacking agent in the cathodic lining and is the main species acting at the reaction front [51,52]. In an industrial reduction cell, the molten electrolyte is in contact with liquid aluminum. This contact initiates several interactions between the metal and the bath. Among these interactions, the formation of sodium at the interface [47] occurs according to the following reaction:

3 NaF_(liq) + Al_(liq) → 3 Na_(g) + AlF₃

(11)

Due to this reaction, metallic sodium is continuously produced at the bath/metal interface and penetrates through the carbon cathode [7,53]. Under the carbon cathode, sodium reacts and reduces oxide materials present in the refractory lining to form sodium aluminates such as alumina, NaAlO₂, nepheline, and silicon. With higher SiO₂ contents (see Figure 6), albite (NaAlSi₃O₈) will be produced at the expense of nepheline (NaAlSiO₄). Since these two reaction products are not stable in contact with sodium, they can further react with sodium. Albite is less stable than nepheline [26,54].

Tschöpe, Schöning et al. [47,51] demonstrated that gaseous Na diffuses into the lining and appears as the sole attacking agent present in the reaction front. They conducted an elemental mapping on the reaction front from three different reduction cells. Linear scans and area mappings across the reaction front were recorded by EPMA. At the reaction front, the sodium gradient is particularly steep and sodium content decreases rapidly. Interestingly, there is no corresponding gradient in fluorine content at the reaction front, and the fluorine concentration is nearly zero throughout the reaction zone. The mapping of the area within the same region indicated no presence of F, and the Na front could be clearly observed. This implies that the attacking agent at the reaction front is sodium, and the degradation mechanism in this zone is not a combination of “wet” (NaF) and “dry” (Na (g)) attack. Na (g) penetrates more rapidly than the components of the bath. Since there is no stable material in contact with Na (g), degradation will continue.

The primary phases in the carbon cathode were cryolite, sodium fluoride, alumina, and a minor amount of calcium fluoride (CaF₂). Under the cathode or in the so-called lens accumulation zone (see Figure 7), α- and β-alumina (α/β-Al₂O₃), cryolite (Na₃AlF₆), sodium fluoride (NaF), nepheline (NaAlSiO₄), and silicon (Si) were detected. In the reacted refractory brick zone, β-alumina, and cryolite disappeared. Finally, near the reaction front, only nepheline (NaAlSiO₄) and silicon (Si) could be detected, along with an amorphous phase. The vitreous phase is primarily composed of sodium aluminosilicate glass [51]. The presence of silicon at the reaction front confirms the reducing conditions due to the presence of Na.

The degradation reactions are summarized in the degradation map created by Tschöpe, Rutlin et al. [51]. The reaction scheme in the degradation map can also be estimated using a ternary phase diagram of the Na₂O-Al₂O₃-SiO₂ system [49]. Since this diagram consists of oxides, the Na content must be converted to Na₂O according to the reaction. This approach is proposed by Pelletier et al. [24,47]. SiO₂ will be reduced by Na to form Na₂O and Si. This conversion reaction is valid for the regions of the degradation map where elemental Si is produced.

According to the percentage of SiO₂ and Al₂O₃ and considering the sodium content of the worn brick near the reaction front, the degradation map is established. This intricate degradation map of “dry” attack based on metallic sodium was inspired by the work of Schøning, Grande et al. [53]. Thus, it is possible to predict the mineralogical evolution of the reaction front in aluminosilicate bricks. The only requirement is knowledge of the Na content and the initial refractory composition [53].

Lambotte and Chartrand [55] developed a thermodynamic model of the Al₂O₃-Na₂O-SiO₂ system, incorporating fluorides such as AlF₃, NaF, and SiF₄ to more accurately reflect the conditions in aluminum electrolysis cells. Their study provides valuable insights into refractory corrosion mechanisms, particularly in environments where fluoride species affect phase stability.

To limit the effects of sodium infiltration on refractory degradation, several protective strategies have been explored. One particularly effective approach involves using a double-layer barrier made of dense refractory bricks, which has been shown to improve the resistance of the cell-bottom lining [56]. Brunk [56] proposed this dual-layer design to reduce sodium penetration, thereby extending the service life of reduction cells. Additionally, the deterioration of refractory linings has a direct impact on the overall efficiency of aluminum reduction cells. Jasim et al. [57] examined the link between refractory degradation and cell performance, highlighting how careful material selection and optimized lining configurations can help minimize downtime and lower maintenance costs.

• Recent Advances

Recent research has provided critical insights into the degradation mechanisms affecting ORBs in aluminum reduction cells. These studies focus on contamination by the cryolitic bath, particularly at lower temperatures (700 °C to 960 °C), where sodium vapor activity diminishes, and the bath becomes the primary reactive agent [8].

3.5.2. On-Field Investigations

In 2024, Ben Salem et al. conducted a comprehensive autopsy of a high-amperage cell to identify the factors contributing to the reduced cell life [8]. The autopsy revealed significant degradation in ORBs, especially in the sidewalls and beneath the cathode blocks. Layers of ORBs displayed distinct structural and mineralogical changes driven by infiltration of bath components. Using X-ray diffraction (XRD), the study identified nepheline (NaAlSiO₄), cryolite (Na₃AlF₆), and silicon (Si) as key phases in the degraded zones. These findings emphasized the aggressive nature of the cryolitic bath and its ability to penetrate and alter refractory materials.

3.5.3. Laboratory-Scale Testing

Complementing the industrial autopsies, laboratory experiments simulated the contamination of ORBs under controlled conditions. Samples were exposed to electrolytic bath components at temperatures ranging from 700 °C to 960 °C [8], enabling a detailed examination of thermal and chemical degradation mechanisms. The results are summarized in

Table 10. Observations from laboratory simulations of ORB degradation.

Aspect	Observation	Significance
Thermal and Chemical Analysis	Thermogravimetry and Differential Thermal Analysis (TG/DTA) revealed mass loss at ~730 °C due to the melting of chiolite (Na5Al3F14) and calcium cryolite (Na2Ca3Al2F14).	The transition to a liquid state facilitated chemical interaction between the bath and ORBs.
Contamination and Phase Formation	At 700 °C, a thin reaction layer formed. At 960 °C, multiple layers were observed—heavily degraded outer zone, intermediate dense layer, and inner unaffected zone.	XRD confirmed the formation of nepheline, albite (NaAlSi3O8), and other sodium-rich phases, consistent with autopsy findings.

.

4. Conclusions

This review highlights the fundamental importance of refractory materials in maintaining the efficiency and longevity of aluminum reduction cells, particularly within the Hall–Héroult process. The presented analysis focuses on the degradation mechanisms induced by the cryolitic bath and sodium vapor, which remain the primary challenges for aluminosilicate refractories. Key findings include the formation of nepheline and albite, which compromise the thermal and mechanical properties of refractories, particularly in high-amperage cells operating at elevated temperatures.

By synthesizing the insights from industrial autopsies and laboratory experiments, this study emphasizes the critical role of bath infiltration in accelerating refractory wear. The results underline the necessity of tailoring refractory compositions to withstand aggressive chemical environments and designing materials with improved resistance to sodium vapor and molten cryolite.

Moving forward, the development of advanced refractory materials should focus on optimizing chemical stability, reducing permeability, and enhancing their overall durability under operational conditions. These improvements, coupled with a deeper understanding of the specific interactions between refractories and bath components, will be pivotal in addressing the challenges posed by modern aluminum reduction technologies. Ultimately, such advancements will contribute to improving the efficiency, sustainability, and economic performance of the aluminum industry.

5. Outlook

This review emphasizes the significant challenges related to the performance and degradation of refractories in aluminum production cells. As modern cells become increasingly prevalent and sustainability takes on greater importance, it is evident that innovative solutions in refractory design and management are urgently needed. Future research should focus on creating advanced materials that can better resist the combined effects of thermal, chemical, and mechanical aggressions. Additionally, combining traditional refractory materials with new technologies could provide hybrid solutions capable of meeting the diverse demands of industrial operations.

Another promising direction is the development of enhanced protective measures. Exploring novel coatings and surface treatments to reduce chemical infiltration and prolong the lifespan of refractories will be essential. Moreover, practical methods to apply these solutions, either during manufacturing or as retrofits, should be prioritized to ensure scalability and industrial adoption.

Improving monitoring technologies is equally critical. The integration of advanced sensors that can continuously track temperature, deformation, and chemical infiltration in real time would provide valuable insights into refractory performance. Pairing these tools with predictive maintenance strategies supported by data analysis and machine learning could significantly minimize unexpected failures and operational disruptions.

Sustainability must also take center stage in future efforts. Research into recycling spent refractories, particularly methods that are cost-effective and capable of recovering valuable materials, is an important step. Equally, the adoption of environmentally friendly production practices will align the industry with global goals to reduce emissions and carbon footprints.

By advancing these areas of research, the development of refractory materials that are not only more durable and efficient but also environmentally sustainable will contribute significantly to the long-term success of aluminum production technologies.