Conversion of Li₃PO₄ Derived from LiFePO₄ Cathode Manufacturing Byproducts to Li₂CO₃ via CaCO₃-Assisted CO₂ Thermal Treatment

Abstract

A sequential lithium-recovery process combining hydrogen reduction, water leaching, CaCO₃-assisted thermal conversion under a CO₂ atmosphere, and a second water-leaching step was investigated using an LFP cathode manufacturing byproduct as the starting material. Hydrogen reduction was first employed to decompose LiFePO₄ and redistribute lithium into a Li₃PO₄-containing phase while Fe and P were predominantly incorporated into Fe-rich phosphide phases. Among the investigated conditions, treatment at 900 °C for 3 h resulted in the disappearance of detectable LiFePO₄ peaks and the formation of Li₃PO₄ and Fe₂P as the predominant crystalline phases. The reduced product was subsequently subjected to water leaching, and a solid-to-liquid ratio of 1:100 with a leaching time of 5 h yielded a Li₃PO₄-rich intermediate with a lithium recovery of 45.06% and a Li content of 17.43 wt.%. The recovered intermediate was then mixed with CaCO₃ and thermally treated under a CO₂ atmosphere to convert lithium phosphate into lithium carbonate while phosphorus was transferred into a Ca–P-rich phase. The most favorable phase-conversion behavior was obtained at 900 °C using a Li₃PO₄-to-CaCO₃ weight ratio of 1:1 and a holding time of 2 h, where Li₂CO₃ and Ca₂P₂O₇ were identified as the predominant crystalline products. Subsequent water leaching separated the Li-containing carbonate fraction from the water-insoluble Ca₂P₂O₇-rich residue. A maximum lithium recovery of 93.59% was obtained at a solid-to-liquid ratio of 1:50 in the solid-to-liquid-ratio experiments, while the leaching-time experiments showed a lithium recovery of 92.54% after 5 h under the same solid-to-liquid ratio. XRD analysis of the final recovered product identified Li₂CO₃ as the predominant crystalline phase. Based on the selected conditions for Stage I and Stage II, the cumulative lithium recovery from the initial LFP feed to the final Li₂CO₃-rich product was approximately 41.70%, indicating that the first water-leaching and Li₃PO₄-intermediate recovery step was the principal limitation of the overall process. These results demonstrate the feasibility of recovering lithium from LFP cathode manufacturing byproducts through controlled phase conversion and water-based separation, while also highlighting the need to improve Stage I lithium recovery and to consider the thermal-energy demand associated with solution concentration and drying during future process development.

1. Introduction

The rapid expansion of lithium-ion batteries (LIBs) for electric vehicles and stationary energy-storage systems has increased the importance of recovering lithium and other resources from end-of-life batteries and manufacturing byproducts [1,2,3,4]. Among commercial cathode materials, lithium iron phosphate (LiFePO₄, LFP) has attracted considerable attention owing to its structural and thermal stability, long cycle life, relatively low cost, and absence of expensive transition metals such as Ni and Co [1,5]. Its olivine structure, stabilized by strong P–O bonds within PO₄ tetrahedra, contributes to its thermal safety [5,6]. However, increasing LFP production also generates manufacturing scrap and process byproducts that represent potentially recoverable lithium resources [7]. Unlike end-of-life batteries, these materials are generated before electrochemical service and can provide a relatively concentrated source of cathode constituents. Their recycling is particularly important because the economic incentive for LFP recovery depends strongly on lithium, given the comparatively low commercial value of Fe and P [3,4,8].

LFP recycling strategies include hydrometallurgical treatment, thermal processing, and direct regeneration [3,4,8,9]. Hydrometallurgical methods dissolve lithium using inorganic or organic reagents, followed by separation and product-recovery operations [10,11,12]. Selective leaching can limit the dissolution of Fe and P while recovering lithium, but reagent use and subsequent solution processing remain important considerations. For example, selective treatment with an H₂SO₄–H₂O₂ system has been reported to achieve 96.85% Li leaching with only 0.027% Fe leaching at 60 °C for 120 min [13]. Direct regeneration, in contrast, aims to restore the electrochemical properties of degraded LFP and is more dependent on feed composition and impurity levels when a purified lithium chemical is the desired product [14,15].

Thermal and roasting-assisted methods offer another approach by converting LiFePO₄ into phases that facilitate subsequent lithium separation. Zhang et al. reported Na₂CO₃-assisted carbothermic treatment, achieving a Li recovery of up to 99.2% after separation [16]. NaHSO₄·H₂O-assisted roasting has also been used to form water-soluble LiNaSO₄, followed by water leaching and lithium-product recovery [17]. In addition, Li et al. demonstrated NaOH-assisted roasting at temperatures as low as 150 °C, with a reported Li recovery efficiency of 96.7% [18]. These studies demonstrate that the conversion reagent influences the lithium-bearing phase formed during treatment and the subsequent separation strategy.

The main characteristics of representative LFP treatment routes are summarized in Table 1. The comparison provides a basis for considering alternative lithium-conversion pathways; however, the reported leaching efficiencies and recovery values should not be regarded as directly equivalent because the studies differ in feed materials, product forms, and recovery definitions.

Table 1. Comparison of representative lithium-recovery routes for LFP materials with the present process.

Treatment Route	Principal Operating Conditions	Reported Li Performance	Separation Characteristics and Practical Considerations
Selective hydrometallurgical leaching [19]	H2SO4–H2O2 system; 60 °C, 120 min	96.85% Li leaching; 0.027% Fe leaching	Selective Li dissolution at relatively low temperature; requires acid, oxidant, and subsequent solution processing
Na2CO3-assisted carbothermic treatment	Sodium-salt-assisted roasting and subsequent separation	99.2% Li recovery, as reported	Formation of separable Li-containing phosphate salts; requires thermal treatment and sodium-salt addition
NaHSO4·H2O-assisted roasting–water leaching [20]	Roasting at 300 °C for 5 h, followed by water leaching at 90 °C for 3 h	100% selective Li recovery by water leaching, as reported	Formation of water-soluble LiNaSO4; requires a sodium-salt additive and subsequent lithium-product recovery
Present study	H2 reduction and CaCO3-assisted CO2 conversion at 900 °C	41.70% estimated cumulative Li recovery	Sequential Li3PO4-rich intermediate recovery and Li2CO3-rich product formation; requires two high-temperature treatments and two water-leaching stages

Although the reported methods demonstrate different approaches to lithium recovery, they also illustrate the influence of reagent selection on phase conversion and downstream separation. In particular, salt-assisted roasting introduces sodium-, sulfate-, or alkaline-derived species into the reaction system, whereas selective hydrometallurgical leaching relies on subsequent solution processing. These characteristics motivate the investigation of alternative phase-control strategies in which lithium is separated from Fe-rich material as a distinct intermediate before conversion into the desired lithium product.

In our previous study, H₂ reduction of LiFePO₄ promoted the formation of Li₃PO₄ and iron phosphides, principally FeP and Fe₂P. Subsequent CaCl₂-assisted thermal chlorination converted the Li₃PO₄-containing fraction into water-soluble LiCl [21]. Although this approach demonstrated a sequential reduction–conversion pathway, LiCl and Li₂CO₃ represent different recovered lithium forms. In particular, Li₂CO₃ is an important refined lithium compound used in battery-material production and other lithium-chemical applications. A route targeting Li₂CO₃ from a separated Li₃PO₄-rich intermediate therefore warrants investigation.

The present study investigates the sequential recovery of lithium from an LFP cathode manufacturing byproduct through H₂-assisted phase redistribution, water leaching of the resulting Li₃PO₄-containing fraction, and CaCO₃-assisted conversion under a CO₂ atmosphere. The resulting Li₂CO₃-rich fraction is subsequently recovered by water leaching and separation from the Ca₂P₂O₇-rich residue. The scientific contribution lies in combining intermediate phase separation with controlled carbonate conversion rather than directly converting the hydrogen-reduced product into a water-soluble chloride or relying on direct acid–oxidant leaching. The effects of thermal-treatment and leaching conditions are evaluated to establish the phase-transformation behavior, quantitative Li recovery, and principal limitations of the proposed process.

2. Materials and Methods

The experimental procedure developed in this study consisted of two sequential stages for the recovery of lithium as Li₂CO₃ from an LFP cathode manufacturing byproduct. In the first stage, the LFP-containing feed material was thermally treated under an H₂ atmosphere to transform the stable LiFePO₄ phase and concentrate lithium into a Li₃PO₄-containing phase. The thermally treated product was subsequently subjected to water leaching and solid–liquid separation to recover a Li₃PO₄-rich intermediate. In the second stage, the recovered Li₃PO₄-rich intermediate was mixed with CaCO₃ and subjected to thermal treatment under a CO₂ atmosphere. This treatment was designed to convert lithium from the phosphate phase into a carbonate phase while transferring phosphorus into a Ca–P-containing phase. The converted product was subsequently subjected to water leaching and filtration to separate the soluble Li-containing fraction from the insoluble Ca–P-rich residue, followed by evaporation and drying to recover Li₂CO₃. The overall experimental procedure is described in detail in the following sections.

2.1. Raw Material and Characterization

An LFP-containing process byproduct generated during the manufacture of LiFePO₄ cathode active material was used as the starting material in this study. Prior to the lithium-recovery experiments, the chemical composition and crystalline phases of the as-received material were characterized to determine the initial Li content and identify the major phases present in the feed material. The chemical composition of the LFP process byproduct was determined using X-ray fluorescence spectroscopy (XRF) and inductively coupled plasma optical emission spectroscopy (ICP-OES). Lithium was independently quantified by ICP-OES on a sample-mass basis, whereas Fe, P, S, Ti, Si, Mn, and Al were analyzed by XRF. The XRF-derived values were normalized to 100% among the elements quantified by XRF, excluding Li and O, and therefore represent relative elemental distributions rather than absolute bulk concentrations. Accordingly, the ICP-OES Li value and the normalized XRF values were obtained on different analytical bases and should not be directly summed to obtain the total bulk composition. Oxygen was not quantitatively determined in the present elemental analysis. Each compositional analysis was repeated five times, and the results are reported as the mean value with the corresponding standard deviation.

The crystalline phases of the starting material were identified by X-ray diffraction (XRD) using an X’Pert³ Powder diffractometer (PANalytical, Almelo, The Netherlands) equipped with a Cu Kα X-ray source. The diffraction patterns were collected over a 2θ range of 10–80° with a step size of 0.02° and a scan rate of 2° min⁻¹. The XRD pattern of the as-received material is shown in Figure 1. The diffraction peaks were predominantly attributable to the LiFePO₄ phase, confirming that LiFePO₄ was the major crystalline phase in the LFP manufacturing byproduct. The analytical results for the starting material are summarized in Table 2. The Li content independently determined by ICP-OES was 3.03 wt.% on a sample-mass basis. On the normalized XRF basis, Fe and P accounted for 68.70 and 29.74 wt.%, respectively, of the elements quantified by XRF, while relatively small amounts of S, Ti, Si, Mn, and Al were also detected. Therefore, the XRF-derived Fe value of 68.70 wt.% should not be interpreted as the absolute Fe concentration of the bulk LFP byproduct. Rather, it represents the relative Fe fraction after normalization of the elements quantified by XRF. This distinction also explains why the normalized XRF composition is not inconsistent with the XRD result showing LiFePO₄ as the predominant crystalline phase.

Figure 1. XRD pattern of the as-received LFP cathode manufacturing byproduct, showing LiFePO₄ as the predominant crystalline phase.

Table 2. Li content determined by ICP-OES and normalized elemental distribution of the XRF-quantified constituents in the LFP cathode manufacturing byproduct.

Element	Content (wt.%)	Standard Deviation
Li	3.03	0.0252
Fe	68.70	1.0585
P	29.74	0.8374
S	0.69	0.0306
Ti	0.38	0.0300
Si	0.18	0.0153
Mn	0.16	0.0208
Al	0.15	0.0252

Note: Li was determined independently by ICP-OES; the remaining elemental values were obtained by XRF and normalized to 100% on the XRF basis.

2.2. Overall Experimental Procedure

The overall process for recovering lithium as Li₂CO₃ from the LFP cathode manufacturing byproduct is schematically illustrated in Figure 2. The process consisted of two sequential stages. In Stage I, the LFP byproduct was subjected to hydrogen reduction to redistribute lithium into a Li₃PO₄-containing phase while Fe was incorporated into Fe-rich phosphide phases, followed by water leaching to recover a Li₃PO₄-rich intermediate. In Stage II, the recovered Li₃PO₄-rich intermediate was reacted with CaCO₃ under a CO₂ atmosphere to promote the formation of Li₂CO₃ and Ca₂P₂O₇, followed by water leaching to separate the Li-containing carbonate fraction from the Ca–P-rich residue. The phase evolution and lithium recovery at each stage were evaluated by XRD and ICP-OES, respectively. Detailed experimental conditions are provided in Section 2.3 and Section 2.4.

Figure 2. Schematic flow diagram of the overall Li₂CO₃ recovery process.

2.3. Hydrogen Reduction of the LFP Byproduct

The LFP cathode manufacturing byproduct was subjected to hydrogen reduction to transform LiFePO₄ into a Li-containing phosphate phase and Fe-containing phosphide phases. The experiments were conducted in a horizontal tube furnace equipped with a gas-supply system, SiC heating elements, a thermocouple, and an alumina crucible positioned in the constant-temperature zone, as illustrated in Figure 3.

Figure 3. Schematic illustration of the horizontal tube furnace used for hydrogen-reduction and CO₂-assisted thermal-conversion experiments: (a) reaction tube, (b) SiC heating elements, (c) thermocouple, and (d) alumina crucible. The red circles represent the SiC heating elements, and the arrows indicate the direction of process-gas flow.

Prior to the reduction experiments, thermogravimetric analysis (TGA) was performed to establish the experimental temperature range. Approximately 2 g of the LFP byproduct was heated from 25 to 990 °C at 5 °C min⁻¹ under flowing H₂ at 300 mL min⁻¹. Based on the mass-change behavior shown in Figure 4, reduction temperatures of 700, 800, and 900 °C were selected.

Figure 4. Thermogravimetric profile of the LFP cathode manufacturing byproduct during heating from 25 to 990 °C under a flowing H₂ atmosphere (300 mL min⁻¹) at a heating rate of 5 °C min⁻¹.

For each hydrogen-reduction experiment conducted to investigate the effects of reaction temperature and holding time on phase transformation, 5.0 g of the LFP byproduct was placed in an alumina crucible and heated at 5 °C min⁻¹ under 99.9% H₂ flowing at 300 mL min⁻¹. The sample was maintained at the selected temperature for 3 h and subsequently cooled at 5 °C min⁻¹ under the same controlled atmosphere. To examine the effect of holding time, additional experiments were conducted at 900 °C for 1, 2, and 3 h under otherwise identical conditions. The resulting products were analyzed by XRD to evaluate the phase transformation. The overall reaction considered for the H₂-assisted transformation is expressed as Equation (1):

3LiFePO₄(s) + 8H₂(g) → Li₃PO₄(s) + Fe₂P(s) + FeP(s) + 8H₂O(g)

(1)

For the separate lithium-recovery evaluation described in Section 3.3, 6.0 g of the initial LFP byproduct was processed under the selected hydrogen-reduction conditions. This feed mass was used as the basis for calculating the initial Li amount and the subsequent Stage I Li recovery. The difference between the 5.0 g and 6.0 g feed masses therefore reflects the distinct purposes of the phase-transformation experiments and the lithium-recovery evaluation, respectively.

2.4. Water Leaching and Recovery of the Li₃PO₄-Rich Intermediate

The hydrogen-reduced product obtained at 900 °C for 3 h was subjected to water leaching to separate the Li-containing phosphate fraction from the Fe-rich phosphide residue. The dissolution of Li₃PO₄ in water is represented schematically by Equation (2):

$${\mathrm{Li}}_3{\mathrm{PO}}_4\left(\mathrm{s}\right)\leftrightarrow 3{\mathrm{Li}}^{+}\left(\mathrm{aq}\right)+{\mathrm{PO}}_4^{3-}\left(\mathrm{aq}\right)$$

(2)

The effect of solid-to-liquid ratio was investigated at ratios of 1:30, 1:50, and 1:100 using distilled water. Each slurry was stirred at 300 rpm for 5 h at room temperature. An additional experiment at a solid-to-liquid ratio of 1:150 was performed to determine whether a further increase in water volume improved Li recovery. Because the hydrogen-reduced product contained Fe-rich phosphide phases, all leaching experiments were conducted in a laboratory fume hood as a precaution against possible gaseous species released during contact with water; PH₃ was not quantitatively monitored in the present study. After leaching, the slurry was separated by vacuum filtration. The insoluble residues were analyzed by XRD, while the Li-containing filtrates were concentrated by evaporation and dried above 90 °C for 24 h to recover Li₃PO₄-rich products. After selection of the solid-to-liquid ratio, the effect of leaching time was examined at a fixed ratio of 1:100 for 1, 3, and 5 h at 300 rpm and room temperature. The recovered products were analyzed for Li content by ICP-OES, and the selected Li₃PO₄-rich intermediate was characterized by XRD before use in the subsequent thermal-conversion experiments.

2.5. CaCO₃-Assisted CO₂ Thermal Conversion of the Li₃PO₄-Rich Intermediate

The recovered Li₃PO₄-rich intermediate was subjected to CaCO₃-assisted thermal treatment under a CO₂ atmosphere to convert the Li-containing phosphate phase into a carbonate phase while incorporating phosphorus into a Ca–P phase. Prior to the conversion experiments, TGA was performed using a mixture of 1 g of Li₃PO₄ and 1 g of CaCO₃. The mixture was heated from 25 to 990 °C at 5 °C min⁻¹ under flowing CO₂ at 300 mL min⁻¹. Based on the mass-change behavior shown in Figure 5, reaction temperatures of 700, 800, and 900 °C were selected.

Figure 5. Thermogravimetric profile of the Li₃PO₄–CaCO₃ mixture during heating from 25 to 990 °C under a flowing CO₂ atmosphere (300 mL min⁻¹) at a heating rate of 5 °C min⁻¹.

For the temperature-dependent experiments, 3 g of the recovered Li₃PO₄-rich intermediate was mixed with 3 g of CaCO₃, corresponding to a Li₃PO₄-rich-intermediate-to-CaCO₃ weight ratio of 1:1. The mixture was heated at 5 °C min⁻¹ under CO₂ (≥99.9%, 300 mL min⁻¹), maintained at the selected temperature for 3 h, and subsequently cooled at 5 °C min⁻¹ under the same atmosphere. After selection of the reaction temperature, the effect of the Li₃PO₄-rich-intermediate-to-CaCO₃ weight ratio was investigated at 900 °C using ratios of 1:0.5, 1:1, and 1:2 for 3 h. The effect of holding time was then examined at 900 °C and a weight ratio of 1:1 for 1, 2, and 3 h. All products were analyzed by XRD to evaluate the phase-conversion behavior. The intended overall conversion is expressed as Equation (3):

2Li₃PO₄(s) + 2CaCO₃(s) + CO₂(g) → 3Li₂CO₃(s) + Ca₂P₂O₇(s)

(3)

The product obtained under the selected thermal-conversion condition was subsequently used for the Stage II water-leaching experiments.

2.6. Water Leaching and Recovery of Li₂CO₃

Following CaCO₃-assisted thermal conversion, water leaching was performed to separate the Li-containing carbonate fraction from the Ca₂P₂O₇-rich residue. The dissolution of Li₂CO₃ is represented schematically by Equation (4):

$${\mathrm{Li}}_2{\mathrm{CO}}_3\left(\mathrm{s}\right)\leftrightarrow 2{\mathrm{Li}}^{+}+{\mathrm{CO}}_3^{2-}\left(\mathrm{aq}\right)$$

(4)

Because the aqueous solubility of Li₂CO₃ decreases with increasing temperature, the leaching experiments were conducted at room temperature. Reported Li₂CO₃ solubilities are approximately 1.54, 1.32, 1.00, and 0.72 g per 100 mL of water at 0, 20, 60, and 100 °C, respectively, whereas Ca₂P₂O₇ is practically insoluble in water [19]. The thermally converted product obtained at 900 °C for 2 h using a 1:1 Li₃PO₄-rich-intermediate-to-CaCO₃ weight ratio was leached with distilled water at solid-to-liquid ratios of 1:10, 1:30, and 1:50. Each slurry was stirred at 300 rpm for 5 h at room temperature and subsequently separated by vacuum filtration. The insoluble residues were analyzed by XRD, while the Li-containing filtrates were concentrated by evaporation and dried above 90 °C for 24 h to recover Li₂CO₃-rich products. After selection of the solid-to-liquid ratio, the effect of leaching time was examined at a fixed ratio of 1:50 for 1, 3, and 5 h under otherwise identical conditions. The recovered products were analyzed for Li content by ICP-OES, and the final product obtained under the selected condition was characterized by XRD. Lithium recovery for each stage was calculated using Equation (5):

$$\mathit{Li} \mathit{recovery}\left(\%\right)={\displaystyle \frac{m_{\mathit{product}}{C}_{\mathit{Li},\mathit{product}}}{m_{\mathit{feed}}{C}_{\mathit{Li},\mathit{feed}}}}\times 100$$

(5)

where the recovered product mass and the corresponding feed mass are expressed in grams, and their respective Li contents are expressed as mass fractions. For Stage I, the initial LFP cathode manufacturing byproduct was used as the feed basis, whereas for Stage II, the recovered Li₃PO₄-rich intermediate introduced into the CaCO₃-assisted thermal-conversion step was used as the feed basis. Accordingly, the calculated Li recovery represents the recovery achieved within each stage rather than the cumulative recovery of the entire process.

2.7. Thermodynamic Analysis

Thermodynamic calculations were performed using HSC Chemistry 6.12 (Outotec Research Oy, Pori, Finland) to evaluate the standard-state thermodynamic tendencies of the two principal thermal reactions involved in the proposed process: the H₂-assisted transformation of LiFePO₄ into Li₃PO₄ and Fe-containing phosphides (Equation (1)) and the CaCO₃-assisted CO₂ conversion of Li₃PO₄ into Li₂CO₃ and Ca₂P₂O₇ (Equation (3)).

HSC Chemistry was used as the thermodynamic calculation software for both reactions. The thermodynamic properties of LiFePO₄ were obtained separately from the literature because this compound was not included in the default HSC database used in this study. The literature-based properties were taken from Churikov et al. [22], who evaluated several thermodynamic data sets for LiFePO₄ and identified a preferred set of standard formation enthalpy, Gibbs free energy, and entropy values. The standard thermodynamic properties of the other species involved in Equation (1) were obtained from the HSC Chemistry database. The reference thermodynamic data and their sources are summarized in Table 3.

Table 3. Reference thermodynamic data for the species involved in Equation (1).

Species	ΔH°f, 298 (kJ mol−1)	S°298 (J mol−1 K−1)	Source
LiFePO4(s)	−1591.23	136.75	Churikov et al. [22]
Li3PO4(s)	−2095.35	104.6	HSC Chemistry 6.12
Fe2P(s)	−130.82	72.51	HSC Chemistry 6.12
FeP(s)	−95.22	50	HSC Chemistry 6.12
H2O(g)	−241.83	188.83	HSC Chemistry 6.12
H2(g)	0	130.68	HSC Chemistry 6.12

For LiFePO₄, the standard enthalpy of formation and standard molar entropy at 298.15 K were −1591.23 kJ mol⁻¹ and 136.75 J mol⁻¹ K⁻¹, respectively [22]. Its temperature-dependent molar heat capacity was represented using the following literature-based expression:

$$C_{p, \mathit{LiFePO}4}^o\left(T\right)=102.68+0.0957T-1.4\times {10}^6{T}^{-2}$$

where T is the absolute temperature in K and $C_p^o$ is expressed in J mol⁻¹ K⁻¹. The temperature-dependent standard enthalpy and entropy were determined from their reference values at 298.15 K using the corresponding heat-capacity integrals. For the remaining species, the thermodynamic functions were evaluated using the applicable temperature-dependent data provided in the HSC database.

The standard Gibbs free-energy change for each reaction was calculated from the stoichiometric differences between the standard Gibbs energies of the products and reactants:

$$∆G_{\mathit{rxn}}^o\left(T\right)=\sum _{\mathit{products}}{v}_i{G}_i^o\left(T\right)-\sum _{\mathit{reactants}}{v}_i{G}_i^o\left(T\right)$$

where v_i represents the stoichiometric coefficient of species i. The standard Gibbs energy of each species was evaluated from its temperature-dependent enthalpy and entropy:

$$G_i^o\left(T\right)=H_i^o-T{S}_i^o\left(T\right)$$

The thermodynamic assessment considered the phase states specified for the individual reactions. For Equation (1), H₂ and H₂O were treated as gaseous species, whereas CO₂ was treated as a gaseous reactant in Equation (3). In the calculation of Equation (3), Li₂CO₃ was specified as a solid product according to the available thermodynamic input data. The calculated standard Gibbs free-energy changes were compared with the experimentally observed phase-transformation behavior.

Because the actual experiments were conducted under flowing gas atmospheres rather than strict standard-state conditions, the calculated ΔG° values were used primarily to interpret the relative thermodynamic tendencies of the reactions with temperature rather than to predict the exact equilibrium conversion.

Two limitations of the thermodynamic input data were considered when interpreting the calculations. First, the temperature-dependent heat-capacity expression adopted for LiFePO₄ was reported for 290–780 K [23]. Its application at the higher temperatures investigated in this study therefore involves extrapolation and introduces uncertainty into the calculated standard Gibbs free-energy changes for Equation (1). Second, the Li₃PO₄ entry in the HSC Chemistry 6.12 database used in this study specifies a temperature interval of 298.15–300 K, which does not cover the thermal-treatment temperatures investigated for Equation (3). In addition, the calculation of Equation (3) treated Li₂CO₃ as a solid product, although pure Li₂CO₃ melts at approximately 993 K and is therefore expected to exist as a liquid at 800 and 900 °C. The calculated standard Gibbs free-energy changes for Equation (3) were consequently interpreted as results of the specified thermodynamic model rather than quantitatively validated predictions of the actual high-temperature conversion. Accordingly, the thermodynamic calculations were interpreted together with, rather than as a substitute for, the experimental phase-identification results.

3. Results and Discussion

3.1. Effect of Hydrogen-Reduction Temperature on Phase Transformation

The effect of hydrogen-reduction temperature on the phase transformation of the LFP cathode manufacturing byproduct was investigated at 700, 800, and 900 °C for 3 h under an H₂ atmosphere. The temperatures were selected based on the thermal behavior observed in the TGA described in Section 2.3. Figure 6 shows the XRD patterns of the products obtained at the investigated temperatures.

Figure 6. XRD patterns of the LFP cathode manufacturing byproduct after hydrogen reduction at 700, 800, and 900 °C for 3 h under an H₂ atmosphere.

At 700 °C, LiFePO₄ remained detectable together with newly formed Li₃PO₄ and Fe₂P, indicating that the transformation of the original LFP phase had begun but remained incomplete within the applied holding time. Increasing the temperature to 800 °C reduced the intensity of the residual LiFePO₄ peaks, while Li₃PO₄ and Fe₂P became more prominent. At 900 °C, LiFePO₄ peaks were no longer detectable, and Li₃PO₄ and Fe₂P were identified as the predominant crystalline phases. These observations indicate progressively more extensive transformation of the original LFP phase with increasing temperature under the investigated conditions. Because XRD was used primarily for qualitative phase identification in this study, the absence of detectable diffraction peaks was not taken as proof of complete conversion or quantitative phase removal.

The significance of the H₂-assisted treatment lies in the redistribution of the constituents of the original LiFePO₄ structure into chemically distinct Li-containing phosphate and Fe-rich phosphide phases. This phase redistribution establishes the basis for subsequent water leaching, during which the Li₃PO₄-containing fraction can be transferred into the aqueous phase while Fe-rich phosphides remain predominantly in the insoluble residue. The temperature-dependent phase evolution may reflect improved reaction progress during the fixed treatment period, although the individual contributions of reaction kinetics and transport processes were not independently determined. Among the temperatures investigated, 900 °C was therefore selected for subsequent experiments because it produced the most extensive detectable transformation of LiFePO₄ into the target phase assemblage within the applied holding time.

3.2. Effect of Holding Time on the Hydrogen-Reduction Behavior

The effect of holding time on the hydrogen-reduction behavior of the LFP cathode manufacturing byproduct was investigated at the selected temperature of 900 °C. Figure 7 shows the XRD patterns of the products obtained after holding times of 1, 2, and 3 h under an H₂ atmosphere.

Figure 7. XRD patterns of the LFP cathode manufacturing byproduct after hydrogen reduction at 900 °C for 1, 2, and 3 h under an H₂ atmosphere.

After 1 h, Li₃PO₄ and Fe₂P were detected together with residual LiFePO₄, indicating that phase transformation had occurred but remained incomplete within the applied holding time. The LiFePO₄ peaks became less pronounced after 2 h and were no longer detectable after 3 h, when Li₃PO₄ and Fe₂P were identified as the predominant crystalline phases. These observations indicate that extending the holding time promoted the progression of the H₂-assisted transformation at a fixed temperature.

The time-dependent phase evolution suggests that the duration of exposure to the reducing atmosphere is an important operating parameter for establishing the Li₃PO₄-rich and Fe-rich phosphide fractions required for subsequent separation. Although longer holding times may allow further gas–solid interaction and phase transformation, the present results do not identify the rate-controlling mechanism or establish quantitative reaction kinetics. Accordingly, 3 h was selected for the subsequent water-leaching experiments because it provided the most extensive detectable transformation of LiFePO₄ among the investigated holding times, rather than because it was demonstrated to be a kinetic or industrial optimum.

3.3. Effect of Solid-to-Liquid Ratio on the Recovery of the Li₃PO₄-Rich Intermediate

The effect of the solid-to-liquid ratio on the separation of the Li₃PO₄-rich fraction from the Fe-rich phosphide residue was investigated by water leaching of the hydrogen-reduced product. As shown in Figure 8, Li₃PO₄ and Fe₂P peaks were detected in the insoluble residues obtained at solid-to-liquid ratios of 1:30 and 1:50, whereas Li₃PO₄ peaks were no longer detectable at 1:100. This trend indicates that increasing the liquid volume promoted the removal of the crystalline Li-containing phosphate fraction from the residue, consistent with the limited aqueous solubility of Li₃PO₄. To determine whether additional water improved the recovery of the isolated intermediate, quantitative Li-recovery experiments were conducted at solid-to-liquid ratios of 1:100 and 1:150.

Figure 8. XRD patterns of the water-insoluble residues obtained after water leaching of the hydrogen-reduced product at solid-to-liquid ratios of 1:30, 1:50, and 1:100.

For the quantitative Li-recovery experiments, 6.0 g of the initial LFP cathode manufacturing byproduct was used, whereas 5.0 g was used in the separate H₂-reduction experiments investigating phase transformation. Based on the initial Li content of 3.03 wt.%, the Li mass in the 6.0 g feed was 0.1818 g. Lithium recovery was calculated according to Equation (5), as defined in Section 2.6. The quantitative recovery results are summarized in Table 4. The measured Li contents of the recovered products were close to the theoretical Li content of stoichiometric Li₃PO₄ (approximately 17.98 wt.%), supporting their identification as Li₃PO₄-rich intermediates.

Table 4. Lithium recovery at different solid-to-liquid ratios during water leaching.

Solid-to-Liquid Ratio	Recovered Product Mass (g)	Li Content (wt.%)	Recovered Li Mass (g)	Li Recovery (%)
1:100	0.48	17.62	0.0846	46.51
1:150	0.49	17.27	0.0846	46.53

Increasing the solid-to-liquid ratio from 1:100 to 1:150 required 50% more water but improved the measured Li recovery by only 0.02 percentage points. The additional liquid volume would also increase the amount of water requiring removal during subsequent product recovery. Accordingly, 1:100 was selected as the laboratory-scale condition that avoided additional water consumption without an appreciable reduction in measured Li recovery. Although the small difference between the two recoveries could not be assessed statistically because replicate experiments were not performed, the results demonstrate that a substantial fraction of the initial Li remained unrecovered in the isolated intermediate. This finding identifies Stage I recovery as an important target for further process improvement without substantially increasing water consumption.

3.4. Effect of Water-Leaching Time on the Recovery of the Li₃PO₄-Rich Intermediate

The effect of leaching time on the separation of the Li-containing phosphate phase from the Fe-rich phosphide residue was investigated at the selected solid-to-liquid ratio of 1:100. Figure 9 shows the XRD patterns of the insoluble residues obtained after leaching for 1, 3, and 5 h. Diffraction peaks corresponding to Li₃PO₄ and Fe₂P were detected after 1 h. Although the intensity of the Li₃PO₄ peaks decreased after 3 h, they remained detectable, indicating that the separation of the crystalline Li-containing phosphate fraction was incomplete at these shorter leaching times.

Figure 9. XRD patterns of the water-insoluble residues obtained after water leaching for 1, 3, and 5 h at a solid-to-liquid ratio of 1:100.

After 5 h, Li₃PO₄ peaks were no longer detectable in the insoluble residue, whereas Fe₂P remained the predominant crystalline phase. This trend indicates that prolonged contact with water promoted the separation of the Li₃PO₄-containing fraction from the Fe-rich residue under the investigated conditions. However, the absence of detectable Li₃PO₄ peaks does not establish complete phase removal because residual amounts below the XRD detection limit cannot be excluded. The Li-containing filtrate obtained after 5 h was concentrated and dried, yielding 0.47 g of a Li₃PO₄-rich intermediate containing 17.43 wt.% Li. Based on the initial Li mass of 0.1818 g in the 6.0 g LFP feed used for the lithium-recovery experiment, the recovered Li mass was approximately 0.0819 g, corresponding to a Stage I Li recovery of 45.06% according to Equation (5). The measured Li content was close to the theoretical value for stoichiometric Li₃PO₄ (approximately 17.98 wt.%), supporting the assignment of the recovered material as a Li₃PO₄-rich intermediate. XRD analysis of the recovered intermediate identified Li₃PO₄ as the predominant crystalline phase, with no detectable Fe₂P peaks (Figure 10). Together, the XRD patterns of the insoluble residue and recovered intermediate support separation of the Li₃PO₄-rich fraction from Fe-rich phosphides. Nevertheless, the measured Li recovery of 45.06% demonstrates that the disappearance of detectable Li₃PO₄ peaks from the residue did not correspond to quantitative recovery of Li in the isolated intermediate. The 5 h condition was therefore selected based on the observed phase separation, rather than a demonstrated maximum in Li recovery, because recovery was not quantified at every leaching time. The recovered Li₃PO₄-rich intermediate was subsequently used for CaCO₃-assisted thermal conversion under a CO₂ atmosphere.

Figure 10. XRD pattern of the Li₃PO₄-rich intermediate recovered after water leaching, filtration, evaporation, and drying.

3.5. Effect of Temperature on CaCO₃-Assisted CO₂ Thermal Conversion

The effect of temperature on the CaCO₃-assisted conversion of the Li₃PO₄-rich intermediate was investigated at 700, 800, and 900 °C under a CO₂ atmosphere. As shown in Figure 11, Li₂CO₃ and Ca₂P₂O₇ were identified in the products obtained at all three temperatures, together with residual Li₃PO₄ at 700 and 800 °C. The intensity of the residual Li₃PO₄ peaks decreased with increasing temperature, and these peaks were no longer detectable after treatment at 900 °C. This phase evolution indicates that higher treatment temperatures promoted the formation of the Li₂CO₃- and Ca₂P₂O₇-containing product assemblage under the investigated conditions.

Figure 11. XRD patterns of the Li₃PO₄-rich intermediate after CaCO₃-assisted thermal conversion at 700, 800, and 900 °C under a CO₂ atmosphere.

The concurrent formation of Li₂CO₃ and Ca₂P₂O₇ is significant for the proposed separation strategy because it indicates the redistribution of lithium and phosphorus from the original phosphate-containing intermediate into chemically distinct carbonate- and calcium-phosphate-containing phases. The formation of the Ca–P-rich phase is consistent with the transfer of phosphorus into a comparatively water-insoluble fraction, while the Li-containing carbonate fraction can be recovered through subsequent water leaching. Thus, the thermal-conversion step serves not only to change the chemical form of lithium but also to establish the phase separation required for downstream Li₂CO₃ recovery.

The thermodynamic assessment of Equation (3) yielded positive standard Gibbs free-energy changes over the investigated temperature range, indicating that the proposed overall conversion is not thermodynamically favored under the phase and standard-state assumptions adopted in the calculation. However, the reliability of these values for representing the actual high-temperature system is subject to two important limitations. First, the Li₃PO₄ entry available in the HSC Chemistry 6.12 database used in this study specifies thermodynamic parameters over a limited temperature interval of 298.15–300 K, which does not cover the investigated treatment temperatures. Second, the calculation treated Li₂CO₃ as a solid product, although pure Li₂CO₃ melts at approximately 993 K and is therefore expected to exist as a liquid at 800 and 900 °C. Consequently, the calculated ΔG° values should be interpreted as indicative results for the specified thermodynamic model rather than quantitatively validated predictions of the actual high-temperature conversion.

At 900 °C, a Li₂CO₃-containing liquid phase may influence contact and mass transport among the reacting components, potentially contributing to the observed temperature-dependent phase evolution. Moreover, the actual starting material was a Li₃PO₄-rich intermediate rather than pure Li₃PO₄, so its phase behavior may differ from that represented by the simplified overall reaction. Nevertheless, neither liquid-phase formation nor a CO₂-rich atmosphere, by itself, demonstrates that the overall Gibbs free-energy change becomes negative. Because the XRD measurements were performed after cooling, the high-temperature phase assemblage and detailed reaction pathway remain unresolved. Equation (3) is therefore treated as a formal overall representation of the observed phase redistribution rather than a thermodynamically verified single-step mechanism. Among the investigated temperatures, 900 °C was selected for subsequent experiments because it produced the most extensive detectable transformation of the Li₃PO₄-rich intermediate, not because it was established as a thermodynamic optimum.

3.6. Effect of Li₃PO₄-to-CaCO₃ Weight Ratio on Phase Conversion

The effect of CaCO₃ addition on phase conversion was investigated at 900 °C for 3 h under a CO₂ atmosphere. The weight ratio of the Li₃PO₄-rich intermediate to CaCO₃ was varied from 1:0.5 to 1:2, while the other reaction parameters were maintained constant. Figure 12 shows the XRD patterns of the products obtained at the investigated mixing ratios. At a ratio of 1:0.5, Li₂CO₃ and Ca₂P₂O₇ were identified together with residual Li₃PO₄, whereas the Li₃PO₄ peaks were no longer detectable at 1:1. Increasing the CaCO₃ addition to 1:2 resulted in the appearance of residual CaCO₃ peaks in addition to the Li₂CO₃ and Ca₂P₂O₇ phases.

Figure 12. XRD patterns of the products obtained after CaCO₃-assisted thermal conversion at 900 °C for 3 h under a CO₂ atmosphere using Li₃PO₄-to-CaCO₃ weight ratios of 1:0.5, 1:1, and 1:2.

The dependence of the product assemblage on CaCO₃ addition indicates that reagent availability influences the redistribution of lithium and phosphorus during thermal conversion. Residual Li₃PO₄ at 1:0.5 suggests insufficient CaCO₃ addition under the investigated conditions, whereas residual CaCO₃ at 1:2 indicates that further reagent addition increased the Ca-containing material remaining in the recovered product. This additional solid material could complicate downstream separation without demonstrating a corresponding improvement in Li recovery. The 1:1 ratio was therefore selected because it produced Li₂CO₃ and Ca₂P₂O₇ as the predominant crystalline phases without detectable residual Li₃PO₄ or CaCO₃. Because the starting material was a Li₃PO₄-rich intermediate rather than pure Li₃PO₄, this ratio represents an empirically selected laboratory condition rather than a stoichiometric optimum.

3.7. Effect of Holding Time on CaCO₃-Assisted CO₂ Thermal Conversion

The effect of holding time on the CaCO₃-assisted thermal conversion of the Li₃PO₄-rich intermediate was investigated at 900 °C under a CO₂ atmosphere. The weight ratio of the Li₃PO₄-rich intermediate to CaCO₃ was maintained at 1:1, and holding times of 1, 2, and 3 h were examined. Figure 13 shows the XRD patterns of the products obtained under these conditions.

Figure 13. XRD patterns of the products obtained after CaCO₃-assisted thermal conversion at 900 °C using a Li₃PO₄-rich intermediate-to-CaCO₃ weight ratio of 1:1 for 1, 2, and 3 h under a CO₂ atmosphere.

After 1 h, Li₂CO₃ and Ca₂P₂O₇ were detected together with residual Li₃PO₄, indicating that the target phase assemblage had formed but the transformation remained incomplete within the applied holding time. After 2 h, Li₃PO₄ peaks were no longer detectable, while Li₂CO₃ and Ca₂P₂O₇ remained the predominant crystalline phases. Extending the holding time to 3 h produced no marked additional changes in the XRD pattern. Thus, the major detectable phase evolution occurred within 2 h under the investigated conditions. The 2 h condition was selected for subsequent experiments because extending the treatment to 3 h did not demonstrate an additional phase-conversion benefit. This selection avoided an additional hour of high-temperature treatment but should not be interpreted as a quantitatively established kinetic or energy optimum.

3.8. Effect of Solid-to-Liquid Ratio on Li₂CO₃ Recovery

The effect of the solid-to-liquid ratio on lithium recovery was investigated by examining the water-insoluble residues and recovered Li₂CO₃-rich products. As shown in Figure 14, Ca₂P₂O₇ was the predominant crystalline phase in the insoluble residues obtained at solid-to-liquid ratios of 1:10, 1:30, and 1:50, whereas Li₂CO₃ peaks were not detectable. This phase assemblage is consistent with the preferential separation of the Li-containing carbonate fraction from the Ca–P-rich residue. Quantitative Li recovery was evaluated separately using the recovered product masses and Li contents.

Figure 14. XRD patterns of the water-insoluble residues obtained after water leaching of the CaCO₃-assisted thermally converted product at solid-to-liquid ratios of 1:10, 1:30, and 1:50.

The Li₃PO₄-rich intermediate used for thermal conversion contained 16.80 wt.% Li, corresponding to an initial Li mass of 0.5040 g in the 3.00 g feed. Lithium recovery was calculated using Equation (5), as defined in Section 2.6. The quantitative results are summarized in Table 5.

Table 5. Effect of solid-to-liquid ratio on lithium recovery during water leaching of the thermally converted product.

Solid-to-Liquid Ratio	Recovered Product Mass (g)	Li Content (wt.%)	Recovered Li Mass (g)	Li Recovery (%)
1:10	3.65	12.60	0.4599	91.25
1:30	3.64	12.65	0.4605	91.36
1:50	3.72	12.68	0.4717	93.59

Increasing the solid-to-liquid ratio from 1:30 to 1:50 required approximately 66.7% more water but increased Li recovery by only 2.23 percentage points. Although 1:50 provided the highest measured recovery and was selected for subsequent laboratory experiments, the additional liquid volume would increase downstream evaporation requirements. Further process optimization should therefore consider the trade-off between Li recovery and water consumption rather than treating the highest measured recovery as an established industrial optimum.

3.9. Effect of Water-Leaching Time on Li₂CO₃ Recovery

The effect of water-leaching time on Li recovery was investigated at the selected solid-to-liquid ratio of 1:50. Figure 15 shows the XRD patterns of the insoluble residues obtained after leaching for 1, 3, and 5 h. Ca₂P₂O₇ was identified as the predominant crystalline phase in all residues, whereas Li₂CO₃ peaks were not detectable. Because the crystalline phase assemblages were similar, the effect of leaching time was evaluated primarily using the quantitative Li-recovery results summarized in Table 6.

Figure 15. XRD patterns of the water-insoluble residues obtained after water leaching for 1, 3, and 5 h at a solid-to-liquid ratio of 1:50.

Table 6. Effect of water-leaching time on lithium recovery.

Leaching Time (h)	Recovered Product Mass (g)	Li Content (wt.%)	Recovered Li Mass (g)	Li Recovery (%)
1	3.71	12.29	0.4561	90.49
3	3.66	12.57	0.4599	91.26
5	3.69	12.64	0.4664	92.54

The Stage II recovery calculations used the initial Li mass defined in Section 3.8. Extending the leaching time from 1 to 5 h increased Li recovery by 2.05 percentage points while increasing the leaching duration fivefold. This relatively modest improvement indicates a trade-off between additional Li recovery and leaching residence time.

The 5 h condition was selected for further laboratory-scale product recovery because it provided the highest measured Li recovery among the investigated conditions. However, the absence of replicate experiments prevents statistical assessment of the observed differences. XRD analysis of the product recovered after evaporation and drying identified Li₂CO₃ as the predominant crystalline phase (Figure 16), supporting the recovery of a Li₂CO₃-rich product through the proposed process without establishing quantitative product purity.

Figure 16. XRD pattern of the Li₂CO₃-rich product recovered after water leaching, filtration, evaporation, and drying under the selected leaching conditions.

3.10. Comparison with the Previously Reported CaCl₂-Assisted LiCl Recovery Route

The present CaCO₃-assisted CO₂ conversion route was compared with our previously reported CaCl₂-assisted LiCl recovery process [19] to clarify the differences in phase-conversion strategy, lithium recovery, and downstream processing (Table 7). Both processes employ H₂ reduction to redistribute lithium from LiFePO₄ into a Li₃PO₄-containing phase; however, their subsequent conversion and separation sequences differ. The previous process directly converts the hydrogen-reduced material into water-soluble LiCl through CaCl₂-assisted thermal chlorination, whereas the present process first isolates a Li₃PO₄-rich intermediate and subsequently converts it into a Li₂CO₃-rich product using CaCO₃ under a CO₂ atmosphere.

Table 7. Comparison of the present CaCO₃-assisted CO₂ conversion route with the previously reported CaCl₂-assisted LiCl recovery route.

Process Parameter	Previous CaCl2-Assisted LiCl Route [18]	Present CaCO3-Assisted CO2 Route
Feed material	LiFePO4 cathode active material	LFP cathode manufacturing byproduct
H2 reduction	900 °C, 1 h	900 °C, 3 h
Intermediate separation	No separate recovery of Li3PO4-rich intermediate	Water leaching and recovery of Li3PO4-rich intermediate
Secondary thermal conversion	CaCl2-assisted chlorination; 900 °C, 3 h	CaCO3-assisted conversion under CO2; 900 °C, 2 h
Main Li conversion product	LiCl	Li2CO3
Main Ca–P-containing phase	Chlorapatite/Ca–P–Cl phases	Ca2P2O7
Water leaching	S/L = 1:10, 25 °C, 24 h	Stage I:S/L = 1:100, 5 h; Stage II:S/L = 1:50, 5 h
Downstream processing	Ca2+ ion exchange and dehydration	Evaporation and drying; no separate ion-exchange step
Final recovered Li form	Anhydrous LiCl	Li2CO3-rich product
Overall Li recovery	71.70%	41.70%
Principal limitation	Chlorapatite formation and LiCl retention in phosphate-containing solids	Low recovery of the Li3PO4-rich intermediate during Stage I

The previously reported route achieved an overall Li recovery of 71.70%, whereas the estimated cumulative recovery of the present process was 41.70%. The lower recovery in the present study was primarily associated with the first water-leaching stage, which recovered 45.06% of the initial Li as a Li₃PO₄-rich intermediate. Although the present route avoids chloride-based secondary conversion and does not require a separate Ca²⁺ ion-exchange step, these differences do not establish an overall process advantage. Rather, the comparison demonstrates that the choice of secondary conversion reagent and the sequence of intermediate separation determine the recovered lithium compound, downstream processing requirements, and principal sources of Li loss.

3.11. Potential Utilization of the Ca₂P₂O₇-Rich Residue and Process-Level Considerations

The Ca₂P₂O₇-rich residue generated during the proposed process may represent a secondary phosphorus resource rather than solely a waste stream. Calcium-phosphate materials recovered from secondary sources have been investigated for phosphorus recycling, fertilizer production, and other industrial applications [23]. Calcium pyrophosphate has also been studied as a component of calcium-phosphate ceramics and glass-ceramics [24,25]. However, the applicability of the residue obtained in the present study remains unverified because its phase purity, impurity content, phosphorus availability, and application-specific properties were not evaluated. In particular, direct agricultural utilization cannot be assumed without further assessment of phosphorus solubility and agronomic performance.

The proposed route differs from conventional acid-leaching-based hydrometallurgical approaches by using thermal phase transformation followed by water leaching to separate Li-containing compounds from Fe-rich phosphide and Ca–P-rich residues. Although this strategy reduces reliance on concentrated acidic or alkaline lixiviants, it requires two high-temperature treatments at 900 °C, H₂ and CO₂ supply, and subsequent evaporation and drying of the Li-containing solutions. The leaching results further indicate that increased water consumption does not necessarily produce a proportional improvement in Li recovery and may increase the downstream evaporation burden, as discussed in Section 3.3 and Section 3.8. Therefore, reduced chemical-reagent consumption should not be interpreted as evidence of lower overall energy demand or operating cost.

The estimated cumulative Li recovery of 41.70% was primarily limited by the relatively low recovery of the Li₃PO₄-rich intermediate during Stage I. Improving this recovery without substantially increasing water consumption is therefore a key requirement for further process development. Because the energy consumption of the thermal treatments, gas supply, evaporation, and drying operations was not quantified, the practical competitiveness of the proposed route relative to conventional hydrometallurgical alternatives remains undetermined. Future work should prioritize improving first-stage Li recovery, reducing thermal-energy demand, and optimizing water use and solution concentration. A detailed mass and energy balance, product-quality assessment, and comparative techno-economic evaluation using comparable feedstocks and product specifications will be necessary to establish the potential for industrial application.

4. Conclusions

This study demonstrated the feasibility of recovering lithium from an LFP cathode manufacturing byproduct through sequential thermal phase conversion and water leaching. The principal scientific contribution is the redistribution of lithium from LiFePO₄ into a Li₃PO₄-rich intermediate and subsequently a Li₂CO₃-rich product, enabling separation from Fe-rich phosphide and Ca–P-rich residues. The experimentally observed phase evolution supports the proposed separation strategy, while the limitations of the available high-temperature thermodynamic data and phase-state assumptions prevent the calculated standard Gibbs free-energy changes from fully describing the actual conversion behavior.

The principal limitation of the proposed process was the relatively low recovery of the Li₃PO₄-rich intermediate during Stage I. Under the selected conditions, Stage I and Stage II Li recoveries were 45.06% and 92.54%, respectively, corresponding to an estimated cumulative recovery of approximately 41.70%. This result identifies Stage I as the main opportunity for improving overall Li recovery. The formation of a Ca₂P₂O₇-rich residue also suggests a potential opportunity for secondary phosphorus-resource recovery, although its suitability for further utilization requires additional characterization.

Although the proposed route reduces reliance on concentrated acidic or alkaline lixiviants during leaching, its practical applicability is constrained by the relatively low cumulative Li recovery, two high-temperature treatments, and the water consumption and evaporation requirements associated with product recovery. The limited recovery improvements obtained by increasing liquid volume further emphasize the need to balance Li recovery against downstream water and energy requirements. Consequently, the present findings do not establish an overall energy, economic, or environmental advantage over conventional hydrometallurgical alternatives. Further development should prioritize improving first-stage Li recovery while minimizing water and energy demand, supported by detailed mass and energy balances, product-quality assessment, and comparative techno-economic evaluation.