Application of Roasting Flotation Technology to Enrich Valuable Metals from Spent LiFePO4 Batteries.

The application of lithium-ion batteries (LIBs) in electric vehicles has attracted wide attention in recent years, especially LiFePO4 batteries that have been extensively used in large electric buses and cars. The increased demand for LIBs has greatly stimulated lithium-ion battery production, which subsequently has led to greatly increased quantities of spent LIBs. From the perspective of environmental protection and resource recovery, the recycling of spent LIBs is of great significance. In this study, the roasting flotation technology was applied to enrich valuable metals from the mixed electrode powder of spent LiFePO4 batteries. Roasting could thoroughly remove the organic outer layer coated on the surface of electrode-active materials, which improved the flotation enrichment efficiency of valuable metals in the mixed electrode powder of spent LiFePO4 batteries. Under the optimum conditions of roasting at 500 °C for 1 h, the enrichment efficiency of Li and Fe reached the best. The recovery and the enrichment ratio of Li were 95.87% and 1.37, respectively, while the recovery and the enrichment ratio of Fe were 95.25% and 1.36, respectively. Roasting flotation was an efficient process to enrich valuable metals from spent LiFePO4 batteries without wasting graphite resources.

Introduction

Because of their high energy density, long serving life, and safe handling, lithium-ion batteries (LIBs) are considered the most important energy-storage systems for our current life and have been extensively applied in portable electronic devices such as mobile phones, laptops, mobile power, and so forth.[1−7] In recent years, with the growing demand for environmental protection and carbon emission reduction, LIBs have become the reasonable choice of driving energy source for vehicles compared to fossil energy.[8,9] Especially, LiFePO4 batteries have been put to wide use in large electric buses and cars in China and have obtained good economic benefits at present.[10,11] However, due to the limitation of service life, LIBs will enter the waste stream after reaching their useful life cycles. It is estimated that up to 340,000 t/y of spent LIBs from electric vehicles every year should be available for recycling by the year 2040.[12] Spent LIBs contain many harmful materials, such as organic electrolytes and heavy metals, and so, the direct disposal of LIBs poses a serious threat to ecosystems and human health.[13] Moreover, there are a large number of valuable metals in spent LIBs, such as Li, Ni, Co, Mn, and Fe.[14−16] Therefore, from the perspective of environmental protection and resource recovery, the recycling of spent LIBs is of great significance.[17]

The recycling methods of spent LIBs are mainly divided into physical methods and chemical methods. Physical methods are usually applied as pretreatment processes of chemical methods to separate electrode materials from other components of LIBs. Chemical methods consist of the pyrometallurgy process[18−22] and the hydrometallurgy process,[23−28] which are directed toward the extraction of valuable metals from cathode materials. It has been proved that after crushing and screening, a fine fraction (−0.25 mm) of spent LIBs mainly consists of anode-active materials (graphite) and cathode-active materials (lithium metal oxide).[15] The pyrometallurgy process burns off carbonaceous substances from the mixed electrode powder of spent LIBs at a high temperature and directly recycles the products of high-temperature treatment.[29,30] Although the pyrometallurgy technology is simple and easy to operate, the high reacting temperature will burn off anode-active materials (graphite) and cause large carbon emissions and environmental pollution. Furthermore, it is not easy to carry out the separation between slags and metals in the pyrometallurgical process. The hydrometallurgy technology involves acid or alkali leaching to dissolve cathode- and anode-active materials, followed by a purification process involving chemical precipitation, solvent extraction, and electrochemical processes.[31−33] Unfortunately, because a large amount of acid solution could enter into the pores of graphite particles, excessive acid consumption is inevitable in the hydrometallurgy process. In addition, the separation of residues from the leach solution is also a complex process.[34]

Therefore, before being fed into metal extraction processes, the separation of the cathode-active materials from graphite becomes a new problem in the recycling process of spent LIBs. Flotation has been considered one of the efficient ways for the separation process of powder minerals in the mining industry because of its advantages of large processing capacity and simple operation. Cathode-active materials of LIBs such as LiCoO2, LiNiCoMn1–O2, and LiFePO4 have good hydrophobicity, while graphite is naturally hydrophobic. In theory, flotation could be a useful method for the separation of cathode and graphite particles. However, the polyvinylidene fluoride (PVDF) binder formed an organic outer layer coated on the surface of electrode-active materials, which decreases the hydrophobicity difference between the cathode and anode particles, leading to the difficulty of conventional flotation to separate cathode and graphite particles.[14] Therefore, the removal of the organic outer layer is essential to the flotation method to separate cathode- and anode-active materials.

It was reported that Fenton reagent has been applied to remove the organic outer layer to enhance the flotation separation for spent LiCoO2 batteries.[35] The results displayed that most of the organic outer layer can be removed in optimum conditions. During the Fenton reaction, the macromolecule material such as PVDF was broken down into small molecules and finally oxidized into CO2 and H2O. However, the application of the Fenton reagent will introduce iron impurities which might disturb the following precious metals extracting processing, while still obtaining a relatively poor removal effect of the organic outer layer and unsatisfactory flotation efficiency. Grinding was also used to enhance the flotation separation for LiCoO2 batteries.[36] Compared to the application of Fenton reagent, grinding had the advantage of not introducing impurities, but it just regenerated some graphite surface without removing the organic outer layer during the grinding process, and so, it failed to obtain desirable enrichment efficiency in the following flotation separation. Roasting has been regarded as an effective method to remove the organic outer layer by researchers,[21,37] and the suitable roasting conditions for the removal of the organic outer layer had been investigated. However, the effects of different roasting conditions on the surface wettability of electrode materials and the subsequent flotation separation were rarely studied. Moreover, it was not clear how roasting affected subsequent flotation separation by changing the properties of electrode-active materials.

In this study, roasting was used to remove the organic outer layer for improving the flotation behavior. The mixed electrode powder below 0.25 mm of spent LiFePO4 batteries was used as raw materials, and the roasting flotation technology was applied to enrich valuable metals from the mixed electrode powder. First, the organic outer layer coated on the surface of electrode-active materials was removed by roasting, and then, froth flotation was used to achieve efficient separation of cathode- and anode-active materials. Surface characteristics of the mixed electrode powder before and after roasting were analyzed, and the effect of roasting conditions on the surface wettability of electrode materials was studied. Based on the fundamental analysis results, flotation tests of the mixed electrode powder after different roasting conditions were conducted to evaluate the effect of roasting on the flotation process.

Materials and Methods

Sampling and Preparation

The spent LiFePO4 batteries used in this study were provided by a local battery collection center. The shells of spent LiFePO4 batteries had been stripped before discharge and part of discharged batteries were dismantled by manual work, and the anode and cathode materials were obtained, respectively, for analyzing their surface wettability. Other discharged batteries were crushed and screened to obtain the mixed electrode powder. The mixed electrode powder below 0.25 mm mainly consisted of cathode-active materials and anode-active materials, and it was chosen as the raw material in this research.

The phase composition of the mixed electrode powder was determined by X-ray diffraction (XRD, Bruker D8 advance, Germany), and the result is shown in Figure . It was found from Figure that the mixed electrode powder was mostly composed of LiFePO4 and graphite. The metal element contents of the mixed electrode powder were measured by an inductively coupled plasma optical emission spectrometer (ICP-OES, ICP5000R, China), and the result is shown in Table . The metal elements in the mixed electrode powder were mainly Fe and Li, and also a few impurities of Al, Cu, and a trace amount of Ni, Co, Mn. Elements of Fe and Li were derived from LiFePO4, while Al and Cu were derived from cathode aluminum foil and anode copper foil, respectively.

Figure 1

XRD pattern of the mixed electrode powder.

Table 1

Metal Element Contents of the Mixed Electrode Powder

element	Fe	Li	Al	Cu	Ni	Mn	Co	others
content/wt %	16.68	2.70	0.45	0.27	0.04	0.03	0.02	79.81

Experimental Procedure

Roasting flotation experiments were carried out in a muffle furnace and a flotation machine. The flowsheet of roasting flotation to enrich valuable metals from spent LiFePO4 batteries is shown in Figure . The roasting process of the mixed electrode powder was carried out in a muffle furnace in an air atmosphere. About 10 g of the mixed electrode powder was put into a corundum boat and transferred to the muffle furnace for roasting at a heating rate of 10 °C/min. An XFG-type flotation machine (Jilin Exploration Machinery Plant, China) was used for the flotation tests of the mixed electrode powder at room temperature, with an impeller speed of 1800 rpm and a pulp density of 40 g/L. Kerosene was used as the collector with the dosage of 300 g/t, while sec-octyl alcohol was used as the frother with the dosage of 300 g/t. 8 g of the mixed electrode powder was used for each flotation test.

Figure 2

Flowsheet of the roasting flotation process.

Characterization of Samples

Thermal gravimetric analysis (TGA) of the mixed electrode powder was conducted by TG-DTA (HCT-2, Beijing Hengjiu Laboratory Equipment Co., Ltd., China) in an air atmosphere. The phase composition of the mixed electrode powder and the metal concentrates were analyzed by XRD (Bruker D8 advance, Germany). To analyze the surface change caused by roasting, a scanning electron microscope (SEM, Phenom ProX, China) coupled with an energy dispersive spectrometer (EDS) was used to display the surface morphology of the mixed electrode powder before and after roasting. The contact angle of anode and cathode materials was measured by a contact angle meter (JY-82B Kruss DSA, Germany) to show the wettability changes. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha+, America) was chosen to determine the surface elemental composition. ICP-OES (ICP5000R, China) was applied to measure the metal contents of samples.

Results

Thermostability Analysis of the Mixed Electrode Powder

The TGA result of the mixed electrode powder is shown in Figure . It can be seen from the thermogravimetry (TG) curve and differential thermogravimetry (DTG) curve that the whole roasting process could be divided into eight stages. Before 155 °C, the mass loss of the mixed electrode powder was due to the evaporation of the remaining water and electrolyte solutions. Between 155 and 188 °C, a slight increase in mass might be due to the absorption of O2 from the air by the residual electrolyte. Further evaporation of residual electrolyte solutions caused mass loss from 188 to 360 °C. The decomposition temperatures of the blinder PVDF and graphite are about 350 and 650 °C, respectively.[9] During 360–407 °C, the mass loss was due to the oxidative decomposition of the binder PVDF. An obvious mass increase during 407–453 °C was attributed to the oxidation of LiFePO4 by O2 in the air, which could be inferred from the first exothermic peak appearing on the differential thermal analysis (DTA) curve. Because of the further oxidative decomposition of PVDF, the mass decreased further at 453–600 °C. Accompanied by the second exothermic peak on the DTA curve, the mass dropped sharply at 600–850 °C, which was due to the oxidation and combustion of graphite. The mass reduced no longer after 850 °C, indicating that the organic materials in the mixed electrode powder had been completely burned out. The mass of the electrode mixture changed a lot during the whole roasting process, and only 60 wt % was left in the end. According to the TGA of the mixed electrode powder, the roasting temperature should not be too high in the case of the combustion of graphite.

Figure 3

Thermal gravimetric analysis of the mixed electrode powder.

Effect of Roasting on the Surface Morphology of the Mixed Electrode Powder

The SEM images under the backscattered electron detector (BSED) mode of the mixed electrode powder before roasting are shown in Figure . All of the following SEM images were also obtained under BSED mode. The bright angular particles were the cathode-active materials (LiFePO4), while the dark fine particles were the anode-active materials (graphite) in Figure . There were a few relatively bigger particles that belonged to LiFePO4, compared to graphite. Most LiFePO4 and graphite were liberated well in the mixed electrode powder, but some fine graphite particles aggregate into large particles, and a small amount of fine LiFePO4 powder adhered to the graphite surface because of the bonding of PVDF.

Figure 4

SEM images of the mixed electrode powder before roasting.

The SEM images of the mixed electrode powder after roasting at different temperatures for 1 h are shown in Figure . Due to the decomposition of PVDF after roasting above 350 °C, there were fewer adhesion phenomena in the roasted mixed electrode powder, and the cathode- and anode-active materials were liberated adequately. However, the cathode-active materials began to ablate when roasting above 600 °C. The cathode-active materials ablated seriously and aggregated with graphite when the roasting temperature reached 700 °C, which is not conducive to subsequent flotation separation.

Figure 5

SEM images of the mixed electrode powder after roasting at different temperatures for 1 h: (a) 400; (b) 500; (c) 600; and (d) 700 °C.

Figure are the SEM images of the mixed electrode powder after roasting at 500 °C for different times. The surface morphology of the mixed electrode powder had little difference after roasting at 500 °C for different times, and there were no obvious adhesion phenomena or ablative phenomena.

Figure 6

SEM images of the mixed electrode powder after roasting at 500 °C for different times: (a) 15; (b) 30; (c) 60; and (d) 120 min.

Effect of Roasting on the Contact Angles of Electrode Materials

For exploration of the effect of roasting conditions on surface wettability of electrode materials, the contact angle tests for anode and cathode materials before and after roasting at different conditions were conducted. The contact angles of anode and cathode materials before and after roasting at different temperatures for 1 h are shown in Table . Due to the organic PVDF layer coated on the surface of electrode-active materials, the difference value of contact angles between anode and cathode materials before roasting was only 6.84°, and so, it was difficult to achieve separation of anode and cathode materials via flotation. With the increasing roasting temperature, both the contact angles of anode and cathode materials became smaller. The difference value was the largest after roasting at 500 °C, indicating that the flotation separation of anode and cathode materials was easier to achieve after roasting at 500 °C.

Table 2

Contact Angles of Samples after Roasting at Different Temperatures for 1 h

roasting temperature/°C	0	400	500	600	700
contact angle of anode materials/deg	71.81	82.94	79.55	71.23	62.80
contact angle of cathode materials/deg	64.97	59.45	47.00	40.35	37.12
difference value/deg	6.84	23.49	32.55	30.88	25.68

The effect of roasting time on the contact angles of cathode and anode materials was investigated, and the result is given in Table . It can be seen from Table that both the contact angles of anode and cathode materials become smaller with the increasing roasting time. The difference value was the largest after roasting at 500 °C for 60 min, which meant that the difference of the wettability between anode and cathode materials was the largest, and that was a good condition for flotation separation. It should be noticed that when the roasting time was longer than 60 min, the contact angle of anode materials decreased sharply due to the excessive oxidation of the exposed graphite after the binder PVDF was removed. The difference value of contact angles between anode and cathode materials showed that the best roasting conditions for subsequent flotation separation were 500 °C and 1 h.

Table 3

Contact Angles of Samples after Roasting at 500 °C for Different Times

roasting time/min	15	30	45	60	120
contact angle of anode materials/deg	83.32	82.48	80.13	79.55	70.82
contact angle of cathode materials/deg	60.56	56.68	50.14	47.00	40.38
difference value/deg	22.76	25.80	29.99	32.55	30.44

Roasting Flotation Tests

Flotation feeds were obtained under different roasting conditions, and then, the flotation experiments were carried out. The mass loss rate (L) was defined by formula 1, which was used to show the mass loss during the roasting process.where m1 is the mass after roasting, m0 is the mass before roasting. The mass loss rate of the mixed electrode powder under different roasting temperatures and time was measured, and the results are shown in Figure . As can be seen from Figure , the mass loss rate changed greatly after 600 °C because the graphite began to oxidize and burn. Therefore, to avoid excessive graphite loss, the roasting temperature should not exceed 600 °C. When roasting at 500 °C for different times, the mass loss rate changed little, and the mass loss rate of the mixed electrode powder within 120 min was no more than 2.89%. Just from the point of view of the graphite loss in the mixed electrode powder caused by roasting time, a roasting time of less than 120 min is acceptable when roasting at 500 °C.

Figure 7

Mass loss rate of the mixed electrode powder under different roasting temperatures and times.

Through the flotation experiments, the influence of roasting parameters on flotation was explored. The enrichment ratio (ER) defined by formula 2 was used to show the concentration level of metals. The metal recovery (R) of roasting flotation was defined by formula 3, which was used to evaluate the recovery efficiency of metals during the whole process. The enrichment efficiency of metal elements was determined by ER and R.where Gm is the grade of metals in the metal concentrate of roasting flotation, G0 is the grade of metals in the mixed electrode powder before roasting, L is the mass loss rate of the mixed electrode powder during the roasting process, and Y is the yield of the metal concentrate of roasting flotation.

The roasting flotation enrichment efficiency of metal elements after roasting at different temperatures for 1 h is presented in Figure . The recovery and the enrichment ratio of Li and Fe reached the maximum at 500 °C. The result was consistent with the analysis of the effect of roasting temperature on the contact angles of cathode and anode materials. When roasted at 500 °C, the binder PVDF wrapped on the surface of electrode particles was decomposed completely and the original surface of cathode- and anode-active materials was exposed. The difference value of contact angles between anode and cathode materials was the largest after roasting at 500 °C, resulting in the high roasting flotation enrichment efficiency of Li and Fe. The effect of roasting time on roasting flotation tests is displayed in Figure . The recovery and the enrichment ratio of Li and Fe reached the maximum after roasting at 500 °C for 60 min. It was in agreement with the result of the difference value of contact angles between anode and cathode materials after roasting at 500 °C for different times. When roasting at 500 °C for 1 h, the difference value of contact angles between anode and cathode materials was the largest and the roasting flotation enrichment efficiency of Li and Fe achieved the best. The recovery and the enrichment ratio of Li were 95.87% and 1.37, respectively, while the recovery and the enrichment ratio of Fe were 95.25% and 1.36, respectively.

Figure 8

Roasting flotation enrichment efficiency after roasting at different temperatures for 1 h: (a) enrichment efficiency of Li; and (b) enrichment efficiency of Fe.

Figure 9

Roasting flotation enrichment efficiency after roasting at 500 °C for different times: (a) enrichment efficiency of Li; and (b) enrichment efficiency of Fe.

The optimum roasting conditions to improve roasting flotation enrichment efficiency were 500 °C and 1 h. Under the optimum conditions of roasting and flotation, the yield of the metal concentrates of roasting flotation was 71.34%. The SEM and EDS images of the metal concentrates are shown in Figure . There were mostly cathode-active materials and also a little graphite in the metal concentrates. The recovered cathode-active materials mainly contained Fe, P, and O elements. The metal element contents of the metal concentrates were detected and listed in Table . The metallic elements of the metal concentrates were chiefly Fe and Li, and also a few Al, Cu, and a trace amount of other metallic impurities. Therefore, further research is needed to remove impurities from the metal concentrates.

Figure 10

SEM and EDS images of the metal concentrates (α and β are the detection positions of EDS analysis at the cathode material surface and the residual graphite surface, respectively).

Table 4

Metal Element Contents of the Metal Concentrates

element	Fe	Li	Al	Cu	Ni	Mn	Co	others
content/wt %	22.76	3.70	0.54	0.32	0.04	0.03	0.02	72.59

Discussion

It can be seen from Figure that the mixed electrode powder before roasting mainly comprised LiFePO4 (cathode-active materials) and graphite (anode-active materials). However, for the existence of the binder PVDF, it formed an organic out layer wrapping on the surface of electrode-active materials, which prevented the original surface of the electrode materials from being exposed. As shown in Table , the difference of contact angles between anode and cathode materials before roasting was less pronounced due to the PVDF layer, leading to the difficulty of conventional flotation to separate cathode and graphite particles. Hence, roasting was applied to remove the PVDF layer for improving the following flotation separation.

The surface properties of the mixed electrode powder changed via roasting. As can be seen from Figures and 5, roasting has a great influence on the surface morphology of the mixed electrode powder. Compared with the mixed electrode powder before roasting, there were fewer adhesion phenomena in the roasted mixed electrode powder after roasting at a suitable temperature. However, the cathode-active materials ablated seriously and aggregated with graphite when the roasting temperature was too high, which is not conducive to subsequent flotation separation.

Roasting had a great influence on the composition and chemical states of the surface elements of the mixed electrode powder. The surface element contents of the mixed electrode powder before and after roasting were detected by XPS, and the result is shown in Table . The main surface elements of the mixed electrode powder before and after roasting at 500 °C for 1 h were both C, O, F, and P. By roasting, the content of the F element decreased from 7.09 to 0.59 at. %, which indicated that the PVDF had been efficiently decomposed and the F element transferred, while the content of the C element increased from 69.27 to 84.70 at. %. This was because more of the graphite surface was exposed when the PVDF organic out layer was removed by roasting. The change of surface elemental composition of the mixed electrode powder demonstrated that roasting could effectively remove the PVDF organic out layer wrapped on electrode-active materials.

Table 5

Surface Element Contents of the Mixed Electrode Powder before and after Roasting

element/at. %	C	O	F	P	Fe
before roasting	69.27	16.36	7.09	5.28	1.99
after roasting	84.70	10.58	0.59	2.27	1.87

The C 1s XPS spectrum of the mixed electrode materials before and after roasting under the optimal roasting parameters is shown in Figure . The C 1s spectrum of the mixed electrode materials before roasting showed that the main peak attributed to graphite was detected at 284.50 eV, while the other main peak at 284.80 eV was from C–C/C–H bonds. Besides the above two main peaks, three small peaks were observed; the oxygen-containing functional group O–C=O at 288.60 eV was related to the ester electrolyte. The peaks at 286.62 and 290.83 eV were derived from the structural units −(CH2–CF2)-n and −(CF2–CH2)-n of organic binder PVDF, respectively. After roasting, the peaks attributed to −(CH2–CF2)-n and −(CF2–CH2)-n from PVDF disappeared indicating that PVDF had decomposed. Some new peaks, including O–C–O at 286.97 eV, CF at 289.40 eV, and C–F at 291.57 eV were detected, indicating that some of the decomposition products and residues remained on the surface of electrode materials.[34]

Figure 11

C 1s XPS spectrum of the mixed electrode materials before and after roasting: (a) mixed electrode materials before roasting; and (b) mixed electrode materials after roasting.

The size of the electrode particles could change by roasting. The particle size distribution of the mixed electrode powder before and after roasting at 500 °C for 1 h is shown in Figure S1 (in the Supporting Information). The average particle sizes before and after roasting were 52.17 and 16.28 μm, respectively. As shown in Figure , some small electrode particles stuck together to form large particles before roasting due to the bonding of PVDF. After the removal of binder PVDF by roasting, the large particles were dispersed into small particles, resulting in a significant reduction in the particle size. The decrease in the average particle size of the roasted mixed electrode powder also indicated that roasting could remove the binder PVDF.

Roasting could not only remove the binder PVDF coated on electrode-active materials but might also affect the properties of electrode-active materials, which affected the following flotation separation greatly. With the increase of roasting temperature, the binder PVDF on the surface of electrode-active materials was decomposed first, and then, the oxidation of graphite and the ablation of cathode-active materials took place. When the roasting temperature was low, the difference value of contact angles between anode and cathode materials was still small because the decomposition of PVDF was incomplete, which led to a poor flotation separation. The PVDF was totally removed as the temperature increased, and the original surface of the cathode- and anode-active materials was entirely exposed and the difference value of contact angles was the maximum, which led to the best flotation separation. Very high roasting temperatures resulted in the fast oxidation of the exposed graphite surface after the PVDF decomposed completely, which reduced the difference value of contact angles, and that was not conducive to the subsequent flotation separation. In addition, very high temperatures led to the thermal ablation of cathode-active materials, which caused the cathode- and anode-active materials to aggregate together, and that was also bad for flotation separation. Therefore, very low or very high roasting temperatures could reduce the flotation efficiency.

Compared to the roasting temperature, the roasting time had an equally important effect on the properties of electrode-active materials. When roasting at 500 °C, the PVDF wrapped on the surface of electrode-active materials began to decompose, while the graphite did not burn. Because of the short roasting time, the decomposition of the binder PVDF was incomplete. As the roasting time increased continuously, the PVDF was removed and the original surface of cathode- and anode-active materials was all exposed, which led to a large difference value of contact angles and resulted in better flotation separation. However, the graphite was gradually oxidized with the increase of roasting time after the PVDF was removed, and then, the surface of graphite turned hydrophilic.[38] Therefore, very long roasting times led to the oxidation of graphite, which reduced the difference in the flotability of cathode- and anode-active materials and thus deteriorated the following flotation separation process.

It has occurred the removal of the binder PVDF, but no excessive oxidation of the graphite or thermal ablation of cathode-active materials after roasting at 500 °C for 1 h. However, the crystal structure of cathode-active materials altered under the optimum roasting conditions. The XRD test result of the mixed electrode powder after roasting at 500 °C for 1 h is shown in Figure S2. Because the roasting process was carried out in the air, the cathode-active materials LiFePO4 reacted with the oxygen in the air and converted to Li3Fe2(PO4)3 and Fe2O3. The oxidation of LiFePO4 is the cause for the obvious mass increase during 407–453 °C in the thermal gravimetric analysis of the mixed electrode powder shown in Figure . The oxidation of LiFePO4 can be presented using the following eq .[37,39,40]

Conclusions

Roasting flotation was an effective process to enrich valuable metals in the mixed electrode powder of spent LiFePO4 batteries. Roasting could remove the PVDF organic outer layer coated on the surface of electrode-active materials, which improved the flotation enrichment efficiency of metallic elements in the mixed electrode powder. The optimum roasting conditions to improve roasting flotation enrichment efficiency were 500 °C and 1 h. Under the optimum roasting conditions, the recovery and the enrichment ratio of Li were 95.87% and 1.37, respectively, and the recovery and the enrichment ratio of Fe were 95.25% and 1.36, respectively. Although valuable metals had been efficiently enriched in the metal concentrates after roasting flotation, a few impurities remained in the metal concentrates, and further research is needed to remove impurities from the metal concentrates.