The establishment of technological approaches for the defluorination of waste fluoropolymers and recovery of eliminated F– may contribute to the development of fluorine recycling routes. In this study, we investigated the effects of alkalinity, phase transfer catalyst (PTC) concentration, reaction temperature, and solvent types on the defluorination of polyvinylidene fluoride (PVDF) by alkaline wet processing. The rate of defluorination of PVDF in 4.0 M sodium hydroxide (NaOH) and 50 mM tetrabutylammonium bromide (TBAB) under aqueous conditions reached 89.2%. In addition, the defluorination reaction proceeded faster in solvents such as diethylene glycol (DEG) and triethylene glycol (TEG) than in water because of the high affinity between PVDF and these diols. To investigate the feasibility of developing a fluorine recycling route, the defluorination of a photovoltaic (PV) backsheet and subsequent CaF2 precipitation from the eliminated F– was examined. A total of 88.3% of F contained in the PV backsheet was recovered as CaF2, which satisfied the quality standards of commercial fluorspar. This study demonstrated that alkaline wet processing is effective for the defluorination of PVDF and that the establishment of a F recycling route along the F supply chain may be feasible.
Notes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
Fluoropolymers are used in the manufacture of various industrial products and systems wherein other types of polymers are not applicable. Fluoropolymers have multiple carbon–F bonds that impart excellent chemical resistance, thermal stability, weather resistance, and electrical insulation properties to the products. Polyvinylidene fluoride (PVDF) is typically used as a component material for cables and valves. Recently, it has been widely used in the development of electronic materials such as backsheets of photovoltaic (PV) modules and binders for Li-ion batteries [1]. The demand for these electronic products, and thus fluoropolymers, is expected to grow in the future owing to the global energy transition to address climate change [2‐4].
The management of waste fluoropolymers is an intractable problem owing to their large demand. Annually, 33,000 tons of fluoropolymers is manufactured in Japan (2021) [5], 52,000 tons in Europe (2015) [6], and 85,000 tons in the USA (2018) [7]. Incineration has the advantage of treating large amounts of waste with various properties; however, toxic gases released upon the incineration of fluoropolymers, such as HF and organo-F substances, have adverse effects on the incineration plants and environment [8]. There is also an environmental risk with landfill disposal, where plasticizers and toxic metals contained in fluoropolymers may leach into soil and groundwater, contaminating the surrounding environment. Degraded fluoropolymer particles also become microplastics and a source of marine plastics [9, 10]. As fluoropolymers are thermoplastics, the conversion of waste fluoropolymers to secondary fluoropolymers through mechanical recycling is an effective waste management approach. However, microcontaminants in the recycling process cause secondary plastics to have lower quality than pure plastics, limiting the number of times the plastics can be mechanically recycled. Therefore, a new plastic recycling method that will address this waste management issues is needed.
Advertisement
Previous studies have reported the development of chlorine recycling through the dechlorination of chloropolymers via alkaline wet processes [11‐15]. Subsequently, they proposed a chlorine-circulation systems through the chlor-alkali industry. Chlorine recycling also produces dechlorinated chloropolymers which are good sources of hydrocarbons. The development of such a recycling route will promote environmentally sound plastic waste management and better management of resources. In this context, further research into the defluorination of waste fluoropolymers and recovery of the eliminated F– may contribute to the better management of waste fluoropolymers and the development of F recycling routes.
This recycling method can be applied to fluoropolymers to extract F– and precipitate it as CaF2, which can be further used to produce new fluoropolymers. Fluorspar, which is mainly composed of CaF2, is used to produce HF, which is then utilized to manufacture fluoropolymers [16]. Promoting the defluorination of waste fluoropolymers and recovering the eliminated F– as CaF2 will contribute to F recycling through the existing F supply chain (Fig. 1). Currently, fluorspar is recognized as a mineral with a global supply risk because its production is highly geographically concentrated [17]. Therefore, the development of a F recycling route based on this study may mitigate the fluorspar supply risk and enhance the stable production of F-containing materials, such as fluoropolymers.
Fig. 1
F supply chain and possible F recycling route
×
In this study, PVDF was chosen as the target fluoropolymer for its prevalent application across industries and projected increase in production [1]. Previous studies reported that alkaline wet processing [18‐22] and pyrolysis [23‐25] facilitated PVDF defluorination. For this study, alkaline wet processing was selected because of its energy efficiency and safety. The pyrolysis of PVDF requires higher energy consumption for defluorination, as temperatures exceeding 440 °C must be used to overcome the heat resistance of PVDF [24]. Pyrolysis also results in the emission of toxic F compounds such as HF, which may corrode recycling equipment and compromise the safety of the working environment. On the other hand, the defluorination of PVDF by alkaline wet processing proceeds at low temperatures, reducing the energy required for this F recycling method. In addition, alkaline wet processing of PVDF is safer than pyrolysis, as the eliminated F– is immediately neutralized with alkaline and stable F compounds.
Mixing PVDF with higher concentrations of NaOH or KOH, followed by heating, promotes the defluorination of PVDF [19‐22]. The addition of a phase transfer catalyst (PTC) has also been reported to be effective in accelerating the defluorination reaction [18, 26]. However, the effects of various parameters, including alkalinity and PTC concentration, reaction temperature, and types of reaction solvent, on the efficiency of defluorination have not been comprehensively investigated. Previous studies monitored the progress of the defluorination reaction by the weight changes in PVDF and the structural change of the polymer; however, such changes were not observed quantitatively through analysis of eliminated F– [19, 20, 26]. Additionally, previous studies carried out PVDF defluorination under aqueous alkaline conditions without investigating the effects of solvent types on defluorination. Studies on the dechlorination of polyvinyl chloride (PVC) and polyvinylidene chloride (PVDC) revealed that alkaline treatment of those polymers in diols promoted dechlorination more effectively than that in aqueous solutions [13‐15, 27]. A change in the solvent type is similarly expected to improve the efficiency of PVDF defluorination by alkaline wet processing. As shown in Fig. 1, the existing F supply chain could be utilized for F recycling when eliminated F– is collected as CaF2. The similar chemical approach has been proposed in previous studies; however, these studies analyzed only the crystal structure of recovered CaF2 and did not perform elemental analysis [21, 25]. Since there are quality standards for commercial fluorspar [28], it is necessary to analyze elemental compositions of recovered CaF2 in order to evaluate the feasibility of this technology.
Advertisement
The objective of this study is to investigate the feasibility of a F recycling route, involving the defluorination of PVDF by alkaline wet processing, and the subsequent conversion of the eliminated F– to CaF2. The effects of various parameters on the defluorination reaction of the PVDF reagent were analyzed. The kinetics of PVDF defluorination were analyzed by the calculation of the activation energy using a model-fitting method. Next, to investigate F recycling from end-of-life (EoL) products, the feasibility of the defluorination of waste PV backsheet and subsequent CaF2 recovery was examined. The quality of CaF2 recovered from the PV backsheet was evaluated using the quality standards applied to commercial fluorspars. Although many EoL products containing PVDF are available, waste PV backsheet was selected in this study because the estimated number of EoL PV modules is expected to increase sharply in the near future, requiring waste PVDF to be managed.
Experimental methods
Materials
PVDF powder (products sold domestically) and PVDF from a waste PV backsheet were used for the defluorination experiments. The waste PV backsheet used in this study was the same as that used in the previous research [29]. Figure 2 illustrates the photograph and scanning electron microscopy (SEM) image (TM4000, Hitachi High-Tech Corporation, acceleration voltage: 15 kV, backscattered electron image, EDX software: Oxford Instruments AZtec) of the prepared reagent PVDF powder and PV backsheet. The F concentrations in the PVDF powder and the backsheet were analyzed using an automatic quick furnace (AQF-2100H, Nittoseiko Analytech, inlet temperature: 900 °C, outlet temperature: 1000 °C), followed by ion chromatography (IC) (Dionex Integrion RFIC, Thermo Scientific, column: IonPac AG19/AS19, suppressor: ADRS-600 4 mm, eluent: 45 mM KOH, eluent flowrate: 1.0 mL/min). The F concentrations in the PVDF powder and backsheet were 55.1 ± 0.2 wt% (n = 3) and 9.3 ± 0.3 wt% (n = 5), respectively.
Fig. 2
a Reagent polyvinylidene fluoride (PVDF) powder, b scanning electron microscopy (SEM) image of the PVDF powder, c shredded photovoltaic (PV) backsheet, d cross-sectional SEM image of the backsheet
×
Defluorination of reagent PVDF powder
The reagent PVDF powder (0.5 g) was placed in a 300 mL SUS304 reactor, to which 50 mL of ion-exchanged water or diol was added as the reaction solvent. Granular NaOH (guaranteed grade, FUJIFILM Wako Pure Chemical Corporation) was added to prepare 1.0, 2.0, and 4.0 M solutions. Tetrabutylammonium bromide (TBAB) (special grade, FUJIFILM Wako Pure Chemical Corporation) was added to the solutions as a PTC at concentrations of 25 and 50 mM. The reactor was heated to 50, 70, and 90 °C in a silicon oil bath, with N2 supplied to the reactor at a flow rate of 100 mL/min. Baffle plates were attached to the interior of the reactor. The reactor was stirred at 100 rpm during the defluorination reaction.
When the solvent used was ion-exchanged water, 200 µL of the reaction solution was sampled after the prescribed reaction time. The F– concentration eliminated from the PVDF powder was analyzed by IC. The defluorination rate under aqueous conditions (\({R}_{\mathrm{FW}}\)) was defined as follows:
where \({W}_{\mathrm{solution}, t}\) [g] and \({W}_{\mathrm{PVDF}, 0}\)[g] represent the weight of F in the solution at time t [min] and initial F weight of the PVDF, respectively.
If diol was used as the solvent, the F– eliminated from the PVDF powder would be precipitated as NaF, making it difficult to trace the defluorination rate by sampling the solution. Instead, the weight of F in the reaction residue after defluorination was analyzed using an automatic quick furnace and IC. Ethylene glycol (EG) (guaranteed grade, FUJIFILM Wako Pure Chemical Corporation), diethylene glycol (DEG) (special grade, FUJIFILM Wako Pure Chemical Corporation), and triethylene glycol (TEG) (1st grade, FUJIFILM Wako Pure Chemical Corporation) were used as the diols. The defluorination rate under diol conditions (\({R}_{\mathrm{FD}}\)) was defined as follows:
where \({C}_{\mathrm{PVDF}, 0}\) [%] represents the initial F concentration of the PVDF, and \({C}_{\mathrm{PVDF}, 240}\) [%] represents the F concentration of the reaction residues after 240 min of the defluorination reaction.
Defluorination of the PV backsheet
Because the PV backsheet contained a large PET layer (Fig. 2d), initial isolation of PVDF from the backsheet was required. PET hydrolysis is a technique to isolate PVDF as a solid material by converting PET into water-soluble substances. The hydrolysis of PET has been widely investigated under acid, neutral, and basic conditions. In this study, alkaline hydrolysis was selected. Although PVDF defluorination generally proceeds in higher alkaline solution, our previous study revealed that a 4.0 M NaOH solution provides the optimal reaction condition for hydrolyzing the PET layer and promoting the separation of PVDF while restricting PVDF defluorination [29]. We placed 10.0 g of the shredded backsheet in a 300 mL SUS304 reactor and added 100 mL of ion-exchanged water and granular NaOH to produce the 4.0 M NaOH solution. The reactor was heated to 90 °C, and the solution was stirred at 100 rpm for 12 h under a N2 flow rate of 100 mL/min to promote alkaline hydrolysis of the PET layer. The solution was filtered to obtain the isolated PVDF as a solid reaction residue and separate PET as a liquid water-soluble sodium terephthalate.
To accelerate the defluorination reaction, the isolated PVDF was shredded (mixer mill, MM500 nano, Retsch) to a particle size that could pass through a 500 μm mesh. The weight of F in the pulverized PVDF was analyzed using an automatic quick furnace and IC. Next, 2.0 g of pulverized PVDF was placed in a 300 mL SUS304 reactor, and 100 mL of ion-exchanged water and granular NaOH was added to produce the 4.0 M NaOH solution. To promote the defluorination reaction, granular TBAB was added at a concentration of 50 mM. The reactor was heated to 90 °C, and the solution was stirred at 100 rpm under a N2 flow rate of 100 mL/min. After the prescribed reaction time had passed, 200 µL of the reaction solution was sampled, and the F concentration was analyzed by IC. The defluorination rate was evaluated using Eq. (1).
CaF2 recovery from PV backsheet
Next, the conversion of the F– eliminated from PVDF into CaF2 was examined. After confirming that the defluorination of the pulverized PVDF had not increased, the reaction solution was filtered to isolate the reaction residues and filtrate. The pH of the filtrate was first adjusted to 6.0 using HCl, before adding CaCl2 at a mole ratio of 1.5 times the F– to precipitate CaF2. The pH adjustment was made to avoid the precipitation of Ca(OH)2 instead of CaF2, which can occur under basic conditions and reduce the purity of the recovered CaF2. The pH adjustment conditions were calculated using the HSC Chemistry software (version 9.9.2.3, Metso Outotec), as shown in Fig. 3. After CaF2 precipitation, a flocculant was added and CaF2 was collected by filtration. X-ray diffraction (XRD) (Ultima IV, Rigaku), wavelength-dispersive X-ray fluorescence (WDXRF) (ZSX Primus II, Rigaku), and inductively coupled plasma optical emission spectroscopy (ICP-OES) (SPECTROGREEN, Hitachi High-Tech Corporation) were used to investigate whether the quality of the recovered CaF2 satisfied the quality standards applied to commercial fluorspar.
Fig. 3
Equilibrium amount of CaF2 and Ca(OH)2. The initial amounts of NaF and CaCl2 used for the calculation were 0.025 and 0.0375 mol, respectively. Calculations were performed with the system state at 25 °C and 1.0 bar
×
Results and discussion
Defluorination of PVDF powder in aqueous conditions
The defluorination rates of the reagent PVDF powders under aqueous conditions [given by Eq. (1)] with different concentrations of NaOH and TBAB are shown in Fig. 4. The reaction temperature was 90 °C. Increasing NaOH and TBAB concentrations accelerated the defluorination reactions and increased the defluorination rate. At a solution concentration of 4.0 M NaOH and 50 mM TBAB, the defluorination rate reached 89.2%. The defluorination reaction did not proceed without TBAB, even at a solution concentration of 4.0 M NaOH, indicating that TBAB is indispensable for the reaction to progress under the conditions examined in this study. The previous study evaluated the weight decrease of PVDF due to the HF elimination during the defluorination reaction, which resulted in a weight loss of 84.5% when PVDF was treated at 90 °C for 5 h in 4.0 M NaOH and 44 mM TBAB [26]. The weight loss corresponds to the defluorination rate obtained by this work.
Fig. 4
Defluorination rates of PVDF under aqueous conditions at 90 °C, as a function of NaOH and TBAB concentrations
×
Subsequently, the effect of different reaction temperatures on the defluorination rate was investigated. Figure 5 illustrates the defluorination rate of the PVDF powder when heated at 50, 70, and 90 °C in 4.0 M NaOH and 25 mM TBAB under aqueous conditions. As with the dechlorination behavior of PVC and PVDC [13, 15], an increase in the reaction temperature enhanced the defluorination rate. The defluorination rate reached 17.5 % after the powder was heated for 240 min at 50 °C, indicating that the defluorination reaction by alkaline wet processing can proceed even under moderate thermal conditions. Defluorination at 50 °C was incomplete at the end of 240 min, indicated by the upward slope of the curve in Fig. 5. Further extension of the reaction time should increase the defluorination rate. These results show that the defluorination of PVDF by alkaline wet processing is more energy efficient than by pyrolysis; alkaline wet processing can be conducted at temperatures under 100 °C, whereas pyrolysis requires temperatures exceeding 400 °C.
Fig. 5
Defluorination rates of PVDF under aqueous conditions in 4.0 M NaOH and 25 mM TBAB, as a function of temperature
×
Kinetic analysis of PVDF defluorination
Next, a kinetic analysis was performed by fitting the plots of the defluorination into the chemical reaction and diffusion models. The pseudo-first-order and intra-particle diffusion models were applied to the kinetic analysis.
The pseudo-first-order kinetic equation is expressed by the following equation:
$$-\mathrm{ln}\left(1-x\right)=kt$$
(3)
where \(x\) is the defluorination rate, \(t\) [min] is the reaction time, and \(k\) [min−1] is the apparent rate constant.
The intra-particle diffusion model is given by the following equation:
$${q}_{t}={K}_{di}\sqrt{t}+A$$
(4)
where \({q}_{t}\) [mg g−1] is the amount of F in the reaction solution at time \(t\),\({K}_{di}\) [mg g−1 min−1/2] is the intra-particle diffusion rate constant, and A [mg g−1] is a constant indicating the thickness of the boundary layer.
As displayed in Fig. 6, the plots for different temperature conditions consist of two linear sections with different slopes in both models. The multilinearity indicates the occurrence of at least two steps in the defluorination reaction of PVDF. Next, the Arrhenius plots of the apparent rate constant were calculated from the slope of the plots of both models (Fig. 7). As for the apparent rate constant, the slope in the first section was used for the calculation. The apparent activation energies under the pseudo-first-order and intra-particle diffusion models were 66.8 and 52.2 kJ mol−1, respectively. This suggests that the rate-controlling step of the PVDF defluorination might be the chemical reactions and that the defluorination rate is dominantly controlled by the alkaline concentration and reaction temperature. The multilinearity of the plots in both models suggests that the reaction rate of PVDF defluorination was varied with changes in porosity and surface area of PVDF powder during the reaction, while a further experiment would be required to fully describe the phenomena.
Fig. 6
a Pseudo-first-order plot and b intra-particle diffusion plot of the PVDF defluorination in 4.0 M NaOH and 25 mM TBAB, as a function of temperature
Fig. 7
Arrhenius plot of the apparent rate constant for a pseudo-first-order model and b intra-particle diffusion model for Fig. 6
×
×
Defluorination of PVDF powder under diol conditions
The rates of defluorination of the reagent PVDF powders under diol conditions [given by Eq. (2)] are presented in Table 1. The PVDF powder was defluorinated in 2.0 M NaOH and 25 mM TBAB at 90 °C in EG, DEG, and TEG. The F concentration of the original PVDF powder (\({C}_{\mathrm{PVDF}, 0}\)) was 55.1%, and the F concentrations of the PVDF residues (\({C}_{\mathrm{PVDF}, 240}\)) were reduced to 47.9%, 17.6%, and 28.8% after defluorination in EG, DEG, and TEG, respectively. Considering the F concentration in the PVDF residue defluorinated under aqueous conditions was 47.7%, the defluorination of the PVDF powder proceeded in the following order of solvents by rate: DEG > TEG > water > EG. Among the diols, significant defluorination was achieved when DEG or TEG was used, whereas using EG did not achieve significant defluorination.
Table 1
PVDF defluorination rate under diol and aqueous conditions
Solution type
RFD [%]
\({C}_{\mathrm{PVDF}, 240}\)(n = 3)
Average [%]
Std dev
EG
13.0
47.9
2.6
DEG
68.1
17.6
0.1
TEG
47.7
28.8
0.7
Water
13.5
47.7
1.4
Reaction in 2.0 M NaOH and 25 mM TBAB at 90 °C for 240 min
The SEM analysis of the PVDF residues revealed that the defluorination reaction did not cause any changes in their shapes or surface states (Fig. 8). In contrast, the EDX analysis showed that the reaction caused changes in the elemental composition of the defluorinated PVDF. Table 2 lists the compositions of constituent elements on the surfaces of the original PVDF powder and defluorinated PVDF powder, calculated by averaging the values from the three-point spectra. Similar to the trend in Table 1, the F concentration was reduced in the order DEG > TEG > EG > water, indicating that the defluorination reaction proceeded faster in DEG and TEG than under aqueous conditions. There was no significant difference in the F concentrations of the reaction residues of PVDF dissolved in EG and water. EDX analysis also showed that the oxygen concentration increased as the reaction progressed. It can be assumed that the hydroxyl groups that reacted with PVDF were incorporated into the polymer, as proposed in a previous study [19].
Fig. 8
SEM image of a original PVDF powder, b PVDF defluorinated in DEG
Table 2
Elemental analysis on the original PVDF powder and PVDF defluorinated in different solvents
Element weight percentage [wt%]
Original PVDF
Defluorinated PVDF
EG
DEG
TEG
Water
C
53.0
59.2
64.8
60.9
58.8
F
42.5
31.0
12.7
23.7
32.4
O
4.2
9.4
22.0
14.4
7.8
Al
0.4
0.3
0.6
1.1
1.0
×
These results revealed that defluorination was more efficient with DEG or TEG as the reaction solvent than with water. Therefore, the difference in the defluorination rate was evaluated based on the affinity between PVDF and each solvent. The composite affinity parameter, RED (relative energy difference), is defined as follows [30, 31]:
$$\mathrm{RED}=\frac{{R}_{a}}{{R}_{0}}$$
(5)
where \({R}_{a}\) is the distance between the polymer and the solvent in the Hansen space, defined by Eq. (6).
where \({R}_{0}\) is the interaction radius of the polymer, \(\delta\) is the solubility parameter, and \({\delta }_{d}\), \({\delta }_{p}\), and \({\delta }_{h}\) are the dispersion, polar, and hydrogen-bonding solubility parameters, respectively. The subscripts P and S represent the polymer and solvent, respectively. The polymer is insoluble in the solvent if RED > 1, partially soluble if RED = 1, and soluble if RED < 1.
Table 3 lists the Hansen solubility parameters of PVDF, water, and each diol, the interaction radius of PVDF, and the RED number calculated using Eqs. (5) and (6). The affinity between PVDF and the solvents followed the order: TEG > DEG > EG > water. We assumed that increasing the affinity between the PVDF powder and the solvents allowed the solvents to permeate into the interior of the PVDF powder, increasing the contact efficiency between the hydroxyl groups and PVDF and promoting the defluorination reaction. The results in Tables 1 and 2 show that the defluorination reaction proceeded most efficiently with DEG, while Table 3 reveals that TEG has the highest affinity for PVDF even though the difference in RED numbers between DEG and TEG is small. Parameters other than affinity, such as TBAB and hydroxyl group activity, might be more influential in achieving a higher rate of defluorination in DEG.
Table 3
Hansen solubility parameters for PVDF and the solvents, and \({R}_{0}\) for PVDF [27, 30]
The shredded PV backsheet had a layered structure of white PVDF and transparent PET (Fig. 9a). The backsheet was heated to 90 °C in 4.0 M NaOH without TBAB. The reaction residues were filtered to obtain isolated PVDF concentrate (Fig. 9b). The PVDF concentrate was then pulverized until it could pass through a 500 µm mesh (Fig. 9c). The F concentration of the pulverized PVDF was 26.9 ± 2.7 wt% (n = 3), which was lower than the theoretical F concentration of pure PVDF. This might be due to the contaminants of undecomposed PET, polymer additives, and waste-derived substances in PVDF.
Fig. 9
a Shredded PV backsheet, b PVDF concentrate obtained from PV backsheet via PET hydrolysis, c pulverized PVDF
×
The defluorination rate of the pulverized PVDF in 4.0 M NaOH and 50 mM TBAB under aqueous conditions [given by Eq. (1)] is shown in Fig. 10. The defluorination rate increased with a constant slope for 360 min before leveling off and finally plateauing at 600 min, reaching 90.4%. The reaction was stopped at this point. Compared to the pulverized PVDF from the PV backsheet, the powder PVDF only took 150 min to reach the same rate of defluorination (Fig. 4). While this can be assumed to be attributable to the different surface areas of the powder and pulverized PVDF particles, further investigation of the influence of particle size on the defluorination of PVDF is required to quantitatively evaluate the relationship between surface area and reaction efficiency.
Fig. 10
Defluorination rate under aqueous conditions in 4.0 M NaOH and 50 mM TBAB at 90 °C
×
CaF2 recovery from PV backsheet
Next, the reaction solution was filtered to separate the reaction residues from the filtrate. Adding CaCl2 to the filtrate produced a white precipitate (Fig. 11a) and a reaction residue containing black PVDF and TiO2 as the polymer additive (Fig. 11b). The XRD pattern of the white precipitate matched that of the reagent CaF2 (special grade, FUJIFILM Wako Pure Chemical Corporation), suggesting CaF2 as the main component of the white precipitate (Fig. 12). These experimental procedures were assumed to have suppressed the formation of Ca salts other than CaF2, because the spectra of the reagents CaCO3 (guaranteed grade, FUJIFILM Wako Pure Chemical Corporation) and Ca(OH)2 (guaranteed grade, FUJIFILM Wako Pure Chemical Corporation) were not detected (Fig. 12). Because the eliminated F– could be separated from the black PVDF and TiO2 residues by filtration, this alkaline wet processing method has the advantages of reducing contaminants and producing high-purity recovered CaF2.
Fig. 11
a Precipitates obtained from the reaction between the filtrate and CaCl2, b reaction residues
Fig. 12
XRD patterns of recovered CaF2 and reagent Ca salts
×
×
Recyclability of recovered CaF2
The recyclability of the recovered CaF2 from the PV backsheet into the supply chain of fluorspar was investigated. Commercial fluorspar is broadly divided into high-purity acid-grade fluorspar and low-purity metallurgical-grade fluorspar. SiO2, S, As, Pb, and CaCO3 are considered impurities [28]. Table 4 compares the elemental composition of the recovered CaF2 with the elemental compositions required of commercial fluorspar. For comparison, all weight percentages of Ca and F measured by WDXRF were determined to indicate CaF2 because the only form of calcium present was CaF2, as suggested by the XRD pattern (Fig. 12). The purity of the recovered CaF2 was 97.5%, which satisfied the quality standards for acid-grade fluorspar. Although SiO2 and As from the glass and Pb from the electrical wiring were expected to be present, the concentrations of these impurities were lower than the maximum levels allowed in acid-grade fluorspar. The origin of S was unknown, but the concentration satisfied the quality standard for metallurgical-grade fluorspar. Therefore, the waste PV backsheet was used as the sample, which contained trace amounts of Fe, Al, and P despite their quality standards not being applied. Acid-grade fluorspar also has a maximum allowable CaCO3 concentration. However, from the XRD pattern (Fig. 12), we determined that no significant amount of CaCO3 was produced in the CaF2 production process. These results showed that the recovered CaF2 satisfied the quality standards for commercial fluorspar. Therefore, a F recycling route incorporating CaF2 recovered from PV backsheet into the supply chain of fluorspar is feasible.
Table 4
Comparison between the quality of recovered CaF2 and the quality standards of commercial grade fluorspar
Analysis results
Standards of commercial fluorspar
Recovered CaF2
Method
Acid grade
Metallurgical grade
CaF2
97.5 [wt%]
WDXRF
> 97 [wt%]
> 80 [wt%]
SiO2
0.52 [wt%]
< 1.0 [wt%]
< 15 [wt%]
S
0.18 [wt%]
ICP-OES
< 0.1 [wt%]
< 0.3 [wt%]
As
3 [ppm]
< 12 [ppm]
–
Pb
15.7 [ppm]
< 550 [ppm]
< 0.5 [wt%]
Fe
511 [ppm]
–
–
Al
45.7 [ppm]
–
–
P
63.7 [ppm]
–
–
CaCO3
–
–
< 1.5 [wt%]
–
Conclusions
The defluorination of PVDF by alkaline wet processing was investigated. Increasing the alkalinity, PTC concentration, and reaction temperature increased the defluorination rate, which reached 89.2% in 4.0 M NaOH and 50 mM TBAB at 90 °C under aqueous conditions. The PVDF defluorination reaction exhibited first-order reaction kinetics. The defluorination reaction proceeded faster in DEG and TEG than under aqueous conditions, and the increase in reaction efficiency was attributed to the high affinity between PVDF and DEG or TEG. The defluorination reaction was completed within at least 150 min, even for the fine particle reagent PVDF powder, under the experimental conditions of this study. To ensure the feasible treatment of large quantities of waste fluoropolymers in the future, more efficient defluorination reaction conditions must be developed. Increasing the reaction temperature and selecting solvents with high affinities for PVDF may be effective approaches to increasing the reaction efficiency. Further research is needed in this regard.
The feasibility of F recycling from waste PV backsheet (as EoL products) was also investigated. The defluorination of a PV backsheet and subsequent CaF2 precipitation from the eliminated F– was performed. Similar to the defluorination behavior of the PVDF powder, a high defluorination rate was achieved, but the reaction took a longer time to complete. F– could be separated from the reaction residues and impurities by filtration, and high-purity CaF2 was recovered. The recovered CaF2 satisfied the quality standards of commercial fluorspar. A total of 88.3% of F contained in the PV backsheet was recovered by the recycling process. This indicates that alkaline wet processing has a high F recycling efficiency; thus, it is an effective approach to manage waste fluoropolymers and aids the development of a F recycling route along the F supply chain.
Acknowledgements
This study was supported by JSPS KAKENHI (Grant Numbers JP20H05708, JP20K12271).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
