Version of Record: https://www.sciencedirect.

com/science/article/pii/S0921344920302184
Manuscript_cce333957b873216a7d4a28c48734393

1 Electrochemical-Assisted Leaching of Active Materials from Lithium Ion Batteries

2
3
4
5
6
7
8 Luis A. Diaza, Mark L. Straussa, Birendra Adhikarib, John R. Klaehnb, Joshua S. McNallyb, and
9 Tedd E. Listera*
10
11
12
13
14
15 a Chemical Processing Department, Idaho National Laboratory, Idaho Falls, ID 83402, USA
16 a Chemical Separations Department, Idaho National Laboratory, Idaho Falls, ID 83402, USA
17 * Corresponding Author. Tel.: +1 208 526 4320. Fax: +1 208 526 4822. E-mail address:
18 [email protected] (T.E. Lister).
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

© 2020 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
47 Abstract

48 The development of a circular economy for lithium-ion batteries (LIB) is essential to realize

49 decarbonization and an electrified energy market. However, the current commercial operations

50 that recover valuable constituents from LIBs require significant energy and reagents, which

51 create toxic emissions and additional waste. We report an electrochemical-based method for

52 leaching valuable metals from the active materials of mixed shredded LIBs. In this process, the

53 use electrons, as green reagent, allows the use and regeneration of Fe2+ in low concentrations as

54 substitute of hydrogen peroxide as reducing agent. Leaching in a membrane separated two-

55 compartment electrochemical cell contributes to decrease the acid requirements as H+ can be

56 generated electrochemically. With this design, leaching efficiencies over 96% for the active

57 metals (Li, Co, Mn, and Ni) were demonstrated at pulp densities up to 240 g/L. Copper present

58 in the active material is electrowon and recovered separately. Preliminary cost analyses

59 demonstrate ca. 80% reduction in energy and chemical costs as compared to traditional

60 hydrometallurgical routes.

61

62 Keywords

63 Electrochemical leaching, lithium ion batteries, recycling

64

65
66 1. Introduction

67 Creating an efficient circular economy for lithium ion batteries (LIB) is crucial to meeting future

68 materials needs for decarbonized energy systems. Significant recycling is already being

69 performed. Roughly 97,000 metric tons of LIBs were recycled globally in 2018—more than 85%

70 of this processing was done in China and Korea (Linnenkoper, 2019). This represents 50% of the

71 worldwide volume for end of life LIBs. As of late 2018, 20 companies were recycling LIBs

72 worldwide and several more are operating at the pilot stage (Jacoby, 2019; Pinegar and Smith,

73 2019) . As LIBs demand increases the need for efficient recycling is becoming more apparent.

74 The state-of-the-art LIB recycling can be divided into three main categories: direct recycling,

75 pyrometallurgy, and hydrometallurgy. Direct recycling seeks comprehensive and efficient re-use

76 of all major battery components with minimum chemical processing (Ciez and Whitacre, 2019).

77 Hence, implementation is contingent upon the manufacturing of batteries that are designed for

78 disassembly. The variability of battery designs and chemistries, and continuously evolving

79 cathode metal oxides formulations have been identified as the main drawback for direct

80 recycling deployment within the circular battery economy (Harper et al., 2019; Sloop et al.,

81 2019).

82 Pyrometallurgical processes are robust and flexible enough to accept all kinds of batteries with

83 minimum pre-processing requirements. As smelting is performed at high temperatures, the

84 combustible qualities of LIBs can be controlled within the furnace (and supply energy).

85 However, pyrometallurgical processes are not stand alone, as hydrometallurgical refining is

86 required to separate and recover the metals of interest (Co, Ni, and Cu) from the produced alloy

87 and Li from the slag (Liu et al., 2019). The carbon footprint of the recycled materials is affected

88 by the use of fossil-based fuels in the smelting process (Chen et al., 2019). Hydrometallurgical
 89 processes are highly selective and efficient in recovering high purity materials (Chen et al.,

90 2019). Although upstream LIB pre-processing and physical beneficiation are required to

91 condition the material for effective leaching (Pinegar and Smith, 2019), mild operating

92 conditions can enable distributed processing of the LIB with considerable lower capital costs

93 than pyrometallurgical processes. In general, hydrometallurgical processes are considered

94 cleaner and less costly than the high temperature processes (Ciez and Whitacre, 2019; Gao et al.,

95 2018; Zhang, X. et al., 2018). However, extensive chemical consumption and waste generation

96 significantly increase the operating costs and overall environmental impact of the

97 hydrometallurgical processes (Meshram et al., 2016). A cradle-to-grave life cycle assessment

98 (LCA) should include the impacts of the production of the required chemicals.

99 Hydrometallurgical leaching of metals from LIBs is dominated by the combination of sulfuric

100 acid (1 to 4 M H2SO4) and hydrogen peroxide (1 to 15 wt.% H2O2) (Zhang, X. et al., 2018).

101 Although organic acids have also been tested as an alternative with lower environmental impact,

102 the reported pulp density operations are low < 60 g/L (Meshram et al., 2020; Zhang, X. et al.,

103 2018). For the H2SO4/H2O2 leachates, high leaching efficiencies >90% are reported for Co, Ni,

104 Mn, and Li from cathode materials at temperatures over 40 ◦C and pulp densities between 30 to

105 200 g/L (Gao et al., 2018; Zhang, X. et al., 2018). H2O2 acts as reducing agent to convert Co+3

106 and Mn+4 into their soluble forms, Co+2 and Mn+2, as shown in Equation 1 (He et al., 2017).

107 Although H2O2 is claimed to be a clean reducing agent due to its decomposition products

108 (oxygen and water) (Pagliaro and Meneguzzo, 2019), synthesis of H2O2 is associated with

109 significant environmental impacts that affect the LCA of hydrometallurgical processes (Li et al.,

110 2019; Pagliaro and Meneguzzo, 2019). Moreover, the costs and risks associated with H2O2

111 transportation and storage cannot be ignored (Moreira et al., 2017).

112 + 1.5 + 1.5

113 → 0.5 + + + + +3 ;

114 + + =1 (1)

115 Alternatives to H2O2 are found in Co mining (sulfur dioxide and/or iron (II) (Fe2+) sulfate,

116 sodium thiosulfate, ascorbic acid), using green solvents, or organic molecules, such as glucose

117 and ethanol (Gao et al., 2019a; Gao et al., 2019b; Harper et al., 2019). In the present work, we

118 describe an electrochemical-assisted leaching (ECL) method that continuously regenerates Fe2+

119 in small concentrations, achieving near complete metal leaching from shredded LIBs.

120 2. Experimental

121 2.1 Materials

122 Leaching solutions were prepared in DI water (18 MΩ-cm) using ACS grade or better chemicals.

123 The LIB metal oxides filter cake (MOFC) was provided by Retriev Technologies. Retriev

124 Technologies uses a proprietary cryogenic milling process followed by physical separations to

125 segregate plastics, steel, and aluminum/copper current collectors, and MOFC from mixed LIBs.

126 The MOFC is primarily a mixture of the anode and cathode powder (active materials) although

127 some other particles such as current collector foils are present. The composition of metals in the

128 MOFC was determined after digestion in aqua regia. Triplicates of the MOFC digestion were

129 performed, and the average composition is presented in Table 1.

130 2.2 Electrochemical cell

131 The electrochemical cell, built in house, consisted of two compartments with a maximum

132 capacity of 60 mL and 250 mL for the anode and cathode, respectively. The transversal area of

133 the cell is of 36 cm2. Anode and cathode compartments were separated by a bipolar membrane
134 Fumasep FBM-PK. The anode consisted on a nickel foam (MTI Corporation) with an immersed

135 geometric area of 36 cm2. 1 M KOH solution was used as anolyte. The cathode consisted on a

136 stainless steel 80 x 80 mesh size with an immersed surface area of 36 cm2. A Biologic SP-300

137 potentiostat was used to provide the required potential and measure the current and cell voltage.

138 A BASi RE-5B Ag/AgCl reference electrode was used. All experiments were performed at room

139 temperature (20 to 23 ⁰C).

140 2.3 Analysis

141 The concentrations of metals in the leachate and digested materials were measured using an

142 inductive coupled plasma mass spectrometer (ICP-MS), Thermo Scientific iCAP Q. Calibration

143 was performed using commercially prepared standards (VGH and Spex). X-ray diffraction

144 (XRD) spectroscopy was performed using a Bruker D8 Advance diffractometer operated at 40

145 kV and 40 mA, with a cobalt (Co) target (K = 1.78897A). Peaks were identified using the

146 software EVA and a 2010 ICDD data base. Thermogravimetric analysis (TGA) was performed

147 on a TA Instrument Q500 TGA in air. TGA was operated with a temperature ramp of 10 ◦C/min

148 from 20 ◦C to 800 ◦C. The commercial graphite measured was from MTI corporation and the

149 leached MOFC was taken from a sample of leached MOFC. 10 mg of sample was used for each

150 run.

151 2.4 Economic analysis

152 A techno-economic model was developed to calculate the overall capital cost, energy cost (both

153 electricity and heating) and reagent costs for ECL and three additional peroxide-based leaching

154 scenarios. Process flow diagrams for the leaching of metals using electrochemical methods and

155 peroxide-based leaching are shown in Figures S4 and S5.

156 For all scenarios, several assumptions are made, including the sizing of the systems. All

157 dependent parameters of the cost calculations depend on the feed input composition. The sizing

158 of the reactors is based on the residence time, flow rate and amount of feed input. The material

159 chosen for the reactor (Teflon was used for the analysis reported here), the sizing of the reactor

160 and pump to simulate agitation through forced recirculation was considered as well as the

161 amount of electrical energy required for the electrochemical leaching and the heat required for

162 the peroxide-based leaching. Subsequently, using the per unit cost of the material and energy,

163 capital cost of the reactor and equipment, energy cost and solvents cost are calculated.

164 3. Results and discussion

165 3.1 Electrochemical-assisted leaching

166 The ECL method presented here employs Fe2+ as a reducing agent. Based on a difference of

167 0.076 V in the standard reduction potential (E0), H2O2 is a slightly stronger reducing agent than

168 Fe2+. Fe2+ (1 g Fe/L in H2SO4) was used for the leaching of Co from the Gecamines mine in

169 Katanga region of the Democratic Republic of Congo (Equation 2) (Kafumbila, 2018). Recently

170 published research has demonstrated its applicability to the leaching of LIBs cathode material

171 (Porvali et al., 2019b).

172 ∗ +3 + 2!" →2 + !" ( # +4 (2)

173 In their work (Porvali et al., 2019b) demonstrated fast extraction kinetics, with Co leaching

174 efficiencies over 90%, when higher than stoichiometric Fe2+/LiCoO2 ratios were used. However,

175 it was recognized that the use of high iron concentration in the leaching is counterproductive as

176 this will affect downstream purification steps. For iron concentrations below stoichiometric

177 proportions, leaching efficiencies are enhanced by adding metallic copper. Although galvanic
178 reactions with metallic copper can directly reduce the cobalt from the cathode material (Equation

179 3), It was found that the regeneration of Fe2+ with copper metal is the dominant mechanism

180 achieve high leaching efficiencies with low Fe concentrations (Equations 2 and 4).

& & & &

181 2 + %+4 =2 + % + +2 (3)

182 !" ( # + %' → % + 2!" (4)

183 In an analogous way, this proof-of-concept demonstrates that electrons, which can be obtained

184 from carbon-free energy sources, can be used to regenerate the reducing agent Fe2+ (Equation 5)

185 and achieve complete leaching of the metals of interest (Li, Co, Mn, and Ni).

&
186 !" + " ( → !" &
(5)

187 A diagram of the ECL system is shown in Figure 1. The process starts with LIB metal oxides

188 filter cake (MOFC) from an industrial battery recycler in North America. The MOFC is a

189 mixture of the anode and cathode powder (active materials) (Pinegar and Smith, 2019). In a

190 typical operation, a mixture of batteries with different chemical compositions and configurations

191 can be simultaneously processed. Hence, the MOFC is a complex mixture of different LIB

192 cathode formulations. Elemental analysis data in Table 1 and XRD diffractograms in Figure 2

193 indicate the as-received material is a mixture of at least five different cathode formulations

194 including LiCoO2, LiMnxCoyO2, LiNixMnyCoZO2, LiCuXMnO2, and LiNixCoxAlzO2, mixed with

195 the graphite anode. Over 30 wt.% of the MOFC corresponds to non-metals, which includes

196 graphite/carbon material, and polymeric bits of battery separators.

197

198 Figure 1. Schematic of ECL of metals from MOFC. leachate, carbon/graphite materials, and

199 electrowon copper can be recovered separately.

200 The core of the technology presented herein consists of a two-compartment, open air,

201 electrochemical cell separated by a bipolar membrane. On the cathode side, a slurry of the

202 MOFC in 200 mL of a (0.5 to 2 M) H2SO4 solution with 10 mM FeSO4 is continuously agitated

203 while a stainless steel (SS) cathode mesh is used to provide the electrons required for the

204 regeneration of the reducing agent. ECL is performed at a controlled cathode potential of -0.3 V

205 vs. Ag/AgCl, determined to be the optimum operating potential through cyclic voltammetry of

206 the FeSO4 solution in H2SO4 (Figure S1). Employing a more negative potential will promote side

207 reactions, such as the evolution of hydrogen and/or the electrodeposition of the leached Ni and

208 Co, whereas a higher potential will decrease reaction rates and the potential may not be sufficient

209 for the Fe3+ reduction to proceed (Figure S1).

210 In the anode chamber, the oxidation of water can produce the balance of protons (H+) necessary

211 to complete the leaching (Equation 6), thereby reducing the acid requirements to nearly
212 stoichiometric ratios. However, the separation of the anode and cathode chambers is important to

213 avoid oxidation of Fe2+ and/or the leached metals such as Mn+2 whose oxidation potential to

214 MnO2 is close to that of water oxidation. A bipolar membrane (BPM) was selected as the best

215 alternative to maintain control of the electrolyte pH, and prevent electrolyte crossover (Li et al.,

216 2018; Vermaas and Smith, 2016). The BPM allows water splitting at the interface of two bonded

217 membranes (anion and cation exchange membranes) that is driven by the difference in potential

218 between anode and cathode (Equation 7). The protons are transferred to the cathode while the

219 hydroxyl ions (OH-) are transferred to the anode. The BPM allows the operation of anode and

220 cathode chambers at different pH regimes. Performing oxygen evolution in alkaline media

221 (Equation 8) avoids the need for platinum group metals anodes, and the thermodynamic cell

222 voltage (E0cathode-E0anode) is reduced by 0.4 V.

& & & & & & &

223 + !" +4 → + + + + !" +2 (6)

( &
224 → + (7)

(
225 4 → + 2 + 4" ( (8)

226 Table 1. Composition of MOFC and undissolved material after ECL.

Undissolved material Undissolved material Undissolved material

MOFC
0.5 M H2SO4, 65 g/L 1 M H2SO4, 120 g/L 2 M H2SO4, 240 g/L
Metal wt % wt % Leaching wt % Leaching wt % Leaching %
Li 2.31% 0.04% 99.45% 0.03% 99.61% 0.01% 99.87%
Co 14.16% 0.36% 99.13% 0.35% 99.19% 1.28% 96.67%
Ni 4.27% 0.10% 99.20% 0.07% 99.44% 0.44% 96.24%
Mn 3.31% 0.03% 99.73% 0.05% 99.49% 0.31% 96.53%
Fe 0.16% 0.24% 47.50% 0.10% 80.28% 0.09% 78.69%
Cu 0.16% 0.09% 79.87% 0.00% 99.32% 0.00% 99.20%
Zn 0.09% 0.04% 86.48% 0.01% 97.84% 0.00% 98.32%
Al 0.48% 0.25% 82.28% 0.78% 47.96% 0.90% 31.64%
Undissolved mass g 4.41 7.74 17.63
Charge C 5808.29 9068.55 19506.60
 Energy W-s 1738.16 2718.89 5844.83
227

228

229

230

231 Figure 2. XRD diffractogram of the as-received MOFC. The diffractograms shows that MOFC
232 consists of a mixture of anode and cathode materials with different chemical formulations. A.U.
233 arbitrary units
234

235 3.2 Metals leaching from MOFC

236 ECL of metals from the MOFC was performed at room temperature using a batch mode

237 operation while maintaining control of the cathode potential. An excess of at least 30% molar

238 H2SO4, based on the composition of Co, Mn, and Ni in the MOFC (Table 1), was added to the

239 catholyte in order to control the final pH. The overall reaction (Equation 9) obtained from

240 Equations 4 to 7 indicates acid consumption during the leaching process. In addition, it was
241 necessary to maintain the pH of the catholyte (by acid addition) below 3 to avoid precipitation of

242 iron during the leaching process. For all three reactions, the final pH of the leachate was above

243 pH 1.

& & & & &

244 +3 → + + + + !" + 3/2 + 1/4 (9)

245 Three different MOFC pulp densities were tested, 65, 120, and 240 g/L in three acid

246 concentrations, 0.5, 1, and 2 M H2SO4, respectively, keeping the anolyte volume at 200 mL and

247 the FeSO4 concentration at 10 mM for all three experiments. A graphical representation of the

248 advantages of the ECL in comparison with other hydrometallurgical processes reported in the

249 literature is presented in Figure 3. This interpretation is based on the comparison presented by

250 (Gao et al., 2018) .

251

252

253 Figure 3. Operation parameters, pulp density vs. acid concentration of literature reported

254 hydrometallurgical processes for the leaching of metals from LIBs (Gao et al., 2018). Solid
255 points represent reported pulp densities at the corresponding acid concentration. While the

256 regions in color are a graphical representation of the operating range.

257 The current profile obtained with the 65 g/L leaching is shown in Figure 3, where the rate of

258 reaction (mass transferred limited current) is controlled by the bulk reaction of Fe2+ with the

259 active material. As metals in the MOFC are depleted as the experiment progresses, the reaction

260 rate decreases to the point that minimal current is observed, which can be identified as the

261 termination point for the leaching.

262

263

264 Figure 4. Current and voltage profiles during the ECL with pulp density of 65 g/L.

265 The potentiostat was programed to apply the required potential for a duration of 30 hours.

266 However, the leaching was considered complete once the net current fell below 10 mA. For the

267 65 g/L experiment, the leaching time was achieved within eight hours. Processing time was not

268 established for the highest pulp densities due to agitation problems that impeded operation of the
269 system. These current profiles are shown in the supporting file, Figure S2. At higher pulp

270 densities, the use of mechanical agitation instead of stir bar can solve these issues.

271 Leaching efficiencies over 96% were achieved for Li, Co, Mn, and Ni (Table 1 and Figure 4). By

272 tracking the cell voltage and current profiles, presented in Figure 4, it was possible to calculate

273 the energy of the ECL process in the three experiments. Specific energy consumptions suggest a

274 linear relationship with the amount of material processed. An average of 34.1 kWh per ton of

275 MOFC, with an 8.3% error, can be estimated as the electrical energy consumption of the process.

276 This average energy only considers the electrical input for the electrochemical process.

277

278

279 Figure 5. Leaching efficiency for metals and specific energy consumption at the tested pulp

280 densities.

281 The Faraday efficiency (FE) of the process was estimated from the information presented in

282 Table 1. FE calculations were based on the moles leached of Co and Mn, which are the metals
283 that are reduced [12]. FE values of 77.2, 82.8, and 92.5 % were calculated for the 64, 120, and

284 240 g/L experiments, respectively. The calculation of FE is presented in the supporting

285 information. An average FE of 83.9% suggests that other processes beyond Fe3+ reduction may

286 have occurred on the cathode. The reduction of Cu on the cathode is a visible side reaction that

287 occurred (Figure 1). Unfortunately, due to the low quantities of Cu plated on the electrode and

288 the presence undissolved carbon/graphite material mixed with copper, it was not possible to

289 measure the quantity of Cu deposited. Moreover, it is not certain that all Cu present in the MOFC

290 was in the metallic state. Pre-treatment processes (shredding and milling) can result in Cu and Al

291 metal particles from the current collectors being present in the MOFC (Porvali et al., 2019a) .

292 Nevertheless, the concentration of Cu in the final leachate did not exceed 5 ppm for all three

293 experiments. The ability to plate Cu during leaching provides a major advantage of ECL, as Cu

294 is recognized as a main impurity in hydrometallurgical process (Peng et al., 2020). Hence,

295 downstream processes for Cu separation can be avoided. Composition of the leachate solutions

296 are presented in Table S1.

297 XRD analysis of the undissolved material (Figure 5) shows that the main crystalline phase is

298 graphitic carbon, with some small peaks suggesting the presence of aluminum oxide phases. This

299 result agrees with the overall aluminum recovery, which featured low leaching efficiencies for

300 the three experiments. Furthermore, Co, Mn, Ni and mixed oxides were not detected.

301 Thermogravimetric analysis (TGA) was performed on the undissolved fraction which was

302 compared against commercial synthetic graphite. TGA analysis (Figure S3) suggests that at least

303 20% of the mass in the undissolved material corresponds to non-graphite material. Graphite is

304 stable up to 600 ◦C, and binders and other organic components in LIBs electrodes are reported to

305 start decomposing above 240 ◦C (Yang et al., 2019; Zhang, J. et al., 2018). Thermal processing
306 of the undissolved material can be used to as a strategy to improve the purity of the recovered

307 graphite.

308

309

310 Figure 6. XRD diffractogram of the undissolved material from the ECL. Inset graph shows a

311 magnification of the diffractogram. A.U. arbitrary units.

312 3.3 Techno-economic considerations

313 To understand the economic benefits of the ECL, a preliminary techno-economic analysis (TEA)

314 was developed to compare peroxide-based leaching with the proposed ECL. For that purpose,

315 three different scenarios of the peroxide based leaching were considered: A) stochiometric ratios

316 of H2SO4 and H2O2 to the moles of transition metals (Ni, Co, and Mn) in the MOFC as presented

317 in Equation 1 (at 60 ◦C, 1 h leaching time and 1727 g/L pulp density), B) 2 M H2SO4 + 6 vol %

318 H2O2 as the leaching agent (at 60 ◦C, 1 h leaching time and 100 g/L pulp density) and C) 2 M

319 H2SO4 + 5 vol % H2O2 as the leaching agent (at 60 ◦C, 2 h leaching time and 100 g/L pulp
320 density). The conditions for scenarios B and C were obtained from reported studies (Kang et al.,

321 2010a; Kim et al., 2014) .

322 For ECL, a mixture of 0.5 M H2SO4 and 0.01 M Fe2SO4 was used as the leaching solution. The

323 residence time per cycle was assumed to be 4 h, 2 h and 1 h for stoichiometric scenarios, A, B,

324 and C, respectively. A constant feed rate of 10000 kg/h was chosen as the benchmark for all

325 scenarios. Feedstock concentration was taken as reported on Table 1 for the MOFC.

326 Process alternatives were compared on two main categories as shown in Figure 7: the cost of

327 reagents in $/tfeed and energy cost required in $/tfeed. The cost of MOFC feedstock was not

328 included. While not considering the cost of the feedstock affects the certainty of the reported

329 operational costs, consistent assumptions are necessary, and still valid, when comparing

330 competing technologies. It is worth mentioning that scenario A is an ideal case (stoichiometric

331 ratios presented in Equation 1), which benchmarks the lowest possible costs that could be

332 achieved with the peroxide-based leaching scenario. In terms of reagent costs, ECL was found to

333 be significantly lower in cost ($227/tfeed) than all three considered peroxide-based leaching

334 scenarios. A reduction in reagent costs of at least 50 % and as high as 84% can be achieved with

335 ECL. Nevertheless, these costs only include the reagents required for the metals leaching.

336 Additional chemical costs incurred in the removal of metals, such as Cu, Fe, and Al, were not

337 considered (Kang et al., 2010b; Porvali et al., 2019a). The higher final pH of the ECL leachate

338 (>1) in comparison with the reported peroxide-based leaching (pH 0.3 (Kang et al., 2010b))

339 suggests an additional cost reduction in chemical consumption (pH adjustment) for downstream

340 metal separation. Process optimization can be performed to achieve better control of the final

341 leachate pH without causing iron precipitation during the leaching (Koppenol and Hider, 2019).
342 Hence, the final pH of the leachate can approach those optimum for metal separation based on

343 solvent extraction (Jose, 1998).

344

345

346 Figure 7. Operational costs (reagent and energy) for ECL and peroxide based alternatives. Inset

347 shows the estimated capital costs.

348 The calculation of energy consumption includes the electrical energy in the case of the ECL

349 (34.1 kWh/tfeed), and the energy required to achieve and maintain the temperature at 60 ◦C for the

350 peroxide-based scenarios. Energy for mixing was simulated with a recirculation pump providing

351 external circulation. The comparison in energy consumption also shows an advantage for ECL

352 over peroxide-based scenarios B and C (Figure 7). Close to an 80% reduction in energy

353 consumption can be achieved with ECL. Scenario A shows a 29% lower cost of energy than

354 ECL, but this is an ideal scenario (stoichiometric ratios). The low energy costs estimated for the

355 scenario A are due to the high pulp density, which is a consequence of lower volumes of slurry

356 to be heated and processed.

357 In addition, we also compared the capital cost of the different scenarios. The size of the reactor

358 vessels was estimated based on the residence time. The material chosen for the electrochemical

359 reactor was Teflon. However, less expensive material such as PVC or polypropylene could be

360 used. A constant feed rate of 10000 kg/h was chosen as the benchmark for all scenarios. Due to

361 the longer reaction times required for ECL, the estimated capital cost is significantly higher than

362 the peroxide-based leaching. Further optimization will be required to decrease leaching times,

363 which could be achieved by increasing the iron concentration and/or improving the agitation of

364 the reaction slurry.

365 4. Conclusions

366 An electrochemical assisted technology was designed and tested for the complete leaching of

367 valuable metals (Li, Co, Mn, and Ni) from recycled lithium ion batteries. The proposed

368 technology is based on the use of electrons as a green reagent for the substitution of chemicals

369 during a hydrometallurgical based leaching process. Hence, an electrified leaching process,

370 which could be powered by low carbon energy sources, fits well with the overall goals to achieve

371 decarbonization and an electrified energy market. The reductant (Fe2+) is already present in the

372 feedstock material and can be regenerated infinitely using far less than stoichiometric quantities.

373 The results show that leaching efficiencies >96% can be achieved for Li, Co, Ni, and Mn at high

374 pulp densities and low concentration of mineral acids. Moreover, a less acidic pH for the

375 leachate product suggests additional decreases in chemical consumption (pH adjustment) for

376 downstream metal separation.

377 The decrease in chemical consumption and the operation at ambient temperature reduce the cost

378 of chemicals and energy ~80% when compared to peroxide based hydrometallurgical leaching.

379 Moreover, electrowon Cu and graphitic carbon can be obtained as by-products of the described
380 metal leaching process. Future optimization is required to reduce the capital costs by means of

381 reducing the leaching times.

382 Acknowledgments

383 This work was supported by the Critical Materials Institute, an Energy Innovation Hub funded

384 by the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy,

385 Advanced Manufacturing Office. The authors would like to acknowledge Retriev Technologies

386 (Richard Schutte and Kathy Bruce) for providing the MOFC materials used in this study as well

387 as technical guidance, and the analytical support provided by Byron White, Arnold Erickson,

388 Michael Jones, and Dr. Dayna Daubaras at Idaho National Laboratory.

389 References

390 Chen, M., Ma, X., Chen, B., et al., 2019. Recycling End-of-Life Electric Vehicle Lithium-Ion

391 Batteries. Joule 3(11), 2622-2646.

392 Ciez, R.E., Whitacre, J.F., 2019. Examining different recycling processes for lithium-ion

393 batteries. Nature Sustainability 2(2), 148-156.

394 Gao, G., He, X., Lou, X., et al., 2019a. A Citric Acid/Na2S2O3 System for the Efficient

395 Leaching of Valuable Metals from Spent Lithium-Ion Batteries. JOM 71(10), 3673-3681.

396 Gao, G., Luo, X., Lou, X., et al., 2019b. Efficient sulfuric acid-Vitamin C leaching system:

397 Towards enhanced extraction of cobalt from spent lithium-ion batteries. Journal of Material

398 Cycles and Waste Management 21(4), 942-949.

399 Gao, W., Liu, C., Cao, H., et al., 2018. Comprehensive evaluation on effective leaching of

400 critical metals from spent lithium-ion batteries. Waste Management 75(2018), 477-485.
401 Harper, G., Sommerville, R., Kendrick, E., et al., 2019. Recycling lithium-ion batteries from

402 electric vehicles. Nature 575(7781), 75-86.

403 He, L.P., Sun, S.Y., Song, X.F., et al., 2017. Leaching process for recovering valuable metals

404 from the LiNi1/3Co1/3Mn1/3O2 cathode of lithium-ion batteries. Waste Management 64, 171-

405 181.

406 Jacoby, M., 2019. It’s time to get serious about recycling lithium-ion batteries. Chemical &

407 Engineering News 97(28).

408 Jose, F., 1998. Cobalt – Manganese Separation : the Extraction of Cobalt ( II ) from Manganese

409 Sulphate Solutions b y C y anex 301. (Ii).

410 Kafumbila, K., 2018. Necessity to use filblast technology for cobalt III leaching performed in

411 agitated open leach tanks.

412 Kang, J., Senanayake, G., Sohn, J., et al., 2010a. Recovery of cobalt sulfate from spent lithium

413 ion batteries by reductive leaching and solvent extraction with Cyanex 272. Hydrometallurgy

414 100(3-4), 168-171.

415 Kang, J., Sohn, J., Chang, H., et al., 2010b. Preparation of cobalt oxide from concentrated

416 cathode material of spent lithium ion batteries by hydrometallurgical method. Advanced Powder

417 Technology 21(2), 175-179.

418 Kim, S., Yang, D., Rhee, K., et al., 2014. Recycling process of spent battery modules in used

419 hybrid electric vehicles using physical/chemical treatments. Research on Chemical Intermediates

420 40(7), 2447-2456.

421 Koppenol, W.H., Hider, R.H., 2019. Iron and redox cycling. Do's and don'ts. Free Radical

422 Biology and Medicine 133(July 2018), 3-10.

423 Li, Y.C., Yan, Z., Hitt, J., et al., 2018. Bipolar Membranes Inhibit Product Crossover in CO2

424 Electrolysis Cells. Adv. Sustainable Syst. 2(4), 1700187.

425 Li, Z., Diaz, L.A., Yang, Z., et al., 2019. Comparative life cycle analysis for value recovery of

426 precious metals and rare earth elements from electronic waste. Resources, Conservation and

427 Recycling 149.

428 Linnenkoper, K., 2019. The unknown success of lithium-ion battery recycling.

429 Liu, C., Lin, J., Cao, H., et al., 2019. Recycling of spent lithium-ion batteries in view of lithium

430 recovery: A critical review. Journal of Cleaner Production 228(1), 801-813.

431 Meshram, P., Abhilash, Pandey, B.D., et al., 2016. Comparision of Different Reductants in

432 Leaching of Spent Lithium Ion Batteries. Jom 68(10), 2613-2623.

433 Meshram, P., Mishra, A., Abhilash, et al., 2020. Environmental impact of spent lithium ion

434 batteries and green recycling perspectives by organic acids – A review. Chemosphere 242,

435 125291.

436 Moreira, F.C., Boaventura, R.A.R., Brillas, E., et al., 2017. Electrochemical advanced oxidation

437 processes: A review on their application to synthetic and real wastewaters. Applied Catalysis B:

438 Environmental 202, 217-261.

439 Pagliaro, M., Meneguzzo, F., 2019. Lithium battery reusing and recycling: A circular economy

440 insight. Heliyon 5(6), e01866-e01866.

441 Peng, C., Lahtinen, K., Medina, E., et al., 2020. Role of impurity copper in Li-ion battery

442 recycling to LiCoO2 cathode materials. Journal of Power Sources 450.

443 Pinegar, H., Smith, Y.R., 2019. Recycling of End-of-Life Lithium Ion Batteries, Part I:

444 Commercial Processes. Journal of Sustainable Metallurgy 5(3).

445 Porvali, A., Aaltonen, M., Ojanen, S., et al., 2019a. Mechanical and hydrometallurgical

446 processes in HCl media for the recycling of valuable metals from Li-ion battery waste.

447 Resources, Conservation and Recycling 142(November 2018), 257-266.

448 Porvali, A., Chernyaev, A., Shukla, S., et al., 2019b. Lithium ion battery active material

449 dissolution kinetics in Fe(II)/Fe(III) catalyzed Cu-H2SO4 leaching system. Separation and

450 Purification Technology 116305(October).

451 Sloop, S.E., Crandon, L., Allen, M., et al., 2019. Cathode healing methods for recycling of

452 lithium-ion batteries. Sustainable Materials and Technologies 22.

453 Vermaas, D.A., Smith, W.A., 2016. Synergistic Electrochemical CO2 Reduction and Water

454 Oxidation with a Bipolar Membrane. ACS Energy Lett. 1(6), 1143-1148.

455 Yang, Y., Song, S., Lei, S., et al., 2019. A process for combination of recycling lithium and

456 regenerating graphite from spent lithium-ion battery. Waste Management 85, 529-537.

457 Zhang, J., Li, X., Song, D., et al., 2018. Effective regeneration of anode material recycled from

458 scrapped Li-ion batteries. Journal of Power Sources 390(April), 38-44.

459 Zhang, X., Li, L., Fan, E., et al., 2018. Toward sustainable and systematic recycling of spent

460 rechargeable batteries. Chemical Society Reviews 47(19), 7239-7302.

461