Recycling Cathode of Lithium-Ion Battery by Using Deep Eutectic Solvents to Extract Cobalt

Recycling Cathode of Lithium-Ion Battery by Using

Deep Eutectic Solvents to Extract Cobalt

Raya Zamil

Master thesis, 30 hp

Supervisor Umeå university: Jean-Francois Boily. Supervisors Uppsala university: 1. Reza Younsi. 2. John Ostrander. Examiner: Madeleine Ramstedt, Christiane Funk, Fredrik Almqvist

I

Abstract

Cobalt is the most critical element used in lithium-ion batteries (LIBs). Therefore, with the fast increase in demands for production of LIBs, recycling of battery materials such as cobalt is becoming more important. Here, A novel method has been tested to extract cobalt from the cathode of Lithium Ion Batteries (LIBs) using two Deep Eutectic Solvents (Urea with Choline Chloride) and (Ethylene Glycol with Choline Chloride). The main principle approach was to thermally dissolve cobalt, then to recover cobalt by chemical precipitation. Variations in temperature (100 °C, 120 °C, 140 °C, 160°C and 180°C only for Ethylene glycol samples) and reaction times (24, 48, 72 h) allowed us to explore conditions to optimize extraction processes. Cobalt concentrations were measured by Inductive Coupled of Plasma- Optical Emission Spectrometry (ICP-OES), and the soluble complexes characterized by ultraviolet-visible spectroscopy (Uv-Vis) and Fourier-Transform Infrared Spectroscopy (FTIR).

III

List of abbreviations

DES Deep Eutectic Solvent Co Cobalt

LCO Lithium Cobalt Oxide EG Ethylene Glycol ChCl Choline Chloride

Reline Mixture of Urea and Choline Chloride

Ethaline Mixture of Ethylene Glycol and Choline Chloride

ICP-OES Inductive Coupled of Plasma- Optical Emission Spectrometry FTIR Fourier-Transform Infrared Spectroscopy

Uv-Vis Ultraviolet-Visible Spectroscopy

V

Table of contents

Recycling Cathode of Lithium-Ion Battery by Using Deep Eutectic Solvents to

Extract Cobalt ... 1

Abstract ... I List of abbreviations ... III 1. Introduction ... 1

Aim of the study: ... 5

2. Popular Scientific Summary Including Social and Ethical Aspects ... 6

2.1 Popular scientific summary... 6

2.2 Social and ethical aspects ... 6

3. Experimental ... 7

3.1 Chemicals ... 7

3.2 Preparation of Deep Eutectic Solvent ... 7

3.3 Digestion by heat ... 7

3.4 Filtration ... 7

3.5 Precipitation of cobalt ... 8

3.6 Instrumentation ... 8

3.6.1 Uv-visible spectroscopy (UV-vis) ... 8

3.6.2 Fourier Transform Infrared spectroscopy (FTIR) ... 9

3.6.3 Inductive coupled of plasma- optical emission spectroscopy (ICP-OES) .... 9

4. Results ... 9

4.1 Digestion ... 9

4.2 Filtration ... 11

4.3 Precipitation. ... 12

4.4 Concentration and Efficiency ... 12

4.5 Characterization ... 13 4.5.1 UV-Vis ... 13 4.5.2 FTIR analysis ... 14 5. Discussion ... 16 6. Conclusion. ... 17 Acknowledgement ... 19 References ... 20

Appendix 1: Experiment’s Pictures ... 22

1

1. Introduction

Lithium ion batteries (LIBs) are one of the most widely used batteries around the world and can easily be found in our daily lives. LIBs have been a large influence behind the IT revolution, functioning as power banks for mobile phones, laptop computers, and electric vehicles. A great effort of scientific research involving solid-state chemistry and physics has been invested to improve the efficiency of LIBs, particularly during the 1970s and 1980s then finally introduced in 1991s. Those efforts started to pay off recently, for example, i) the 2019 Nobel Prize in Chemistry devoted to the development of lithium-ion batteries (© The Royal Swedish Academy of Sciences, no date) and ii) and major plants to construct giga-factories to LIBs production. LIBs have become more popular due to several advantages over other rechargeable batteries. Some advantages of lithium include i) its light weight in comparison with other batteries of similar size ii) lithium is a highly reactive element. This means that a high amount of energy can be stored in its atomic bonds resulting in a remarkably high energy density (Ulvestad, 2018).

One of the leading technologies in reducing CO2 emissions is electric vehicles,

2

Figure 1: High and low demand scenarios for cobalt used in electric cars in the EU 2019-2030. Low scenario: estimates electric vehicles to make up 4-5 percent of the car market by 2020 and further increase to 20 percent by 2030. High scenario: estimates electric vehicles to make up 8-10 percent of the

car market by 2020 and further increase to 70 percent by 2030. (Statista Research Department, 2020)

Battery recycling can assist in reducing the harmful effects of batteries on the environment and open new doors for reclaiming or repurposing metals that are not abundant in nature like cobalt and nickel. According to the European Commission, cobalt is classified as one of the Critical Raw Material (CRM) in Europe. See Figure (2).

Figure 2: Position of cobalt as one of the critical raw material, black circle. (European Commission, 2017)

3

waste have been tested. These methods can achieve the goal of recovering metals, but the challenge here is to come up with a novel method that has a limited impact on the environment, especially in terms of emissions and energy consumption.

The primary methods currently used for the extraction process are pyrometallurgy, hydrometallurgy, biometallurgy, or a combination of these methods. Closely examining each of these approaches would provide a better insight into each. When comparing those techniques and their impact, ideas for newer methods that may meet the requisite criteria e.g. minimal environmental impact can be developed.

Pyrometallurgical technique is depending on heating and smelting. This is the traditional approach applied in the chemical industries nowadays. Pyrometallurgical processes involve the application of high temperatures (1400°C or higher) to extract metals by reducing component metal oxides to alloys. The operation of this method is quite simple and does not require any pretreatment. However, this approach consumes a high amount of energy due to the use of high temperatures that results in high costs and low recovery of the desired material. Additionally, this technique has high relative risks to the environment since it can release harmful gases during the decomposition of LIB materials, therefore off-gas treatment is mandatory (Tuncuk et al., 2012)(Ijadi Bajestani, Mousavi and Shojaosadati, 2014).

Hydrometallurgical processes can also be considered as one of the most common and widely used methods for extracting metals from LIBs cathodes. In order to extract metals, acid leaching is applied by using an inorganic strong reagent like hydrochloric acid, nitric acid, or sulfuric acid (Meshram, Pandey and Mankhand, 2015). Besides that, some reductants like NH3OH and H2O2 are used to increase the efficiency and the speed

of the leaching reactions. According to some research, the high efficiency of this approach resulted in high levels of cobalt extractions, around 99%. This can be achieved by using HCl. Similarly, the high efficiency of 95% can be achieved by using H₂SO₄ without an additive agent (Wang et al., 2016). Although leaching metals with inorganic acids is a straightforward and effective method, it poses two potential problems. The first one that it produces acidic wastewaters that are difficult to handle, and that can emit harmful gases such as Cl2, SO2, and NOx thus releasing a number of environmental

pollutants (Yao, Feng and Xi, 2015).The second problem is that this approach requires large amounts of leaching reagents and increases the risk of incidents threatening the health of workers. On the other hand, this method requires less energy in comparison with pyrometallurgy. Many organic acids are introduced as an alternative to using inorganic acids to minimize the harmful effects, utilizing less or non-hazardous materials like citric acid, formic acid, oxalic acid, etc. (Wang et al., 2016) (Nayaka et al., 2016). However, these acids are facing many challenges to overcome their negative environmental impact, for instance, they need to use additional reducing reagents or higher temperature to accelerate the process.

4

(Liu, Li and Ge, 2016) (Zeng et al., 2013). Biometallurgy method has shown outstanding results because of its many advantages (Ijadi Bajestani, Mousavi and Shojaosadati, 2014) (Lambert et al., 2015), such as i) it is much more environmentally friendly in comparison with pyrometallurgy and hydrometallurgy methods. The applied microorganism in mineral oxidation processes are autotrophs (organisms that can produce their own food using light, water, carbon dioxide, or other chemicals). That is why carbon dioxide during smelting operation (Olson, Brierley and Brierley, 2003) (Habibi, Shamshiri Kourdestani and Hadadi, 2020) (Ruan et al., 2014) ii) it is simple and cost-effective, requiring less energy and minimal maintenance. However, biometallurgy suffers from some serious drawbacks such as high cost and time-consuming processes. The process can take up to around 10 days which is longer than pyrometallurgy and hydrometallurgy processing times. Also, nickel and cobalt are difficult to separate which dictates the need for more solvent-extraction steps (Harper et al., 2019). Another drawback is the difficulty in controlling bio-reactions (Habibi, Shamshiri Kourdestani and Hadadi, 2020).

Recently, Tran et al (Tran et al., 2019) presented a novel method using green solvents called Deep Eutectic Solvents (DESs) to leach various metals from the cathode of LIBs. This method is promising to be highly efficient for cobalt recycling, and more environmental-friendly by having lower energy consumption. DES are homogenous mixtures of two or more components where the melting point of the mixture is lower than the melting point of each individual component (Smith, Abbott and Ryder, 2014). In Figure (3) a phase diagram of two-component forming deep eutectic solvent is shown.

5

To date, these solvents are synthesized by mixing quaternary ammonium halide salts which acts as hydrogen bond acceptor (HBA), one example is choline chloride (ChCl) and a hydrogen bond donor (HBD) like urea, carboxylic acids (e.g. malonic acid), or alcohols such as glycerol or ethylene glycol. These interactions between the HBD and HBA will form a strong hydrogen bond that reduces the melting point of the mixture, Figure (4) illustrates this point clearly: the interaction between NH2 group in

the urea as HBD and the anion in HBA like chloride in choline chloride is the main feature of DES’s structure (Perkins and Perkins, 2013). The molar ratio between HBD and HBA differs depending on the purpose of using DES. Previous research has established that DESs are biodegradable, non-toxic, relatively easy, and quick to prepare. DESs are generally low-cost materials and available in large quantities. After transition metals dissolve in DES different complexes can be formed. A study by (Fogel et al., 1964) investigates the species of metal salt in different DES, showed that cobalt maybe forms cobalt chloride complex with a different formula of different DES solvents.

Figure 4: The proposed mechanism of chemical reaction of in preparation of Reline

From various research within the filed, it can be inferred that urea and ethylene glycol are capable of dissolving transition metals like cobalt effectively similar to that of HCl (Abbott et al., 2006) (Tran et al., 2019). Time and temperature are the driving factors of the energy consumption of the process and hence are the determining factor in assessing to what extent the method is environmentally friendly. Accordingly, this report is built on the work of previous researches and further investigate the ability of urea and ethylene glycol in extracting cobalt from LCO cathode material and define the optimal conditions of the extraction process.

Aim of the study:

6

2. Popular Scientific Summary Including Social and Ethical

Aspects

2.1 Popular scientific summary

LIBs are something we use on a daily basis. Those batteries are easily found in devices like laptops, mobile phones, and electric vehicles. The increase in LIBs consumption will lead to an increase in the volume of battery waste which might turn into hazardous waste. This waste will have a negative impact on the environment in terms of global warming, ecological toxicity, and depletion of resources. Also, some of the elements used in batteries like cobalt and nickel are not naturally abundant everywhere, so there is increased risk of having access to those elements. Therefore, finding a solution is urgent. Recycling LIBs material is crucial in reducing the harmful effect and in reclaiming metals that are non-abundant in nature. In this study, a new method has been tested using deep eutectic solvents (DES) to recycle the cathode of LIBs by extracting cobalt. The results show the ability to reclaim cobalt by applying heat for a longer time. Additionally, it is to be considered an environmentally friendly method since it neither used nor produced any harmful products.

2.2 Social and ethical aspects

Ethics in chemistry research is of great importance due to the uniqueness of this field. This specialty manifests in a number of ways, firstly, chemists produce a high number of substances every year, and a number of these substances might have a harmful effect on the environment or the health of human beings. Additionally, chemistry is linked more to the practical application than other sciences. The high number of chemists who are active in the industry, hence again, a higher impact on the daily life cycle can consolidate this (Kovac, 2015). This study aims in helping to enhance the world we live in by introducing a new method for the extraction of cobalt from LIBs. This will allow a reduction in depleting the cobalt resources in the world.

This work can be a support in the current transition of the transportation sector from fuel-driven vehicles to electrically driven ones. This sector was among the top contributors to the CO2 emissions in 2017 according to the IEA, (IEA Data and

Statistics, 2020). Similarly, this study can be considered a step towards more advanced techniques within electricity storage development. All these aspects together will indeed have a positive impact on the environment by reducing the green gas house emissions from transportation and energy sectors.

7

3. Experimental

Lithium cobalt (III) oxide (LiCoO2 99,8%, Sigma-Aldrich), Choline chloride

(CH3)3N(Cl)CH2CH2OH > 99% (Source), Urea (NH2CONH2, Merck), an-hydrous

ethylene glycol (HOCH2CH2OH, Sigma-Aldrich), and deionized water were used

without any further purification . For chemical precipitation: sodium hypochlorite (Bleach) (COLGATE-PALMOLIVE) was used.

ICP analysis: concentrated hydrochloric acid fuming 37% (Merck), concentrated nitric acid 65% (Merck), water blank-ASTM type water 18 Megohm (SPEX CertiPrep).

3.2 Preparation of Deep Eutectic Solvent

In this study, two different Deep Eutectic Solvent (DES) are prepared for cobalt extraction. The first solvent is ethaline, (a mixture of choline chloride (ChCl) and ethylene glycol (EG)) with molar ratio (1:2) respectively. The second solvent is reline (a mixture of choline chloride (ChCl) and urea(U)) with molar ration (1:2) respectively. The mixture of two solvents is then heated to 80°C for a couple of hours forming a mixture that remains liquid at room temperature when cooled.

3.3 Digestion by heat

5 (ml) of both DES solvent ethaline and reline is mixed separately with 0,1 (g) of LCO in closed vials or test tubes to be heated by oil bath on a hot plate equipped with a temperature sensor using different digestion time intervals and temperatures. The plan was to use an oil bath for all temperatures. For samples at 140°C we attempted using sand instead of the oil bath, but inconsistent results among samples so it was not used again. Each sample has a specific digestion time (24, 48, and72) hours at a specific temperature (100, 120, 140, and 160°C) for both solvents. The (Urea-ChCl) solvent is very thick, viscous, and hard to work with for analysis, especially at the higher temperature, therefore, 4% of water was added to the samples of (Urea-ChCl) at 160 °C (Du et al., 2016). The samples of (EG-ChCl) have an additional temperature (180 °C) to investigate, at this temperature, a specific amount of water is also added (2%, 5%) to possibly improve the efficiency of leaching cobalt and to avoid crystallization of the DES which takes place at very low water contents. See Figure (14) in Appendix (1).

3.4 Filtration

After the digestion of LCO in both ethaline and reline, some unreacted LCO powder was left in the bottom of the vials. This powder is firstly filtered and then measure how much of LCO powder is remaining. The purpose of applying filtration is to estimate how much of LCO powders do not react with the solvents at each experiment conditions (time and temperature).

8

grade 50-70 mm quantitative cellulose with pore size 2,7µm) are set up using in-house vacuum lines. Filter papers are dried in the oven at 60° for approximately 24 hours to remove any residual moisture before sample filtration to obtain more accurate weight. After all residues are full, all samples are dried at 70° and some negative pressure is applied (around 830 bars) for about 4 hours. Samples are then re-weighed to obtain the amount of residual LCO. The amount of unreacted LCO is given by the following equation:

𝑈𝑛𝑟𝑒𝑎𝑐𝑡𝑒𝑑𝐿𝐶𝑂 = 𝑊2 − 𝑊1

𝑊o ∗ 100 Equation (1)

where:

W1: the weight of dried filter paper without unreacted LCO. W2: the weight of dried filter paper with unreacted LCO. Wo: the original amount of LCO (0.1g).

3.5 Precipitation of cobalt

Precipitation is widely applied to leached metals like cobalt. It is characterized by high recovery effectiveness because it is based on the valence discrepancy of metal ions. Precipitation is accomplished by adding sodium hypochlorite NaClO (aq) solution to digested samples to precipitate cobalt. However, both NaOH and H2O2 were also

tested before selecting sodium hypochlorite but they did not work well. The procedure starts with obtaining a specific volume of digested samples, usually about 500µl or less for (EG-ChCl). While for (Urea-ChCl) solvent, the volume was 100µl. Then, the sample is mixed with 100µl of water to increase the ability of any cobalt to precipitate, then sodium hypochlorite is added until precipitant cobalt is formed. The amount of hypochlorite varied depending on the cobalt content in the sample, with as little as 200µL being required to remove available cobalt from solution. The samples are then centrifuged for about 3 minutes at 2500 rpm. This helps to separate the cobalt from the mixture more quickly.

3.6 Instrumentation

3.6.1 UV-visible spectroscopy (UV-Vis)

9

3.6.2 Fourier Transform Infrared spectroscopy (FTIR)

All samples measured by Attenuated Total Reflection (ATR)-FTIR spectroscopy on Bruker Tensor 27 system with a diamond platinum ATR unit. All the spectra are recorded in absorbance mode in the (400-4000) cm-1 with a resolution of 4 cm-1 and are the result of 32 scans. All samples were sampled directly without additional preparation.

3.6.3 Inductive coupled of plasma- optical emission spectroscopy (ICP-OES)

This part of the experiment has to be performed by a licensed machine lab technician, so a licensed John Ostrander was recruited for this. The precipitation of cobalt is used to quantify the concentration of cobalt by using a PerkinElmer, Avio 200 ICP. Before starting the analysis, some sample preparations must be done, as the ICP used is not capable of handling organic compounds such as the DES used here, the cobalt precipitant must be treated by acid digestion with concentrated HNO3 and concentrated

HCl with ratio (3:1) respectively. The HNO3 is good for digesting inorganic component,

while HCl effectively digests any organic residues in the sample. Acid digestion requires one day at least. The next step is diluting the samples to 10 ml with ultrapure water, and then filtering the sample prior to analysis using 0,2µm pore filtration. Unfortunately, samples with a low amount of cobalt did not precipitate and were therefore not analyzed.

The leaching efficiency is defined by the equation below:

ƞ = (𝐶 × 𝑉

𝑀_𝑥 ) × 100% 𝐸𝑞𝑢𝑎𝑡𝑖𝑜𝑛 (2) where:

ƞ: final percentage of how much extracting cobalt. C: final concentration of the metal (mg/1)

V: the volume of the initial leaching solution in (l) Mx: the initial mass of LCO powder in (mg)

4. Results

10

Figure 5: Dissolved cobalt in reline (Urea-ChCl) and ethaline (EG-ChCl). All samples are tested at 100°C within a different time interval.

Figure 6: Lighter samples are dissolved cobalt in ethaline (EG-ChCl) and darker samples are dissolved cobalt in reline (Urea-ChCl). All samples are tested at 160°C within a different time interval

11

4.2 Filtration

The percentage of unreacted LCO using Equation (1) is shown in Figure (7) and Figure (8). The variation of unreacted LCO amount can be noticed depending on the variation of temperature and time. The results for EG-ChCl samples showed the the amount of unreacted cobalt was quite high at low temperatures of 100 °C and 120 °C and that the amount remained more of less the same independent of time interval. However, at relatively high temperature of 180 °C and when water was present in the mixture, the amount of unreacted LCO decreased. The best result was achieved at 180 °C after 72 h when 5% H2O was present in the solution.

For Urea-ChCl sample, the influence of time and temperature became already significant at 120 °C after 72 h time interval. The best result for Urea-ChCl samples was achieved at the relatively high temperature of 160 °C with water added to the solution.

Figure7: Filtration results of (EG-ChCl) samples at different temperatures and time intervals 77,90%85,80% 86,60% 90,80% 88,10% 94,20% 74,80% 68,90% 60,90% 78,90% 64,80% 54,10% 0,00% 10,00% 20,00% 30,00% 40,00% 50,00% 60,00% 70,00% 80,00% 90,00% 100,00% Perc en ta ge %

(EG-ChCl) samples at different time and temperature

Precentage of unreacted LCO

12

Figure8 : Filtration results of (Urea-ChCl) samples at different temperature and time intervals 4.3 Precipitation.

The (EG-ChCl) samples at low temperatures (100-140) °C do not precipitate, while at 160 °C, a small amount of precipitate is starting to form. By increasing the temperature beyond 160 °C, the amount of precipitation is increased. On the other hand, (Urea-ChCl) samples show an extraordinarily small amount when reacted at 100 °C for 72h and increase gradually with longer reaction time and higher temperatures.

The final precipitant for both solvents has an opaque brown chocolate color, with a powdery texture.

4.4 Concentration and Efficiency

The concentration of cobalt precipitated from the DES samples for both solvents have been measured by ICP-OES, where the efficiencies are calculated by applying Equation (2). The results are presented in Table (1), (2).

Table 1: EG-ChCl samples with their measured concentration and calculated efficiencies

Samples Name Concentration PPM (mg/l) Efficiency % 1.EG 160°C 24h 4 0,02% 2.EG 160°C 48h 5 0,03% 3.EG 160°C 72h 524 2,75% 4.EG 180°C 2%H2O 24h 1869 8,90% 5.EG 180°C 2%H2O 48h 93 0,45% 6.EG 180°C 2%H2O 72h 129 0,60% 7.EG 180°C 5%H2O 24h 1459 7,25% 8.EG 180°C 5%H2O 48h 2900 14,48% 9.EG 180°C 5%H2O 72h 133 0,62% 92% 90,65% 88,32% 78,60% 69,10% 57,65% 83,11% 90% 84,88% 23,93% 7,61% 17,44% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 100°C 24h 100°C 48h 100°C 72h 120°C 24h 120°C 48h 120°C 72h 140°C 24h 140°C 48h 140°C 72h 4% H2O 160°C 24h 4% H2O 160°C 48h 4% H2O 160°C 72h p erce n ta ge %

(Urea-ChCl) samples at different temperture and time

Precentage of unreacted LCO

13

Table 2: Urea-ChCl samples with their measured concentration and calculated efficiencies

Samples Name Concentration PPM (mg/l) Efficiency % 1.Urea 100°C 72h 508 3% 2.Urea 120°C 24h 551 3% 3.Urea 120°C 48h 8116 40% 4.Urea 120°C 72h 8099 40% 5.Urea 140°C 24h No precipitation - 6.Urea 140°C 48h 233 1% 7.Urea 140°C 72h 404 2% 8.Urea 4%H20 160°C 24h 9029 43% 9.Urea 4%H20 160°C 48h 6042 29% 10.Urea 4%H20 160°C72h 5551 31% 4.5 Characterization 4.5.1 UV-Vis

This study used UV-Vis spectrophotometry analysis to characterize cobalt complexes formed in each solvent. Figure (9) shows two spectra of (Urea-ChCl) samples, both samples are identical and have two peaks observed in the visible region (580-700 nm). To determine the formula of this cobalt complex in Urea -ChCl solvent, Fogel et al investigated that cobalt can form cobalt chloride complex with formula [CoCl3(OD)]-d, where OD refers to oxygen donor. The first coordination consists of

lighter atoms such as oxygen and nitrogen, which may come from either water or urea. This could be related to the lower activity of chloride in the Urea-based liquid.

Figure 9 : Uv-Vis spectra of digested cobalt in Urea ChCl samples form [CoCl3(OD)]-d

For EG-ChCl samples, only one cobalt complex is expected, but Figure (10) shows two complexes forming in EG-ChCl solvents in the visible region for temperatures 160°C and 180°C. Those complexes are i) tetracholorocobalate (II)

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 300 400 500 600 700 800 A b sor b an ce wavelength (nm)

14

(CoCl4)2-, with 3 districting bands in the visible range 600-700 nm (630, 667, 696)nm

and ii) hexaaquacobalt(II) chloride [Co(H2O)6]2+with district band in the 500-550 nm

range, see Figure (11 ). At 140°C, no peak is present which indicates that no cobalt is formed and this because of the failure in sand bath as it is discussed in the discussion section below.

Figure10: Uv-Vis spectra of digested cobalt in EG- ChCl samples form (CoCl4)

2-Figure 11: Uv-Vis spectra of digested cobalt in EG- ChCl samples form [Co(H2O)6]2+

4.5.2 FTIR analysis

From Figure (12), it described that all samples that dissolved cobalt by (Urea-ChCl) solvent have the same functional group albeit with different intensity. The main functional groups are in the region of NH-OH stretching modes that are observed between 3600 cm–1 and 3000 cm–1 and NH2 bending and C=O stretching bands which is observed between 1700 cm–1 and 1600 cm–1, while the region 400 cm-1 _{to 600 cm}-1

refers to the Cl- _{binding to the metal.}

Figure (13) indicates also that all samples of (EG-ChCl) at different times and temperatures have the same functional groups. The main regions are located in: i)

0 0,5 1 1,5 2 2,5 300 400 500 600 700 800 A b sor b an ce wavelength (nm)

Uv-vis spectra of EG-ChCl samples

15

between 3500 cm-1 to 3000 cm-1 and indicates the OH functional group, ii) two peaks in the region 2800 cm-1 to 3000 cm-1 have corresponded to C-H, CH3 and CH2 stretching

in the EG, iii) region 408cm-1 to 600 cm-1 referring to the Cl- binding to the metal. The FTIR analysis indicates there is no new functional group is formed meaning there are no new toxic or harmful components are formed by mixing LCO with DES at the examined reaction times and temperatures.

Figure12 : FTIR spectra of urea-ChCl samples at different time and temperature

Figure13: FTIR spectra of EG-ChCl samples at different time and temperature

16

5. Discussion

The DES solvents ethaline and reline were both successful in extracting cobalt from LCO cathode materials, albeit with different efficiencies When LCO is digested in each of these DES solvents, a noticeable variation in color is observed at different temperatures and times. In this case, the color variation is considered as an indicator of the cobalt because cobalt is one of the transition metals that have unpaired electrons in the valence layer. Those valence electrons are responsible for forming colorful complexes. This is consolidated by the obtained efficiencies values, Table (1), and Table (2). Urea-ChCl samples that show blue color samples from the beginning have higher efficiencies to leaching cobalt compared to EG-ChCl samples that have lower leaching efficiencies. The EG-ChCl samples at 100 °C are showing grey-white color. This is true until applying a temperature value of 160 °C where a colored solution is formed with different shades from light green to dark blue. On the other hand, Urea-ChCl samples form a blue solution at a low temperature as 100 °C, which indicates that cobalt complexes are already forming, see Figure (5) and Figure (6).

Time is an important factor in the cobalt leaching process. As seen in the filtration results Figure (7) and Figure (8) the amount of unreacted LCO decreases with increasing time interval from 24h to 72h in both solvents, For instance, for Urea-ChCl samples at 120 °C, the amount of unreacted LCO is 78% at 24h while, but it decreased to 57%.after 72h intervals However, at 140 °C, and as mentioned earlier, a sand bath was used at this temperature instead of an oil bath for both solvents. It is evident from the results that sand bath is not as effective as oil for this experiment. The problem might be that the sand had used without heating it up which is normally the case. Time impact is also evident in the concentration values. For instance, the concentration of the Urea-ChCl sample at 120 °C reacted for 24h is 508 ppm but it steps up to 8099 ppm at 72h. Temperature is equally important to heating time and is the other parameter that influences the leaching of cobalt. Temperature affects the rate of the chemical reaction by increasing the collision frequency among the corresponding ions. As shown in Table (2), the efficiency increased from 3% to 40% in Urea-ChCl samples when increasing the temperature from 100 °C 120 ° C for the same time interval (72 h). Additionally, (EG-ChCl) samples have shown no leaching until reaching higher temperatures. In Figure (8), the Urea-ChCl sample at 100 °C for 72 h heating time shows that unreacted LCO is 88,32% where at 120 °C for the same heating time shows 57% of unreacted LCO. Therefore, the temperature has a great influence on leaching cobalt.

It is concluded that Urea-ChCl is more effective in extracting cobalt from LIB than EG-ChCl. However, the optimal experimental conditions for extracting cobalt from LIBs using Urea-Chcl needs further analysis in terms of comparing energy consumption for different conditions (reaction time and temperature) as well as repeating the experiment by using oil bath for 140 °C.

17

precipitation are not tested by ICP-OES and hence need another instrument to carry out the analysis, such as Atomic Absorption Spectroscopy (AAS). Likewise, for lower temperature samples, a pink-brown color was clearly visible, even after leaving the samples overnight with a suitable amount of sodium hypochlorite (NaClO) for precipitation, no appreciable amount had formed, and the supernatants were still had the pink-brown color. Those samples are listed in Appendix 1, Figure (21), and Figure (22).

This study used UV-Vis spectrophotometry analysis to characterize the cobalt complex formed. For (EG-ChCl) samples, Figure (10) and Figure (11) show that both complexes are Co2+_{. This suggests that EG acts as a reducing agent role by changing the}

oxidation state of cobalt in LCO from Co3+ to Co2+ which form two different cobalt chloride complexes. Regarding (Urea - ChCl) samples, Figure (9), shown different spectra form (EG-ChCl) spectra. This might be a cobalt chloride complex form. The chemical formula of the cobalt chloride complex of urea samples could be [CoCl3(OD)] -d_{. The nature of the ligand bonded to cobalt has a dramatic effect on the stability of the}

two oxidation states of cobalt, for instance, Co3+ _{has a great affinity for nitrogen donors}

especially ammonia or amine. From this study’s perspective, this might refer to urea-ChCl samples having Co3+ and the reason behind the rapid dissolution of cobalt at a lower temperature in comparison with EG-ChCl solvents. This idea is merely a theory based on the previous paper and is outside the scope of this study. UV-Vis is actually quite suitable to quantify the amount of cobalt that dissolved in DES but the way that it works is very specific to has a calibration curve. The spectrum can change with solvent and species of interest. Therefore, for each kind of sample, a different calibration is needed. In this case, we were not able to establish a calibration curve in the time we had, for various reasons, such as the complex that formed in urea samples is unknown. Additionally, the viscosity of urea samples is high making them hard to handle (see Figure 23 in Appendix1).

To determine to what degree, the method is environmentally friendly, two factors must be taken into account. First, the energy consumption by unite volume of extracted cobalt should be calculated and compared to other methods in the industry. Second, FTIR analysis has reported what functional groups are formed after reacting LCO with both DES solvents. In Figure(13) and Figure(14), there are no new functional groups formed, as stated in chapter 4.5.2. FTIR Analysis. This means that no toxic or harmful compounds are formed. However, as previously mentioned in the introduction, the main feature of DES is forming a strong hydrogen bond between the HBD and HBA Figure(4). Since dissolving cobalt in DES lead to forming different cobalt chloride complexes, this means the chloride anion in both DES solvents binds to cobalt when LCO is added and forms cobalt chloride complexes. This reveals that a change in the formation of the DES structure happened and both DES solvents are not reusable.

6. Conclusion.

18

19

Acknowledgement

I would like to thank all people who are working at the structural chemistry department at Uppsala university starting with my supervisors Reza Younsi and John Ostrander. I would give a special thanks to my Colleagues Erik Östlund and Yonas Tesfamhret who kept supporting me, especially in the lab work.

I would also express my gratitude to my professors and colleagues at Umeå University. You have made those two years unforgettable for me.

Special thanks go also to the best family in the world, Mom and sister Rand. You have always been my friends and companions in joy and sorrow. I love you both!

Last but not least, to the dearest person to my heart. My husband Abdul Mouez who went through those tough times with me and was the supporting person I needed along the whole journey. Also, academic thank you to him for revising this work.

20

References

© The Royal Swedish Academy of Sciences (no date) Nobel prize in chemistry 2019, october 2019.

Abbott, A. P. et al. (2006) ‘Solubility of metal oxides in deep eutectic solvents based on choline chloride’, Journal of Chemical and Engineering Data, 51(4), pp. 1280–1282. doi: 10.1021/je060038c.

Deng, D. (2015) ‘Li-ion batteries: Basics, progress, and challenges’, Energy Science and Engineering, 3(5), pp. 385–418. doi: 10.1002/ese3.95.

Du, C. et al. (2016) ‘Effect of water presence on choline chloride-2urea ionic liquid and coating platings from the hydrated ionic liquid’, Scientific Reports. Nature Publishing Group, 6(July), pp. 1–14. doi: 10.1038/srep29225.

European Commission (2017) COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS on the 2017 list of Critical Raw Materials for the EU. Available at:

https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52017DC0490#document1.

Fogel, N. et al. (1964) ‘Behavior of Ammonium Tetrachlorocobaltate(II) on Hydration’, Inorganic Chemistry, 3(5), pp. 720–726. doi: 10.1021/ic50015a027.

Habibi, A., Shamshiri Kourdestani, S. and Hadadi, M. (2020) ‘Biohydrometallurgy as an environmentally friendly approach in metals recovery from electrical waste: A review’, Waste Management and Research, 38(3), pp. 232–244. doi: 10.1177/0734242X19895321.

Harper, G. et al. (2019) ‘Recycling lithium-ion batteries from electric vehicles’, Nature. Springer US, 575(7781), pp. 75–86. doi: 10.1038/s41586-019-1682-5.

IEA Data and Statistics (2020). Available at:

https://www.iea.org/data-and-statistics?country=WORLD&fuel=Energy supply&indicator=Total primary energy supply (TPES) by source.

Ijadi Bajestani, M., Mousavi, S. M. and Shojaosadati, S. A. (2014) ‘Bioleaching of heavy metals from spent household batteries using Acidithiobacillus ferrooxidans: Statistical

evaluation and optimization’, Separation and Purification Technology. Elsevier B.V., 132, pp. 309–316. doi: 10.1016/j.seppur.2014.05.023.

Johansson, A. and Nyström, M. (2019) Fordonsstatistik. Available at: https://www.scb.se/. Kovac, J. (2015) ‘Ethics in Science: The Unique Consequences of Chemistry’, Accountability in Research. Taylor & Francis, 22(6), pp. 312–329. doi: 10.1080/08989621.2015.1047709. Lambert, F. et al. (2015) ‘Copper leaching from waste electric cables by biohydrometallurgy’, Minerals Engineering. Elsevier Ltd, 76, pp. 38–46. doi: 10.1016/j.mineng.2014.12.029. Liu, R., Li, J. and Ge, Z. (2016) ‘Review on Chromobacterium Violaceum for Gold

Bioleaching from E-waste’, Procedia Environmental Sciences. Elsevier B.V., 31, pp. 947–953. doi: 10.1016/j.proenv.2016.02.119.

Meshram, P., Pandey, B. D. and Mankhand, T. R. (2015) ‘Recovery of valuable metals from cathodic active material of spent lithium ion batteries: Leaching and kinetic aspects’, Waste Management. Elsevier Ltd, 45, pp. 306–313. doi: 10.1016/j.wasman.2015.05.027.

Nayaka, G. P. et al. (2016) ‘Use of mild organic acid reagents to recover the Co and Li from spent Li-ion batteries’, Waste Management. Elsevier Ltd, 51, pp. 234–238. doi:

10.1016/j.wasman.2015.12.008.

Olson, G. J., Brierley, J. A. and Brierley, C. L. (2003) ‘Bioleaching review part B: Progress in bioleaching: Applications of microbial processes by the minerals industries’, Applied

Microbiology and Biotechnology, 63(3), pp. 249–257. doi: 10.1007/s00253-003-1404-6. Perkins, S. L. and Perkins, S. L. (2013) ‘by’, (August).

Ruan, J. et al. (2014) ‘A new strain for recovering precious metals from waste printed circuit boards’, Waste Management. Elsevier Ltd, 34(5), pp. 901–907. doi:

10.1016/j.wasman.2014.02.014.

21 https://www.statista.com/statistics/1015584/cobalt-demand-scenarios-for-electric-vehicles-eu/. Till Bunsen et al. (2019) ‘Global EV Outlook 2019 to electric mobility’, OECD iea.org, p. 232. Available at: www.iea.org/publications/reports/globalevoutlook2019/.

Tran, M. K. et al. (2019) ‘Deep eutectic solvents for cathode recycling of Li-ion batteries’, Nature Energy. Springer US, 4(4), pp. 339–345. doi: 10.1038/s41560-019-0368-4.

Tuncuk, A. et al. (2012) ‘Aqueous metal recovery techniques from e-scrap: Hydrometallurgy in recycling’, Minerals Engineering. Elsevier Ltd, 25(1), pp. 28–37. doi:

10.1016/j.mineng.2011.09.019.

U.S. EPA (2013) ‘Application of Life-Cycle Assessment to Nanoscale Technology: Lithium-ion Batteries for Electric Vehicles’, United States Environmental ProtectLithium-ion Agency, pp. 1– 119. Available at:

https://www.epa.gov/sites/production/files/2014-01/documents/lithium_batteries_lca.pdf.

Ulvestad, A. (2018) ‘A Brief Review of Current Lithium Ion Battery Technology and Potential Solid State Battery Technologies’. Available at: http://arxiv.org/abs/1803.04317.

Wang, F. et al. (2016) ‘Recovery of cobalt from spent lithium ion batteries using sulphuric acid leaching followed by solid-liquid separation and solvent extraction’, RSC Advances. Royal Society of Chemistry, 6(88), pp. 85303–85311. doi: 10.1039/c6ra16801a.

Yao, L., Feng, Y. and Xi, G. (2015) ‘A new method for the synthesis of

LiNi1/3Co1/3Mn1/3O2 from waste lithium ion batteries’, RSC Advances. Royal Society of Chemistry, 5(55), pp. 44107–44114. doi: 10.1039/c4ra16390g.

22

Appendix 1: Experiment’s Pictures

Figure14: Crystals form in the EG-ChCl samples that have low water content.

Figure15: EG-ChCl samples at 180 °C and 2%, 5% of water content

23

Figure 17: EG-ChCl and Urea -ChCl samples at 100 °C with different heating time

Figure18: EG-ChCl and Urea -ChCl samples at 120 °C with different heating time

24

Figure20: All Urea-ChCl samples at different times and temperatures.

Figure 21: Precipitation of ethaline samples

25

26

Appendix 2: Complete Calculations

Sample Name Orginal amount g Weight of the paper Weight of paper

after drying Difference g percentage%

2. LCOin EG 100C 24h 0,1017 0,3519 0,4312 0,0793 78,0 3.LCO in EG 100C 72h 0,1075 0,3575 0,4498 0,0923 85,9 4.LCO in EG 120C 24h 0,1005 0,3392 0,4263 0,0871 86,7 5.LCO in EG 120C 48h 0,1002 0,3353 0,4263 0,091 90,8 6.LCO inEG 120C 72h 0,104 0,3593 0,4509 0,0916 88,1 7.LCO inEG 140C 48h 0,1004 0,3639 0,4585 0,0946 94,2

8.LCO inEG 180C48h 5%water 0,1001 0,3501 0,415 0,0649 64,8