Optimizing the Leaching Parameters and Studying the Kinetics of Copper Recovery from Waste Printed Circuit Boards.

The study of copper (Cu) recovery is crucial for the entire recovery process of waste printed circuit boards (WPCBs), and Cu can be leached efficiently via a sulfuric acid-hydrogen peroxide (H2SO4-H2O2) system. To achieve high Cu recovery, it is important to evaluate the parameters of the leaching process and understand the Cu leaching kinetics. Applying statistical and mathematical techniques to the leaching process will further benefit the optimization of the Cu leaching parameters. Moreover, the leaching kinetics of Cu in the H2SO4-H2O2 solution is yet to be fully understood. Hence, in the present work, process parameters, such as temperature, H2SO4 and H2O2 concentrations, solid-liquid ratio, particle size, and stirring speed, were optimized statistically by the response surface methodology (RSM). The results showed that the leaching kinetics conformed to the Avrami model. The maximum Cu leaching efficiency was 99.47%, and it was obtained based on the following optimal conditions: 30.98 °C, 2.6 mol/L H2SO4, 1.87 mol/L H2O2, a solid-liquid ratio of 0.05 g/mL, 135 mesh, and 378 rpm. RSM was used for the optimization of the process parameters, and the leaching kinetics in this system was clarified. This study provides an important pathway for the investigation of other metal recoveries from WPCBs.

Introduction

With the advancement of technology and the development of society, the lifespan of electronic and electrical equipment (EEE) has been foreshortened. This has resultantly caused an alarming increase in the number of waste EEE (WEEE). Waste printed circuit boards (WPCBs), the core component of WEEE, which were originally manufactured to provide reliable electrical connections for electronic components, are also largely produced.[1] These WPCBs contain abundant high-value metals (e.g., gold, silver, copper, etc.),[2−4] and copper, which predominantly makes up the content of the conductive circuits of printed circuit boards, is the most abundant metal.[5,6] This makes WPCBs important secondary copper sources, and the recovery of copper from WPCBs is of great economic and environmental benefits.

Several studies have been carried out to recover copper from WPCBs based on physical processes,[7,8] pyrometallurgy,[9] hydrometallurgy,[2,10−13] and biometallurgy.[14−17] Compared with other recovery processes, hydrometallurgy has attracted extensive attention due to its cost effectiveness, ease of operation, and being relatively environmentally friendly.[18−21] Common copper leaching solutions include inorganic acids, ionic liquids, ammonia-based solutions, ethylenediaminetetraacetic acid (EDTA), and citrate. Among them, ammonia-based solutions and H2SO4 are widely used due to their simple operation and low cost.[22,23] However, the toxicity and volatility of ammonia is a cause of environmental and operational safety concerns. Meanwhile, the low solubility of copper in the H2SO4 solution due to the poor oxidation of H2SO4 is the major limitation of its application.[22,24,25] Therefore, the search for appropriate oxidants toward the improvement of the leaching efficiency of copper in the H2SO4 solution is of high necessity.

Hydrogen peroxide (H2O2), as an efficient oxidant, has been used in combination with H2SO4 for copper recovery. The literature studies showed that the temperature, H2SO4 concentration, H2O2 concentration, a solid–liquid ratio, particle size,[24,26,27] and the stirring speed have significant effects on the copper leaching process. To the best of our knowledge, for the H2SO4–H2O2 leaching system, only the effect of a single parameter on the leaching efficiency was investigated in the complex solid–liquid mixed system, thus ignoring the influence of the interactions of experimental parameters playing different roles in the process. This therefore creates an information gap that needs to be espied. Therefore, the influence of the interactions of key experimental parameters on the leaching efficiency still needs to be studied comprehensively.

Response surface methodology (RSM), which is a powerful statistical analysis technique that analyzes the influence of the interactions of several variables, has been widely applied in deciding optimum conditions for chemical or physical processes.[28−30] For instance, by modeling the relationships between different experimental parameters, the RSM can optimize the desirable ones.[31−33] RSM has also been used in metal leaching processes, such as gold leaching in thiosulfate solution,[34] and the leaching of Cu, Fe, Ni, and Ag in the H2SO4–CuSO4–NaCl solution.[35] A survey of literature proves that there is no available information on the application of the RSM for copper leaching in the H2SO4–H2O2 system. Therefore, as a powerful tool, RSM has the potential of being applied to copper leaching in the H2SO4–H2O2 system, which has not yet been researched.

Leaching kinetic studies are very meaningful for understanding leaching mechanisms and the improvement of recovery rates. Han et al.[10] used glycine with H2O2 to leach 94.08% of copper from WPCBs, and the leaching kinetics were also studied with a shrinkage core model. The result showed that the leaching kinetics conformed to diffusion and chemically controlled reactions. Jadhao et al.[2] used chelation technology with EDTA to recover 83.8% of copper, and the rate-determining step was the diffusion-controlled process based on the shrinkage core model. However, in the H2SO4–H2O2 system, the leaching kinetic of copper is still unclear.

In this research, copper from WPCBs was leached using the H2SO4–H2O2 solution. The effects of various parameters (temperature, H2SO4 and H2O2 concentrations, solid–liquid ratio, particle size, and stirring speed) on the copper leaching efficiency were evaluated. Furthermore, the influence of the interactions of experimental parameters on the Cu leaching process was comprehensively studied by the Box–Behnken design (BBD) of the RSM, and a combination of multiple parameters generating an optimal response was identified. The kinetics of copper leaching in the H2SO4–H2O2 system was also elucidated based on the experimental data and model fitting. This study provides a comprehensive understanding of the copper leaching process and would contribute to the recovery of other metals in WPCBs.

Results and Discussion

Leaching Studies

Figure illustrates the effect of various experimental parameters on the copper leaching efficiency. The initial leaching efficiency of copper was high and reached equilibrium above 60 min as shown in Figure a. The leaching efficiency increased with temperature increment below 30 °C. As the temperature increased, the reactivities of the metals also increased, which promoted the leaching of other active metals competing with copper. In addition, a high temperature promotes the rapid decomposition of hydrogen peroxide producing oxygen. However, the solubility of oxygen in aqueous phases decreases with increasing temperature, which further inhibits copper leaching.[36] Therefore, the leaching efficiency decreased with increasing temperature, which may be attributed to the decomposition of H2O2 at high temperatures. The highest leaching efficiency of copper was obtained when the optimum leaching temperature was 30 °C.

Figure 1

Effect of variables on the copper leaching efficiency: (a) temperature, (b) H2SO4 concentration, (c) H2O2 concentration, (d) solid–liquid ratio, (e) particle size, and (f) stirring speed.

The effect of the H2SO4 concentration on the copper leaching efficiency is shown in Figure b. The leaching efficiency of copper, which is sensitive to the H2SO4 concentration, first increased and then decreased with increasing H2SO4 concentration. As the H2SO4 concentration increased, a high hydrogen ion concentration promoted copper leaching, and the leaching rate reached 99.47% at 2.5 mol/L. When the H2SO4 concentration was higher than 2.5 mol/L, more active metals were leached consuming more acid. The metal leaching process generates hydrogen, which would adsorb on the surface of copper, thereby reducing its leaching efficiency. In addition, high metal leaching efficiencies lead to increases in the viscosities of leaching solutions, which consequently hinder leaching processes.[37] Therefore, the optimum concentration of H2SO4 was 2.5 mol/L.

The effect of the H2O2 concentration was studied in the concentration range of 1–2.5 mol/L. As shown in Figure c, copper leaching was first enhanced and then decreased with increasing H2O2 concentration, and the leaching efficiency reached a maximum of 99.47% at 2.0 mol/L H2O2. As a strong oxidant, the increase of the H2O2 concentration improved the leaching efficiency of copper, and a high H2O2 concentration can limit the reduction or precipitation of copper species. The growth rate of the leaching efficiency decreased over time, which may be due to the self-decomposition of H2O2.[38] At a high H2O2 concentration, the active metals (Sn, Fe, and Ni) were leached preferentially,[23] which catalyzed the decomposition of H2O2, generating a large number of oxygen bubbles. The oxygen bubbles would adsorb on the copper surface and hinder the contact between the WPCB powder and the leaching solution, thus reducing the leaching efficiency.[39,40] Therefore, the optimum H2O2 concentration was 2 mol/L.

Under the condition of 30 °C, 2.5 mol/L H2SO4, 2 mol/L H2O2, and a leaching time of 180 min, the effect of a solid–liquid ratio was studied, and the results showed that the pulp density had an adverse effect on the copper leaching process (Figure d). At a solid–liquid ratio of 0.05 g/mL, the highest leaching rate of 99.47% was attained. However, only 55.30% of copper was leached at 0.1 g/mL. As a solid-to-liquid ratio increases, the increase in the quantity of nonmetallic materials would promote the agglomeration of raw materials, causing collisions and frictions between the WPCB powders during the leaching process, which would impede the full contact of metals with leaching solutions, thereby reducing the leaching efficiency. In addition, the contents of active metals increase in the raw materials with the increase in the solid–liquid ratio, which reduces copper ions in the leaching solution through replacement reactions, hence reducing the leaching efficiency of copper. Therefore, the optimum solid–liquid ratio was 0.05 g/mL.

The effect of particle size on the leaching efficiency is shown in Figure e. With a decrease in particle size between 18 and 150 mesh, the leaching efficiency of copper increased and then decreased. The morphologies of the different particle sizes are shown in Figure . Copper was present in the middle layer of the WPCBs. When the particle size was large (such as 18 mesh), copper was still covered with the organic matter layer as shown in Figure a and exhibited poor liberation resulting in low leaching efficiency.[41−43] With a decrease in particle size, the liberation degree of the metal and nonmetal materials was improved, as shown in Figure b–e, which conduced to the contact between the metal and leaching solution promoting copper leaching.[44] However, the leaching efficiency was decreased when the particle size was less than 300 mesh because of entrainment and agglomeration.[45] Therefore, the optimum particle size should be 150 mesh according to the leaching efficiency.

Figure 2

Stereo microscope observation of different size fractions: (a) >18 mesh, (b) 18–35 mesh, (c) 35–75 mesh, (d) 75–100 mesh, (e) 100–150 mesh, and (f) >300 mesh.

The stirring speed had a positive effect on the copper leaching efficiency as shown in Figure f. The leaching efficiency was only 79.08% at 100 rpm, and it increased rapidly when the stirring speed was more than 300 rpm. This may be due to the complex composition of WPCBs. At low speeds, the WPCB powder does not suspend completely, and the heavier metal particles are deposited at the bottom of the beaker and are covered by the lighter nonmetal particles, thereby hindering the full contact of the metal with the leaching reagent and decreasing the leaching efficiency. Increasing the stirring speed will promote the diffusion rate of the metal particles in the leaching solution, hence increasing the leaching efficiency.[46] When the stirring speed reached 400 and 500 rpm, the leaching efficiencies increased to 99.47 and 99.62%, respectively. Therefore, a further increase in the stirring speed cannot increase the leaching efficiency significantly.

Modeling and Statistical Analysis of Data

Based on the abovementioned results, the key experimental parameters, such as temperature, concentrations of H2SO4 and H2O2, solid–liquid ratio, particle size, and stirring speed, were optimized by the BBD. Table shows the results produced by the BBD including the synergistic effects of combining different parameters and their corresponding leaching efficiencies. The results show that the leaching efficiencies of copper were in the range of 27.27–99.57%, and the maximum yield (99.57%) was obtained at the medium value of each experimental parameter.

Table 1

BBD Design Arrangement and Results

	independent variable						responses (%)
run	temperature (°C)	[H2SO4] (mol/L)	[H2O2] (mL/L)	solid–liquid ratio (g/mL)	particle size (mesh)	stirring speed (rpm)	actual	predicted
1	40	2	1.5	0.075	200	400	62.35	55.39
2	40	3	2	0.075	200	300	53.27	59.43
3	50	2.5	1.5	0.075	150	500	40.37	45.80
4	40	2.5	2	0.075	150	400	60.58	62.93
5	30	3	2	0.1	150	400	53.32	50.08
6	50	2.5	2.5	0.075	150	500	42.57	33.40
7	40	2.5	2.5	0.05	150	300	67.32	64.96
8	40	2.5	2.5	0.1	150	300	37.32	32.05
9	50	2	2	0.05	150	400	61.45	63.74
10	30	2.5	2	0.1	100	400	52.40	52.55
11	30	3	2	0.05	150	400	95.44	95.72
12	50	2.5	2.5	0.075	150	300	33.28	31.69
13	50	2.5	2	0.1	100	400	27.34	30.76
14	40	2.5	2	0.075	150	400	61.37	62.93
15	30	2.5	2	0.05	100	400	99.42	96.95
16	30	2.5	1.5	0.075	150	500	72.32	75.73
17	30	2.5	2.5	0.075	150	300	52.15	48.54
18	50	2.5	2	0.1	200	400	37.27	38.79
19	40	2.5	1.5	0.1	150	500	47.15	47.69
20	40	2.5	2	0.075	150	400	60.75	62.93
21	30	2	2	0.1	150	400	53.17	57.02
22	40	2.5	2	0.075	150	400	73.35	62.93
23	40	2.5	1.5	0.05	150	300	80.75	79.40
24	40	3	2.5	0.075	200	400	35.86	42.16
25	50	2.5	1.5	0.075	150	300	40.15	38.13
26	50	2.5	2	0.05	100	400	63.25	67.88
27	30	2.5	1.5	0.075	150	300	60.47	67.82
28	40	3	2.5	0.075	100	400	30.32	37.28
29	40	2.5	2.5	0.1	150	500	27.27	30.44
30	40	2.5	2.5	0.05	150	500	68.47	70.23
31	40	3	1.5	0.075	100	400	52.86	51.80
32	40	2.5	1.5	0.05	150	500	87.17	90.62
33	40	2.5	2	0.075	150	400	60.28	62.93
34	40	2.5	2	0.075	150	400	61.27	62.93
35	40	2	2.5	0.075	200	400	37.17	38.22
36	50	3	2	0.1	150	400	35.28	34.09
37	30	2	2	0.05	150	400	92.37	94.51
38	40	2	2	0.075	200	500	58.19	60.31
39	30	2.5	2.5	0.075	150	500	50.28	50.48
40	40	2.5	1.5	0.1	150	300	43.28	43.34
41	50	3	2	0.05	150	400	77.23	72.44
42	40	2	2.5	0.075	100	400	32.27	34.83
43	40	3	1.5	0.075	200	400	57.25	54.70
44	40	2	2	0.075	100	300	55.28	57.85
45	30	2.5	2	0.1	200	400	60.17	56.49
46	40	3	2	0.075	200	500	69.25	66.68
47	40	3	2	0.075	100	300	55.35	53.24
48	40	2	2	0.075	200	300	60.38	62.56
49	40	2	2	0.075	100	500	66.37	60.21
50	50	2	2	0.1	150	400	32.86	33.53
51	40	2	1.5	0.075	100	400	60.27	53.97
52	40	3	2	0.075	100	500	67.28	65.10
53	30	2.5	2	0.05	200	400	99.57	95.21
54	50	2.5	2	0.05	200	400	69.43	70.23

According to the experimental results from Table , the model was fitted by multiple linear regressions. A second-order regression model with a coefficient of R2 = 0.9521 is derived from eq . From the equation, the response (leaching efficiency, γ) at any coded level for various experimental parameters can be predicted.The reliability of the regression model was investigated by analysis of variance (ANOVA), as shown in Table . The significance test was performed at a 95% confidence level using p-values. The model is said to be statistically significant if the p value is less than 0.05.[28,47] Accordingly, the p-value (<0.0001) of the regression model is indicative of the adequacy of the model at a 95% confidence level (Table ). The terms with p-values greater than 0.05 have insignificant effects on predicted responses.[29] The statistical significance of the linear and quadratic terms of the various parameters as well as their interactions is shown in Table . The terms with underscore are statistically insignificant at a confidence level of 95%.

Table 2

ANOVA of the Response Surface Quadratic Model for Copper Leaching Efficienciesa

souce	sum of square	df	mean square	F value	p-value
model	16371.99	27	606.37	19.13	<0.0001	significant
A-temperature	3280.68	1	3280.685	103.48	<0.0001
B-H2SO4 concentration	4.66	1	4.66	0.1471	0.7044
C-H2O2 concentration	1505.91	1	1505.91	47.5	<0.0001
>D-solid–liquid ratio	8627.56	1	8627.56	272.13	<0.0001
E-particle size	59.38	1	59.38	1.87	0.1829
F-stirring speed	138.67	1	138.67	4.37	0.0464
AB	0.048	1	0.048	0.00278	0.9587
AC	157.5	1	157.5	9.18	0.009
AD	0.018	1	0.018	0.00102	0.007
AE	8.38	1	8.38	0.2645	0.6114
AF	0.0276	1	0.0276	0.0009	0.9767
BC	0.21	1	0.21	0.012	0.9131
BD	0.012	1	0.012	0.000705	0.02
BE	2.19	1	2.19	0.0691	0.7947
BF	45.17	1	45.17	1.42	0.2434
CD	3.67	1	3.67	0.21	0.023
CE	1.97	1	1.97	0.0621	0.8051
CF	35.52	1	35.52	1.12	0.2996
DE	16.16	1	16.16	0.5097	0.4816
DF	23.63	1	23.63	0.7454	0.3958
EF	10.65	1	10.65	0.3359	0.5672
A2	121.8	1	121.8	7.1	0.0185
B2	0.041	1	0.041	0.0024	0.9616
C2	944.28	1	944.28	55.04	<0.0001
D2	45.58	1	45.58	2.66	0.1254
E2	25.56	1	25.56	0.8064	0.3774
F2	35.48	1	35.48	1.12	0.2999
residual	824.30	26	31.70
lack of fit	693.24	21	33.01	1.26	0.4340	not significant
pure error	131.06	5	26.21
cor total	17196.29	53

R2 = 0.9521, Radj2 = 0.9023.

The predicted vs practical plot for copper leaching efficiencies is shown in Figure . There is a good linear relationship between the predicted values and the actual values. This result shows that the regression model obtained through the BBD is consistent with the experimental results and can be used to optimize the experimental parameters.

Figure 3

Linear correlation between observed and predicted values for the copper leaching efficiency.

To study the effect of the relationships between variables on the response, three-dimensional (3D) response surface plots of the regression model were constructed. Figure shows the 3D response surface plots of the experimental parameters with high reciprocity, and others are shown in Figure S1. As shown in Figure , the interaction of AD, BD, CD, DE, and DF has a significant effect on response, and the effect of D (solid–liquid ratio) exhibited a greater influence on the copper leaching efficiency than A (temperature), B (H2SO4 concentration), C (H2O2 concentration), E (particle size), and F (stirring speed), which is the same with the F-value results in Table .

Figure 4

Response surface plots reflecting the simultaneous effects of dual parameters on the leaching efficiency of copper (third parameters are held at the center level). (a) Temperature and [H2O2], (b) temperature and solid–liquid ratio, (c) [H2SO4] and [H2O2], (d) [H2SO4] and solid–liquid ratio, (e) [H2O2] and solid–liquid ratio, (f) [H2O2] and particles size, (g) [H2O2] and stirring speed, (h) solid–liquid ratio and particle size, and (I) solid–liquid ratio and stirring speed.

From the results, the predicted model shows a good reflection of the relationships between the experimental and predicted results. Therefore, RSM was also used to optimize the leaching conditions. The obtained optimal experimental parameters were 30.98 °C, 2.6 mol/L H2SO4, 1.87 mol/L H2O2, a solid–liquid ratio of 0.05 g/mL, 135 mesh, and 378 rpm. The predicted leaching efficiency was 99.62% while the experimental result at the optimum experimental parameters was 99.47%. Due to the closeness between both results, the experimental parameters can be optimized by the RSM.

Kinetic Analysis of Copper Leaching

Kinetic Model

Leaching kinetics is expressed by homogeneous or heterogeneous models. In heterogeneous models, the leaching process usually includes the following steps: (i) diffusion through boundary layers (external diffusion), (ii) diffusion through solid product layers (internal diffusion), (iii) surface chemical control reactions, and (iv) mixed reactions. The shrinking core model (SCM) is the most commonly used kinetic equation. The rate equation of SCM controlled by a chemical reaction, diffusion reaction, and mixed reactions is shown in eqs –4.[48]where X is the leaching efficiency; kr, kd, and kM are the chemical reaction rate constant, diffusion reaction rate constant, and mixed control reaction rate constant, respectively; and t is the leaching time. In addition, homogeneous models could also be used to determine leaching kinetics as in the following eqs –7.[49]where X is the leaching efficiency, k is the apparent kinetic constant, t is the leaching time, and n is the feature parameter.

Kinetic Studies

The leaching of copper from WPCBs in the H2SO4–H2O2 system is a complicated solid–liquid reaction process. The leaching kinetics of copper was examined by varying the H2SO4 and H2O2 concentrations, temperature, solid–liquid ratio, particle size, and stirring speed. First of all, the leaching data obtained at different temperatures was fitted with different standard kinetic equations, and the fitting results are shown in Figure S2, while the corresponding fitting correlation coefficients are shown in Table S1. As shown in Figure S2 and Table S1, the fitting results of the different kinetic equations had large errors. Due to the high initial leaching efficiency, the leaching process was found to fit the Avrami model most satisfactorily as shown in Figure a. The slope and intercept are denoted as n and ln k, respectively. The different n and ln k corresponding to the different temperatures are summarized in Table . The value of n is almost constant with an average value of 0.1695. Therefore, the Avrami model is given in eq .According to the fitting results of Figure a, the plots of lnk versus ln T are shown in Figure b. Therefore, the kinetic equation on the effect of temperature on the copper leaching efficiency is obtained from eq .In addition, the relationship between the leaching rate constant and various factors were also studied. According to the experimental results shown in Figure , −ln (1 – X) vs t0.1695 is plotted to obtain the fitting equations at different H2SO4 concentrations, H2O2 concentrations, solid–liquid ratios, particle size and stirring speeds. The leaching efficiency showed good linear relationships at different leaching times as shown in Figure . The slope k of each fitted straight line is the rate constant of the different experimental conditions. The plots of ln k vs ln B, ln C, ln E, and ln F for the different experimental parameters are shown in Figure . The kinetic equations on the influence of H2SO4 concentration, H2O2 concentration, solid–liquid ratio, particles size, and stirring speed on the copper leaching efficiency are shown in eqs –14. By combining these equations with eq , the leaching efficiency (X) can be predicted. The copper leaching kinetics in the H2SO4–H2O2 system is clarified as well.

Figure 5

Relationship (a) between ln[−ln(1 – X)] and ln t and (b) between ln k and 1/T at different temperatures.

Table 3

Values of n and ln k at Different Temperatures

T (K)	n	ln k	R2
293.15	0.1798	–0.6597	0.97706
303.15	0.1749	0.05778	0.94413
313.15	0.1641	–0.17542	0.97419
323.15	0.1593	–0.6597	0.98726

Figure 6

Plots of −ln(1 – X) vs t0.1695 at different experimental conditions: (a) H2SO4 concentration, (b) H2O2 concentration, (c) solid–liquid ratio, (d) particle size, and (e) stirring speed.

Figure 7

Relationship of rate constants with different experimental parameters: (a) ln k vs ln B, (b) ln k vs ln C, (c) ln k vs ln D, (d) ln k vs ln E, and (e) ln k vs ln F.

Conclusions

In this paper, the leaching of Cu from WPCBs using the H2SO4–H2O2 system has been comprehensively studied, and the influence of various parameters on the leaching efficiency of copper was experimentally investigated. The results showed that the temperature, solid–liquid ratio, and H2O2 concentration had significant effects on the leaching efficiency. BBD based on the RSM was used to study the effects and reciprocity of various parameters on the leaching efficiency and also to optimize the experimental parameters. From the response surface plots, the interactive relationships between each of the following pairs of A (temperature)-C (H2O2 concentration), A (temperature)-D (solid–liquid ratio), B (H2SO4 concentration)-D (solid–liquid ratio), C (H2O2 concentration)-D (solid–liquid ratio), E (particles sizes)-D (solid–liquid ratio), and F (stirring speed)-D (solid–liquid ratio) showed significant effects on the copper leaching efficiency. Based on the results of the leaching experiments, the leaching mechanism of copper in the H2SO4-H2O2 system has been established. The copper leaching process and data were well fitted to the Avrami model. Moreover, the kinetic equations of various parameters were established.

Experimental Section

Materials

WPCBs derived from end-of-life mobile phones of different brands (Figure a,b) were disassembled, cut, and crushed to small sizes. The size distribution range of the particles is shown in Figure c, and particles with sizes between 150 and 200 mesh were used for the leaching experiments. The WPCB particles used in this study were obtained after leaching tin and lead with hydrochloric acid (HCl). Typically, 1.0 g of WPCB powder was digested in aqua regia at 100 °C, and the contents of the main metals were determined by inductively coupled plasma atomic emission spectrometry (ICP-AES). The metal contents of the WPCBs and HCl leached residue are shown in Tables and 5, respectively.

Figure 8

Experimental raw materials: (a) waste mobile phone, (b) WPCBs, and (c) size distribution of particles. (Photograph courtesy of “Juanjuan Hao”. Copyright 2020. Publicly published on the internet and free domain.).

Table 4

Metal Content of WPCB Powders (wt %)

metal	Au	Ag	Cu	Fe	Ni	Sn	Pb	Zn
content	0.017	0.13	88.49	1.06	0.59	5.21	1.63	1.71

Table 5

Metal Composition of WPCB Powder after HCl Pretreatment (wt %)

element	Au	Ag	Cu	Fe	Sn	Pb	Ni
wt %	0.019	0.15	97.02	0.48	0.47	0.15	1.72

Leaching Experiments

The leaching experiments were conducted in a 250 mL beaker. The temperature of the experiment was maintained using a water bath. Briefly, 100 mL of fresh leaching solution for each experiment was prepared by adding certain concentrations of H2SO4 and 30 vol % H2O2. All of the solutions were prepared with deionized water. The leaching experiments were designed to understand the effects of the H2SO4 concentration (1.5–3.0 mol/L), H2O2 concentration (1.0–2.5 mol/L), solid–liquid ratio (0.05–0.1 g/mL), and temperature (20–60 °C) on copper leaching efficiency. Liquid samples, about 5–6 mL each, were taken at regular intervals, and their compositions were analyzed by ICP-AES. Each experiment was carried out three times and the average values were reported. This was done to prevent experimental errors. The leaching efficiency of copper is calculated by the following eq .where X is the leaching rate of the i-th sampling, wt %; V0 is the total volume of the leaching solution, L; V is the volume of the solution taken out of every time, L; C is the concentration of copper of solution taken out of every time, g/L; and M is the total mass of copper in raw materials, g.

Optimization of Experimental Parameters

RSM is a powerful tool for the investigation of the effects of several independent parameters on a response, and the Box–Behnken design (BBD), which is a typical method of the RSM, is used to optimize the experimental parameters. Based on the leaching experimental results, a reasonable range for each experimental parameter was chosen for the experimental design. The total number of experiments (N) are determined from eq .where k is the number of experimental parameters and n0 is the number of repetitions at the center points. The response (γ) was estimated using a second-order mathematical model based on the experimental data (eq ).where γ is the predicted response; α0 is the model constant; x1, x2, x3, and x4 represent the experimental parameters (temperature, H2SO4 concentration, H2O2 concentration, solid–liquid ratio, particle size, and stirring speed); α is a line coefficient; α is an interaction coefficient, and α is the quadratic coefficient (i = 1–6, j = 1–6). Each experimental parameters have three levels (i.e., low, medium, and high) with equally spaced intervals shown in Table . Design expert 8.0 software was applied for statistical analysis.

Table 6

Parameters and Corresponding Levels in Optimization Experiments

		levels
parameters		low	medium	high
		–1	0	1
A	temperature (°C)	30	40	50
B	H2SO4 concentration (mol/L)	2.0	2.5	3
C	H2O2 concentration (mol/L)	1.5	2	2.5
D	solid–liquid ratio (g/mL)	0.05	0.075	0.1
E	particle size (mesh)	100	150	200
F	stirring speed (rpm)	300	400	500