Optimization Studies on Recovery of Metals from Printed Circuit Board Waste

The aim of the study was to recover copper and lead metal from waste printed circuit boards (PCBs). The electrowinning method is found to be an effective recycling process to recover copper and lead metal from printed circuit board wastes. In order to simplify the process with affordable equipment, a simple ammonical leaching operation method was adopted. The selected PCBs were incinerated into fine ash powder at 500°C for 1 hour in the pyrolysis reactor. Then, the fine ash powder was subjected to acid-leaching process to recover the metals with varying conditions like acid-base concentration, electrode combination, and leaching time. The relative electrolysis solution of 0.1 M lead nitrate for lead and 0.1 M copper sulphate for copper was used to extract metals from PCBs at room temperature. The amount of lead and copper extracted from the process was determined by an atomic absorption spectrophotometer, and results found were 73.29% and 82.17%, respectively. Further, the optimum conditions for the recovery of metals were determined by using RSM software. The results showed that the percentage of lead and copper recovery were 78.25% and 89.1% should be 4 hrs 10 A/dm2.


Introduction
Recycling of e-waste is an important subject not only from the point of waste treatment but also from the recovery aspect of valuable materials [1][2][3][4]. Among the resources in e-waste, metals contribute more than 95% of the materials market value. Hence, the recovery of valuable metals is the inherent motive in e-waste disposal. In the past decades, many techniques for recovering valuable metals from e-waste have been developed such as gravity separation, magnetic separation, and electrostatic separation [5] synthesis of CuCl with e-waste, separation of PCBs with organic solvent method [6,7], cyanide and noncyanide lixiviants leaching methods, ammonium persulfate leaching bioleaching methods [8][9][10], or a combination of these approaches. Among those methods, hydrometallurgical methods are more accurate, predictable, and controllable [11]. erefore, hydrometallurgical techniques are most active in the research of valuable metal recovery from electronic scraps in the past two decades. However, traditional hydrometallurgical methods are acid dependent, time-consuming, and inefficient for simultaneous recovery of precious metals. Remarkably, a large amount of corrosive or toxic reagents, such as aqua regia, nitric acid, cyanide and halide, are consumed, producing large quantities of toxic and corrosive fumes or solution [12,13]. erefore, it is necessary to seek a more environmental friendly method for the recovery of valuable metals from e-wastes. Hydrometallurgical methods are used in the upgrading and refining stages of the recycling chain [14][15][16]. In this research article, the recovery of lead and copper metals from e-waste is widely investigated. e PCBs were converted into fine ash powder and subjected to electrowinning process for the recovery of metals. e experimental results were determined by EDS and AAS, respectively. Furthermore, the experimental results are validated through RSM software at different parameters like acid-base concentration, electrode combination, and leaching time [17][18][19][20][21][22].

Materials.
e computer PCBs were collected from various sources for the recovery of metals. e collected PCBs were crushed using roll crusher and powdered by a hammer mill. e crushed PCBs were incarnated through pyrolysis to avoid side reaction in the leaching process with the electrolyte solution.
e optimum condition of the pyrolysis reactor was 500°C in atmospheric pressure for 1 h where the epoxy resins and polymers were volatized at the temperature less than 500°C. e volatized contents were condensed and collected separately. e ferrous materials present in the obtained ash were separated by a magnetic separator.

Electrowinning Process.
e fine ash powder was treated with aqua regia solution (3 : 1 ratio of HCl and HNO 3 ) in the incineration chamber in order to avoid the liberation of toxic fumes.
en the precipitated salts obtained from the leaching was analyzed by EDS to determine the composition of metal present in the salts (Figure 1). e electrowinning setup consists of bath arrangement and amplifier. e bath having two slots for the anode and cathode fixing and the electrode is connected with amplifier, and the current density was varied through the amplifier ( Figure 2).

Extraction Process of Lead.
About 25 g of incinerated fine ash was added into the acid bath followed by the addition of ammonical electrolyte solution. e current density was set to 1 to 10 (A/dm 2 ). e solution was agitated at regular interval to get an effective electrodeposition: After the stipulated time of operation, pure lead was deposited on lead cathode. e deposited elements were scrapped and stored in an air tight container. e recovered lead quantitated from the EDS method. e spent acid left with mud filtered at pH 6-10 was stored in a glass container for further treatment.

Extraction Process of Copper.
About 25 g of incinerated fine ash was added into the acid bath followed by the addition of ammonical electrolyte solution. e current density was set to 1 to 10 (A/dm 2 ). e solution was agitated at regular intervals to get an effective electrodeposition. After the stipulated time of operation, pure copper (cupric) was deposited on the cathode and impure copper (cuprous ion) were deposited on the anode. e deposited elements were scrapped and stored in an air tight container. e recovered copper quantitated from the EDS method. e spent acid left with mud (nonleached elements) was filtered (pH-8.4) and were stored in a glass container for further treatment ( Figure 3): e spent solution collected from the electrodeposition was neutralized to 6.9 for the safe disposal as per the standard. Moreover, the presence of any metal in the spent solution was analyzed by Fourier-transform infrared spectroscopy. e results ( Figure 4) show that the metallic traces were found to be absent which confirms that all the metals recovered from the ashes deposited on the electrode.

RSM for Lead.
e response surface methodology (RSM) is a statistical modeling technique employed for multiple regression analysis using quantitative data obtained from designed experiments to solve multivariable equations (Table 1). e response surfaces can be visualized as threedimensional plots that exhibit the response as a function of two factors while keeping the other factors constant. In this above plot, the red zone corresponds to the extract percentage above 85%, yellow zone shows 60 to 70%, and the blue zone confirms below 40% extraction of lead ( (3) e model as a function of coded factor could be utilized to predict the response of each parameter within the given limit. Here, the maximum limit of process parameters (factors) is termed (coded) as +1 and minimum limit is terms (coded) as −1. (4) Equation (4) in terms of process parameters could be utilized to predict the response for the provided levels of each parameter (Table 2). In this equation, the original units of each parameters should be considered for each levels. In order to evaluate the comparative effect of each factor, the above equation should not be considered since the coefficients are balanced to embrace the units of each parameters. Also, the intercept does not fall at design space center.

Analysis of Variance (ANOVA).
Analysis of variance is used to determine the significant effects of process variables on current efficiency (Table 3)     fit is significant. ere is only a 0.08% chance that a lack of fit F value could be large that could occur due to noise. e coefficient represents the expected change in response per unit change in the factor value, when all remaining factors were constant. e intercept in an orthogonal design is the overall average response of all the runs. e coefficients are adjustments around the average factor settings. When the factors are orthogonal, the variance inflation factors (VIFs) are 1; VIFs greater than 1 indicate multicolinearity; the higher the VIF, the more severe the correlation of factors. As a rough rule, VIFs less than 10 are tolerable. Hence, from the data obtained (Table 4), the VIF values of lead are found to be tolerable.

Model Terms.
For a standard deviation of 1, the power calculations are performed using response type "continuous," and parameters are Δ � 2 and σ � 1. e power is evaluated over −1 to +1 coded factor space. From (Table 5), the standard errors should be similar to each other in a balanced design. e ideal VIF value should be 1, VIFs above 10 are cause for concern, and VIFs above 100 are cause for alarm, indicating coefficients are poorly estimated due to multicolinearity, where ideal Ri 2 is 0.0. High Ri 2 means terms are correlated with each other, possibly leading to poor models. If the design has multilinear constraints, then multicolinearity will exist to a greater degree. is inflates the VIFs and the Ri 2 , rendering these statistics would not perform well. Hence, FDS could be used. Power is an inappropriate tool to evaluate response surface designs. Use prediction-based metrics provided in this program via fraction of design space (FDS) statistics.

Fit Statistics.
A negative predicted R 2 implies that the overall mean may be a better predictor of the response than the current model. In some cases, a higher order model may also predict better. Adeq. precision measures the signal to noise ratio. A ratio greater than 4 is desirable. e ratio of 5.915 indicates an adequate signal. is model can be used to navigate the design space. e optimization of current efficiency is shown in Figure 6. From the results, it is observed that 69% of lead extract is obtained at current density � 10 A dm −2 , solvent ratio � 5 : 2, and the electrolysis time � 4 hours (Figures 7 and 8). e significance of regression coefficients were analyzed using the p-test and t-test. e p values are used to check the effect of interaction among the variables. A larger magnitude of t-value and a smaller magnitude of p value are significant in the corresponding coefficient term. e coefficient of current efficiency and the corresponding t and p values are shown in Table 6. Finally, the coefficients in the interaction terms for current density-electrolysis time is significant compared to current density-solvent ratio, and current density-electrolysis time.

RSM for Copper.
e regression equation for the RSM data plots for the copper is in terms of coded factors form as follows: e model (Equation 5) as a function of coded factor could be utilized to predict the response of each parameter within the given limit. Here, the maximum limit of process parameters (factors) is termed(coded) as +1 and minimum limit is termed (coded) as −1.
e modified equation or coded equation is very much useful in order to find the comparative effect of the process parameters by relating the coefficient of factors (Table 7). e final equation in terms of actual factors is Equation (5) in terms of process parameters could be utilized to predict the response for the provided levels of each parameter. In this equation, the original units of each parameters should be considered for each levels. In order to evaluate the comparative effect of each factor, the above equation should not be considered since the coefficients are balanced to embrace the units of each parameters. Also, the intercept does not falls at design space center (Table 8). In this contour plot, the red zone indicates extract percentages above 85%. And yellow and blue zones indicate 60 to 70% and below 40% extraction of copper (Figures 9 and 9(a)).

Analysis of Variance (ANOVA).
Analysis of variance is used to determine the significant effects of process variables on current efficiency along with the factor coding. e sum of squares is found to be Type III-partial derived from the ANOVA quadratic model. e model F value of 155.08 in the Table 9 implies the model is significant. A minimum value of 0.01% is possible for the F value due to noise. P values less than 0.0500 indicate model terms are significant.     In this case, A, B, C, AC, A 2 , and B 2 are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. If there are many insignificant model terms (not counting those required to support hierarchy), model reduction may improve the model. e lack of fit F value is nil that implies the lack of fit is significant. e coefficient represents the expected change in response per unit change in factor value, when all remaining factors were constant. e intercept in an orthogonal design is the overall average response of all the runs. e coefficients are adjustments around the average factor settings. When the factors are orthogonal, the VIFs are 1; VIFs greater than 1 indicate multicolinearity; the higher the VIF, the more severe the correlation of factors. As a rough rule, VIFs less than 10 are tolerable. Hence, from the data obtained (Table 10), the VIF Values of lead are found to be tolerable.

Fit Statistics.
A predicted R 2 implies that the overall mean may be a better predictor of the response than the current model. In some cases, a higher order model may also predict better. Adeq. precision measures the signal to noise ratio. A ratio greater than 4 is desirable. A ratio of 44.9 indicates an adequate signal. is model can be used to navigate the design space. e optimization of current efficiency is shown in Figure 10. e optimum extraction of 69% Cu is obtained at current density � 19 A dm −2 , solvent ratio � 5 : 2, and electrolysis time � 4 hour (Figures 11 and  12). e significance of regression coefficients was analyzed using the p-test and t-test. e p values are used to check the effect of interaction among the variables. A larger magnitude of t-value and a smaller magnitude of p value are significant in the corresponding coefficient term. e coefficient of      (Table 12). Finally, the coefficients in the interaction terms for current density-electrolysis time is significant compared to current density-solvent ratio and current density-electrolysis time.

Conclusion
e ammonia-lead nitrate and ammonia-copper sulphate system have been employed as a leaching agent for recovery of lead and copper from scraped printed circuit board wastes. A two-stage leaching was employed, wherein the first stage consisted of leaching the scrap board with 0.1 M Pb (NO 3 ) 2      dissolution of lead and copper leaching rate, and other metals was found in lower amounts, respectively. e undissolved residue portion from the leaching stage containing nickel, tin, and silica were leached out in respective treatments. e current efficiency was found to increase with current density and concentration ratio with the contact time in acid bath. Hence, 73.29% lead and 82.17% copper have been successfully recovered from the electrolysis process. And, also by RSM Software prediction, the recovery of lead and copper are as 78.25% and 89.1%, respectively. In addition to the quadratic model equation, ANOVA, model terms, and fit statistics were also tested for the experimental conditions.

Data Availability
All the data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.