Environ. Sci. Technol. 2009 43, 8974–8978
A Novel Process for Recovering Valuable Metals from Waste Nickel-Cadmium Batteries KUI HUANG, JIA LI, AND ZHENMING XU* School of Environmental Science and Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, People’s Republic of China
Received June 05, 2009. Accepted October 26, 2009.. Revised manuscript received October 13, 2009
The environment is seriously polluted due to improper and inefficient recycling of waste nickel-cadmium (Ni-Cd) batteries in China. The aim of this work is aimed to seek an environmentally friendly recycling process for resolving the negative impacts on environmental and human health resulting from waste Ni-Cd batteries. This work investigates the fundamentals of waste Ni-Cd batteries recycling through vacuum metallurgy separation (VMS) and magnetic separation (MS). The results obtained demonstrate that the optimal temperature, the addition of carbon powder, and heating time in VMS are 1023 K, 1 wt %, 1.5 h, respectively. More than 99.2 wt % Cd is recovered under the optimal experimental condition, and the Cd purity is 99.98%. Around 98.0 wt % ferromagnetic materials are recovered through MS under 60 rpm rotational speed and the recovery ratios of Fe, Ni and Co are 99.2 wt %, 96.1 wt %, and 86.4 wt %, respectively. The composition of ferromagnetic fractions in the residue after VMS increases from 82.3 to 99.6%. Based on these results, a process (including dismantling and crushing, VMS and MS) for recycling of waste Ni-Cd batteries is proposed. This novel process provides a possibility for recycling waste Ni-Cd batteries in a large industrial scale.
1. Introduction Waste nickel-cadmium (Ni-Cd) batteries were classified as a hazardous waste due to their high Cd concentration (1, 2). Although Ni-Cd batteries have already been forbidden in China, the previously produced Ni-Cd batteries were accumulated and posed a serious environment problem resulting from lacking of effective recycling technologies. Significant environmental and human health risk presented by electronic waste (e-waste, including waste batteries, waste printed circuit boards, waste liquid crystal display, etc.) aroused the attention of many researchers and strong reaction of the government in China (3-7). Cd and Ni released from waste Ni-Cd batteries were one of the major resources of Cd and Ni contamination. Therefore, it is urgent to develop a recycling technology with high performance and without negative impact to the environment to resolve the problems resulting from waste Ni-Cd batteries. The existing processes of recycling waste Ni-Cd batteries are pyrometallurgy, hydrometallurgy, and biometallurgy. Biometallurgy is a promising method, but the cycle for biometallurgy is too long and currently the known bacteria which are suitable for the treatment for waste Ni-Cd batteries * Corresponding author phone: +86 21 54747495; Fax: +86 21 54747495; e-mail:
[email protected]. 8974
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 23, 2009
are scarce and hard to culture (8, 9). Hydrometallurgy, with the benefit of low energy requirements, is an attractive solution. It does not pollute the environment and can be easily applied at industrial scale. It has been well developed by researchers such as Nogueira, Freitas, and Pietrelli, et al. (10-15). However, the recovered materials through hydrometallurgy were metallic compound, and the recovered product should be produced further. The traditional pyrometallurgy is extremely hazardous to the environment as a large part of the metals vapors is released into the atmosphere and contaminates distant areas. Vacuum metallurgy separation (VMS) has the advantages of high efficiency and better environmental properties. VMS process without atmosphere involving in can reduce pollution dramatically. This means vacuum metallurgy does not need secondary off-gas or wastewater treatment (16). It can be considered as a complementary approach to hydrometallurgy. Therefore, VMS was applied in our studies in order to directly obtain high purity metal rather than middle products. Some fundamentals on vacuum metallurgy have been established in recent years (17, 18). However, the previous studies focused on the fundamentals of vacuum distillation, and it is necessary to do some in-depth theoretical research to form a complete industry chain of waste Ni-Cd batteries recycling. In addition, the remnants after VMS were a complicated mixture of various materials, which should be further purified. Currently there is no correlative study on the further separation after VMS, and these materials were sent to steelworks directly. In order to increase the remnants purity after VMS, magnetic separation (MS) was adopted in this study. In this study, the fundamentals of waste Ni-Cd batteries recycling through VMS and MS were investigated. The objective of this work was to seek an environmentally friendly and efficient recycling process to resolve the environment problem caused by waste Ni-Cd batteries in a large industrial scale.
2. Experimental Section 2.1. Samples Preparation and Characterization. A total of 255 waste Ni-Cd batteries used in this work were provided by Foshan Nanhai VRUN Battery Factory, Guangdong province. These batteries were evenly divided into 17 groups. Initially all batteries were striped iron shells by self-designed machine. Then plastic, paper, and the electrolyte (KOH solution) were separated and collected to avoid secondary pollution. The remaining materials were crushed to a mixture of scraps by a self-designed shearer after doing previous separation. A schematic diagram of a dismantled waste Ni-Cd battery is shown in Figure 1. Chemical composition of the initial electrode materials and the residues after VMS and MS were determined by inductively coupled plasma atomic emission spectrometry
FIGURE 1. Schematic diagram of a dismantled waste Ni-Cd battery. 10.1021/es901659n CCC: $40.75
2009 American Chemical Society
Published on Web 11/02/2009
TABLE 1. Chemical Analysis of Ni-Cd Batteriesa element
mean content (%)
Cd Ni Fe Co Zn Li Cu a
23.2 16.4 35.0 1.1 0.5 0.1 0.3
Note: Fe includes the iron shell.
lgP*Me )
A + BlgT + CT + D T
(1)
Where A, B, C, D are constants determined for each metal element. It is reported that initially Cd presented in the form of hydroxide in the Ni-Cd batteries (11). During the VMS process, the decomposition of Cd and Ni monoxides occurred first, and the products of such decomposition were the oxides of these elements (17, 18). Cd(OH)2 ) CdO + H2O
(2)
Ni(OH)2 ) NiO + H2O
(3)
As the heating continued, there existed the possibility of Cd oxide decomposition causing the formation of Cd vapor (17, 18). This decomposition is possibly through reduction of the Cd vapor pressure in the atmosphere or the presence of a reduction agent. However, the boiling point of Cd oxide is over 2000 K. Energy consumption will be huge in heating process if it is recovered efficiently. Tenorio and Espinosa (17) used coal as reducing agent in their studies. As carbon powder was a cheap material, carbon powder (particle size < 100 mesh) was added in this study. The reactions in the system are shown as follows: CdO + C ) Cd(s) + CO
(4)
Cd(s) ) Cd(g)
(5)
FIGURE 2. Relationship between the saturation pressure and temperature of Cd and other metals. (ICP-AES, IRIS Advangtage 1000, THERMO, U.S.) after appropriately dissolved with a mixture 1:1 (v/v) of 37% HCl and 68% HNO3. The mean values of the main chemical elements found in the representative waste Ni-Cd batteries are presented in Table 1. After the VMS processing, all condensed materials were characterized and analyzed by ICP-AES and X-ray powder diffraction (XRPD, D/MAX-2200/PC, Rigaku Corporation, Japan) with Cu KR radiation, operating at 40 kV and 30 mA, where l ) 0.15418 nm for the Cu KR line. A scan speed of 2° (2θ) min-1 was utilized. 2.2. Apparatus. An exploratory vacuum resistance furnace and a magnetic pulley were fabricated and used in this experiment as shown in Supporting Information (SI) Figure S1. The vacuum resistance furnace consisted of furnace body, control cabinet, cooling system and vacuum pump team. The dried materials were placed in alumina crucible in furnace body. The magnetic separator consisted of separator body and control cabinet. The drum length, drum diameter, total length, maximum magnetic intensity of the magnetic separator is 475 mm, 160 mm, 1050 mm, and 250 mT, respectively.
3. Results and Discussion 3.1. Vacuum Metallurgy Separation and Recovery of Cd. The difference in vapor pressures of various metals at certain temperature is the principle of separating pure metals from mixed metallic materials by VMS. As a result, the metal with high vapor pressure and low boiling point can be separated from the mixed metals through distillation or sublimation, and then it can be recycled through condensation under certain temperature (16, 19). According to eq 1 (16, 19), Cd will melt at about 538 K when the saturation pressure decreases to below 1.33 Pa as shown in Figure 2, which curve is obtained from eq 1, while the saturation pressure is much higher than Fe and Ni at the same temperature. This supplies the thermodynamic foundation for separating Cd and other high vapor pressure metals from Fe and Ni particles.
The most important technological variables in the Cd distillation process included temperature, pressure, the addition of carbon powder, and heating time. The chamber pressure was maintained a level between 1 and 3.1 Pa during the distillation process. Based on the recovery ratios of Cd, a series of tests were adopted to derive the optimal parameters. The examined temperatures were 573, 675, 773, 873, 973, 1023, 1073, and 1173 K, heating times were 0.5, 1, 1.5, 2, and 2.5 h, and the addition of carbon powder were 0, 0.5, 1, 1.5, and 2 wt %, respectively. The Cd recovery ratios at the different tests were calculated by the following equation: Cdrecovery )
Mi × wt%Cdi - Mf × wt%Cdf Mi × wt%Cdi
(6)
Where Mi is the loaded initial mass, wt%Cdiis the initial Cd concentration, Mfis the mass of the material remaining at the crucible after the test, and wt%Cdf is the final Cd concentration in the crucible. Initial Cd quantity was considered to be 23.2% according to the characterization tests as shown in Table 1. Figure 3 presents Cd recovery ratios as a function of temperature, heating time, and the addition of carbon powder, respectively. All the data have been repeated three times. The relationship between Cd recovery ratio and temperature is shown in Figure 3a. Figure 3a shows that lower recovery ratios for Cd were obtained below 773 K, indicating that the Cd could not be effectively recovered. This is attributed to the fact that the saturated vapor pressure of metallic Cd is significantly lower than the operation pressure. For example, at 310 K temperature, the saturated vapor pressure of metallic Cd is merely about 1.33 × 10-8 Pa. But at above 773 K, the reduced metallic Cd could immediately evaporate and be recovered since the boiling point of Cd is about 761 K at 1.33 × 103 Pa. The result is also in accordance with Figure 2. The Cd recovery ratios increased dramatically with the increase of temperature and a plateau value was reached when the temperature was VOL. 43, NO. 23, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8975
FIGURE 3. Factors influencing Cd recovery ratios: (a) temperature; (b) the addition of carbon powder; (c) heating time. 1023 K. In addition, an increase in the temperature also caused a strong decrease in the heating time for the decomposition of Cd oxide. The relationship between Cd recovery ratio and the addition of carbon powder is shown in Figure 3b. It shows that even though carbon powder was not used as reducing agent, about half of Cd was still recovered, which indicates some components of waste Ni-Cd batteries may be employed as reducing agent in the VMS process. According to a report by Tenorio (18), plastic can act as a reduction agent, and so can the fabric used as a base for the electrolyte. Oxide reduction observed in tests even without the addition of carbon powder indicates that some components of waste Ni-Cd batteries, which were not separated during the dismantling and crushing process and remains attached to the electrode, act as a reduction agent, probably hydrocarbons, hydrogen, or CO originated from the pyrolysis of plastics or fabric in waste Ni-Cd batteries. However, the intrinsic carbon contained in waste Ni-Cd batteries was insufficient to reduce all oxides. The use of some materials to consume the oxygen inside the furnace was a way to reduce the partial oxygen pressure during vacuum distilling (17). Thus, it is still necessary to add a reduction agent in order to obtain metallic Cd. Figure 3b shows that the Cd recovery ratios were almost the same when more than 1 wt % carbon powder were added, which indicates that 1 wt % reducing agent in the reactions in this system was sufficient. Consequently, a plateau value was reached, and the optimal addition of carbon powder is 1 wt %. The relationship between Cd recovery ratio and heating time is shown in Figure 3c. It shows that Cd recovery ratios increased sharply with the increase of heating time below 1 h. This may be attributed to the fact that the evaporation of moisture, elimination of volatile materials contained in the sample and the decomposition of hydroxides occurred 8976
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 23, 2009
initially in the heating process. There is no Cd oxide decomposition during this first step. Thus, enough heating time is necessary to the decomposition of all Cd oxide and then evaporation in the VMS process. Figure 3c indicates that the heating time below 1 h was too short to recover Cd. However, when the heating time was 1.5 h, recovery ratio of Cd corresponded closely to the values gained, but the energy consumption was much lower. Consequently, the optimal heating time is 1.5 h. More than 99.2 wt % Cd was recovered in this heating time. In a similar way, the optimal temperature is 1023 K. The condensed Cd formed a thin film and then was collected as shown in Figure 4a. According to the ICP-AES and XRPD analyzed results, more than 99.2 wt % Cd was recovered under the proposed optimal experimental conditions (temperature ) 1023 K, the addition of carbon powder ) 1 wt %, and heating time ) 1.5 h, respectively). The condensed product was characterized as metallic Cd as shown in Figure 4b, the Cd purity was 99.98%. However, the Cd purity decreased to 98.5% when the examined temperature was 1173 K. The other elements, occupying less than 1.5 wt %, were mainly Zn (1.4 wt %), and the other contaminants were iron and copper, which are intrinsic to the laboratory process and that can be avoided by using industrial process. The presence of Zn in the condensed materials was expected since both positive and negative electrodes had Zn in their composition. According to Table 1, there was 0.5 wt % Zn in waste Ni-Cd batteries. Similarly to Cd, Zn shows volatility at the studied temperatures as shown in Figure 2 (16). This indicates that a lower pressure is not the better at any time. There is a pressure was called critical pressure at certain temperature. Different material has different critical pressure, which rises with increasing temperature. High vacuum not only highly depends on equipments investments,
FIGURE 4. Image of (a) the condensed materials and (b) X-ray diffraction pattern. Radiation: Cu (Kr)
TABLE 2. Chemical Analysis for The Residue in Crucible after VMSa element Cd Ni Fe Co Zn Li Cu
mean content of residue (%) 0.3 25.4 55.3 1.6 0.7 0.1 0.4
a Note: mean values of the residues in crucible for several chemical analysis results.
but also induces the distillation of other high vapor pressure metals, which can be condensed with Cd and decreases its purity. Consequently, appropriate pressure is crucial and the fractional condensation may be a valuable alternative choice. 3.2. Magnetic Separation and Recovery of Ferromagnetic Materials. Fe, Co, and Ni are ferromagnetic materials. Therefore, MS was adopted to recover Fe, Co and Ni based on the fact that different materials are extremely different in magnetic property in nonuniform magnetic field. After VMS step, the residues were magnetically separated by a magnetic pulley as shown in SI Figure S2. Table 2 presents the results of the chemical analysis for the residue in crucible after VMS under the proposed optimal condition. Table 2 reveals that the ferromagnetic materials represent around 82.3 wt % in the residue. The compositions of Fe, Ni and Co in ferromagnetic materials were 67.2 wt %, 30.9 wt %, and 1.9 wt %, respectively. In MS step, 1303.5 g ferromagnetic fractions and 285.7 g nonferromagnetic fractions were obtained, which is equivalent to 49.9 and 19.9 wt % of the initial weight, respectively. An effective separation and recovery of ferromagnetic materials, which reached 98.0 wt %, was obtained under 60 rpm rotational speed of the magnetic separator. The recovery ratios of Fe, Ni, and Co were 99.2 wt %, 96.1 wt %, and 86.4 wt %, respectively. Therefore, the ferromagnetic fractions were further purified. According to the ICP-AES analyzed result, the composition of ferromagnetic fractions increased from 82.3 to 99.6%. The other elements were mainly Cu (0.2 wt %). As can be observed, the remaining Cd concentrated in the powdered fraction of the materials. A Ni-Co alloy containing less than 0.01 wt % Cd was also obtained as products of the recycling when the temperature reached 1173 K. The mass of remaining Cd and the alloy can be negligible. These ferromagnetic fractions, as second raw materials, are able to be transported to steelworks to manufacture Ni-Fe alloy or stainless steel.
4. Novel and Environmentally Friendly Process 4.1. Environmental Impact Assessment. With the conception of improving efficiency of resource-utilization, the simple recycling process for waste Ni-Cd batteries has a negative effect on the recovery of the valuable metals (such as Ni, Co, Zn, Li). Consequently, the process (including dismantling and crushing, VMS and MS) for recycling of waste Ni-Cd batteries is proposed based on the above results as shown in Figure 5. This process is to recover major heavy metals, particularly Cd and Ni, from the waste Ni-Cd batteries and convert them into harmless or reusable products. Some pollution control measures were adopted to treat the waste paper, plastics, and electrolyte in the experiment process. Then these waste materials were sent to hazardous waste management agency for further treatment after dismantling and crushing process. The Cd and Ni were effectively recycled, and the recovered materials were high value-added products. Therefore, the negative impacts on ecology quality and human health resulting from waste Ni-Cd batteries can be resolved effectively through the proposed process. 4.2. Mass Balance. The results of the chemical analysis by ICP-AES reveal that Cd, Fe, and Ni represent around 75 wt % of the waste Ni-Cd batteries. Therefore, the effective recycling of these metals, not only brings benefits in environmental terms, but also brings profit in economical terms. The initial weight of the 255 batteries was 2612.5 g. After the previous separation, 321.3 g nonmetallic materials (including paper, electrolyte, plastic, etc.) were obtained and these waste materials were sent to hazardous waste management agency for further treatment after dismantling processing. The other samples (2236.4 g) were crushed and then were sent to the furnace. Through VMS process, 599.6 g high vapor pressure metals were obtained and the residues in VMS were separated through MS. In MS process, 285.7 g nonferromagnetic fractions and 1303.5 g ferromagnetic fractions were obtained. The mean compositions of each fraction are shown in SI Figure S3. 4.3. Economic Assessment. Waste batteries recycling are motivated for environmental reasons. Therefore, waste Ni-Cd batteries, which is high Cd concentration, must be collected and recycled even the recycling process is not profitable. The revenues from recycling of waste batteries are attained by selling recycled metals, and the revenues can offset the cost of the recycling system. In the proposed process, the electric energy consumption of VMS (including heating and vacuum process) is the main economic impact. Table 3 presents the results of the power and the operation time in each process if the proposed process is applied to an industrial scale. The energy costs were calculated by the following equation: Ccost ) P × T × Ce VOL. 43, NO. 23, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
(7)
9
8977
FIGURE 5. The process proposed for recycling of waste Ni-Cd batteries.
TABLE 3. Estimate Energy Costs for Recycling in an Industrial Scale process crush VMS MS
power (kw)
operation time(h)
estimated cost ($/t)
20 500 10
1 3 0.5
3 225 0.75
Where Ccost is the cost of each process, P is the power of each process, T is the operation time of each process, and Ce is the unit price of electricity in China (the average power charge applying to industry is $0.15/kWh). Consequently, the energy cost to process 1 t waste Ni-Cd batteries in an industrial scale is at maximum $228.75 as shown in Table 3. Around 230 kg Cd and 520 kg ferromagnetic products produced by the proposed process worth at least $1390.5 and $146.4, respectively ($1536.9 in total). Additionally, the government provides subsidies for recycling waste batteries (including sorting, collecting and transporting batteries, labor, capital, etc.). The results demonstrate that there are at least $1308.1 revenues in 1 t waste Ni-Cd batteries recycling through the proposed process. The proposed process consumes highly energy, however, there are also large returns due to the recovered materials are high valueadded product. Recycling process operating with economic benefits strong depends on the quality (in particular the purity) of the recovered products and on the flexibility of the process (14). VMS and MS process have the advantages of high efficiency, flexibility, and easy operation. The proposed process with high efficiency of the recovery of valuable metals is in accordance with cyclical economy concept. Therefore, this novel process for recycling of waste Ni-Cd batteries will be benefits in environmental terms and profits in economical terms.
Supporting Information Available Figures showing the experimental apparatus, Figures showing the magnetic separator function and the image of the separated materials, and mass balance. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Ministry of Environmental Protection of the People’s Republic of China. Technical Policy on the Prevention and Control of Pollution by Waste Battery; EPA: Beijing, 2003. (2) The European Parliament and the Council of European Union. New Directive 2006/66/EC on Batteries and Accumulators and Wastes Batteries and Accumulators; Official Journal of the European Union: Brussels, 2006.
8978
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 23, 2009
(3) Leung, A. O. W.; Duzgoren-Aydin, N. S.; Cheung, K. C.; Wong, H. Heavy metals concentrations of surface uust from e-waste recycling and its human health implications in southeast China. Environ. Sci. Technol. 2008, 42, 2674–2680. (4) Wong, C. S. C.; Wu, S. C.; Duzgoren-Aydin, N. S.; Aydin, A.; Wong, M. H. Trace metal contamination of sediments in an e-waste processing village in China. Environ. Pollut. 2007, 145, 434–442. (5) Wang, F.; Leung, A. O. W.; Wu, S. C.; Yang, M. S.; Wong, M. H. Chemical and ecotoxicological analyses of sediments and elutriates of contaminated rivers due to e-waste recycling activities using a diverse battery of bioassays. Environ. Pollut. 2009, 157, 2082–2090. (6) Zhang, J.; Min, H. Eco-toxicity and metal contamination of paddy soil in an e-wastes recycling area. J. Hazard. Mater. 2007, 165, 744–750. (7) Leung, A. O. W.; Cai, Z. W.; Wong, M. H. Environmental contamination from electronic waste recycling at Guiyu, southeast China. J. Mater. Cycles Waste Manag. 2006, 8, 21–33. (8) Zhao, L.; Yang, D.; Zhu, N. Bioleaching of spent Ni-Cd batteries by continuous flow system: Effect of hydraulic retention time and process load. J. Hazard. Mater. 2008, 160, 648–654. (9) Zhao, L.; Zhu, N.; Wang, X. Comparison of bio-dissolution of spent Ni-Cd batteries by sewage sludge using ferrous ions and elemental sulfur as substrate. Chemosphere 2008, 70, 974–981. (10) Nogueira, C. A.; Margarido, F. Leaching behaviour of electrode materials of spent nickel-cadmium batteries in sulphuric acid media. Hydrometallurgy 2004, 72, 111–118. (11) Nogueira, C. A.; Margarido, F. Chemical and physical characterization of electrode materials of spent sealed Ni-Cd batteries. Waste Manage. 2007, 27, 1570–1579. (12) Freitas, M. B. J. G.; Rosale´m, S. F. Electrochemical recovery of Cd from spent Ni-Cd batteries. J. Power Sources 2005, 139, 366–370. (13) Freitas, M. B. J. G.; Penha, T. R.; Sirtoli, S. Chemical and electrochemical recycling of the negative electrodes from spent Ni-Cd batteries. J. Power Sources 2007, 163, 1114–1119. (14) Pietrelli, L.; Bellomo, B.; Montereali, M. R. Characterization and leaching of NiCd and NiMH spent batteries for the recovery of metals. Waste Manage. 2005, 25, 221–226. (15) Pietrelli, L.; Bellomo, B.; Fontana, D.; Montereali, M. R. Rare earths recovery from NiMH spent batteries. Hydrometallurgy 2002, 66, 135–139. (16) Zhan, L.; Xu, Z. Application of vacuum metallurgy to separate pure metal from mixed metallic particles of crushed waste printed circuit board scraps. Environ. Sci. Technol. 2008, 42, 7676–7681. (17) Espinosa, D. C. R.; Tenorio, J. A. S. Recycling of nickel-cadium batteries using coal as reducing agent. J. Power Sources 2006, 157, 600–604. (18) Espinosa, D. C. R.; Tenorio, J. A. S. Fundamental aspects of recycling of nickel-cadmium batteries through vacuum distillation. J. Power Sources 2004, 135, 320–326. (19) Hu, G.; Cai, X.; Rong, Y. Fundamentals of Materials Science, 2nd ed.; Shanghai Jiao Tong University Press: Shanghai, 2006 (in Chinese).
ES901659N
