- Email: [email protected]
Recycling indium from waste liquid crystal display panel by vacuum carbon-reduction
Journal of Hazardous Materials 268 (2014) 185–190
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Recycling indium from waste liquid crystal display panel by vacuum carbon-reduction Yunxia He, En Ma, Zhenming Xu ∗ School of Environmental Science and Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, People’s Republic of China
h i g h l i g h t s • The recovery and reuse of indium in waste LCD panel is essential for conservation of resources and environmental protection. • We recycled indium from waste liquid crystal display panel by vacuum carbon-reduction. • There is no hazardous materials produced in this process.
a r t i c l e
i n f o
Article history: Received 27 September 2013 Received in revised form 16 December 2013 Accepted 7 January 2014 Available online 13 January 2014 Keywords: LCD In2 O3 Indium Vacuum Carbon
a b s t r a c t This study investigated the recovery of indium from waste liquid crystal display (LCD) panel using vacuum carbon-reduction. First of all, high purity In2 O3 was investigated. The results indicated that indium can be reclaimed from In2 O3 using vacuum carbon-reduction in thermodynamics and dynamics. The conditions of 1223 K, 50 wt% carbon addition, 30 min, and 1 Pa were confirmed as the optimal conditions for pure In2 O3 and high purity indium could be selectively recovered on condensing zone. Based on this, the experiment of the recovery of indium from waste LCD power was performed. The best parameters were confirmed as 1223 K and 1 Pa with 30 wt% carbon addition for 30 min. The recovery rate of indium from LCD powder could reach to 90 wt%. No hazardous materials produced in this process. Therefore, this technique provides the possibility of reutilization of LCD in an environmentally friendly way. © 2014 Elsevier B.V. All rights reserved.
1. Introduction With the advantages of small volume, light quality and low power consumption, LCD has replaced cathode ray tube (CRT) display in most of appliances. The total global TV shipments were 233 million in 2012, while LCD TV sheared 87.3%, plasma display panel (PDP) TV and CRT TV shared 5.7%, 6.9%, respectively, according to the latest finding published in the NPD Display Search [1]. Besides, the increase in demand for high-end products, such as smart phone, tablet PC and digital product, is considered to drive the flat panel display shipments. Taking account of the short life-cycle, 3–5 using years in general, and the endless requirement for new products, a large number of LCD products are coming into discarding period. Indium–tin oxide (ITO), a solid solution of indium(III) oxide (In2 O3 ) (90–95%) and tin(IV) oxide (SnO2 ) (10–5%), is widely used as transparent conductive film in LCD because of its significant properties, such as electrical conductivity and optical transparency. Indium consumption in ITO accounts for over 70% of its total
∗ Corresponding author. Tel.: +86 21 54747495; fax: +86 21 54747495. E-mail address: [email protected] (Z. Xu). 0304-3894/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2014.01.011
consumption and the greatest demand on ITO is LCD nowadays. However, the global reserve of indium is about 16–19 thousand tons and is only 1/6 of that of gold. In addition, the average content of indium in Zinc blende as the most important indium-carrier mineral ranges from less than 1–100 ppm, while that is about 250 ppm in LCD [2]. Therefore, the recycling and reuse of indium from waste LCD become particularly important. Various methods for recycling of indium from secondary sources containing indium, particular ITO sputtering waste, have been investigated, mainly based on the hydrometallurgy [3,4], which involves leaching, concentrate and separation, and electrorefining. Among them concentrate and separation involve precipitation, cementation, and solvent extraction. In general, the recovery rate of indium is high through hydrometallurgy. However, the using of various solvents, including corrosive acid and hazardous extraction substances, increases the potential environmental risks. The recycling of indium scrap has been investigated from ITO through pyrometallurgy, but few from end products. In2 O3 can be reduced to metallic indium under high temperature condition with reducing atmosphere [5], such as H2 , C. Alloy of indium and tin can be recovered. Pure metallic indium could be separated form alloy through vacuum. Satoshi and Katsuya [6] conducted
186
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
laboratory scale experiments for the recycling of metallic indium from ITO scrap. The process consisted of two stages, reduction of ITO with CO at 1073 K and vaporization of In–Sn alloy at 1373 K under vacuum. Metallic indium drops could be recovered on condensing zone. This method requires high temperature for the separation of indium from In-Sn alloy. Besides the use of CO is a potential safety risk as its asphyxiating. In addition to leaching-extraction and pyrometallurgy, some researchers have reported the chlorination process of waste LCD panel for the recycling of indium. Takahashi et al. [7] presented a method of recovering indium from waste LCD panel by chlorideinduced vaporization. In this process, crushed LCD panel was first treated in HCl solution for altering the indium(III) oxide into a chloride-induced indium compound. Then the chloride-induced indium compound was vaporized at relatively low temperature (673 K) in nitrogen atmosphere. The vaporization rate of indium could reach 84.3 wt%. In addition, chlorinated separation of indium from ITO scrap or LCD panel was also investigated by other researchers [8,9]. Comparing with pyrometallurgy, chlorinated separation has relative lower temperature. However, it needs a lot of additive and subsequent purification procedure for metallic indium. These two methods are still in laboratory stage. Therefore, in order to directly recover metallic indium from waste LCD panel, environmentally friendly methods and high efficiently methods for achieving high purity metallic indium are desired. Vacuum metallurgy has been widely used in non-ferrous metal smelting [10]. In addition, this method is also reported to separate heavy metal lead from funnel glass [11]. However, there is no study was reported on how this technique is used to recover indium from waste LCD panel. In this paper, we study the feasibility of recycling of metallic indium from In2 O3 through vacuum carbon-reduction at relative low temperature. Based on the feasibility experiments, the method of vacuum carbon-reduction is applied to recycle indium from discard LCD panel in experimental scale. Here coke power is chosen as reducing agent due to its safety, low price and high calorific value.
2. Materials and methods 2.1. Materials LCD panels were firstly dismantled from discarded computers by hand, and then polymer films (mainly contain (polyvinyl alcohol) PVC and (Triacetyl Cellulose) TAC) were also removed through manual from these panels. Liquid crystal between two glasses was dissolved by acetone. Take TFT-LCD for example, the main structure of LCD panel is shown in Fig. 1. Last remained glass panels were crushed into power and sieved to smaller than 0.3 mm. Coke powder (0.8 mm, carbon content >80%) were prepared as reductant. Besides In2 O3 (99.99 mass% purity) were also prepared in this experiment. 2.2. Apparatus Experiments were carried out in a self-assemble vacuum furnace. The schematic diagram of experiment system is shown in Fig. 2. The main body consisted of tubular electric resistance furnace (chamber dimensions is Ø40 mm × 600 mm), quartz tube reactor (Ø35 mm × 900 mm) and vacuum pump (power is 0.25 kW). Middle part of furnace is heating zone and in the outside of it temperature is gradually decrease with the distance from heating zone, and the end of furnace is condensing zone. 2.3. Methods Sample and carbon powder were blend well with a certain mass ratio in a quartz boat, and then the quartz boat with mixture was put into a quartz tube. The quartz tube was pushed into furnace heating area, and heated up to experimental temperature in nitrogen atmosphere. After system reached set temperature, it was evacuated to 1 Pa by vacuum pump. Temperatures examined were 1073, 1123, 1173, 1223 and 1273 K. After some definite time, the sample was cooled to room temperature under vacuum atmosphere and
Fig. 1. Structure of TFT-LCD panel.
Fig. 2. Schematic diagram of the experimental setup.
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
187
Fig. 3. Relations between recovery rate of indium and (a) temperature, (b) carbon adding amount and (c) heating time for pure In2 O3.
identified before and after experiment. The difference quality of indium before and after experiment was considered as the result of reduction with carbon and vaporization of indium. Recovery rate of indium was calculated by the following formula: R=
M0 − M1 × 100% M0
(1)
where M0 and M1 are the initial amount and the remaining amount of indium, respectively. Recovery rate of indium is as the function of temperature, carbon adding amount, and heating time. 2.4. Analysis The quantitative analysis of indium in residues was measured by inductively coupled plasma optical emission spectrometer (ICPOES, icap6300, US), after being completely dissolved with a 1:1 mixture (v/v) of 37 wt% HCl and 68 wt% HNO3 . The crystalline phase of products was identified by X-ray polycrystalline diffractometer (XRD, D8 ADVANCE, Germany) using Cu K␣ radiation. 3. Results and discussion 3.1. Vacuum carbon-reduction for In2 O3 Pure In2 O3 was firstly studied considering that In2 O3 is the target oxide and the quality of it in LCD panel is tiny. The effects of temperature, addition of carbon powder and heating time on
recovery rate of indium were studied. After the best conditions were confirmed, the products produced under optimal condition were analyzed. 3.1.1. Effects of temperature and vacuum on reduction of In2 O3 with carbon In the process of vacuum carbon-reduction, In2 O3 is considered to be firstly reduced to metallic indium, and then indium evaporated into gaseous phase and condensed on low temperature zone. Therefore, temperature and vacuum degree are key factors for reduction of In2 O3 and evaporation of indium. Fig. 3a shows the relation between recovery rate of indium and temperature. The recovery rate of indium increased sharply with the increase of temperature at below 1173 K (from 30 wt% to 98 wt%) and a plateau value (100 wt%) is reached at above 1173 K. The reaction between In2 O3 and C, and the Gibbs free energy (G) at 700 K are listed below: In2 O3 + 3C = 2In(s) + 3CO(g) G700 K = 0
(2)
Relative thermodynamic data were obtained through thermodynamic handbook [12]. The redox reaction between In2 O3 and C takes place when G is 0. At 1 Pa condition, In2 O3 can be reduced to In by carbon powder when the temperature is higher than 700 K according to the equation (2). Therefore, In2 O3 is considered to be reduced to metallic indium since the examined temperatures in the experiments are higher than 700 K (from 1073 K to 1273 K). However, lower recovery rate of indium was obtained
188
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
3.1.3. Effect of heating time The relationship between recovery rate of indium and time is shown in Fig. 3c, showing that the recovery rate of indium can reach almostly 100 wt% at heating time of 10 min and reach 100 wt% at 30 min. Therefore, heatng time for indium recycling is considered as 30 min. In general, the recovery rate of indium is increased with the increasing of temperature, addition of carbon powder and heating time under certain condition. The above experiments show that the recovery rate of indium can reach 100 wt% at the conditions of pressure = 1 Pa, temperature = 1223 K, addition of carbon power = 50 wt% and heating time = 30 min. Therefore, these conditions were considered as the optimal reaction conditions for the recycling of metallic indium from pure In2 O3 powder.
Fig. 4. Relation between vapor pressure of indium and temperature.
within 1073–1173 K. Therefore, the vapor pressure of indium may be the key limiting factor for indium recovery at below 1173 K. Fig. 4 shows the relation between vapor pressure of indium and temperature (date source [13]), indicating saturate vapor pressure of indium is about 1 Pa while the temperature is about 1223 K. Therefore, reduced metallic indium could immediately evaporate and be recovered at above 1223 K under 1 Pa. However, it is merely about 0.06 Pa at 1173 K. This contributes the low recovery rate of indium at below 1173 K. Based on this, following reduction experiment for recycling indium was conducted at 1223 K and 1 Pa. 3.1.2. Effect of carbon adding amount The relationship between recovery rate of indium and addition of carbon powder is shown in Fig. 3b. It shows that indium recovery increased dramaticly with the increase of addition of carbon powder under 50 wt%, and a plate is reached in later. When amount of carbon additon is 30 wt%, the recovery rate of indium is only 65 wt%. This indicates that indium could not sufficiently reduced with blow 30 wt% carbon additon. The possible reactions between In2 O3 and C are listed below when carbon is not enough for reducing In2 O3 to In: In2 O3 (s) + C(s) = 2InO(s) + CO(g) G1223 K = −79 kJ/mol
(3)
In2 O3 (s) + 2C(s) = In2 O(g) + 2CO(g) G1223 K = −333 kJ/mol (4)
As we can see that InO and In2 O can be generated at 1223 K under 1 Pa. Therefore, the recovery rate of indium is low when C is not enough since indium exists in InO solid form. In addition, the evaporation of suboxide In2 O may cause the lose of indium. But there is no In2 O be found in product (this will be discussed in the following section). Recovery rate of indium can reach 50 wt% without the addition of carbon powder. This possibly results from the evaporation of suboxide In2 O and In2 O3 . In2 O3 may breake into evaporable In2 O and O2 , and the boiling point of In2 O3 is about 1123 K. In high vacuum conditon, carbon is condsered as the main reductant, therefore, the increased of carbon powder adding contributed to the increasing of indium recovery rate. The recovry rate of indium reaches 100 wt% at 50 wt% carbon powder amount. This percentage was considered to be the optimum carbon adding amount.
3.1.4. Analysis of condensed products for pure In2 O3 Products produced under optimal conditon were collected and analyzed by XRD. The distribution of condensed product and temperature is shown in Fig. 5a. It shows that metallic indium was condensed on the inner-wall of the quartz tube (Fig. 5b) with the temperature range of 833–573 K. In addition, the XRF shows that the indium has a 100% purity. In B zone with the temperature range of 433–573 K, XRD pattern (Fig. 5c) demonstrates that the major constitue of products is In2 O3 with the remainder is metallic In. This could be the oxidation of In2 O produced from the decomposition and evaporation of In2 O3 . Besides, a part of In2 O may react with reducing agent such as CO, that contributed to the metallic indium in B zone. In2 O3 in B zone accounts for less than five percent of total quality of products and it is recoverable. Therefore, the loss of indium caused by vaporation and decomposition is negligibal. The vapor of indium reduced from In2 O3 at 1223 K and 1 Pa cooled to indium solid with airflow and adhered to the inner wall of quartz tube. The temperature of quartz tube gradually decreased from the heating zone to the end of tube as shown in Fig. 6. It shows that the temperature dropped from 1223 K at heating zone to 400 K at 428 mm distance from the center of heating zone. Therefore, we can learn that the condensing zone of metallic indium is 350–388 mm off the center of furnace and temperature of condensing zone ranges from 573 K to 833 K according to Figs. 5(a) and 6. 3.2. Vacuum carbon-reduction for LCD glass powder The experiments for pure In2 O3 have confirmed that the recovery of metallic indium from In2 O3 is feasible, and optimal conditions of 1223 K, 50 wt% carbon adding amount, 30 min heating time and 1 Pa, were also determined for pure In2 O3 . However, in practice, In2 O3 content is only about 0.2 g/kg (LCD glass) and the major ingredient of LCD glass is SiO2 . These elements lead to the differences between pure In2 O3 and LCD glass for recycling indium. In addition, SnO2 should be considered since it could be reduced to impurity Sn. Based on above elements, carbon adding amount was considered for the investigation of indium recycling from waste LCD panel. 3.2.1. Effect of carbon adding amount Fig. 7 shows the relation between recovery rate of indium and carbon adding amount for LCD powder. As we can see that the recovery rate of indium is increased with the increasing of carbon powder amount at blow 30 wt% addition of carbon powder. It reaches 90 wt% when the addition of carbon powder is 30 wt% and it does not continue to increase with the increasing of carbon powder amount. This indicates that the consumption of carbon powder is very large for the indium recycling from LCD powder. Two reasons could lead to the large consumption of carbon powder. One reason
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
189
Fig. 5. (a) Distribution of condensed product and temperature in one end of quartz tube for pure In2 O3 . (b) Photo for metallic indium inside A zone. (c) X-ray diffraction patterns of In2 O3 and In in B zone.
is a large number of glass powders from glass substrate hinder the immediate contact of In2 O3 and carbon powder. Another reason is impurity substance react with carbon powder lead to the increase of carbon consumption. Impurity SnO2 is discussed here since ITO is composed of SnO2 and In2 O3.
3.2.2. Effect of impurity SnO2 The possible reaction between SnO2 and C, and the corresponding Gibbs energy at 1 Pa are listed below: SnO2 (s) + 2 C(s) = Sn(s) + 2CO(g) G1223 K = −4 × 105 kJ/mol (5)
Fig. 6. The distribution of temperature of the quartz tube at 1223 K.
Fig. 7. Relation between recovery rate of indium and carbon adding amount for LCD powder.
190
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
carbon-reduction. The best parameters were confirmed as 1223 K and 1 Pa with 30 wt% carbon addition for 30 min, and the recovery rate of indium from LCD powder could reach to 90 wt%. In addition, there is no hazardous materials produced in this process. Acknowledgements
Fig. 8. Distribution of condensed product and temperature for LCD powder in one end of quartz tube.
This work was supported by the National Natural Science Foundation of China (51178262, 51278293), and the Scientific Research Projects of Science and Technology Commission of Shanghai Municipality (11DZ2282800). References
The Gibbs energy is less than 0 at 1223 K. Therefore, Sn can be generated at 1223 K. Results show that content of SnO2 in LCD panel is decreased from 196 ppm to 110 ppm and the recovery rate of tin is only about 15 wt%. The vapour pressure of Sn (is only 0.002 Pa) is much lower than that of In (is about 1 Pa) at 1223 K. Therefore, the recovery rate of Sn is low due to its low saturate vapour pressure. This also significantly reduces the effect of Sn on In product. 3.2.3. Analysis of condensed products for LCD powder 250 g LCD powder were investigated under the conditions of 1223 K, 1 Pa, 30 min, and 30 wt% addition of carbon. The distribution of condensing products and temperature are shown in Fig. 8. There is a rainbow circle in the quartz tube and the inside layer of rainbow nearing 833 K is silver-white. What’s more, the condensing temperature of these substances agrees with that of metallic indium in experiment for pure In2 O3 (Fig. 5a). Therefore, it is assumed that indium can be reduced and separated from LCD powder. 4. Conclusions This study investigated the recovery of indium from waste LCD panel using vacuum carbon-reduction. An environmentally friendly vacuum carbon-reduction method for recycling of indium from waste LCD panel has been developed. The results indicated that the indium can be reclaimed from LCD panel by vacuum
[1] Display Search, Global LCD TV shipments fall for the first time in 2012; outlook cautious for 2013, http://www.displaysearch.com/cps/rde/xchg/ displaysearch/hs.xsl/130321 global lcd tv shipments fall for the first time in 2012.asp (last accessed: 02.09.2013). [2] J. Yang, T. Retegan, C. Ekberg, Indium recovery from discarded LCD panel glass by solvent extraction, Hydrometallurgy 137 (2013) 68–77. [3] Y. Li, Z. Liu, Q. Li, Z. Liu, L. Zeng, Recovery of indium from used indium–tin oxide (ITO) targets, Hydrometallurgy 105 (2011) 207–212. [4] S. Virolainen, D. Ibana, E. Paatero, Recovery of indium from indium tin oxide by solvent extraction, Hydrometallurgy 107 (2011) 56–61. [5] J. Chen, J. Yao, Y. Zhou, Z. Chen, X. Wang, J. Huang, Recovery indium from waste ITO target, Chin. J. Rare Metal 27 (2003) 101–103. [6] I. Satoshi, M. Katsuya, Recoveries of metallic indium and tin from ITO by means of prometallurgy, High Temp. Mater. Proc. 30 (2011) 317–322. [7] K. Takahashi, A. Sasaki, G. Dodbiba, J. Sadaki, N. Sato, T. Fujita, Recovering indium from the liquid crystal display of discarded cellular phones by means of chloride-induced vaporization at relatively low temperature, Metall. Mater. Trans. A 40 (2009) 891–900. [8] O. Terakado, D. Iwaki, K. Murayama, M. Hirasawa, Indium recovery from indium tin oxide, ITO, thin film deposited on glass plate by chlorination treatment with ammonium chloride, Mater. Trans. 52 (2011) 1655–1660. [9] E. Ma, Z. Xu, Technological process and optimum design of organic materials vacuum pyrolysis and indium chlorinated separation from waste liquid crystal display panels, J. Hazard. Mater. 263 (2013) 610–617. [10] Y. Dai, B. Yang, Vacuum Metallurgy of Nonferrous Metals, 2nd ed., Metallurgical Industry Press, Beijing, 2009. [11] M. Chen, F. Zhang, J. Zhu, Lead recovery and the feasibility of foam glass production from funnel glass of dismantled cathode ray tube through pyrovacuum process, J. Hazard. Mater. 161 (2009) 1109–1113. [12] D. Ye, Practical Inorganic Thermodynamic Date Manual, Metallurgical Industry Press, Beijing, 1981. [13] C. Xu, Vacuum Engineering Technology, Chemical Industry Press, Beijing, 2006.
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Recycling indium from waste liquid crystal display panel by vacuum carbon-reduction Yunxia He, En Ma, Zhenming Xu ∗ School of Environmental Science and Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, People’s Republic of China
h i g h l i g h t s • The recovery and reuse of indium in waste LCD panel is essential for conservation of resources and environmental protection. • We recycled indium from waste liquid crystal display panel by vacuum carbon-reduction. • There is no hazardous materials produced in this process.
a r t i c l e
i n f o
Article history: Received 27 September 2013 Received in revised form 16 December 2013 Accepted 7 January 2014 Available online 13 January 2014 Keywords: LCD In2 O3 Indium Vacuum Carbon
a b s t r a c t This study investigated the recovery of indium from waste liquid crystal display (LCD) panel using vacuum carbon-reduction. First of all, high purity In2 O3 was investigated. The results indicated that indium can be reclaimed from In2 O3 using vacuum carbon-reduction in thermodynamics and dynamics. The conditions of 1223 K, 50 wt% carbon addition, 30 min, and 1 Pa were confirmed as the optimal conditions for pure In2 O3 and high purity indium could be selectively recovered on condensing zone. Based on this, the experiment of the recovery of indium from waste LCD power was performed. The best parameters were confirmed as 1223 K and 1 Pa with 30 wt% carbon addition for 30 min. The recovery rate of indium from LCD powder could reach to 90 wt%. No hazardous materials produced in this process. Therefore, this technique provides the possibility of reutilization of LCD in an environmentally friendly way. © 2014 Elsevier B.V. All rights reserved.
1. Introduction With the advantages of small volume, light quality and low power consumption, LCD has replaced cathode ray tube (CRT) display in most of appliances. The total global TV shipments were 233 million in 2012, while LCD TV sheared 87.3%, plasma display panel (PDP) TV and CRT TV shared 5.7%, 6.9%, respectively, according to the latest finding published in the NPD Display Search [1]. Besides, the increase in demand for high-end products, such as smart phone, tablet PC and digital product, is considered to drive the flat panel display shipments. Taking account of the short life-cycle, 3–5 using years in general, and the endless requirement for new products, a large number of LCD products are coming into discarding period. Indium–tin oxide (ITO), a solid solution of indium(III) oxide (In2 O3 ) (90–95%) and tin(IV) oxide (SnO2 ) (10–5%), is widely used as transparent conductive film in LCD because of its significant properties, such as electrical conductivity and optical transparency. Indium consumption in ITO accounts for over 70% of its total
∗ Corresponding author. Tel.: +86 21 54747495; fax: +86 21 54747495. E-mail address: [email protected] (Z. Xu). 0304-3894/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2014.01.011
consumption and the greatest demand on ITO is LCD nowadays. However, the global reserve of indium is about 16–19 thousand tons and is only 1/6 of that of gold. In addition, the average content of indium in Zinc blende as the most important indium-carrier mineral ranges from less than 1–100 ppm, while that is about 250 ppm in LCD [2]. Therefore, the recycling and reuse of indium from waste LCD become particularly important. Various methods for recycling of indium from secondary sources containing indium, particular ITO sputtering waste, have been investigated, mainly based on the hydrometallurgy [3,4], which involves leaching, concentrate and separation, and electrorefining. Among them concentrate and separation involve precipitation, cementation, and solvent extraction. In general, the recovery rate of indium is high through hydrometallurgy. However, the using of various solvents, including corrosive acid and hazardous extraction substances, increases the potential environmental risks. The recycling of indium scrap has been investigated from ITO through pyrometallurgy, but few from end products. In2 O3 can be reduced to metallic indium under high temperature condition with reducing atmosphere [5], such as H2 , C. Alloy of indium and tin can be recovered. Pure metallic indium could be separated form alloy through vacuum. Satoshi and Katsuya [6] conducted
186
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
laboratory scale experiments for the recycling of metallic indium from ITO scrap. The process consisted of two stages, reduction of ITO with CO at 1073 K and vaporization of In–Sn alloy at 1373 K under vacuum. Metallic indium drops could be recovered on condensing zone. This method requires high temperature for the separation of indium from In-Sn alloy. Besides the use of CO is a potential safety risk as its asphyxiating. In addition to leaching-extraction and pyrometallurgy, some researchers have reported the chlorination process of waste LCD panel for the recycling of indium. Takahashi et al. [7] presented a method of recovering indium from waste LCD panel by chlorideinduced vaporization. In this process, crushed LCD panel was first treated in HCl solution for altering the indium(III) oxide into a chloride-induced indium compound. Then the chloride-induced indium compound was vaporized at relatively low temperature (673 K) in nitrogen atmosphere. The vaporization rate of indium could reach 84.3 wt%. In addition, chlorinated separation of indium from ITO scrap or LCD panel was also investigated by other researchers [8,9]. Comparing with pyrometallurgy, chlorinated separation has relative lower temperature. However, it needs a lot of additive and subsequent purification procedure for metallic indium. These two methods are still in laboratory stage. Therefore, in order to directly recover metallic indium from waste LCD panel, environmentally friendly methods and high efficiently methods for achieving high purity metallic indium are desired. Vacuum metallurgy has been widely used in non-ferrous metal smelting [10]. In addition, this method is also reported to separate heavy metal lead from funnel glass [11]. However, there is no study was reported on how this technique is used to recover indium from waste LCD panel. In this paper, we study the feasibility of recycling of metallic indium from In2 O3 through vacuum carbon-reduction at relative low temperature. Based on the feasibility experiments, the method of vacuum carbon-reduction is applied to recycle indium from discard LCD panel in experimental scale. Here coke power is chosen as reducing agent due to its safety, low price and high calorific value.
2. Materials and methods 2.1. Materials LCD panels were firstly dismantled from discarded computers by hand, and then polymer films (mainly contain (polyvinyl alcohol) PVC and (Triacetyl Cellulose) TAC) were also removed through manual from these panels. Liquid crystal between two glasses was dissolved by acetone. Take TFT-LCD for example, the main structure of LCD panel is shown in Fig. 1. Last remained glass panels were crushed into power and sieved to smaller than 0.3 mm. Coke powder (0.8 mm, carbon content >80%) were prepared as reductant. Besides In2 O3 (99.99 mass% purity) were also prepared in this experiment. 2.2. Apparatus Experiments were carried out in a self-assemble vacuum furnace. The schematic diagram of experiment system is shown in Fig. 2. The main body consisted of tubular electric resistance furnace (chamber dimensions is Ø40 mm × 600 mm), quartz tube reactor (Ø35 mm × 900 mm) and vacuum pump (power is 0.25 kW). Middle part of furnace is heating zone and in the outside of it temperature is gradually decrease with the distance from heating zone, and the end of furnace is condensing zone. 2.3. Methods Sample and carbon powder were blend well with a certain mass ratio in a quartz boat, and then the quartz boat with mixture was put into a quartz tube. The quartz tube was pushed into furnace heating area, and heated up to experimental temperature in nitrogen atmosphere. After system reached set temperature, it was evacuated to 1 Pa by vacuum pump. Temperatures examined were 1073, 1123, 1173, 1223 and 1273 K. After some definite time, the sample was cooled to room temperature under vacuum atmosphere and
Fig. 1. Structure of TFT-LCD panel.
Fig. 2. Schematic diagram of the experimental setup.
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
187
Fig. 3. Relations between recovery rate of indium and (a) temperature, (b) carbon adding amount and (c) heating time for pure In2 O3.
identified before and after experiment. The difference quality of indium before and after experiment was considered as the result of reduction with carbon and vaporization of indium. Recovery rate of indium was calculated by the following formula: R=
M0 − M1 × 100% M0
(1)
where M0 and M1 are the initial amount and the remaining amount of indium, respectively. Recovery rate of indium is as the function of temperature, carbon adding amount, and heating time. 2.4. Analysis The quantitative analysis of indium in residues was measured by inductively coupled plasma optical emission spectrometer (ICPOES, icap6300, US), after being completely dissolved with a 1:1 mixture (v/v) of 37 wt% HCl and 68 wt% HNO3 . The crystalline phase of products was identified by X-ray polycrystalline diffractometer (XRD, D8 ADVANCE, Germany) using Cu K␣ radiation. 3. Results and discussion 3.1. Vacuum carbon-reduction for In2 O3 Pure In2 O3 was firstly studied considering that In2 O3 is the target oxide and the quality of it in LCD panel is tiny. The effects of temperature, addition of carbon powder and heating time on
recovery rate of indium were studied. After the best conditions were confirmed, the products produced under optimal condition were analyzed. 3.1.1. Effects of temperature and vacuum on reduction of In2 O3 with carbon In the process of vacuum carbon-reduction, In2 O3 is considered to be firstly reduced to metallic indium, and then indium evaporated into gaseous phase and condensed on low temperature zone. Therefore, temperature and vacuum degree are key factors for reduction of In2 O3 and evaporation of indium. Fig. 3a shows the relation between recovery rate of indium and temperature. The recovery rate of indium increased sharply with the increase of temperature at below 1173 K (from 30 wt% to 98 wt%) and a plateau value (100 wt%) is reached at above 1173 K. The reaction between In2 O3 and C, and the Gibbs free energy (G) at 700 K are listed below: In2 O3 + 3C = 2In(s) + 3CO(g) G700 K = 0
(2)
Relative thermodynamic data were obtained through thermodynamic handbook [12]. The redox reaction between In2 O3 and C takes place when G is 0. At 1 Pa condition, In2 O3 can be reduced to In by carbon powder when the temperature is higher than 700 K according to the equation (2). Therefore, In2 O3 is considered to be reduced to metallic indium since the examined temperatures in the experiments are higher than 700 K (from 1073 K to 1273 K). However, lower recovery rate of indium was obtained
188
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
3.1.3. Effect of heating time The relationship between recovery rate of indium and time is shown in Fig. 3c, showing that the recovery rate of indium can reach almostly 100 wt% at heating time of 10 min and reach 100 wt% at 30 min. Therefore, heatng time for indium recycling is considered as 30 min. In general, the recovery rate of indium is increased with the increasing of temperature, addition of carbon powder and heating time under certain condition. The above experiments show that the recovery rate of indium can reach 100 wt% at the conditions of pressure = 1 Pa, temperature = 1223 K, addition of carbon power = 50 wt% and heating time = 30 min. Therefore, these conditions were considered as the optimal reaction conditions for the recycling of metallic indium from pure In2 O3 powder.
Fig. 4. Relation between vapor pressure of indium and temperature.
within 1073–1173 K. Therefore, the vapor pressure of indium may be the key limiting factor for indium recovery at below 1173 K. Fig. 4 shows the relation between vapor pressure of indium and temperature (date source [13]), indicating saturate vapor pressure of indium is about 1 Pa while the temperature is about 1223 K. Therefore, reduced metallic indium could immediately evaporate and be recovered at above 1223 K under 1 Pa. However, it is merely about 0.06 Pa at 1173 K. This contributes the low recovery rate of indium at below 1173 K. Based on this, following reduction experiment for recycling indium was conducted at 1223 K and 1 Pa. 3.1.2. Effect of carbon adding amount The relationship between recovery rate of indium and addition of carbon powder is shown in Fig. 3b. It shows that indium recovery increased dramaticly with the increase of addition of carbon powder under 50 wt%, and a plate is reached in later. When amount of carbon additon is 30 wt%, the recovery rate of indium is only 65 wt%. This indicates that indium could not sufficiently reduced with blow 30 wt% carbon additon. The possible reactions between In2 O3 and C are listed below when carbon is not enough for reducing In2 O3 to In: In2 O3 (s) + C(s) = 2InO(s) + CO(g) G1223 K = −79 kJ/mol
(3)
In2 O3 (s) + 2C(s) = In2 O(g) + 2CO(g) G1223 K = −333 kJ/mol (4)
As we can see that InO and In2 O can be generated at 1223 K under 1 Pa. Therefore, the recovery rate of indium is low when C is not enough since indium exists in InO solid form. In addition, the evaporation of suboxide In2 O may cause the lose of indium. But there is no In2 O be found in product (this will be discussed in the following section). Recovery rate of indium can reach 50 wt% without the addition of carbon powder. This possibly results from the evaporation of suboxide In2 O and In2 O3 . In2 O3 may breake into evaporable In2 O and O2 , and the boiling point of In2 O3 is about 1123 K. In high vacuum conditon, carbon is condsered as the main reductant, therefore, the increased of carbon powder adding contributed to the increasing of indium recovery rate. The recovry rate of indium reaches 100 wt% at 50 wt% carbon powder amount. This percentage was considered to be the optimum carbon adding amount.
3.1.4. Analysis of condensed products for pure In2 O3 Products produced under optimal conditon were collected and analyzed by XRD. The distribution of condensed product and temperature is shown in Fig. 5a. It shows that metallic indium was condensed on the inner-wall of the quartz tube (Fig. 5b) with the temperature range of 833–573 K. In addition, the XRF shows that the indium has a 100% purity. In B zone with the temperature range of 433–573 K, XRD pattern (Fig. 5c) demonstrates that the major constitue of products is In2 O3 with the remainder is metallic In. This could be the oxidation of In2 O produced from the decomposition and evaporation of In2 O3 . Besides, a part of In2 O may react with reducing agent such as CO, that contributed to the metallic indium in B zone. In2 O3 in B zone accounts for less than five percent of total quality of products and it is recoverable. Therefore, the loss of indium caused by vaporation and decomposition is negligibal. The vapor of indium reduced from In2 O3 at 1223 K and 1 Pa cooled to indium solid with airflow and adhered to the inner wall of quartz tube. The temperature of quartz tube gradually decreased from the heating zone to the end of tube as shown in Fig. 6. It shows that the temperature dropped from 1223 K at heating zone to 400 K at 428 mm distance from the center of heating zone. Therefore, we can learn that the condensing zone of metallic indium is 350–388 mm off the center of furnace and temperature of condensing zone ranges from 573 K to 833 K according to Figs. 5(a) and 6. 3.2. Vacuum carbon-reduction for LCD glass powder The experiments for pure In2 O3 have confirmed that the recovery of metallic indium from In2 O3 is feasible, and optimal conditions of 1223 K, 50 wt% carbon adding amount, 30 min heating time and 1 Pa, were also determined for pure In2 O3 . However, in practice, In2 O3 content is only about 0.2 g/kg (LCD glass) and the major ingredient of LCD glass is SiO2 . These elements lead to the differences between pure In2 O3 and LCD glass for recycling indium. In addition, SnO2 should be considered since it could be reduced to impurity Sn. Based on above elements, carbon adding amount was considered for the investigation of indium recycling from waste LCD panel. 3.2.1. Effect of carbon adding amount Fig. 7 shows the relation between recovery rate of indium and carbon adding amount for LCD powder. As we can see that the recovery rate of indium is increased with the increasing of carbon powder amount at blow 30 wt% addition of carbon powder. It reaches 90 wt% when the addition of carbon powder is 30 wt% and it does not continue to increase with the increasing of carbon powder amount. This indicates that the consumption of carbon powder is very large for the indium recycling from LCD powder. Two reasons could lead to the large consumption of carbon powder. One reason
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
189
Fig. 5. (a) Distribution of condensed product and temperature in one end of quartz tube for pure In2 O3 . (b) Photo for metallic indium inside A zone. (c) X-ray diffraction patterns of In2 O3 and In in B zone.
is a large number of glass powders from glass substrate hinder the immediate contact of In2 O3 and carbon powder. Another reason is impurity substance react with carbon powder lead to the increase of carbon consumption. Impurity SnO2 is discussed here since ITO is composed of SnO2 and In2 O3.
3.2.2. Effect of impurity SnO2 The possible reaction between SnO2 and C, and the corresponding Gibbs energy at 1 Pa are listed below: SnO2 (s) + 2 C(s) = Sn(s) + 2CO(g) G1223 K = −4 × 105 kJ/mol (5)
Fig. 6. The distribution of temperature of the quartz tube at 1223 K.
Fig. 7. Relation between recovery rate of indium and carbon adding amount for LCD powder.
190
Y. He et al. / Journal of Hazardous Materials 268 (2014) 185–190
carbon-reduction. The best parameters were confirmed as 1223 K and 1 Pa with 30 wt% carbon addition for 30 min, and the recovery rate of indium from LCD powder could reach to 90 wt%. In addition, there is no hazardous materials produced in this process. Acknowledgements
Fig. 8. Distribution of condensed product and temperature for LCD powder in one end of quartz tube.
This work was supported by the National Natural Science Foundation of China (51178262, 51278293), and the Scientific Research Projects of Science and Technology Commission of Shanghai Municipality (11DZ2282800). References
The Gibbs energy is less than 0 at 1223 K. Therefore, Sn can be generated at 1223 K. Results show that content of SnO2 in LCD panel is decreased from 196 ppm to 110 ppm and the recovery rate of tin is only about 15 wt%. The vapour pressure of Sn (is only 0.002 Pa) is much lower than that of In (is about 1 Pa) at 1223 K. Therefore, the recovery rate of Sn is low due to its low saturate vapour pressure. This also significantly reduces the effect of Sn on In product. 3.2.3. Analysis of condensed products for LCD powder 250 g LCD powder were investigated under the conditions of 1223 K, 1 Pa, 30 min, and 30 wt% addition of carbon. The distribution of condensing products and temperature are shown in Fig. 8. There is a rainbow circle in the quartz tube and the inside layer of rainbow nearing 833 K is silver-white. What’s more, the condensing temperature of these substances agrees with that of metallic indium in experiment for pure In2 O3 (Fig. 5a). Therefore, it is assumed that indium can be reduced and separated from LCD powder. 4. Conclusions This study investigated the recovery of indium from waste LCD panel using vacuum carbon-reduction. An environmentally friendly vacuum carbon-reduction method for recycling of indium from waste LCD panel has been developed. The results indicated that the indium can be reclaimed from LCD panel by vacuum
[1] Display Search, Global LCD TV shipments fall for the first time in 2012; outlook cautious for 2013, http://www.displaysearch.com/cps/rde/xchg/ displaysearch/hs.xsl/130321 global lcd tv shipments fall for the first time in 2012.asp (last accessed: 02.09.2013). [2] J. Yang, T. Retegan, C. Ekberg, Indium recovery from discarded LCD panel glass by solvent extraction, Hydrometallurgy 137 (2013) 68–77. [3] Y. Li, Z. Liu, Q. Li, Z. Liu, L. Zeng, Recovery of indium from used indium–tin oxide (ITO) targets, Hydrometallurgy 105 (2011) 207–212. [4] S. Virolainen, D. Ibana, E. Paatero, Recovery of indium from indium tin oxide by solvent extraction, Hydrometallurgy 107 (2011) 56–61. [5] J. Chen, J. Yao, Y. Zhou, Z. Chen, X. Wang, J. Huang, Recovery indium from waste ITO target, Chin. J. Rare Metal 27 (2003) 101–103. [6] I. Satoshi, M. Katsuya, Recoveries of metallic indium and tin from ITO by means of prometallurgy, High Temp. Mater. Proc. 30 (2011) 317–322. [7] K. Takahashi, A. Sasaki, G. Dodbiba, J. Sadaki, N. Sato, T. Fujita, Recovering indium from the liquid crystal display of discarded cellular phones by means of chloride-induced vaporization at relatively low temperature, Metall. Mater. Trans. A 40 (2009) 891–900. [8] O. Terakado, D. Iwaki, K. Murayama, M. Hirasawa, Indium recovery from indium tin oxide, ITO, thin film deposited on glass plate by chlorination treatment with ammonium chloride, Mater. Trans. 52 (2011) 1655–1660. [9] E. Ma, Z. Xu, Technological process and optimum design of organic materials vacuum pyrolysis and indium chlorinated separation from waste liquid crystal display panels, J. Hazard. Mater. 263 (2013) 610–617. [10] Y. Dai, B. Yang, Vacuum Metallurgy of Nonferrous Metals, 2nd ed., Metallurgical Industry Press, Beijing, 2009. [11] M. Chen, F. Zhang, J. Zhu, Lead recovery and the feasibility of foam glass production from funnel glass of dismantled cathode ray tube through pyrovacuum process, J. Hazard. Mater. 161 (2009) 1109–1113. [12] D. Ye, Practical Inorganic Thermodynamic Date Manual, Metallurgical Industry Press, Beijing, 1981. [13] C. Xu, Vacuum Engineering Technology, Chemical Industry Press, Beijing, 2006.