The Effect of the Redox Potential of Aqua Regia and Temperature on the Au, Cu, and Fe Dissolution from Wpcbs †

recycling

Article The Effect of the Redox Potential of Aqua Regia and Temperature on the Au, Cu, and Fe Dissolution from WPCBs †

Heini Elomaa ID , Sipi Seisko, Tero Junnila, Tuomas Sirviö, Benjamin P. Wilson, Jari Aromaa ID and Mari Lundström *

Laboratory of Hydrometallurgy and Corrosion, Department of Chemical and Metallurgical Engineering, School of Chemical Engineering, Aalto University, P.O. Box 16200, FI-00076 Espoo, Finland; heini.elomaa@aalto.fi (H.E.); sipi.seisko@aalto.fi (S.S.); tero.junnila@aalto.fi (T.J.); tuomas.sirvio@aalto.fi (T.S.); ben.wilson@aalto.fi (B.P.W.); jari.aromaa@aalto.fi (J.A.) * Correspondence: mari.lundstrom@aalto.fi † This paper is an extended version of our paper published in 6th International Conference Quo Vadis Recycling, High Tatras, Slovak Republic, 6–9 June 2017.

Received: 22 August 2017; Accepted: 31 August 2017; Published: 1 September 2017

Abstract: Constant growth in waste electrical and electronic equipment (WEEE) levels necessitates the development of new, commercially viable recycling processes. Waste printed circuit boards (WPCBs) are a sub-group of WEEE that are of increasing interest due to their relatively high level of valuable metal content including Au, Ag, and platinum group metals (PGMs). Currently, precious metals like gold are mainly recycled from WEEE streams through copper smelting/refining; however, the possibility to peel gold from WPCBs prior to smelting, could offer advantages for recycling. In this study, the suitability of aqua regia for selective or partially selective gold leaching from un-crushed WPCBs was investigated. The redox potential of aqua regia solutions and the dissolution efficiencies of Au, Cu, and Fe from WPCBs were investigated at different temperatures (40–80 ◦C) and concentrations (2–32%) in batch leaching tests. The redox potential of aqua regia solution was found to depend on concentration and temperature. It is suggested that Au dissolution in aqua regia requires dissolved Cu2+ ions originating from the WPCB material to work. Au extraction (>50%) was shown to require a redox potential >700 mV with [Cu2+] > 2500 ppm, as a potential >850 mV alone was insufficient without cupric ions. Significant amounts of Au and Cu could be dissolved with only minor Fe dissolution at ≥8% aqua regia at 80 ◦C. Results suggest that leaching of uncrushed WPCBs in 8% aqua regia (T = 80 ◦C) can provide the opportunity for partial Au recovery prior to further processing.

Keywords: waste printed circuit boards; gold leaching; copper leaching

1. Introduction In the past decades, the amount of waste electric and electronic equipment (WEEE) has been constantly growing, which influences the development of recycling processes. WEEE can be divided in many categories, which are defined in the EU/2012/19 [1] directive. The most valuable sub-group within WEEE are waste printed circuit boards (WPCBs) that contain significant amounts of valuable metals such as Au, Ag, and PGMs, increasing the interest towards the recycling of WPCBs and the recovery of valuable metals. The difficulty of recycling WEEE lies in the heterogeneity of the materials. Aqua regia is the traditional medium for dissolving gold and platinum group metals in acid digestion, which is a common method of analyzing valuable metals [2]. In acid digestion, precious metals, such as gold,

Recycling 2017, 2, 14; doi:10.3390/recycling2030014 www.mdpi.com/journal/recycling Recycling 2017, 2, 14 2 of 9 are ﬁrst leached in boiling aqua regia and after ﬁltration analyzed by atomic absorption spectroscopy (AAS) or inductively coupled plasma mass spectroscopy (ICP). The drawback of pure aqua regia, however, is the aggressive and corrosive nature of the concentrated lixiviant. Veit et al. [3] concluded that aqua regia is usually used on the laboratory scale, but not industrially widely applied as hydrometallurgical process media. Additionally, Yannopoulos [4] stated that the dissolution of gold in aqua regia is used in analytical chemistry for either volumetric or gravimetric determinations of the soluble gold. Hydrochloric acid in the presence of oxidants such as nitric acid, oxygen, cupric or ferric ions and manganese dioxide can dissolve gold [4]. Aqua regia is the combination of three parts concentrated hydrochloric to one part concentrated nitric acid and is well known to vigorously attack gold. The reactions for gold leaching in aqua regia can be described as in Equations (1)–(3) [4,5]:

2HNO3 + 6HCl → 2NO + 4H2O + 3Cl2 (1)

2Au + 11HCl + 3HNO3 → 2HAuCl4 + 3NOCl + 6H2O (2)

Au + 4HCl + HNO3 → H[AuCl4] + 2H2O + NO. (3) Concentrated aqua regia has been proven to be efficient lixiviate for the simultaneous leaching of gold, silver, and palladium [6]. However, research by Zhang and Xu [7] suggested that aqua regia leaching of WEEE is not selective and can aggressively digest both base and precious metals concurrently. According to Lekka et al. [8], aqua regia was found to be an efficient leaching solution for metals from PCB powders. In addition, Sheng and Etsell [9] investigated the leaching of gold from computer circuit boards using aqua regia after pre-leaching the base metals by nitric acid. Furthermore, Park et al. investigated the dissolution of gold, silver, and palladium from PCBs, and they suggested a method for recovering precious metals by aqua regia leaching. For gold, a liquid–liquid extraction with toluene and recovery as nanoparticles using dodecanethiol and sodium borohydrate was suggested. Park et al. stated that aqua regia is efficient leachant for PCBs as gold, silver, and palladium can be separated at the same time [6]. This study aims at investigating diluted aqua regia as a potential media for selective or partially selective leaching of visible gold from PCBs. Simultaneously the dissolution of copper and iron is investigated.

2. Experimental

2.1. Material In this study, the dissolution of gold available for leaching on WPCBs was investigated in aqua regia solutions batch leaching tests. Additionally, the redox potential of aqua regia solutions was studied to ascertain any correlation between the potential and gold leaching phenomenon. The challenge in non-crushed WPCB leaching is the non-homogeneous nature of the raw material. In this study, WPCBs originating from industrial telecommunication center PCBs were used. This type of raw material provides relatively homogeneous nature as one side of the WPCBs is entirely gold coated, Figure1a). Larger electronic components such as capacitors and transistors were removed from the surface of the WPCB, thus standardizing the raw material. For the batch leaching tests, the raw materials was not crushed, but cut to ca. 5 cm × 5 cm squares. This in order investigate whether gold can be dissolved selectively or semi-selectively from the WPCBs, simultaneously minimizing base metal dissolution from the WPCB structure. Total gold and silver content in the raw material was analyzed using a Pb-ﬁre assay (Labtium, Outokumpu, Finland), whereas other metal contents were determined by the total digestion. The chemical analysis of the gold in raw material (Parallel Samples 1 and 2) is presented in Table1. Two samples were analyzed by the ﬁre-assay method, and the results showed remarkable variation in precious metals, even with two parallel GFAAS analysis, due to challenges in the analysis of this kind of secondary raw material. In addition, the crushing conducted prior to Recycling 2017, 2, 14 3 of 9 analysis is suggested to decrease the sample precious metal content, as the precious metals are known Recycling 2017, 7, 14 3 of 9 to enrich into dust fraction, remaining partially in the crusher [10]. Table 1. Parallel chemical analysis of waste printed circuit board (WPCB) raw material (Samples 1 Table 1. Parallel chemical analysis of waste printed circuit board (WPCB) raw material and 2) used in leaching experiments, element amount presented as mg/kg of WPCB. For Sample 2, (Samples 1 and 2) used in leaching experiments, element amount presented as mg/kg of WPCB. only Au and Ag were analyzed (n.a. = not analyzed). For Sample 2, only Au and Ag were analyzed (n.a. = not analyzed). Au Ag Cu Fe Sn Al Ni Pb Zn Sample 1 372Au 570 Ag250,000 Cu 37,300 Fe 11,670 Sn 49,800 Al 8830 Ni 30,000 Pb 14,600 Zn SampleSample 2 1 421372 245 570 n.a.250,000 n.a.37,300 n.a. 11,670 49,800n.a. n.a. 8830 30,000n.a. 14,600 n.a. Sample 2 421 245 n.a. n.a. n.a. n.a. n.a. n.a. n.a.

(a) (b)

Figure 1. The WPCB raw material investigated: ( (aa)) the the gold-coated side; side; ( (bb)) the side where larger electronic components have been removed.

2.2. Methods The gold leaching experiments were carried out in aqua regia solutions with varying concentrations, thethe molarmolar ratioratio ofof thethe nitric nitric to to hydrochloric hydrochloric acid acid kept kept constantly constantly at at 1:3 1:3 with with dilution dilution to distilledto distilled water. water. Table Table2 outlines 2 outlines the concentration the concentr of theation aqua of regia the solutionsaqua regia investigated. solutions Additionally,investigated. theAdditionally, amounts ofthe free amounts chlorides of free in solution—calculated chlorides in solution—calculated based on the based hydrochloric on the hydrochloric acid molarity—is acid alsomolarity—is shown. also shown.

Table 2. AquaAqua regia regia concentrations concentrations and and corresponding corresponding chloride chloride concentration concentration in the in redox the redoxtest series test series(R1–R1) (R1–R1) and in and the inbatch the batchleaching leaching test series test series used usedin WPCB in WPCB leaching leaching (L1–L5). (L1–L5). In addition, In addition, the thetemperature temperature range range investigated investigated in the in thetest testseries series R1–R10 R1–R10 and and L1–L5 L1–L5 is presented. is presented.

Aqua Regia Batch Leaching of WPCB, − Aqua Regia ConcentrationRedox Redox Test Test Series Series Batch Leaching of WPCB, Test Series ChlorideChloride [Cl ], M[Cl−], M Concentration0.01% R1 (25–80 ◦C)Test Series 0.003 0.1% R2 (25–80 ◦C) 0.026 0.01% 0.5% R1 (25–80 R3 (25–80 °C)◦C) 0.129 0.003 1% R4 (25–80 ◦C) 0.257 0.1% 2% R2 (25–80 R5 (25–80 °C)◦C) L1 (40–80 ◦C) 0.515 0.026 0.5% 4% R3 (25–80 R6 (25–80 °C)◦C) L2 (40–80 ◦C) 1.030 0.129 5% R7 (25–80 ◦C) 1.286 1% 8% R4 (25–80 R8 (25–80 °C)◦C) L3 (40–80 ◦C) 2.058 0.257 16% R9 (25–80 ◦C) L4 (40–80 ◦C) 4.116 2% 32% R5 (25–80 R10 (25–80 °C)◦C) L1 L5 (40–80(40–80◦C) °C) 8.233 0.515 4% R6 (25–80 °C) L2 (40–80 °C) 1.030 The redox5% potential of aqua R7 regia(25–80 solutions °C) (R1–R10) was studied at eight different 1.286 temperatures: 25, 30, 35, 40,8% 50, 60, 70, and R8 80 (25–80◦C. The °C) solution redox L3 (40–80 potential °C) was measured 2.058 by immersing a reference electrode16% (vs. Ag/AgCl, R9 (25–80 Red °C) Rod ref 201) L4 in (40–80 solution °C) with a platinum 4.116 wire. A Lauda Immersion Thermostat32% A100 R10 provided (25–80 °C) temperature control. L5 (40–80 For °C) the batch leaching 8.233 experiments (L1–L5), WPCBs were cut into ca. 5 × 5 cm pieces, and the experiments were conducted at 40 ◦C and 80The◦C redoxin a 1potential dm3 glass of aqua reactor, regia heated solutions by Aqaline(R1–R10) AL was 25 studied water bathat eight with different thermostatic temperatures: control. The25, 30, rate 35, of 40, oxygen 50, 60, sparging 70, and (180 NL/min)°C. The soluti andon rotational redox potential speed (100 was rpm) measured were keptby immersing constant ina everyreference experiment. electrode Leaching(vs. Ag/AgCl, time wasRed 2Rod h. Solution ref 201) samplingin solution was with conducted a platinum with wire. a glass A pipetteLauda atImmersion time intervals Thermostat of 1, 3, A100 5, 10, provided 20, 30, 60, temperature and 120 min control. for subsequent For the batch analysis leaching by atomic experiments absorption (L1– L5), WPCBs were cut into ca. 5 × 5 cm pieces, and the experiments were conducted at 40 °C and 80 °C in a 1 dm3 glass reactor, heated by Aqaline AL 25 water bath with thermostatic control. The rate of oxygen sparging (1 NL/min) and rotational speed (100 rpm) were kept constant in every experiment. Leaching time was 2 h. Solution sampling was conducted with a glass pipette at time Recycling 2017, 7, 14 4 of 9

intervals of 1, 3, 5, 10, 20, 30, 60, and 120 min for subsequent analysis by atomic absorption Recyclingspectroscopy2017, 2, 14 (AAS, Thermo Scientific iCE 3000 Series, Waltham, MA, USA). Solutions samples4 ofwere 9 filtered, and 2 drops of 50% nitric acid were added to the samples in order to prevent sedimentation. spectroscopySolution (AAS, additions Thermo were Scientiﬁc performe iCEd 3000in order Series, to Waltham,maintain the MA, liquid/solid USA). Solutions ratio and samples mitigate were the ﬁltered,effects and of evaporation. 2 drops of 50% The nitric evaporation acid were was added determined to the samples during in ordera 2 h toleaching prevent test sedimentation. in all studied temperatures,Solution additions and the were evaporating performed volume in order of liquid to maintain addedthe at 10 liquid/solid min intervals. ratio All and redox mitigate potentials the effectswere ofmeasured evaporation. using The Ag/AgCl evaporation in saturated was determined KCl as reference during a electrode 2 h leaching (Red test Rod in allref studied201) and temperatures,platinum as andthe counter the evaporating electrode. volume of liquid added at 10 min intervals. All redox potentials were measured using Ag/AgCl in saturated KCl as reference electrode (Red Rod ref 201) and platinum as3. the Results counter and electrode. Discussion

3.3.1. Results Redox and Potential Discussion Figure 2 shows the measured redox potentials in investigated solutions (R1–R10) at 3.1. Redox Potential temperatures between 25 and 80 °C. The results show that the redox potential increases both with the increaseFigure 2in shows aqua theregia measured concentration redox potentialsand with inincreases investigated in temperature. solutions (R1–R10) The redox at temperatures potential was betweenshown 25to andrange 80 from◦C. The540 resultsto 900 mV show vs. that Ag/AgCl the redox and potential obeyed the increases Nernstboth equation—i.e., with the increase the well- inknown aqua regia relationship concentration between and reduction with increases potential, in temperature. temperature, The and redox concentration potential was(activity) shown of to the rangeelectroactive from 540 species—except to 900 mV vs. at Ag/AgCl the most anddilute obeyed point, 0.01% the Nernst aqua regia, equation—i.e., as demonstrated the well-known by the data relationshippresented betweenin Figure reduction 2. potential, temperature, and concentration (activity) of the electroactive species—except at the most dilute point, 0.01% aqua regia, as demonstrated by the data presented in Figure2.

° 900 25 C 30 °C 35 °C 850 40 °C 50 °C 800 60 °C 70 °C 750 80 °C

700

650

600

REDOX POTENTIAL (mV vs. Ag/AgCl) vs. (mV REDOX POTENTIAL 550

0 5 10 15 20 25 30 35 AQUA REGIA CONCENTRATION (%)

Figure 2. The redox potential of aqua regia solutions at temperatures of 25, 30, 35, 40, 50, 60, 70, andFigure 80 ◦C 2. as The a function redox potential of aqua regia of aqua concentration. regia solutions at temperatures of 25, 30, 35, 40, 50, 60, 70, and 80 °C as a function of aqua regia concentration. 3.2. Batch Leaching 3.2. Batch Leaching During the batch leaching tests, the redox potential was measured at each solution sampling time toDuring observe the the batch oxidative leaching power tests, of thethe solutionredox pote as antial function was measured of WPCB leachingat each solution time (Figure sampling3). Thetime redox to observe potential the of oxidative the aqua regiapower leaching of the solution solution as varied a function in the of range WPCB 300–900 leaching mV vs.time Ag/AgCl (Figure 3). afterThe the redox raw materialpotential exposure. of the aqua It can regia be leaching seen that solution the redox varied potential in the of therange leaching 300–900 solution mV vs. generally Ag/AgCl declinedafter the quickly raw material after the exposure. samples wereIt can immersed be seen intothat thethe leachingredox potential solution of (aqua the leaching regia ≤ 16% solution at T =generally 40 ◦C and declined aqua regia quickly≤ 4% after at T the = 80 samples◦C). This were indicates immersed that the into initial the leaching leaching reactionssolution (aqua consumed regia ≤ the16% oxidative at T = species40 °C and present aqua in regia the solution.≤ 4% at T As = 80 the °C). dissolution This indicates progressed that the the initial metal concentrationleaching reactions in theconsumed solution increased. the oxidative The appliedspecies airpresent purging in oxidizedthe solution. the dissolved As the dissolution metallic species progressed in the solutionthe metal resultingconcentration in redox in potential the solution increase increased. with time. The The applied opposite air phenomenon,purging oxidized redox the potential dissolved remaining metallic inspecies the same in levelthe solution or slightly resulting increasing in redox from potential the initial increase aqua regiawith potentialtime. The wasopposite typical phenomenon, at higher concentrationsredox potential and remaining temperatures in the (aqua same regia level 32% or slightly at T = 40increasing◦C and aqua from regia the initial≥ 8% aqua at T =regia 80 ◦ potentialC). RecyclingRecycling 2017 2017, ,7 7, ,14 14 55 of of 9 9 waswas typical typical at at higher higher concentrations concentrations and and temperatures temperatures (aqua (aqua regia regia 32% 32% at at T T = = 40 40 °C °C and and aqua aqua regia regia ≥ ≥ Recycling 2017, 2, 14 5 of 9 8%8% at at T T = = 80 80 °C). °C).

((aa)) ((bb))

FigureFigure 3. 3. AquaAqua Aqua regia regiaregia redoxredox redox potential potential prior prior to to raw raw material material exposure exposure (t (t = = 0) 0) and and duringduring during the the WPCB WPCB batch batch leachingleaching tests tests (L1–L5) (L1–L5) at at temperatures temperatures (( a(a)) ) 4040 40 ◦°C °CC and and ( b(b) ) 80 80 ◦°C. °C.C.

Figure 4 presents the PCB pieces after leaching experiments in 8% aqua regia solution at FigureFigure4 4presents presents the the PCB PCB pieces pieces after after leaching leaching experiments experiments in in 8% 8% aqua aqua regia regia solution solution at at temperatures of 40 °C (a) and 80 °C (b), respectively. As can be observed, after leaching at 40 °C temperaturestemperatures ofof 4040◦ °CC (a) (a) and and 80 80◦ C°C(b), (b),respectively. respectively. AsAs cancan bebe observed,observed, afterafter leachingleaching atat 4040◦ °CC (Figure 4a) gold is still visible on the PCB surface. After leaching in 8% aqua regia solution at 80 °C (Figure(Figure4 a)4a) gold gold is is still still visible visible on on the the PCB PCB surface. surface. After After leaching leaching in in 8% 8% aqua aqua regia regia solution solution at at 80 80◦ °CC (Figure 4b), leaching of visible Au from the WPCB material was evident and no similar shiny gold (Figure(Figure4 b),4b), leaching leaching of of visible visible Au Au from from the the WPCB WPCB material material was was evident evident and and no no similar similar shiny shiny gold gold could be seen. couldcould bebe seen.seen.

((aa) ) ((bb)) Figure 4. The WPCB pieces after leaching experiment L3 in 8% aqua regia at temperatures (a) 40 and FigureFigure 4. 4.The The WPCBWPCB piecespieces afterafter leachingleaching experimentexperiment L3L3 inin 8%8% aquaaqua regiaregia atat temperaturestemperatures (a(a)) 40 40 and and (b) 80 °C. (b(b)) 80 80◦ °C.C.

Two-hourTwo-hour batch batch leaching leaching experiments experiments (L1–L5) (L1–L5) were were conducted conducted with with aqua aqua regia regia leaching leaching solution solution Two-hour batch leaching experiments (L1–L5) were conducted with aqua regia leaching solution concentrationsconcentrations 2–32% 2–32% for for the the WPCB WPCB material. material. The The diss dissolutionolution of of Cu Cu and and Au Au is is presented presented in in Figure Figure 5 5 concentrations 2–32% for the WPCB material. The dissolution of Cu and Au is presented in Figures5 andand Figure Figure 6 6 as as normalized normalized to to the the dissolution dissolution of of these these elements elements in in 32% 32% aqua aqua regia regia at at 80 80 °C °C (L5, (L5, the the and6 as normalized to the dissolution of these elements in 32% aqua regia at 80 ◦C (L5, the most mostmost aggressiveaggressive leachingleaching media).media). ThisThis inin orderorder toto definedefine thethe metalmetal extractionextraction efficiencyefficiency intointo thethe aggressive leaching media). This in order to deﬁne the metal extraction efﬁciency into the solutions solutionssolutions compared compared to to maximum maximum extraction extraction into into the the solution. solution. compared to maximum extraction into the solution. FigureFigure 5 5 shows shows that, that, generally, generally, an an increase increase in in te temperaturemperature increases increases Cu Cu extraction extraction in in the the solution. solution. Figure5 shows that, generally, an increase in temperature increases Cu extraction in the solution. AtAt 32% 32% aqua aqua regia regia (T (T = = 80 80 °C), °C), 30 30 min min was was enough enough for for maximum maximum Cu Cu dissolution dissolution (Figure (Figure 5a); 5a); however, however, At 32% aqua regia (T = 80 ◦C), 30 min was enough for maximum Cu dissolution (Figure5a); however, thethe maximum maximum Cu Cu dissolution dissolution could could al alsoso be be also also achieved achieved at at 16% 16% (T (T = = 80 80 °C) °C) after after 120 120 min min of of leaching leaching the maximum Cu dissolution could also be also achieved at 16% (T = 80 ◦C) after 120 min of leaching (Figure(Figure 5c). 5c). The The lower lower temperature temperature (T (T = = 40 40 °C) °C) was was shown shown to to leach leach only only minor minor amount amount of of Cu Cu with with t t≤ ≤ (Figure5c). The lower temperature (T = 40 ◦C) was shown to leach only minor amount of Cu with 6060 min.min. WhenWhen solutionsolution analysisanalysis waswas comparedcompared toto thethe originaloriginal CuCu contentcontent inin thethe WPCBWPCB solidsolid rawraw t ≤ 60 min. When solution analysis was compared to the original Cu content in the WPCB solid raw materialmaterial (Table(Table 1),1), thethe maximummaximum CuCu recoveryrecovery thatthat couldcould bebe achievedachieved inin anyany ofof thethe leachingleaching material (Table1), the maximum Cu recovery that could be achieved in any of the leaching experiments experimentsexperiments waswas 61%.61%. TheThe CuCu recoveryrecovery ofof 61%61% cancan bebe explainedexplained byby thethe factfact thatthat onlyonly partpart ofof thethe was 61%. The Cu recovery of 61% can be explained by the fact that only part of the copper is available coppercopper is is available available for for leaching leaching in in the the non-crushed non-crushed WP WPCBCB material, material, whereas whereas a a big big part part of of Cu Cu is is in in the the for leaching in the non-crushed WPCB material, whereas a big part of Cu is in the laminated WPCB laminatedlaminated WPCB WPCB structure structure not not exposed exposed to to the the leaching leaching solution. solution. structure not exposed to the leaching solution. Recycling 2017, 2, 14 6 of 9 Recycling 2017, 7, 14 6 of 9

t = 30 min t = 60 min ° 100 40 C 100 40 °C ° 80 C 80 °C 80 80

60 60

40 40 Cu EXTRACTION, (%) Cu EXTRACTION, Cu EXTRACTION, (%) 20 20

0 0

0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35 AQUA REGIA CONCENTRATION, (%) AQUA REGIA CONCENTRATION (%) (a) (b) t = 120 min 4000

100 80 ° C (%) 40 ° C (%) 80 3000 80 ° C (mg/l) 40 ° C (mg/l) 60 2000

40 Cu (mg/l)

Cu EXTRACTION, (%) 1000 20

0 0 0 5 10 15 20 25 30 35 AQUA REGIA CONCENTRATION (%) (c)

FigureFigure 5. 5. TheThe extractions extractions of of copper copper at at t t = = 30 30 min min ( (aa);); t t = = 60 60 min min ( (bb);); and and t t = 120 120 min min ( (c).). Extraction Extraction presentedpresented as as % % at at left left yy-axis-axis ( (aa––cc)) and and mg/L mg/L at at right right yy-axis-axis (c (c).).

The determination of the Au recovery was challenging, as the recoveries calculated (solution The determination of the Au recovery was challenging, as the recoveries calculated analysis vs. initial solid analysis in Table 1) gave Au extraction values >100%. This is due to the error (solution analysis vs. initial solid analysis in Table1) gave Au extraction values >100%. This is margins related to the WPCB-type material pre-treatment, during which the sample is crushed and due to the error margins related to the WPCB-type material pre-treatment, during which the sample part of the Au can also be lost to the crusher as gold dust [10]. Additionally, the gold analysis from is crushed and part of the Au can also be lost to the crusher as gold dust [10]. Additionally, the gold this type of secondary raw material is not straightforward. Table 1 shows remarkable variation, even analysis from this type of secondary raw material is not straightforward. Table1 shows remarkable between two identical samples. Therefore, the Au extraction was observed as the relative recovery, variation, even between two identical samples. Therefore, the Au extraction was observed as the i.e., versus the maximum dissolved amount of gold achieved. Figure 6 shows that, generally, an increase relative recovery, i.e., versus the maximum dissolved amount of gold achieved. Figure6 shows that, in temperatures increases Au extraction in the solution. At 32% aqua regia (T = 80 °C), 30 min was generally, an increase in temperatures increases Au extraction in the solution. At 32% aqua regia sufficient for maximum Au dissolution. Equally high Au extraction could not be achieved in any of the (T = 80 ◦C), 30 min was sufﬁcient for maximum Au dissolution. Equally high Au extraction could more dilute aqua regia solutions. Additionally, leaching times ≤60 min (Figure 6a,b) did not provide not be achieved in any of the more dilute aqua regia solutions. Additionally, leaching times ≤60 min maximum Au dissolution in any other conditions than at 32% aqua regia. The lower temperature (T = 40 (Figure6a,b) did not provide maximum Au dissolution in any other conditions than at 32% aqua regia. °C) was shown not to leach Au. However, it can be seen that 8% aqua regia at T = 80 °C was able to The lower temperature (T = 40 ◦C) was shown not to leach Au. However, it can be seen that 8% aqua dissolve over half of the maximum leachable gold at t = 120 min. regia at T = 80 ◦C was able to dissolve over half of the maximum leachable gold at t = 120 min. Recycling 2017, 2, 14 7 of 9 Recycling 2017, 7, 14 7 of 9

t = 30 min t = 60 min

100 40 °C 100 40 °C 80 °C 80 °C 80 80

60 60

40 40 Au EXTRACTION,(%) Au EXTRACTION, (%) EXTRACTION, Au 20 20

0 0

0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35 AQUA REGIA CONCENTRATION, (%) AQUA REGIA CONCENTRATION (%) (a) (b) t = 120 min 14 ° 100 80 C (%) 40 ° C (%) 12 80 ° C (mg/l) 80 40 ° C (mg/l) 10

60 8

6 40 Au (mg/l) Au 4 Au EXTRACTION, (%) EXTRACTION, Au 20 2

0 0 0 5 10 15 20 25 30 35 AQUA REGIA CONCENTRATION (%) (c)

FigureFigure 6.6. The extractionsextractions ofof goldgold atat ((aa)) tt == 30 min; ((b)) tt == 60 min; and (c) t = 120 120 min. min.

When Fe solution analysis was compared to thethe original Fe content in the WPCB solid raw material (Table1 1),), itit waswas observedobserved thatthat onlyonly minorminor FeFe dissolved.dissolved. TheThe maximummaximum FeFe extractionextraction inin anyany of the experiments experiments (L1–L5) (L1–L5) was was low, low, ~3% ~3% (max (max 30.1 30.1 mg/L mg/L in solution in solution in 16% in 16% aqua aqua regia regia at 80 at°C). 80 This◦C). Thissuggests suggests that leaching that leaching of Cu ofand Cu Au and from Au non-crushed from non-crushed WPCBs WPCBs is selective is selective to Fe and to that Fe and Fe present that Fe presentin the WPCB in the is WPCB mostly is in mostly the laminated in the laminated structure, structure, not available not available for leaching. for leaching. When Figure 33 is is comparedcompared toto FiguresFigures5 5and and6, 6, it canit can be be suggested suggested that that a higha high Au Au extraction extraction (>50% of maximum extraction) required aa redoxredox potentialpotential >700>700 mVmV vs.vs. Ag/AgClAg/AgCl (L3–L5, (L3–L5, T = 80 °C)◦C) and that the the maximum maximum Au Au extraction extraction was was achieved achieved at a at redox a redox potential potential >850 >850 mV vs. mV Ag/AgCl vs. Ag/AgCl (L5, T (L5,= 80 T°C). = 80 However,◦C). However, a high aredox high potential redox potential (>850 mV (>850 vs. mVAg/AgCl) vs. Ag/AgCl) alone was alone not enough was not to enough dissolve to dissolvegold (L5, goldT = 40 (L5, °C). TIt =seems 40 ◦C that). It Au seems dissolution that Au was dissolution heavily supported was heavily by dissolved supported Cu by2+ ions, dissolved acting Cuas gold2+ ions, oxidant acting [11], as high gold Au oxidant extractions [11], (> high50%) Au being extractions achieved (>50%) in solutions being with achieved [Cu2+] in> 2500 solutions ppm. withThe results[Cu2+] indicate > 2500 ppm that. 8% The aqua results regia indicate (T = 80 that °C, 8%t = 120 aqua min) regia can (T provide = 80 ◦C, a tprocess = 120 min) window, can provide where a57% process of the window, visible gold where can 57%be dissolved. of the visible Simultaneously, gold can be 44% dissolved. of the Cu Simultaneously, is dissolved. The 44% dissolution of the Cu isfrom dissolved. non-crushed The dissolutionWPCBs is selective from non-crushed towards iron. WPCBs These isprocess selective conditions towards provide iron. These the recycling process conditionsoperator a possibility provide the of partially recycling dissolving operator aAu possibility and Cu prior of partially to further dissolving processing Au via, and e.g., Cu a primary prior to furthercopper refinery processing route via, [12]. e.g., a primary copper reﬁnery route [12].

4. Conclusions The redox potential characteristics of aqua regia solutions were determined in this study and were found to increase as a function of both increased concentration and temperature. Results showed that redox potential ranged from 540 to 900 mV vs. Ag/AgCl in the studied range (aqua regia concentration = 0.01 to 32% and temperature = 40–80 ◦C. In addition, the redox potential measured in a pure aqua regia solution obeyed the Nernst equation with the exception of the lowest aqua regia concentration of 0.01%. The redox potentials varied between 300 and 900 mV vs. Ag/AgCl after the WPCB exposure into the aqua regia solution. The batch leaching tests for WPCB raw material suggested that high redox (>850 mV vs. Ag/AgCl) potential alone did not allow Au dissolution, but Au dissolution was heavily supported by aqua regia concentration and dissolved Cu2+ ions, acting as gold oxidant [11]. High Au extractions (>50%) were achieved in solutions with [Cu2+] >2500 ppm at redox potentials >700 mV vs. Ag/AgCl. The results indicate that 8% aqua regia (T = 80 ◦C, t = 120 min) can provide a process window, where most (57%) of the visible gold can be dissolved with simultaneous Cu dissolution (44%). In the investigated leaching environment, the dissolution of Au and Cu is selective towards iron. It can be concluded that the results provide systematic data about the use of aqua regia for Au, Cu, and Fe leaching from the WPCB material. Aqua regia can be seen as a potential hydrometallurgical pre-treatment step to separate partially gold and copper from non-crushed WPCB with only minimal iron dissolution.

Acknowledgments: The authors would like to thank the Emil Aaltonen Foundation (Ympäristöystävällistä kultaa—project funding), the CMEco project (7405/31/2016), and the ARVI project funded by Finnish innovation agency TEKES for financial support. “RawMatTERS Finland Infrastructure” (RAMI) and NoWASTE-project (297962) supported by the Academy of Finland and are greatly acknowledged. Author Contributions: Heini Elomaa and Sipi Seisko conceived and designed the experiments. Tero Junnila and Tuomas Sirviö performed the experiments. Heini Elomaa analyzed the data. Jari Aromaa and Sipi Seisko contributed to the design of experiments and analysis of the data. Heini Elomaa wrote the paper and significant contribution to writing was made by Benjamin P. Wilson and Mari Lundström. Mari Lundström is the corresponding author. Conflicts of Interest: The authors declare no conflict of interest.

References

1. European Commission. Directive 2012/19/EU of the European Parliament and of the Council of 4 of July on Waste Electrical and Electronic Equipment; Ofﬁcial Journal of the European Union: Strasbourg, France, 2012. 2. Marsden, J.; House, I. The Chemistry of Gold Extraction, 2nd ed.; The Society of Mining, Metallurgy and Exploration: Littleton, CO, USA, 2006; p. 650, ISBN 10-0-87335-240-8. 3. Veit, H.M.; Bernardes, A.M.; Ferreira, J.Z.; Tenório, J.A.S.; de Fraga Malfatti, C. Recovery of copper from printed circuit boards scraps by mechanical processing and electrometallurgy. J. Hazard. Mater. 2006, 137, 1704–1709. [CrossRef][PubMed] 4. Yannopoulos, J.C. The Extractive Metallurgy of Gold; Van Nostrand Reinhold: New York, NY, USA, 1991; p. 281, ISBN 0-442-31797-2. 5. Cui, J.; Zhang, L. Metallurgical recovery of metals from electronic waste: A review. J. Hazard. Mater. 2008, 158, 228–256. [CrossRef][PubMed] 6. Park, Y.J.; Fray, D.J. Recovery of high purity precious metals from printed circuit boards. J. Hazard. Mater. 2009, 167, 1152–1158. [CrossRef][PubMed] 7. Zhang, L.; Xu, Z. A review of current progress of recycling technologies for metals from waste electrical and electronic equipment. J. Clean. Prod. 2016, 127, 19–36. [CrossRef] 8. Lekka, M.; Masavetas, I.; Benedetti, A.V.; Moutsatsou, A.; Fedrizzi, L. Gold recovery from waste electrical and electronic equipment by electrodeposition: A feasibility study. Hydrometallurgy 2015, 157, 97–106. [CrossRef] 9. Sheng, P.P.; Etsell, T.H. Recovery of gold from computer circuit board scrap using aqua regia. Waste Manag. Res. 2007, 25, 380–383. [CrossRef][PubMed] Recycling 2017, 2, 14 9 of 9

10. Saulny, M. The Behavior of Printed Circuit Boards in Knife Shredder and Its Effects on the Recycling Process. Master’s Thesis, Aalto University, Espoo, Finland, 2015. 11. Lampinen, M.; Seisko, S.; Forsström, O.; Laari, A.; Aromaa, J.; Lundström, M.; Koiranen, T. Mechanism and kinetics of gold leaching by cupric chloride. Hydrometallurgy 2017, 169, 103–111. [CrossRef] 12. Hagelüken, C. Recycling of electronic scrap at Umicore’s integrated metals smelter and reﬁnery. Erzmetall 2006, 59, 152–161.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).