Sage Journals: Discover world-class research

Abstract

Adsorption of the gold–thiosulfate complex ion ( $Au (S 2 O 3) 23 -$ ) on silver ferrocyanide (AgFC)-impregnated activated carbon in aqueous solution has been studied in order to find an effectual adsorbent for the thiosulfate extracting gold from ores. This study was performed using AgFC-impregnated activated carbon (AC-Ag-R-FC: AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide) and an artificial aqueous solution of $Au (S 2 O 3) 23 -$ . Gold–thiosulfate complex adsorption kinetics and isotherm studies were carried out at pH = 9.0 on modified materials. It has been also found that the adsorption fits the intraparticle diffusion and Freundlich isotherm well. In order to understand the adsorption mechanism, raw and modified materials were characterized by N₂ adsorption–desorption measurements at 77 K, scanning electron microscopy and X-ray photoelectron spectroscopy. The maximum adsorption capability of $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC is 3.55 kgt⁻¹. Clearly, the extraordinary adsorption capacity of AC for $Au (S 2 O 3) 23 -$ offers a new approach to address challenging gold–thiosulfate complex separation and could promote the future development of thiosulfate leaching gold process.

Keywords

Modified activated carbon silver ferrocyanide gold thiosulfate adsorption

Introduction

Activated carbons (ACs) are highly porous carbonaceous materials that possess a unique set of properties: a well-developed reproducible microporous structure, a large internal surface area, a high degree of surface reactivity, and an adsorption capacity that makes them effective and versatile adsorbents (Hu et al., 2000; Lee et al., 2006; Poinern et al., 2011). In the gold hydrometallurgy industry, AC is used for the adsorption of Au(I) in the form of $Au (CN) 2 -$ ions that are present in the cyanide leach liquors that normally contain approximately 10 mgl⁻¹ of Au(I). There are two AC-based technologies that are widely used in the gold industry, and 80% of the gold production in the world employs one of these methods. The first is the carbon-in-pulp (CIP) process, and the second is the carbon-in-leach (CIL) process. The gold mining industry has used AC to extract gold (Au) from crushed ore for many years to maximize extraction rates and hence reduce processing costs. However, due to mounting environmental concerns (such as the leakage of sodium cyanide in the Tianjin explosion accident in China) and unresolved practical difficulties in using cyanide to extract gold from copper–gold, refractory, and carbonaceous ores, the thiosulfate system has been widely accepted by researchers around the world over the last three decades as a potential alternative lixiviant for the leaching and recovery of gold (Abbruzzese et al., 1995; Fotoohi and Mercier, 2014; Zipperian et al., 1988). Jeffrey et al. (2003) and Lam and Dreisinger (2003) showed that it is essential to maintaining adequate copper(II), ammonia, and thiosulfate concentrations and thus a higher oxidation rate of gold dissolution. A possible mechanism for the leaching reaction has been proposed as

Au + 5 S 2 O 32 - + Cu (NH 3) 42 + \to Au (S 2 O 3) 23 - + 4 NH 3 + Cu (S 2 O 3) 35 -

(1)

4 Cu (S 2 O 3) 35 - + 16 NH 3 + O 2 + 2 H 2 O \to 4 Cu (NH 3) 42 + + 12 S 2 O 32 - + 4 OH -

(2)

The thiosulfate system extracts gold from carbonaceous ores; however, the gold–thiosulfate complex does not effectively adsorb onto AC. This renders the widely accepted CIP or CIL technology useless for the thiosulfate system (Zhang and Dreisinger, 2002). Early research also confirmed that AC (in its pristine form) does not adsorb the gold–thiosulfate complex. Mohansingh (2000) studied the adsorption of gold onto AC by varying parameters and varying the concentration of different species in solution and found that gold from copper–ammonium–thiosulfate solution did not adsorb well onto AC and the maximum adsorption capability of gold on AC was 0.267 kgt⁻¹ AC by calculation. Kononova et al. (2001) also studied sorption recovery of gold from thiosulfate solutions with a concentration of 9.5–17.9 mgl⁻¹ after leaching of products. The maximum recovery percentage of gold on AC was 16.9%, possibly because of the high anionic charge of the thiosulfate complex and the bulkiness of the thiosulfate ions compared to gold cyanide. Yu et al. (2015) fabricated AC impregnated with cupric ferrocyanide species to recover the gold from thiosulfate solution. However, the whole adsorption capacity was lower and the adsorption mechanism is not clear, which is not ideal for commercial applications. In this paper, adsorption of the gold–thiosulfate complex ion ( $Au (S 2 O 3) 23 -$ ) on silver ferrocyanide (AgFC)-impregnated AC in aqueous solution has been studied in order to find a effectual adsorbent for the gold thiosulfate process for extracting gold from ores and make adsorption mechanism clear.

Experimental

Materials

The silver ferrocyanide-impregnated AC used in this work, AC-Ag-R-FC (AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide), was prepared in our laboratory. The details of the preparation of this adsorbent have been described later. A commercial coconut shell AC was also used in this work for the adsorption of $Au (S 2 O 3) 23 -$ ) in aqueous solution, in order to compare the AC-Ag-R-FC with conventional AC.

The reagents of sodium thiosulfate ( $Na 2 S 2 O 3$ ) for the preparation of the complex ion $Au (S 2 O 3) 23 -$ ) solution, sodium hydroxide ( $NaOH$ ), and hydrochloric acid ( $HCl$ ) for adjusting pH, used in this work, were of analytic grade. A standard gold powder from the Shanghai Reagent Company in China with 99.999% purity was used to prepare the complex ion $Au (S 2 O 3) 23 -$ ) solution. The water used in this work was deionized water. Other reagents used were of analytic grade.

Methods

Preparation of AC (AC-Ag-R-FC-modification)

There grams of adsorbent were placed in a 50 ml beaker with 20 ml of 1–100 m mol l⁻¹ $AgNO 3$ solution prepared with deionized water. A period of impregnation, the supernatant silver solution was decantation. The resulting particles were washed eight times with deionized water and separated by filter, then put in the crucible, roast for 2 h in muffle furnace under 900℃. Here, we designated this AC treated with a silver solution as AC-Ag-R.

Then, 1–20 m mol l⁻¹ $K 4 Fe (CN) 6$ solution prepared with deionized water was placed in the beaker of AC-Ag and allowed to stand for 1–4 h at ambient temperature. The particles obtained were washed with deionized water and separated by decantation. This process was repeated several times. The precipitate was allowed to dry for 12 h at 106℃ in air. Here, we designated this AC treated with a silver solution as AC-Ag-R-FC. Two reactions may occur on the AC as following equation (Bellomo, 1970)

AC + 4 Ag + + Fe (CN) 64 - \to AC - Ag 4 Fe (CN) 6

(3)

Preparation of $Au (S 2 O 3) 23 -)$

First, 1 mol l⁻¹ $Na 2 S 2 O 3$ aqueous solution and the standard gold solution (pure gold dissolved in aqua regia) wer $NaOH$ e prepared. The standard gold solution has been evaporation of the solution to near dryness to eliminate the remaining acid. At this point, 0.05 mol l⁻¹ solution was added to achieve pH values of 9.0 and then 1 mol l⁻¹ $Na 2 S 2 O 3$ aqueous solution was added to prepare $Au (S 2 O 3) 23 -$ . For each adsorption test, the $Au (S 2 O 3) 23 -$ solution was prepared with a given gold concentration. This method of preparation of $Au (S 2 O 3) 23 -$ solution has been described elsewhere.

Adsorption tests

The tests of the adsorption of $Au (S 2 O 3) 23 -$ onto AC-Ag-R-FC from $Au (S 2 O 3) 23 -$ solution were carried out by the following procedure. First, 1 g of AC-Ag-R-FC or other adsorbents were added to a conical flask that had already been filled with 100 ml of $Au (S 2 O 3) 23 -$ solution at a given initial gold concentration. Then, the conical flask was put into a water bath in order to have a constant temperature of 25℃ throughout the adsorption with mechanical agitation. At 24 h, part of the solution was withdrawn for the chemical analysis of gold concentration using a PE AA300 atomic absorption spectrophotometer. The recovery percentage of gold adsorption onto the AC was calculated on the basis of the expression

R = c 0 - c t c 0 \times 100 %

(4)

where R is the recovery percentage of gold on AC and c₀ and c_t are the gold concentration in the

Au (S 2 O 3) 23 -

solution at 0 (initial) and t time, respectively.

pH

The effects of variable pH on adsorbent performance was conducted with 1.0 g of each adsorbent added to 100.00 ml of 10 mgl⁻¹ $Au (S 2 O 3) 23 -$ solutions. The variable pH solutions were prepared by adding dilute $NaOH$ or $HCl$ dropwise to achieve pH values of 5.0–11.0. Bottles were shaken at 200 r min⁻¹ for 24 h at an ambient temperature.

Point of zero charge (PZC) measurements

The pH_pzc (pH value at which the surface has a zero net charge) was determined by the so-called pH drift method (Guedidi et al., 2013). The pH of deoxygenized $NaCl$ aqueous solutions (50 ml at 0.01 mol l⁻¹) was adjusted to successive initial values between 2 and 12 by adding either 0.1 mol l⁻¹ $HCl$ or $NaOH$ . Suspensions were prepared by adding AC (0.15 g) to each solution. After stirring for 48 h under N₂, the final pH was measured and plotted versus the initial pH. The pH_pzc was determined to be the value at which pH_final = pH_initial.

Apparatus

Chemical analysis

AAS-300 (PerkinElmer, USA) atomic absorption spectrometry with deuterium background corrector and hollow cathode lamp was used. The operating conditions were as follows: Au: wavelength, 328.07 nm; slit width, 2.71 nm; lamp current, 10 mA; Ag: wavelength, 328.1 nm; slit width, 1.3 nm; lamp current, 10 mA.

Morphology

Scanning electron microscopy (SEM) was used to characterize the morphology. Samples were deposited on graphite plates and examined by a LEO 1530 electron microscope.

The surface area and porosity properties of the absorbents were determined by nitrogen ( $N 2$ ) adsorption–desorption at −196℃ (77K) with a relative pressure (P/P_o) using an automated gas sorption system (Autosorb-1MP, Quantachrome Ins, USA). The Brunauer–Emmett–Teller (BET) surface area, cumulative pore volume, and average pore width were obtained from the N₂ adsorption isotherms. The pore size distribution (PSD) was analyzed by using the nonlocal density functional theory (Zheng et al., 2014). The cumulative pore volume was calculated by measuring the amount of N₂ adsorbed at a relative pressure of 0.964 (Cesano et al., 2012).

X-ray photoelectron spectroscopy (XPS)

XPS measurements were performed using a ULVAC-PHI Inc. XPS instrument. This system used a focused monochromatic Al–Kα X-ray source for analysis. Initially, a survey scan identified all the detectable elements, through a high generation energy and a short dwell time (Boudou et al., 2003). Next, the instrument conducted narrower and more detailed scans of selected peaks, so as to quantify chemical state information with a low generation energy and a long dwell time. XPS data have provided information on the surface chemical nature of the functional groups through analysis of the C 1s, N 1s, O 1s, Fe 2p, S 2p, Ag 3d, and Au 4f spectra. After baseline subtraction, the curve fitting was performed assuming a mixed Gaussian–Lorentzian peak shape (the ratio of Gaussian to Lorentzian form was equal to 0.3). The C 1s electron binding energy corresponding to graphitic carbon was referenced at 284.8 eV for the calibration (Guedidi et al., 2013).

Results and discussion

The preparation of AC

Through preliminary exploration experiments that design a typical synthesis, 3 g AC was placed in 20 m mol l⁻¹ $AgNO 3$ solution for 2 h, placed in a crucible, and heated for 2 h in a muffle furnace at 900℃. Subsequently, AC-Ag was placed in 50 m mol l⁻¹ $K 4 Fe (CN) 6$ solution prepared with deionized water and allowed to stand for 3 h at ambient temperature. The particles were dried for 12 h at 106℃ in air. We designated this AC as AC-Ag-R-FC. This material enables the fast adsorption of $Au (S 2 O 3) 23 -$ on AC in an aqueous solution under mild conditions (50℃). (For more information about modified process you can refer to Yu et al. (2015)).

To compare the effects on gold adsorption of AC with different degrees of modification, we designed five different modification processes: AC-Ag, AC-Ag-R, AC-Ag-R-FC, AC-AgFC, AC-FC (AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide). From Figure 1, the results of these experiments show that the recovery rate of gold on the carbon in AC-Ag-R-FC was the highest (up to 70.89%). The AC treated with $AgNO 3$ or $K 4 Fe (CN) 6$ alone did not affect the recovery of gold from solution. However, AC-Ag and AC-AgFC both possess the ability to partially absorb the gold–thiosulfate complex, which indicates that the process of heating can significantly improve the recovery percentage of gold.

Figure 1.

The effects of the AC modification process on the recovery percentage of gold (0.01 mol l⁻¹ $AgNO 3$ , 0.05 mol l⁻¹ $K 4 Fe (CN) 6$ , temperature of 900℃, 10 mgl⁻¹ of $Au (S 2 O 3) 23 -$ ).

Characterization of ACs

Figures 2 and 3 show the N₂ adsorption–desorption isotherms and PSDs of carbon samples, respectively. Surface area and pore volume properties are summarized in Table 1. From the N₂ isotherms (Figure 3), it is evident that the different processing approaches do not affect the isotherm shape; they are typical of microporous–mesoporous adsorbents (type I and IV in the IUPAC classification) with a hysteresis loop (H4 types) in the desorption branch at relative pressure ratios of approximately 0.5. According to the classification of the IUPAC, the diameter of adsorbent pores is as follows: less than 2 nm is micropores, between 2 and 50 nm is mesopores, and more than 50 nm is macropores (Njoku and Hameed, 2011). The average pore width of the absorbents was found to be approximately 2.8 nm (Table 1), which indicates mesoporosity. The PSD of the absorbents displayed in Figure 3 indicates that most of the pores are in the micropore and mesopore range.

Figure 2.

Adsorption–desorption isotherms of N₂ at 77.4 K on activated carbon prepared at different conditions.

Figure 3.

Pore size distribution for activated carbon prepared at different conditions.

Table 1.

Textural characteristics of activated carbons prepared at different conditions.

Material	S_BET (m² g⁻¹)^a	V_T (cm² g⁻¹)^b	Average pore diameter (nm)
AC	1946	1.1525	2.583
AC-AgFC	958.4	0.6825	2.848
AC-FC	902.7	0.643	2.849
AC-Ag	830.1	0.5935	2.860
AC-Ag-R	1019	0.7695	3.021
AC-Ag-R-FC	1044	0.7463	2.860

AC: activated carbon; BET: Brunauer–Emmett–Teller; Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

Surface area of BET.

Total pore volume of BET.

The N₂ isotherms at 77 K exhibited a considerable decrease of the BET surface area (Table 1, S_BET = 958.4 m² g⁻¹ for AC-AgFC, 902.7 m² g⁻¹ for AC-FC, 830.1 m² g⁻¹ for AC-Ag, 1019 m² g⁻¹ for AC-Ag-R, and 1044 m² g⁻¹ for AC-Ag-R-FC) of the modified carbon compared to the pristine AC (S_BET = 1946 m² g⁻¹). As can be observed from SEM (Figure 4), the decrease in the specific surface areas after modification was related to reagent deposition on the surface of AC or pore collapse. Additionally, the specific surface area of heated AC is larger than that observed for other types of sorbents. As shown in Table 1, heating can increase both the average pore diameter of the AC and the specific surface area.

Figure 4.

Scanning electron microscope (SEM) micrograph of activated carbon at different conditions (AC-Ag-R-FC-Au: the AC-Ag-R-FC after adsorption of gold). (a) AC, (b) AC-Ag-R-FC, and (c) AC-Ag-R-FC-Au. AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

Figure 4 illustrates the exposed porosity and surface morphology as measured by SEM. SEM images of AC, AC-Ag-R-FC, and AC-Ag-R-FC-Au show that the pores of AC-Ag-R-FC collapse and deposit particles of native silver (Ag) and $Ag 4 Fe (CN) 6$ (confirmed by XPS) on the surface; after adsorption of the gold–thiosulfate complex, some of the substances that consisted of gold complexes and other compounds (confirmed by XPS) are still deposited on the surface of AC-Ag-R-FC.

Adsorption tests

Effects of the pH in artificial aqueous solution

As can be observed from the results shown in Figure 5(a), the pH variation affected gold adsorption (i.e. the recovery percentage of gold was severely promoted at the lower pH values) and then started to level off at pH ≥ 7.0. Figure 5(b) shows that the pH_zpc value of 7.25 was obtained for AC-Ag-R-FC, so that for pH values lower than pH_pzc, the adsorbent presented a positive surface charge that promoted the adsorption of $Au (S 2 O 3) 23 -$ . This behavior explains the high adsorption capacity of the adsorbent for $Au (S 2 O 3) 23 -$ at pH 5.0 (Guedidi et al., 2013; Radovic et al., 1997). Additionally, the final pH values of the adsorbate solutions always reached 8.5 regardless of the initial pH value of the solution (see the inset in Figure 5(a)). We demonstrate the phenomenon as follows.

Figure 5.

The effects of the pH in artificial aqueous solution on the recovery percentage of gold (10 mg l⁻¹ of $Au (S 2 O 3) 23 -$ , 25℃).

Effect of gold concentration and temperature

We performed experiments at different temperatures and initial gold concentration with the purpose of determining the adsorption capacity of AC-Ag-R-FC for $Au (S 2 O 3) 23 -$ . As can be observed from Figure 6, the initial gold concentration and temperature have a significant influence on the adsorption of gold. It is clear that the higher the initial concentration of gold is the lower the recovery percentage will be. However, the adsorption capacity of gold increased as the initial gold concentration or temperature increased. This result can be understood by assuming that an adsorption equilibrium exists between AC and the solution and that this equilibrium was influenced by variations in temperature and the initial gold concentration. In particular, the adsorption capacity of gold increased as the initial gold concentration or temperature increased. The adsorption capability of $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC was 3.55 kgt⁻¹ in 100 mgl⁻¹ gold at 50℃.

Figure 6.

The effects of gold concentration on the recovery percentage of gold at different temperatures in artificial aqueous solution (pH = 8.5).

Adsorption isotherm

To examine the relationship between adsorbed (q_e) and aqueous concentration (c_e) at equilibrium, sorption isotherm models are widely employed for fitting the data, of which Langmuir and Freundlich equations are most widely used. The Langmuir model supposes that adsorbate substances was monolayer adsorption, and no interaction between adsorbed substances on homogeneous surface (Gonen et al., 2004). Freundlich model is suitable for nonideal adsorption on heterogeneous surfaces (Al-Othman et al., 2012). The heterogeneity is caused by the presence of different functional groups on the surface and various adsorbent–adsorbate interactions. In order to obtain the equilibrium data, sorption experiments on AC-Ag-R-FC, in which initial $Au (S 2 O 3) 23 -$ concentrations were varied while the adsorbate for both samples was kept constant, were conducted with the equilibrium duration of 24h.

Each Adsorption isotherm testing was implemented after 1.0 g of AC-Ag-R-FC was added into 100 ml of solution, which contains 20, 30, 50, and 100 mgl⁻¹ of $Au (S 2 O 3) 23 -$ , respectively. Adsorption isotherm data were fitted to Langmuir model and Freundlich model as shown below, where c_e is the $Au (S 2 O 3) 23 -$ concentration at equilibrium, q_e is the amount adsorbed at equilibrium, q_max and b are constants, q_max is the maximum adsorption capacity, and k is the adsorption constant (Njoku and Hameed, 2011)

Langmuiradsorption : q e = q \max bc e 1 + bc e

(5)

Freundlichmodel : q e = kc e 1 / n

(6)

Figure 7 shows the plots of 1/c versus 1/q_e for the $Au (S 2 O 3) 23 -$ adsorption isotherm. From the slope and intercept of the correlation line, q_max and b were calculated to be shown in Table 2 at temperatures of 30, 40, 50℃, respectively. The poor linear correlation (R²) demonstrated the low adaptability of the $Au (S 2 O 3) 23 -$ adsorption isotherm to the Langmuir equation.

Figure 7.

Adsorption isotherm of the complex ion $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC at temperatures 30, 40, 50℃ (Langmuir model). AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

Table 2.

Adsorption isotherm model parameters of the complex ion $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC.

Temperature (℃)	Langmuir parameters			Freundlich parameters
Temperature (℃)	R ²	q_max (mg g⁻¹)	b (l g⁻¹)	R ²	k	n
30	0.958	1.212	0.052	0.978	0.164	2.342
40	0.993	5.102	0.014	0.999	0.111	1.318
50	0.991	4.717	0.039	0.993	0.336	1.704

AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

Figure 8 shows the plots of logc_e versus logq_e for the $Au (S 2 O 3) 23 -$ adsorption isotherm. The good linear correlation (R²) demonstrated the high adaptability of the $Au (S 2 O 3) 23 -$ adsorption isotherm to the Freundlich model. From the slope and intercept of the correlation line, n and k were calculated to be shown in Table 2 at temperatures of 30, 40, 50℃, respectively.

Figure 8.

Adsorption isotherm of the complex ion $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC at temperatures 30, 40, 50℃ (Freundlich model). AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

As can be observed from Figures 7 and 8 and Table 2, the $Au (S 2 O 3) 23 -$ adsorption on AC-CuFC isotherm followed the Freundlich isotherm. These isotherms showed the relationship between the amounts of $Au (S 2 O 3) 23 -$ adsorbed (q_e) and its equilibrium concentration (c_e) in solution. It is also clear from these figures that the adsorptivity of $Au (S 2 O 3) 23 -$ increased with increase in temperature. This suggested that $Au (S 2 O 3) 23 -$ adsorption from aqueous solutions on AC-Ag-R-FC was endothermic process. The increase in the adsorption capacity may be due to the chemical interaction between adsorbate and adsorbent, creation of some new adsorption sites, or the increased rate of intraparticle diffusion of $Au (S 2 O 3) 23 -$ ions into the pores of the adsorbent at higher temperature which indicated that adsorption was not monolayer physical adsorption but multimolecular chemical adsorption. As noted in Table 2, the values of n in Freundlich model suggested the adsorption of $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC was an easily adsorption process.

Adsorption kinetics

Adsorption kinetics can provide valuable information about the reaction pathways and mechanism of the reactions. Pseudo-first-order, pseudo-second-order models, and the intraparticular diffusion model were tested for their abilities to describe the process of $Au (S 2 O 3) 23 -$ adsorption (Al-Othman et al., 2012; Duranoğlu et al., 2012)

Pseudo-first-orderkineticmodel : \log (q e - q t) = \log q e - k 1 2.303 t

(7)

Pseudo-second-orderkineticmodel : t q t = 1 k 2 q e 2 + 1 q e t

(8)

Theintraparticulardiffusionmodel : q t = k p t 1 / 2 + c

(9)

where q_e and q_t are

Au (S 2 O 3) 23 -

adsorption at equilibrium and time t; k₁, k₂, and k_p are adsorption rate constants; and c is the intercept (mg g⁻¹).

Figure 9(a) to (c) shows the pseudo-first-order kinetic model, pseudo-second-order kinetic model, and intraparticle diffusion model of $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC at gold concentrations of 20, 30, 50, 100 mgl⁻¹ under 30℃, respectively. Table 3 shows the kinetic model parameters which indicated that intraparticle diffusion model was more suitable for adsorption kinetics compared to pseudo-first-order kinetic model or pseudo-second-order kinetic model. The maximum adsorption capacity of gold on AC-Ag-R-FC was 4.608 kgt⁻¹ at temperature of 50℃ according to pseudo-second-order kinetic model.

Figure 9.

Adsorption kinetic model of $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC at gold concentration 20, 30, 50, 100 mgl⁻¹ (30℃) ((a) pseudo-first order, (b) pseudo-second order, (c) intraparticle diffusion kinetic model). AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

Table 3.

Adsorption kinetics model parameters of the complex ion $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC.

Kinetic models	Temperature	Parameters	Concentration of gold (mg l⁻¹)
Kinetic models	Temperature	Parameters	20	30	50	100
Pseudo-first-order kinetic model	30℃	R ²	0.661	0.799	0.873	0.923
		k₁ (h⁻¹)	0.067	0.06	0.064	0.064
		q_e,c (mg g⁻¹)	0.288	0.424	0.689	0.902
	40℃	R ²	0.736	0.86	0.927	0.778
		k₁(h⁻¹)	0.06	0.069	0.09	0.064
		q_e,c (mg g⁻¹)	0.473	0.75	1.253	1.816
	50℃	R ²	0.923	0.953	0.975	0.952
		k₁ (h⁻¹)	0.048	0.051	0.076	0.058
		q_e,c (mg g⁻¹)	0.962	1.38	2.399	3.266
Pseudo-second-order kinetic model	30℃	R ²	0.992	0.983	0.994	0.986
		k₂ (g mg⁻¹ h⁻¹)	0.982	0.431	0.197	0.142
		q_e,c (mg g⁻¹)	0.504	0.622	0.946	1.215
	40℃	R ²	0.983	0.987	0.996	0.986
		k₂ (g mg⁻¹ h⁻¹)	0.439	0.244	0.194	0.119
		q_e,c (mg g⁻¹)	0.736	1.052	1.634	2.762
	50℃	R ²	0.935	0.931	0.977	0.964
		k₂ (g mg⁻¹ h⁻¹)	0.072	0.039	0.023	0.022
		q_e,c (mg g⁻¹)	1.399	2.066	3.413	4.608
Intraparticle diffusion model	30℃	R ²	0.996	0.989	0.977	0.993
		k_p (mg g⁻¹ h^−1/2)	0.045	0.079	0.142	0.181
		C (mg g⁻¹)	0.254	0.167	0.116	0.133
	40℃	R ²	0.992	0.988	0.962	0.996
		k_p (mg g⁻¹ h^−1/2)	0.083	0.137	0.212	0.312
		C (mg g⁻¹)	0.261	0.265	0.475	0.980
	50℃	R ²	0.992	0.991	0.962	0.992
		k_p (mg g⁻¹ h^−1/2)	0.210	0.307	0.513	0.704
		C (mg g⁻¹)	−0.015	−0.103	−0.199	−0.062

AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

Generally, the intraparticle diffusion models possess two lines that attributed to phenomena such as initial surface adsorption and subsequent intraparticle diffusion. It was reported that sole limiting step of intraparticle diffusion needs respective plot of q_t versus t^1/2 pass through the origin (C value to be zero) (Ghaedi et al., 2013). The values of C (intercept) highly depend on the amount of adsorbent and initial gold concentration. Figure 9 and Table 3 (the values of C) suggested the linear portion of the curve does not pass through the origin. It may be concluded that the adsorption mechanism of this adsorbate is a rather complex process and that the intraparticle diffusion was not the only rate-controlling step (Al-Othman et al., 2012).

Adsorption mechanism

To understand the adsorption mechanism, surface chemical groups of the AC were also characterized by XPS. Figure 10 presents the C 1s, N 1s, O 1s, Fe 2p, Ag 3d, S 2p, and Au 4f spectra of treated AC before and after adsorption of the gold–thiosulfate complex. Deconvolution of the C 1s, N 1s, O 1s, Fe 2p, Ag 3d, S 2p, and Au 4f spectra resulted in several peaks as shown in Figure 10 and the signal attribution shown in Table 4.

Figure 10.

XPS analysis of AC-Ag-R-FC and AC-Ag-R-FC-Au. AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

Table 4.

The XPS analysis of AC-Ag-R-FC and AC-Ag-R-FC-Au.

Chemical element	AC-Ag-R-FC		AC-Ag-R-FC-Au		Reference(s)
Chemical element	Energy (eV)	Signal attribution	Energy (eV)	Signal attribution	Reference(s)
C 1s	283.0	$Fe (CN) 64 -$	284.8	Graphite carbon	Franz et al. (2000)
	284.8	Graphite carbon	286.4	$Fe (CN) 63 -$
	291.4	–O–C = O	288.6	–C = O
	293.0	–C–Cl₃
O 1s	530.3	–C = O	531.9	$S 2 O 32 -$ , –C = O	Tseng et al. (2010)
			533.5	$- NO 3 -$ , –OH
			536.2	Chemisorbed O
Fe 2p	710.5	$Fe (CN) 64 -$	712.1	$Fe (CN) 63 -$	Cao et al. (2010) and Lukkari et al. (1999)
N 1s	396.4	Nitrides	400.8	$Fe (CN) 63 -$	Franz et al. (2000)
N 1s	398.1	$Fe (CN) 64 -$	406.8	–NO₃	Franz et al. (2000)
Ag 3d	366.7	Ag⁺	368.5	Ag	Malik et al. (2005) and Ristova and Ristov (2001)
Ag 3d	368.4	Ag	368.5	Ag	Malik et al. (2005) and Ristova and Ristov (2001)
Au 4f	Not detected		84.74	Au	Makhova et al. (2007)
S 2p	Not detected		161.7	$S 2 O 32 -$ , $S 2 -$ (sulfide)	Makhova et al. (2007) and Mikhlin et al. (2010)
			164.3	–C–S-(thiophene)
			168.6	$SO 42 -$
			169.6	$SO 42 -$

AC: activated carbon, Ag: silver nitrate, R: heating, FC: potassium ferrocyanide.

As shown in Figure 10 and Table 4, comparison of the XPS core-level detailed scans of carbon before and after use in the solution showed that the spectra of C 1s, N 1s, and Fe 2p were altered and that the energy increased; this may be caused by the transformation of $Fe (CN) 64 -$ into $Fe (CN) 63 -$ . In regards to Ag 3d, one of the spectrum peaks (366.7 eV, which was attributed to the Ag⁺) disappeared and shifted to zero valence silver. The reasons for the disappearance of Ag⁺ peaks may be that the silver ion transfer into the solution or the silver ion concentration adsorbed on AC is lower not to be detected. This result suggested that oxidation/reduction occurred during the process of adsorption.

Harish et al. (2011) reported that a redox reaction occurs between gold(III) chloride and potassium hexacyanoferrate(II) leading to the formation of a charge transfer band and the conversion of hexacyanoferrate(II) to hexacyanoferrate(III) (as evidenced by the emergence of new absorption peaks in UV–Vis spectra) and that the Au complex may be , which is not feasible via a simple chemical or electrochemical reaction (Machado et al., 2011). In the adsorption process at lower pH (pH < 7.2), $Ag 4 Fe (CN) 6$ and $Au (S 2 O 3) 23 -$ possibly form intermediate $KFe x [Au (CN) 2] y$ and are then spontaneously reduced to Au(0), as indicated by the spectrum of Au 4f. According to equations (10) and (11), the production of hydroxyl caused the final pH values of the solution to be as high as 8.5. Burdinski and Blees (2007) also found that $Au (S 2 O 3) 23 -$ can be reduced to Au(0) by $Fe (CN) 64 -$ . Under alkaline conditions, Yonezu et al. (2007) found that $Au (S 2 O 3) 23 -$ complex ions disproportionately occur on the surface of alumina and that Au(0) and Au(III) species ( $[AuCl 4 - n OH n] -$ ) should be formed. The $[AuCl 4 - n OH n] -$ complex ion has a large oxidation power and Au(III) species are spontaneously reduced to Au(0) due to a change in structure by the formation of an $Ag - O - Au$ bond (Burdinski and Blees, 2007). As a result, once $Au (S 2 O 3) 23 -$ complex ions disproportionately exist on the surface of AC-Ag-R-FC, they are all converted to Au(0), as described by equations (11) to (13)

K + + Ag 4 Fe (CN) 6 + 3 Au (S 2 O 3) 23 - + H 2 O + 8 e - ⇋ KFe x [Au (CN) 2] y + 2 S + 5 S 2 O 32 - + Ag + 6 OH -

(10)

KFe x [Au (CN) 2] y + 2 e - ⇋ K + + yAu 0 + xFe (CN) 63 -

(11)

3 Au (S 2 O 3) 23 - + (4 - n) Cl - + nOH - ⇋ [AuCl 4 - n OH n] - + 6 S 2 O 32 - + 2 Au 0

(12)

4 [AuCl 4 - n OH n] - + 2 Ag + O 2 + 12 e - ⇋ 2 Au 0 + (16 - 4 n) Cl - + 4 nOH - + 2 Ag - O - Au

(13)

In equations (10) and (11), x = 1, y = 3. The chloride ion of equations (12) and (13) which acted as catalytic is from the preparation of gold–thiosulfate complex.

This adsorption mechanism might occur by two cases: At first, the gold–thiosulfate complex concentration on surface of modified AC increased by surface forces, then (i) under acidic conditions, the adsorbed $Au (S 2 O 3) 23 -$ at the surface of AC-Ag-R-FC reduces to Au(0) by $Fe (CN) 64 -$ as described by equations (10) and (11), and the production of hydroxyl increases the final pH values of the solution to as high as 8.5; and (ii) under alkaline conditions, in addition to the (i) case, a disproportionate amount of the adsorbed $Au (S 2 O 3) 23 -$ at the surface of AC-Ag-R-FC occurs and forms Ag–O–Au, as described in equations (12) and (13); the consumption of hydroxyl thus increases the final pH values of the solution to 8.5. The experimental results suggest that $Ag 4 Fe (CN) 6$ plays an important role in the concentration of gold(I) complex ions and the subsequent reduction of gold during the adsorption process.

Conclusions

The test results and mechanism of adsorption predict that AC-Ag-R-FC is a promising material for use in $Au (S 2 O 3) 23 -$ separation. The adsorption capability of $Au (S 2 O 3) 23 -$ on AC-Ag-R-FC is 3.55 kgt⁻¹ at a gold concentration of 100 mgl⁻¹ at 50℃, which is an important property of the new AC. Clearly, the extraordinary adsorption capacity of chemically modified AC for $Au (S 2 O 3) 23 -$ offers a new approach to tackling challenging gold–thiosulfate complex separation and could promote the future development of the thiosulfate leaching gold process.

Footnotes

Authors’ contribution

FZ and XH contributed equally to this work.

Acknowledgements

Talent Startup Project of Kunming University of Science and Technology (14118602, KKSY201307056); Research Center for Analysis and Measurement, Kunming University of Science and Technology (20150308); Young Scholar of the graduate student in Yunnan province in 2015.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was financially supported by the National Natural Science Foundation of China (51674128).

References

Abbruzzese

Fornari

Massidda

et al. (1995) Thiosulfate leaching for gold hydrometallurgy. Hydrometallurgy 39(1–3): 265–276.

Al-Othman

Ali

Naushad

(2012) Hexavalent chromium removal from aqueous medium by activated carbon prepared from peanut shell: Adsorption kinetics, equilibrium and thermodynamic studies. Chemical Engineering Journal 184: 238–247.

Bellomo

(1970) Formation of copper(II), zinc(II), silver(I) and lead(II) ferrocyanides. Talanta 17(11): 1109–1114.

Boudou

Chehimi

Broniek

et al. (2003) Adsorption of H₂S or SO₂ on an activated carbon cloth modified by ammonia treatment. Carbon 41(10): 1999–2007.

Burdinski

Blees

(2007) Thiosulfate- and thiosulfonate-based etchants for the patterning of gold using microcontact printing. Chemistry of Materials 19(16): 3933–3944.

Cao

Liu

Zhang

et al. (2010) In situ controllable growth of Prussian blue nanocubes on reduced graphene oxide: Facile synthesis and their application as enhanced nanoelectrocatalyst for H₂O₂ reduction. Acs Applied Materials and Interfaces 2(8): 2339–2346.

Cesano

Rahman

Bertarione

et al. (2012) Preparation and adsorption properties of activated porous carbons obtained using volatile zinc templating phases. Carbon 50(5): 2047–2051.

Duranoğlu

Trochimczuk

Beker

(2012) Kinetics and thermodynamics of hexavalent chromium adsorption onto activated carbon derived from acrylonitrile-divinylbenzene copolymer. Chemical Engineering Journal 187: 193–202.

Fotoohi

Mercier

(2014) Recovery of precious metals from ammoniacal thiosulfate solutions by hybrid mesoporous silica: 1-Factors affecting gold adsorption. Separation and Purification Technology 127: 84–96.

10.

Franz

Arafat

Pinto

(2000) Effect of chemical surface heterogeneity on the adsorption mechanism of dissolved aromatics on activated carbon. Carbon 38(13): 1807–1819.

11.

Ghaedi

Ansari

Sahraei

(2013) ZnS:Cu nanoparticles loaded on activated carbon as novel adsorbent for kinetic, thermodynamic and isotherm studies of Reactive Orange 12 and Direct yellow 12 adsorption. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 114: 687–694.

12.

Gonen

Kabasakal

Ozdil

(2004) Recovery of cyanide in gold leach waste solution by volatilization and absorption. Journal of Hazardous Materials 113(1–3): 231–236.

13.

Guedidi

Reinert

Lévêque

J-M

et al. (2013) The effects of the surface oxidation of activated carbon, the solution pH and the temperature on adsorption of ibuprofen. Carbon 54: 432–443.

14.

Harish

Joseph

Phani

KLN

(2011) Interaction between gold (III) chloride and potassium hexacyanoferrate (II/III)—Does it lead to gold analogue of Prussian blue? Electrochimica Acta 56(16): 5717–5721.

15.

Srinivasan

(2000) Preparation of mesoporous high‐surface‐area activated carbon. Advanced Materials 12(1): 62–65.

16.

Jeffrey

Breuer

Chu

(2003) The importance of controlling oxygen addition during the thiosulfate leaching of gold ores. International Journal of Mineral Processing 72(1–4): 323–330.

17.

Kononova

Kholmogorov

Kononov

et al. (2001) Sorption recovery of gold from thiosulphate solutions after leaching of products of chemical preparation of hard concentrates. Hydrometallurgy 59(1): 115–123.

18.

Lam AE and Dreisinger DB (2003) The importance of the Cu (II) catalyst in the thiosulfate leaching of gold. In: Hydrometallurgy 2003: 5th International Symposium Honoring Professor Ian M. Ritchie, pp.195–211.

19.

Lee

Kim

Hyeon

(2006) Recent progress in the synthesis of porous carbon materials. Advanced Materials 18(16): 2073–2094.

20.

Lukkari

Meretoja

Kartio

et al. (1999) Organic thiosulfates (Bunte salts): Novel surface-active sulfur compounds for the preparation of self-assembled monolayers on gold. Langmuir 15(10): 3529–3537.

21.

Machado

Bergmann

Fernandes

et al. (2011) Adsorption of Reactive Red M-2BE dye from water solutions by multi-walled carbon nanotubes and activated carbon. Journal of Hazardous Materials 192(3): 1122–1131.

22.

Makhova

Mikhlin

Romanchenko

(2007) A combined XPS, XANES and STM/STS study of gold and silver deposition on metal sulphides. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 575(1): 75–77.

23.

Malik

Kulesza

Wlodarczyk

et al. (2005) Formation of ultra-thin Prussian blue layer on carbon steel that promotes adherence of hybrid polypyrrole based protective coating. Journal of Solid State Electrochemistry 9(5): 403–411.

24.

Mikhlin

Likhatski

Tomashevich

et al. (2010) XAS and XPS examination of the Au–S nanostructures produced via the reduction of aqueous gold (III) by sulfide ions. Journal of Electron Spectroscopy and Related Phenomena 177(1): 24–29.

25.

Mohansingh R (2000) Adsorption of gold from gold copper ammonium thiosulfate complex onto activated carbon and ion exchange resins. Master’s thesis, University of Nevada, Reno.

26.

Njoku

Hameed

(2011) Preparation and characterization of activated carbon from corncob by chemical activation with H₃PO₄ for 2,4-dichlorophenoxyacetic acid adsorption. Chemical Engineering Journal 173(2): 391–399.

27.

Poinern

GEJ

Senanayake

Shah

et al. (2011) Adsorption of the aurocyanide, complex on granular activated carbons derived from macadamia nut shells – A preliminary study. Minerals Engineering 24(15): 1694–1702.

28.

Radovic

Silva

Ume

et al. (1997) An experimental and theoretical study of the adsorption of aromatics possessing electron-withdrawing and electron-donating functional groups by chemically modified activated carbons. Carbon 35(9): 1339–1348.

29.

Ristova

Ristov

(2001) XPS profile analysis on CdS thin film modified with Ag by an ion exchange. Applied Surface Science 181(1): 68–77.

30.

Tseng

T-F

Yang

Y-L

Lin

Y-J

et al. (2010) Effects of electric potential treatment of a chromium hexacyanoferrate modified biosensor based on PQQ-dependent glucose dehydrogenase. Sensors 10(7): 6347–6360.

31.

Yonezu

Yokoyama

Okaue

et al. (2007) Concentration of gold(I) thiosulfate complex ions on the surface of alumina gel and their change in chemical state: Preliminary experiment in the elucidation of the formation mechanism of epithermal gold deposits. Resource Geology 57(4): 400–408.

32.

et al. (2015) Adsorption of the gold–thiosulfate complex ion onto cupric ferrocyanide (CuFC)-impregnated activated carbon in aqueous solutions. Hydrometallurgy 154: 111–117.

33.

Zhang

Dreisinger

(2002) The adsorption of gold and copper onto ion-exchange resins from ammoniacal thiosulfate solutions. Hydrometallurgy 66(1–3): 67–76.

34.

Zheng

Z-Q

Xia

H-Y

Srinivasakannan

et al. (2014) Utilization of Crofton weed for preparation of activated carbon by microwave induced CO₂ activation. Chemical Engineering and Processing: Process Intensification 82: 1–8.

35.

Zipperian

Raghavan

Wilson

(1988) Gold and silver extraction by ammoniacal thiosulfate leaching from a rhyolite ore. Hydrometallurgy 19(3): 361–375.

Adsorption of gold from thiosulfate solutions with chemically modified activated carbon

Abstract

Keywords

Introduction

Experimental

Materials

Methods

Preparation of AC (AC-Ag-R-FC-modification)

Preparation of Au ( S 2 O 3 ) 2 3 - )

Adsorption tests

pH

Point of zero charge (PZC) measurements

Apparatus

Chemical analysis

Morphology

X-ray photoelectron spectroscopy (XPS)

Results and discussion

The preparation of AC

Characterization of ACs

Adsorption tests

Effects of the pH in artificial aqueous solution

Effect of gold concentration and temperature

Adsorption isotherm

Adsorption kinetics

Adsorption mechanism

Conclusions

Footnotes

Authors’ contribution

Acknowledgements

Declaration of Conflicting Interests

Funding

References

Preparation of $Au (S 2 O 3) 23 -)$