Experimental procedures

Testing material and chemicals

Cemented WC-Co tool bits scraps

Cemented WC-Co tool bits were supplied by M/s Bharat Futuristic Pvt. Ltd, Bangalore, India. Most of the scraps had some brazing material left over on their surface. So, before the electrochemical studies, they were kept in an aqua regia solution, for 4 h, to dissolve all the unwanted brazing material. Then, the tool bits were cleaned by ultrasounds, for 30 min, using deionized water. Fig. 1 shows the digital image of various WC-Co scraps.

The digital image of the as-received WC-Co scrap materials used in the present study is shown in Fig. 2.

The cemented WC scrap sample contained 81.2% W, 5.2% C, 10.15% Co, 0.98% Fe, 0.062% Al and 0.005% Ti (all in wt.%). W, Co, and C are the major constituents in WC-Co. Impurities, such as Al, Ti and Mo, were present in trace amounts, including about 1.0% Fe.

The XRD pattern of the sample is shown in Fig. 3(e). The XRD peaks showed that the major phases were WC, Co₆W₆C and Co (JCPDS files: 25-1047, 23-0939 and 05-0727, respectively).

Chemicals

The NH₃ electrolytes were prepared by proper dilution of an AR grade NH₃ solution. All the dilutions were done using laboratory-grade deionized water processed by the Millipore Mili-Q system. The AR grade NH₄Cl, (NH₄₂)CO₃ and (NH₄)₂SO₄ were used as additives in the PDP and electrodissolution studies. All the abovementioned substances were procured from Merck Specialities Private Limited (Mumbai).

Electrodissolution methods

Initially, the WC-Co sample was subjected to PDP tests, so as to obtain an electrochemical response (dissolution and passivation), using a Gamry Echem Analyst potentiostat. WC-Co dissolution was subsequently studied by electrodissolution experiments, with the help of an electrolytic cell, by turning the sample into an anode in an electrochemical cell. A rectifier supplied the DC, procured from M/s Gayatri Enterprises, Pune. W and Co dissolved in the electrodissolution experiments were recovered by the precipitation method. Details of all the experimental procedures are given in the following sections.

PDP studies

A Gamry Echem Analyst potentiostat was used for PDP studies. A three electrode cell sell setup comprised WC-Co scraps, SC and a Pt foil, which were used as WE, RE and CE, respectively. The WE was connected, with the help of the Pt wire, to the power supply. The exposed surface area of the WC-Co sample was ~1cm², and it was coated with the chemically inert resin. The sample was polished so as to produce a new mirror surface, with the help of different grits of emery paper and a polishing disc, and rinsed with deionized water, before each experiment. The working cell consisted of a 150 mL glass beaker containing 100 mL of an electrolyte solution, WE (WC-Co scraps) and CE. This cell was filled with fresh electrolytes for each PDP experiment. The SC RE was placed in another glass beaker (reference cell) containing a 100 mL saturated KCl solution. A salt bridge was used to connect both cells.

To prepare a salt bridge, 2.5 gm agar-agar were thoroughly mixed into a solution of 25 g KCl in 150 mL deionized water, and heated at 90 ºC. After mixing, the solution was partially cooled and sucked through one end of a ‘U’ shaped glass tube (salt bridge), until it reached the other end. The salt bridge was stored by dipping both ends in a saturated KCl solution in a glass beaker. During PDP studies, one end of the salt bridge was placed close to the WE, while the other was placed near the RE ⁽17⁾. In the PDP study, the applied E was increased with a programmed SR, while the current was continuously monitored. Then, J vs. applied E was drawn. Such as in a typical PDP scan, the initial and final E were set from -1.5 V to 7 V, with respect to the RE, at a SR of 1.5 mV/s. All the PDP experiments were carried out at room temperature. In some of them, the calculated quantity of a few chemicals was dissolved as an additive into the electrolyte, before the scan.

Electrodissolution studies

Electrodissolution experiments of the WC scraps were carried out in a 500 mL glass beaker cell. The cell, filled with 275 mL of electrolyte, was equipped with a water recirculation system, in order to cool the electrolytes during electrodissolution. WC scraps were connected to the Pt wire, and suspended in the electrolyte, as the anode. A 304 grade SS sheet was employed as cathode. Both anode and cathode were immersed into the electrolyte solution, and DC power was supplied by a rectifier manufactured by Gayatri Enterprises, Pune (model: Power Tron). In the electrodissolution experiments, the WC anode was connected to a positive power supply, and the negative power supply was connected to the SS cathode. In all the tests, there were efforts to control the cooling water temperature, by recirculation. During electrodissolution, the E (from 5 to 15 V) was applied by the rectifier, and the corresponding dissolution current was manually noted every 5 min, and the changes in temperature were measured by a thermometer.

Three sets of electrodissolution experiments were carried out, so as to obtain optimal conditions. In the first, different additives (NH₄Cl, ((NH₄)₂SO₄ and (NH₄)₂CO₃) were taken (5% w/v) in 150 g/L NH₃, at an applied potential of 10 V, to see their effect on J, and on W and Co recovery. The second set of experiments aimed to evaluate the impact of different NH₃ concentrations (100, 150 and 200 g/L) in 5% NH₄Cl, at an applied E of 10 V. The third set of experiments was run by varying the applied E (5, 10 and 15 V) and the additive concentrations (2, 5, 10 and 15% w/v) in a 150 g/L NH₃ electrolyte.

It is worth mentioning that all the electrodissolution experiments were carried out during 2 h. After each investigation, the cathode and anode were rinsed in the deionized water, dried in the air for 1 h, and weighed in an electronic balance, so as to calculate the anodic dissolution and cathodic deposition percentages, using the following formulas:

where W_a and w_a are the weight (grams) of the anode (WC sample), before and after the experiment, respectively, and w_Ci and w_Cf are the initial and final weight (grams) of the cathode. After the experiments, the electrolyte was analyzed, in order to determine W and Co concentrations. It is worth mentioning that each electrochemical measurement was performed four times, so as to ensure the repeatability of the experiments.

Characterization

The distribution of the samples morphology and surface elements was studied by SEM (Hitachi-S3400N scanning electron microscope), coupled with EDX. XRD patterns of WC-Co scraps and obtained WO₃ powder were recorded on a Siemens D500 X-ray diffractometer, using Ni filtered CuK_α radiation (30 kVA, 15 mA), at a scanning speed of 2º(2θ)/min. The different phases were identified from the XRD pattern, with the help of JCPDS files.

Observation of the dissolved WC-Co scraps was also performed using AFM, so as to study the extent of the typical dissolution behaviour and explain its mechanisms, using the topographic image of the attacked surface, after the PDP and electrodissolution experiments. In order to study the surface characteristics, dissolved WC-Co samples were subjected to ultrasonic cleaning using acetone, for removing the loosely bounded undissolved and passivated materials from the anode surface. All the topographical studies were carried out in the contact mode, with asylum research AFM (MFP3D), at a SR of 0.5 Hz.

WO₃ recovery from the electrolytic solution

After the WC-Co dissolution, the concentrations of W, Co and other elements were estimated by ICP-OES. The spectrometer was calibrated using different dilutions of an individual analyte standard solution procured from Merck, India. After that, the samples were run through a capillary into the ICP, so as to determine the emission, which was automatically processed by the spectrometer, giving a final concentration value of a particular analyte.

The electrolyte obtained after the optimum electrodissolution experiment was further processed, in order to recover dissolved W as YTO. 50 mL of concentrated HCl were added drop by drop to the electrolyte containing H₈N₂O₄W (Fig. 13), by constant stirring it, which produced a white milky solution. After HCl addition, the white solution was heated at 50 ºC, for 1 h, for obtaining a yellow color precipitate of H₂WO₄ (Fig. 13(b)), which was separated by a centrifuge (5000 rpm). The yellow H₂WO₄ precipitate was heated for 1 h, at 900 ºC, in a muffle furnace, so as to obtain YTO, which was subjected to XRD analysis.

Results and discussions

WC-Co scraps characterization

The scraps chemical composition was: 81.2% W, 5.2% C, 10.15% Co, 0.98% Fe, 0.062% Al, and 0.005% Ti (all in wt.%). W, Co, and C were the principal constituents in the WC-Co tool bits. However, the impurities (Al, Ti and Mo) were also present in trace amounts, including about 1.0% Fe.

In order to remove the brazed material, WC-Co scraps were dipped into the aqua regia solution, for 4h, at 50 ºC. The samples were then cleaned, for performing the microstructural and XRD analyses. Fig. 3 (a) shows the micrographs of the WC-Co scraps in the BSE mode. Grey-colored fragmented WC grains were observed in the dark-colored Co matrix.

The WC-Co tool bit was scrapped material obtained from mining sites. Since tool bits (as-received) faced extreme environmental conditions, fragmented WC grains were observed in the Co matrix. Different shapes and sizes and disintegrated small WC grains are shown in Fig 3(a).

Mapping results have confirmed the presence of various elements (O, C, W and Co) in the materials. The EDS map shows the homogeneous distribution of all the elements (Fig. 3(b-e)). The sample XRD pattern is shown in Fig. 3(f). XRD peaks showed that the major phases were WC (JCPDS file: 25-1047), Co₆W₆C (JCPDS file: 23-0939) and Co (JCPDS file: 05-0727).

PDP tests

The anode dissolution extent, its inherent mechanisms and passivation behavior can be well understood with the help of polarization tests, in terms of the OCP (Table 2) and J values (Fig. 4). The OCP value gives information about the thermodynamic stability or nature of the very thin passive film (EDL) formed at the WC-Co and NH₃ solution interface. However, maximum anodic J and OCP higher negative values provide useful information about the active dissolution reaction, rather than about passivation. It is worth mentioning that the OCP value was obtained after 1 h of exposure tests (Table 2). Therefore, the EDL was formed at the electrode/electrolyte interface. Consequently, all samples have been polarized from the OCP value.

The OCP values increased with higher NH₃ solution concentrations (50, 100 and 150 g/L), because the tendency to dissolve WC-Co materials was higher in 150 g/L NH₃. Also, the pseudo-passive plateau region (from -1 to 1 V_SCE) has been observed for the 150 g/L NH₃ solution, indicating an active dissolution tendency (Fig. 4(a)).

However, in the 50 and 100 g/L NH₃ solution, a stable passive region could be seen in the same potential range (-1 to 1 V_SCE), showing the regular passive tendency on the WC-Co electrode surface. The anodic J values of the 50 and 100 g/L NH₃ solutions, at the same E range, were observed to be lower than that of the 150 g/L solution. This means that higher concentrations (150 g/L) of the NH₃ solution inhibited the passivation tendency.

Maximum anodic J at minimum E is a favored condition to achieve higher anodic dissolution rates, which is only possible by avoiding the anode passivation, with a higher NH₃ solution concentration (Fig. 4(a)).

Hence, maximum anodic current led to the maximum dissolution rate of WC-Co samples ^17,38. However, the overall maximum anodic J was significantly lower in the NH₃ solution without additives, and further increase in the E may have strengthen the anodic current. Still, it would consume more energy, and increase the probability of formation of a passive oxide layer on the electrode surface.

Therefore, the addition of chemical additives can avoid the formation of any passivation (Fig. 4(b)). Nevertheless, the NH₃ solution was initially not enough conducting to conduct the dissolved ions fastly in between the electrode. Thus, the dissolution rate was found to be very low, due to decreased anodic J.

The addition of compounds, such as NH₄Cl, (NH₄)₂SO₄ and (NH₄)₂CO₃, has not only contributed to increase the solution conductivity (by adding the additional ions), but has also avoid the possibility of any stable passive film formation on the WC-Co materials exposed surface (Fig. 4 (b and c)).

When it is likely that a stable passive layer is formed, the Cl^- ions turn out to be highly aggressive, and penetrate through it, by attacking W and Co (which is responsible for local passivation), forming complex products and decreasing the possibility of the layer formation. Similarly, SO₄ ^-2 and CO₃ ^-2 have done the same thing, but they were less effective than Cl^- ions ³⁹.

Therefore, such additives (containing aggressive ions) decreased the possibility of a stable passive layer being formed, and the resultant maximum anodic dissolution current was obtained at minimum E (Table 2).

However, WC dissolution was still hindered by the same optimal concentration of 150 g/L NH₃ solution, with 1% NH₃ Cl additives (Fig. 4(b)). So, an increase in NH₄Cl up to 10 %, in a 150 g/L NH₃ solution, enhanced the anodic dissolution.

SO₄ ^-2 and CO₃ ^-2 have a similar behaviour, but they are less effective than the Cl^- ions ³⁹.

In summary, if the NH₃ solution concentration increases from 50 to 150 g/L, the corresponding OCP value decreases. This means that the NH₃ concentration makes the surface thermodynamically more active for dissolution, while the anodic current drops, because Co tries to passivate the surface in the absence of these aggressive ions, by forming a stable complex (Fig. 4(c)).

Furthermore, a more negative OCP value (Table 2) represents a higher tendency for WC dissolution in NH₄OH. This indicates that NH₃ concentrations can limit the anodic dissolution rate. However, the OCP values were still very low, even after the addition of 1% NH₄Cl, (NH₄)₂SO₄ and (NH₄)₂CO₃. Moreover, OCP tended to shift towards the noble direction, with 1 to 10% NH ₄Cl. It is interesting to note that this behavior was not maintained in the presence of Cl^- ions (after the addition of 5 to 10% NH₄Cl), which probably reduced the tendency for a passive layer being formed on the exposed surface. Thereby, a stronger dissolution tendency has been observed at higher concentrations (150 g/L NH₃ with 10% NH₄Cl) of Cl^- ions, at minimum E.

Hence, it is inferred that the additives played a significant role in the WC materials de-passivation, and in the solution conductivity increase. However, W driving force and maximum solubility limit, in the NH₃ solution fixed concentration, became a rate-limiting step for WC-Co dissolution. This could explain why OCP value has shifted towards a noble direction, even after the addition of 1% NH₄Cl, (NH₄)₂SO₄ and (NH₄)₂CO₃, and an increase in the NH₄Cl additives concentration, from 1 to 10%. Since the electrolyte volume (100 mL) was fixed in all polarization tests, the same J value (819.1 A/cm²) has been noticed at different NH₄C concentrations (5 and 10 %). However, the applied E was lower (1.98 V_SCE), in the case of 10% NH₄Cl. Although the OCP value has shifted to a noble direction, WC-Co dissolution tendency could be strongly limited by the NH₃ solution concentration (without agitation). Also, maximum J value has been similarly obtained, because Co could not form a passive film on the surface, due to a high concentration of Cl^- ions (in both cases).

In general, W was more anodically active in the NH₃ medium than Co. Thereby, the formation of a W-based passive layer is less probable than that of Co, with higher NH₃ concentrations. Hence, once the maximum Cl^- ions concentration was reached (in the same volume of the NH₃ solution), Co also ceased being able to form a passive layer, due to their aggressive nature. Therefore, the anodic J value has reached its maximum value at the minimum E.

During the electrodissolution experiments, the WC-Co dissolution progressed, by forming WO₄ ²⁻ and Co²⁺. However, W standard dissolution or oxidation E (0.09 V) was lower than that of Co (0.28 V). Hence, W tended to dissolve first, forming WO₄ ²⁻ ions. And then, the anode became passivated when E reached a higher value, due to Co dissolution and formation of a passive layer ⁴⁰. Therefore, W formed WO₄ ²⁻, through the anodic oxidation reaction in the NH₃ solution, or due to the NaOH solution high pH (~11). Further, WO₄ ²⁻ reacted with NH₄ ⁺, which is the soluble product, and the following chemical reaction can be expected ^39,40.

The formation of WO₄ ^2- was thermodynamically more feasible with a higher NH₃ concentration (150 g/L), at a given E range. Hence, this reaction moved forward, and became more thermodynamically active when the NH₃ content increased.

On the other hand, the Co binder anodic dissolution also occurred through the oxidation reaction, by forming Co²⁺. However, these tended to form a Co-based amine complex. Co has a higher tendency to form an octahedral amine-based complex (⁽Co(NH₃)₆ ⁾⁺²), at a high dissolution E (0.28 V), in which NH₃ acts as a ligand, and the following reaction can be expected ⁴¹:

Although Co²⁺ can be passivated by forming Co-hydroxides, which hinders further oxidation, this reaction depends on the NH₄OH concentration that WO₄ ^2- ions can also consume, and the following reaction is expected to occur:

This concentration (150 g/L) and the NH₃ solution pH (~11) significantly affected the WC-Co surfaces dissolution and passivation tendency. However, the Co⁺² oxidation E was very high (0.28 V). Hence, at higher E and pH values, ⁽Co(NH₃)₆ ⁾⁺² complex became more stable than Co³⁺, which is also relatively stable at lower pH values ⁴¹. Hence, at higher E and low pH (~ 7) values, Co⁺² ions also oxidized into the very low oxidation E of Co⁺³ ions (-1.82 V), forming the unstable amine-based complex ⁽Co(NH₃)₆ ^{)+3 (39}.

150 g/L NH₃ with the aggressive anions Cl^-, SO₄ ^2- and CO₃ ^2- (separately obtained in the presence of NH₄Cl, (NH₄)₂SO_4, and (NH₄)₂CO_3,) have significantly influenced the abovementioned reaction, by avoiding the stable Co passive layer formation. Therefore, enhanced anodic current was due to the higher anodic dissolution, by the formation of ⁽Co(NH₃)₆ ⁾Cl₂, ⁽Co(NH₃)₆ ⁾SO₄ and ⁽Co(NH₃)₆ ⁾CO₃ complexes, and the following chemical reaction can be expected ³⁹.

WC-Co dissolution mechanisms

After the PDP tests, the AFM micrographs were used to obtain the dissolution behavior of the polarized WC-Co surface, which was caused by the aggressiveness of the different concentrations of NH₃ electrolytes with various additives. The AFM topographical micrographs from the corroded WC-Co surface helped to examine the anode materials dissolution tendency, based on the average R_a. The dissolution intensity, which is a localized attack on the polarized surface, can vary accordingly to different conditions, which depend on the ions (Cl^-, SO₄ ^-2 and CO₃ ^-2) aggressiveness, and on the correspondingly obtained maximum anodic J. Hence, the attacked surface was evaluated after the PDP tests, in terms of the average R_a value Fig. 5 shows the WC-Co anode topographical image, after electrochemical dissolution tests in a freely aerated 150g/L NH₃ solution with 1% (NH₄)₂CO₃.

A more severe dissolution can be seen at the WC grains and Co matrix interface. Since the WC grains are nobler than the Co matrix, maximum dissolution can be seen in the binder phase (Co) regions. The dissolution differential that occurs between the WC grains and the Co electrode is due to their different E values, which lead to the cathode and anode formation. Hence, the dissolution behavior is mainly influenced by the formation of galvanic couples between the WC grains noble cathode and the Co phase active anode ^2,42. The average R_a value was 0.387 ± 0.017µm, in the presence of (NH₄)₂CO₃ additives.

Fig. 6 shows the WC-Co anode topographical image, after the electrochemical dissolution tests, in a freely aerated 150g/L NH₃ solution with 1% (NH₄)₂SO₄. The average R_a was obtained at 0.503 ± 0.013µm. However, the dissolution behavior was identical.

Fig. 7 shows the WC-Co anode topographical image, after electrochemical dissolution tests, in a freely aerated 150g/L NH₃ solution with 1% NH₄Cl. The lowest (0.387 ± 0.017µm), intermediate (0.503 ± 0.013µm) and highest (0.704 µm ± 0.083µm) R_a values were obtained with 1% each (NH₄)₂CO_3, NH₄Cl and (NH₄)₂SO_4, respectively.

Fig. 8 shows the polarized WC-Co anode topographical image in a 150 g/L NH₃ solution with 5% NH₄Cl. Since the maximum dissolution current was observed in the presence of 5% NH₄Cl additives, it can be seen in terms of the average R_a value (0.842 µm ± 0.051µm). The stronger attack led to WC-Co maximum dissolution depth, which resulted in a rougher surface, and in its limited tendency to form an adherent or protective layer on the anode surface. In fact, the formation of a stable passive layer can reduce the dissolution attack, in the presence of (NH₄)₂CO₃ and (NH₄)₂SO₄ additives. Therefore, the anode susceptibility to dissolution, with different additives in the NH₃ solution, is strongly influenced by the obtained maximum anodic J and the presence of different ions.

WC-Co electrodissolution tendency

Combined effect of NH ₃ and additive concentrations

It is worth mentioning that the time of all the electrodissolution experiments was exactly 2 h, and the electrolyte was at ambient temperature. Co cathodic deposition, and the average anodic J obtained after the WC-Co scraps dissolution in freely aerated solutions, using different parameters, are given in Table 3.

However, electrodissolution studies in the NH₃ medium without additive were not carried out, since PDP studies without an additive gave rise to a much weaker anodic current. Therefore, a study in the NH₃ medium (150 to 200 g/L) with 5% NH₄Cl additives was performed, to ensure the active dissolution and the deposition effect with the solution higher concentrations.

It can be seen, from Fig. 9(a), that the anodic dissolution that was achieved at a NH₃ concentration of 200 g/L, was the highest, while the cathodic one was the lowest. This was perhaps because some loosened anode material, caused by Co partial dissolution, settled at the bottom of the electrolyte without being dissolved (later, it could be recovered by electrowinning methods). However, the highest J value was achieved with a 150 g/L NH₃ solution. Thus, this NH₃ concentration with 5% NH₄Cl was considered an optimum condition, and further experiments were carried out with this parameter. Fig. 9(b) also shows the correlation among the anode material (WC-Co tool bits) dissolution (%), Co deposition (%) at the cathode and average J, achieved by the electrodissolution experiments, in the 150 g/L NH₃ solution with 5% each of (NH₄)₂CO₃, (NH₄)₂SO₄ and NH₄Cl, at 10 V. Those parameters achieved their maximum values with 5% each NH4Cl, followed by (NH₄)₂SO₄ and (NH₄)₂CO₃. The highest cathodic deposition (11.72%) and the maximum average J (0.5995 A/cm²) were observed, because Cl^- ions accelerated the dissolution rate, by avoiding the passivation in the 150 g/L NH₃ electrolyte with 5% NH₄Cl. Therefore, it was found that NH₄Cl was the best option, and, hence, further experiments in the NH₃ medium were carried out only with this additive.

During the studies on the NH₃ concentration effect on the WC-Co tool bits electrodissolution, it was observed that some Cl evolution took place at the cathode, with the NH₃ lower concentrations (50 and 100 g/L), which may have influenced Co deposition behavior on the SS cathode. At higher NH₃ concentrations (150 and 200 g/L), no Cl evolution was observed. It seems that, with higher NH₃ concentrations, whatever Cl amount was evolved at the anode, in NH₄Cl electrolysis, got converted to HCl, which reacted with NH₃, producing NH₄Cl ⁴³.

NH ₄ Cl and impressed voltage synergistic effect

The effect of varied (2, 5, 10 and 15%) NH₄Cl amounts, in the 150 g/L NH₃ electrolyte, on the WC-Co tool bits electrodissolution, at 5, 10 and 15 V, is shown in Fig. 10 (a), (b) and (c), respectively. It can be seen that the electrodissolution was stronger as the NH₄Cl amount increased from 2% to 15%.

Fig. 10 (a) shows the correlation among the WC-Co scraps dissolution (%), Co cathodic deposition (%) and average J values, with NH₄Cl different concentrations (%) in a 150g/L NH₃ solution, at an operating voltage of 5 V, at ambient temperature. It can be seen that the average J values increased with higher NH₄Cl amounts.

Additionally, the NH₃ electrolyte became more conductive with higher levels of Cl^- ions. The maximum dissolution achieved by the WC-Co tool bits was 26.92%, with 15% NH₄Cl, because the average anodic J increased with higher contents of this additive, from 2 to 15%, at 5 V.

The cathodic deposition values in these experiments are also plotted in Fig. 10(a), which shows that it decreased with higher additive amounts, reaching its maximum (4.36%) with 2% NH₄Cl.

The average J values with NH₄Cl different amounts, at 10 V, are plotted in Fig. 10(b). They were much higher at 10 V than at 5 V. Maximum J, with 15% of the additive, was 1.66 A/cm², at 10 V, while, at 5 V, it was 0.34 A/cm².

The effect of NH₄Cl addition, at this higher voltage, on the WC-Co tool bits dissolution (%) is shown in Fig. 10(b). It was seen that the anodic dissolution was stronger with increasing additive percentages, and maximum dissolution (48.3%) was obtained with 15% NH₄Cl.

It was observed that the maximum cathodic deposition (11.72%) was obtained with 5% of NH₄Cl, and, with higher percentages, it drastically decreased (Fig. 10 (b and c)), due to the formation of Cl₂ gas bubbles near the cathode.

The electrodissolution studies were carried out at higher voltage (15 V), with 2, 5 and 10% NH₄Cl. It can be seen, from Fig. 10(c), that the WC-Co tool bits anodic dissolution increased with higher NH₄Cl amounts. About 50% of the anode dissolution occurred with 5% NH₄Cl. With 10% NH₄Cl, the dissolution was just 51%. Therefore, experiments with higher NH₄Cl amounts were not carried out. The cathodic deposition values with different percentages of the additive, in the electrodissolution experiment, at 15 V, are shown in Fig. 10 (c). Here, the maximum cathodic deposition was achieved with 5% NH₄Cl. It is worth mentioning that the increase in voltage (10 to 15 V) with higher NH₄Cl percentages (10 to 15%) led to stronger anodic currents that caused the anode decomposition and fragmentation. Hence, less Co dissolution in an electrolyte may lead to lower deposition.

Fig. 11 shows that the average J values and corresponding WC-Co scrap dissolution increased at higher operating voltages, with concentrations of 2, 5 and 10% NH₄Cl. However, the cathodic deposition reached its maximum at 10 V, with those concentrations of the additive. Therefore, 5% NH₄Cl in the 150 g/L NH₃ solution, at 10 V, was an optimum parameter for WC-Co scraps dissolution (%), Co deposition on the cathode surface and average maximum J.

It is worth mentioning that all the electrodissolution studies have been carried out for two hours, and it is expected that, with a longer duration of the experiment, higher recovery could be achieved.

Dissolved WC-Co surface morphology

Resuming, the optimum parameter for the stronger WC-Co scraps dissolution (%) and average maximum J was found to be with 15 % NH₄Cl in the 150 g/L NH₃ solution, at 10 V. The extent of WC-Co dissolution can be clearly observed by AFM micrographs. Fig. 7 shows the WC-Co anode materials topographical image after the electrochemical dissolution tests, in a freely aerated 150 g/L NH₃ solution with 15% NH₄Cl. The maximum average R_a value (1.82 ±0.085 μm) was obtained. It is worth mentioning that, before the AFM analysis, the WC-Co remaining anode, obtained after electrodissolution, was subjected to 20 min of ultrasonic cleaning, in an acetone medium, so as to remove its loosely bounded residues, and enhance the surface topography image.

Fig. 12 (a) shows the number of WC grains that were dissolved or coming out from the matrix. However, the dissolution could have started in the WC grains and Co interface. However, undissolved WC grains could also be seen after the Co phase dissolution.

W recovery from the electrolytic solution

W recovery in the electrodissolution experiments was calculated based on the anode chemical analysis and the metal dissolved in the electrolyte. The recovery values, at 5 V, with different NH₄Cl percentages, are given in Table 4. It can be seen that W recovery increased with higher NH₄Cl percentages, reaching a maximum value of 37.51% with 15% NH₄Cl. At 10 V, W recovery also increased with higher NH₄Cl percentages in the 150 g/L NH₃ electrolytes, reaching a maximum value of 65.26% with 15% NH₄Cl.

With different concentrations of NH₄Cl in 150 g/L NH₃ electrolytes, W recovery in the electrodissolution was quite high, at 15 V. It was 61.10% with 2% NH₄Cl, and it increased to 69.72% with 5% NH₄Cl. With 10% NH₄Cl, there was no significant change in W recovery (65.53%). Since W recovery with 5 and 10% NH₄Cl was almost the same, no further experiments with higher amounts of the additive were made.

Although the highest W recovery in the electrodissolution study was obtained with 5% NH₄Cl, at 15 V, the electrolyte temperature increased more (75 ^oC) than at 10 V (45 ^oC). It may be noted that the experiments at 10 V yielded 65.1% of W recovery, with 15% NH₄Cl, which was very near to its highest amount (69.72%), at 15 V. However, since the electrodissolution at 10 V consumed less energy than at 15 V, it seems that the optimal parameters for WC-CO tool bits electrodissolution, were 10 V and 15% NH₄Cl in 150 g/L NH₃. W recovery in the electrolyte solution was 65.1%, for 2 h, and it was further treated to produce saleable YTO.

Products characterization

After the electrodissolution experiments, the NH₃ electrolyte containing W, as H₈N₂O₄W, was acidified with HCl and heated, for 1 h, at 50 ºC. The precipitated H₂WO₄ was separated by centrifugation, and heated at 900 ºC, to obtain YTO. An image of YTO prepared from WC-Co scraps is shown in Fig. 13.

Fig. 13(a) shows the H₈N₂O₄W solution obtained after the electrodissolution tests. Fig. 13(b) shows the precipitated H₂WO₄ obtained after the H₈N₂O₄W solution acidification. The acidification with HCl was followed by mild heating at 50 ºC, for 1 h. Fig. 13(c) shows the YTO powder produced from the precipitated H₂WO₄ that was obtained after centrifugation, followed by heating at 900 ºC, in a tube furnace. The purity of W and Co recovered from the WC-Co scraps was also a significant concern of the present study. Hence, YTO powder was also compared to the standard WO₃ XRD pattern, as shown in Fig. 13(d). The XRD pattern of the YTO prepared in the present work matched well with the standard WO₃ JCPDS pattern (43-1035) shown in Fig. 13(e), which confirmed the nature of the product. Co deposited on the cathode was manually separated and washed with deionized water, for the XRD characterization. A SEM image of Co cathodic deposition (inset of Fig. 13(f)), in the form of separated chips, and the corresponding EDS spectra, are shown in Fig. 13(f) and (g), respectively.

In the present paper, PDP studies on WC-Co scraps have been carried out, since they reveal better their anodic dissolution behavior, in order to recover those metal values. PDP studies also provided information on suitable electrolytes, and found out the maximum anodic current at selective electrolytes, with or without additives. The maximum anodic current produced the WC-Co scraps stronger dissolution, without any passivation. During the electrodissolution experiments, it was initially found that the anodic current was very high, and, after some time, it continuously decreased, causing a reduction in the anode dissolution. This phenomenon is called passivation, and it occurred due to the formation of the anode passive layer. However, it was seen that the anodic passivation retarded the scraps dissolution, and adversely affected the metals recovery. So, in order to minimize the passivity and increase the anodic dissolution, some chemicals were used as additives. NH₄Cl, (NH₄)₂CO₃ and (NH₄)₂SO₄ were added in different concentrations to the NH₃ electrolyte. The studies revealed that 5% NH₄Cl were suitable for obtaining the maximum anodic dissolution (current) in 150 g/L NH₃. Therefore, the WC-Co tool bits electrodissolution in NH₃ effectively dissolved both W, as H₈N₂O₄W, and Co, as a Co-amine complex. Further, Co was deposited at the cathode during the electrolysis, while H₈N₂O₄W remained in the electrolyte, from where W could be recovered as H₂WO₄ or WO₃.

Therefore, WC-Co scrap produced by the mining industries was electrochemically dissolved, producing H₈N₂O₄W, and Co collected from the cathode. H₈N₂O₄W was converted into APT crystal, by filtering and evaporating it into white APT. On the other hand, H₈N₂O₄W electrolytes could be converted into W acid, by neutralization, using HCl, and later, they could be recovered in the form of YTO_. Hence, the present recycling technique has more advantages than those from the reported recycling techniques, such as single-step direct processes, high purity saleable products (APT, WO₃ and Co), low energy consumption and low cost environmentally friendly recycling methods.

Process flowsheet

Based on the electrodissolution studies, a flowsheet was proposed and illustrated (Fig. 14), for explaining the unit operations.

For example, Co further treatment by electrorefining may increase its purity. The newly developed methods for recycling WC-Co scraps through electrodissolution routes, and recovering saleable APT and WO₃ very pure, along with cathodically deposited Co, have an advantage compared to conventional processes, such as energy-intensive oxidation for eliminating WC-Co, since they have just a few industrial requirements, limited processing steps, recovered saleable products in simple unit operations, etc. ³⁸.