The removal of toxic metals from liquid effluents by ion exchange resins. Part V: Nickel(II)/H⁺/Dowex C400

Francisco José Alguacil^a,*

^a Centro Nacional de Investigaciones Metalúrgicas (CENIM-CSIC), Avda. Gregorio del Amo 8, 28040 Madrid, Spain

*Corresponding author: fjalgua@cenim.csic.es

1. INTRODUCTIONTOP

Though nickel is considered a hazardous element, there is not a clear evidence that soluble nickel salts alone are responsible for cancer, or whether when they appeared in conjunction with insoluble nickel form. In fact, this element does not appear in the list of harmful elements in primary and secondary drinking water standards (USEPA, 2017).

The overall consideration of nickel toxicology has lead to a recent number of reports about the removal of this metal from aqueous solutions using different technologies (El-Bahy and El-Bahy, 2016; Guan et al., 2016; Jain et al., 2016; Melo et al., 2016; Otrembska and Gega, 2016; Taha et al., 2017; Yousef et al., 2016; Zhang and Chen, 2016; Moghbeli et al., 2017). Next in the series of articles published by the author about the use of ion exchange technology in the removal of toxic metals (Alguacil, 2002; Alguacil et al., 2002; Alguacil, 2003; Alguacil, 2017), the present work reports about the use of the cationic exchange resin Dowex C400 on the removal of nickel(II) from aqueous solutions and under various experimental conditions, and different mathematical models were used to fit the various aspects in relation of the cationic exchange process.

2. EXPERIMENTALTOP

Dowex C400 (Fluka) having the characteristics shown in Table 1 were used to investigate the removal of nickel(II) from solutions. Other chemical used in the investigation were of AR grade, whereas the experimental procedure used here was the same as described in previous works (Alguacil, 2002; Alguacil et al., 2002; Alguacil, 2003; Alguacil 2017). The multiwalled carbon nanotubes presented the characteristics described elsewhere (Alguacil et al., 2016), whereas the carboxylic functionalized multiwalled carbon nanotubes were obtained from the Aldrich catalogue, ref. 755125.

All the loading experiments were carried out using 200 mL of the aqueous solution, 800 rpm and 20 ºC, except when these variables were investigated.

Metals in solution were analysed by AAS spectrometry, whereas metals load in the resin was calculated by the mass balance.

3. RESULTS AND DISCUSSIONTOP

3.1. Nickel loading onto the resinTOP

Being Dowex C400 a cationic exchanger, in which the sulfonic group is the active one, the equilibrium which describes the removal of nickel(II) from the solutions in the 0-5 pH range is represented by (Eq. (1)):

where, R is the non-reactive part of the resin, and r and aq represented the resin and the aqueous solution, respectively.

In batch ion exchange-adsorption experiments, the stirring speed could play a key-role in the performance of the system, since the correct stirring speed allows to gain a maximum in the metal loaded in the corresponding solid resin-adsorbent; however, this variable is often neglected in many scientific reports (AlOmar et al., 2017; Drasinac et al., 2017; Kim and Kwak, 2017; Ogden et al., 2017; Wang and Liu, 2017). In the present work, aqueous solutions containing 0.01 gL^-1 Ni(II) at pH 5 was put into contact with 0.5 gL^-1 of the resin, and the system was stirred at different stirring speeds ranging 200-1000 rpm. Results obtained from this set of experiments are shown in Fig. 1. The maximum in adsorption loading is reached in the 800-1000 rpm range, and thus, the thickness of the aqueous solution layer reached a minimum and a maximum in nickel(II) uptake is accomplished.

The influence of the temperature on metal load onto the resin was investigated using a resin dose of 0.25 gL^-1 and aqueous solution of 0.01 gL^-1 Ni(II) at pH 5. The results from Table 2 showed that the values of the nickel distribution coefficient, defined as (Eq. (2)):

where [Ni]_r and [Ni]_aq represented the nickel(II) concentrations at equilibrium in the resin and in the aqueous solution, respectively, increases as the operational temperature increases.

Thermodynamic parameters were estimated using conventional relationships, indicating that the reaction is endothermic (ΔHº 65 kJmol^-1), and ΔSº and ΔGº are 111 kJmol^-1K and 33 kJmol^-1, respectively. The positive ΔGº value indicated that the exchange process is not spontaneous, and the also positive ΔSº value is an indication of increasing randomness at the liquid-resin interface during the exchange process.

The influence of the aqueous pH, ranging 0-5, on metal load onto the resin was investigated using 0.25 gL^-1 of resin and an aqueous solution containing 0.01 gL^-1 Ni(II). As can be seen seen from Fig. 2 and Table 3, the pH of the aqueous solution has little influence in the metal uptake in the pH range 1-3, though some decrease in the metal load can be observed as the pH shifted to more acidic values. However, this nickel uptake completely falls off when the pH of the solution reached values near 0, although this behaviour will be useful (see below) when the nickel(II) elution is considered.

The influence of the resin dosage on nickel uptake onto the resin was also investigated, when the resin dosage was changed between 0.5 and 0.05 gL^-1, and the results obtained are summarized in Table 4. As it is expected, nickel adsorption increases with the resin dosage until reaching, under the present experimental conditions, a metal uptake in the 110 mg.g^-1 range.

The aqueous ionic strength is another experimental variable which could influence the metal loading onto the resins, and its effect on the present system was investigated by the use of different concentrations of lithium chloride in the aqueous solution. After 3 h of contact between 0.25 gL^-1resin and an aqueous solution containing 0.01 gL^-1Ni(II), the results presented in Table 5 showed a continuous decrease in the percentage of nickel uptake, or nickel concentration in the resin, as the aqueous ionic strength is increased, being this a point to consider when real solutions need to be treated since they have normally ionic strengths greater than zero.

It is important to remark that when real solutions are considered is that very often the targeted metal is accompanied by a miriade of other metals, thus, the selectivity of the present system is next investigated using binary solutions containing nickel(II) and another metal in equimolar concentrations. Values of metal distribution coefficients (see Eq. (2)) and separation factors β, calculated as in Eq. (3), are shown in Table 6:

It can be seen from this Table 6, that the separation of nickel(II) from the accompanying metals, β>1, is only possible in the case of zinc(II) and cobalt(II), being manganese(II) and copper(II) more or less equally loaded onto the resin than nickel(II) (not separation possible here), whereas indium(III) is by far loaded preferably to nickel (β<1). It is worth to mention here that in all the experiments using binary solutions, the loading curve versus time for nickel(II), as represented in Fig. 3, always responded to the same pattern and with practically the same quantitative results for each time, which also corresponded to the data obtained when only nickel(II) is present in the aqueous solution (nickel uptake of 39 mg.g^-1 at equilibrium versus 39.2 mg.g^-1 for binary solutions) this means that apparently the population or crowding effect showed in other separations technology, i.e. liquid membranes, did not appear here (Alonso et al., 2006).

It is worth to brief mention here, that when the population or crowding effect, in a given system, appeared, it means that the value of the metal uptake of a target element in, i.e. the binary solution, decreased in comparison with the same value obtained when the given element is alone in solution.

The performance of Dowex C400 was compared with results obtained in the removal of nickel(II) by other smart adsorbents like multiwalled carbon nanotubes (MWCN) and carboxylic functionalized multiwalled carbon nanotubes (C-MWCN) are. Results for these series of experiments are shown in Table 7. At the present experimental conditions, the resin has a better performance, in terms of nickel removal from the aqueous solution, that either type of carbon nanotubes tested in this work.

The adsorption data were used to examine their fit to the Langmuir or Freundlich models (Alguacil et al., 2016). The results of these tests show that the adsorption is best fit to the Freundlich model (r²: 0.993):

with 1/n 0.33 and ln K_f 4.1. In the above equation, [Ni]_r,e and [Ni]_aq,e represented the nickel concentrations at equilibrium in the resin and in the aqueous solution, respectively, whereas K_f is a Freundlich constant.

Adsorption or exchange kinetics was also evaluated. The pseudo-first and pseudo-second order rate equations, Eqs. (5 and 6), respectively:

were used to estimate their fit to the experimental data. In the above equations, [Ni]_r,e represented the nickel concentration in the resin at equilibrium, [Ni]_r,t represented the metal concentration in the resin at an elapsed time, t is the elapsed time and k₁ and k₂ are the respective rate constant for each model. Numerical results, Table 8, showed that the kinetics is slightly best explained within the pseudo-first order model.

Moreover, the exchange process was evaluated by the use of the models described elsewhere (López Díaz-Pavón et al., 2014). Results of the experimental data fitting to the models, showed that the rate law governing (best fitting) the nickel exchange within the resin is a liquid-solid phase reaction which includes a moving boundary process described by:

with rate constant k 3.5x10^-3 min^-1 and r² 0.988. In the above equation, the fractional attainment to the equilibrium, F, was calculated by the next relationship:

where, [Ni]_r,t and [Ni]_r,e are the nickel concentrations in the resin at and elapsed time t and at equilibrium, respectively.

3.2. Metal elution from the loaded resinTOP

As it is mentioned above, and since nickel is poorly loaded onto the resin at pH values in the zero range, it seemed logical the use of mineral acids to accomplish both the nickel elution from the resin and the resin regeneration in the same step. Results obtained from these series of experiments are summarized in Table 9. It can be seen, that, under the same experimental conditions, a slightly best percentage of nickel elution is obtained when sulphuric acid solution is used as eluant, whereas the change in the amount of resin/volume of eluant solution also slightly influenced the percentage of nickel recovered from the loaded resin (98.5% versus >99%). These results lead to obtain aqueous solutions with concentrations, at least, one magnitude order greater than the original aqueous feed (0.1 gL^-1 in the eluted solution versus 0.01 gL^-1 in the original feed). Beside the elution of nickel, the use of these acidic solutions result in the resin regeneration by shifting the equilibrium shown in Eq. (1) to the left.

4. CONCLUSIONSTOP

ACKNOWLEDGEMENTTOP

To the CSIC (Spain) for support. To Mr. J.M. Medina for assistance in part of the experimental work.

REFERENCESTOP