Preparation of cellulose/activated carbon cells: application to the adsorption of cobalt from stagnant waters

2.1 Cellulose/activated carbon cells preparation

Activated carbon used in this study was a homemade material prepared as described in our previous work. ²⁷ Thus, two activated carbons were prepared by chemical activation using ZnCl₂ and H₃PO₄ that lead respectively to ACZn and ACP samples.

The cellulose was obtained from sanitary paper which was produced from pure cellulose by a local Algerian company. Hence, a mass of 5 g of this material was put in a volume of 1 L of distilled water and mixed into a domestic blinder for 5 min at 12,000 rpm.

At the end of this operation, 5 g of activated carbon was added to the former mixture that subsequently took the appearance of an emulsion. After homogenization of the whole by mechanical stirring, it was poured into a homemade square shaped polyester grid (15 cm side). The content of the latter was dried in air and then unmolded. The final product was then dried overnight in an oven at 100 °C. The cell cellulose/activated carbon prepared from ACZn and ACP carbon samples were respectively named C-ACZn and C-ACP. The so-prepared cells were weighed in order to determine the exact mass of activated carbon loaded. The results showed that for the particle size used, the percentage of loaded activated carbon was around 80 % for both activated carbons. Additionally, the same procedure was carried out to prepare virgin cells free of activated carbons and called SP.

2.2 Textural characterization

Nitrogen adsorption–desorption isotherms are recorded at −196 °C on an ASAP 2020 (Micromeritics) equipment. The samples were outgassed under vacuum at 120 °C for 24 h. The specific surface area (S_BET) was estimated by applying the Brunauer, Emmett and Teller equation, the total pore volume (V_t) was calculated by measuring the adsorbed amount of N₂ at a relative pressure of 0.98; while the micropore volume (V_µ) was determined using the Dubinin-Radushkevich equation. Then, the volume of mesopores (V_mes) is deduced by subtracting the micropore volume from the total pore volume.

2.3 Optical microscope imaging

Optical microscope images of the prepared cells were recorded in order to observe their morphological aspects. In this respect, a Leica DM2500M microscope was used.

2.4 Tensile tests

The purpose of the tensile tests was to emphasize the effect of incorporating activated carbon granules into the cellulose on the mechanical behavior of the latter. This will partly explain the behavior of the obtained membranes in an aqueous medium.

Hence, the prepared cells were subjected to tensile tests under the conditions of the ISO 1924-2-2008 standard for papers and cardboards. Two series of tests were carried out: on dry and on wet cells. For the wet cells, they were first immersed in distilled water for 18 h then drained before having their ends dried so they can be hold between the grips of the machine. All the tests were performed in triplicates.

2.5 Thermogravimetric analysis

Thermal stability of cellulose materials is an important factor in assessing their performance and use making it essential to evaluate their degradation behavior before use. The thermal properties of the various samples were therefore studied by thermogravimetric analysis which was performed on a Perkin Elmer TG 8000 under a nitrogen atmosphere at a heating rate of 10 K min⁻¹ from ambient temperature to 500 °C.

2.6 Fourier-transform infrared spectroscopy

Pellets of sample: KBr were prepared with the same sample weight in a mass ratio sample: KBr of 1:300. The pellet was introduced into a FTIR-8400 Shimadzu spectrometer equipment to be analyzed in the spectral range 4,000–400 cm⁻¹ using 60 scans and 1 cm⁻¹ resolution.

2.7 Adsorption tests

Adsorption tests were carried out in non-sealed beakers without any agitation to simulate stagnant waters. Co(II) solutions were prepared from Co(NO₃)₂, 6H₂O. Two series of tests were performed; the former consisted of putting 150 mg of dried activated carbon in a non-sealed beaker containing 50 mL of a 50 ppm Co(II) solution. In the second series, a dried cell prepared as described in Section 2.1 and cut in dimensions allowing to enclose 150 mg of activated carbon was put in contact with the Co(II) solution as mentioned above. Both series of tests were conducted at different contact times between the contaminated solution and the adsorption device. Once the desired contact time is reached, the solution is filtered and the filtrate was analyzed by means of air/acetylene flame atomic absorption spectrophotometry (AAS) using a Schimadzu AA-6300 apparatus to quantify the remaining metal ion amount. The adsorbed amount was determined in percentage (R) and in mg per g of activated carbon (q) using Equation (1).

where C₀: the initial concentration of the solution [ppm], C: the concentration of the solution after adsorption [ppm].

3.1 Textural characterization

Nitrogen adsorption–desorption isotherms (Figure 1) and the textural parameters deduced from them (Table 1) show that the prepared activated carbons are highly porous with both micropores and mesopores. In addition, these types of carbons have shown their efficiency for the retention of heavy metals ions in fixed bed process. ²⁷ This supports the choice of testing them as adsorbent materials in cellulose-based cells intended to the adsorption of Co(II) from stagnant waters.

Figure 1:

N₂ sorption isotherms measured at −196 °C of prepared activated carbons by ZnCl₂ (ACZn) and H₃PO₄ (ACP).

3.2 Optical microscope imaging

The images recorded by optical microscope are given in Figure 2 with a scale of 50 µm. They show, as expected, that the prepared cells consist of a random assembly of cellulose fibers of different sizes (indicated by rectangles in Figure 2). Nevertheless, what is unexpected is the arrangement of the activated carbon granules within the cell. Indeed, these granules are not stuck between the fibers but they seem to stick on the fibers (Indicated by circles in Figure 2). This phenomenon is noticed for granules of different sizes.

Figure 2:

Images obtained by optical microscope (50 µm scale) of the prepared cells: (a) C-ACZn: cellulose/activated carbon cell loaded by activated carbon prepared by ZnCl₂, and (b) C-ACP: cellulose/activated carbon cell loaded by activated carbon prepared by H₃PO₄. Rectangles and circles refer to cellulose fibers and activated carbons grains, respectively.

3.3 Tensile tests

The standardized tests carried out allowed the calculations of the breaking stress that are given in Table 2.

According to these results, the presence of activated carbon granules decreases the mechanical resistance of the produced cells. This suggests that the activated carbon grains are inserted between the cellulose fibers, reducing the interaction between the latter and consequently the mechanical strength of the whole. This finding is in accordance with the results collected by optical imaging. The same tendency is observed after wetting the samples where the cellulose sheet alone (SP) has better mechanical resistance than the ones combined with carbons. One can see that after wetting the materials for 18 h, their mechanical strength is reduced however without making them completely unusable.

3.4 Thermogravimetric analysis

Thermogravimetric and its derivative curves of the prepared cells are given in Figure 3. All three samples appear stable up to 200 °C, with a slight mass loss of around 1 %–3 % recorded around 100 °C due to moisture evaporation. ^{28 , 29} A higher mass loss takes place between 250 °C and 390 °C, due to cellulose degradation processes (e.g. dehydration, depolymerization, decarboxylation and decomposition of glycosyl units), followed by the formation of a carbonized residue. ²⁸

Figure 3:

Thermograms and their derivatives of: SP: cellulose sheet without activated carbon loading. C-ACZn: cellulose/activated carbon cell loaded by activated carbon prepared by ZnCl₂. C-ACP: cellulose/activated carbon cell loaded by activated carbon prepared by H₃PO₄.

One can note that cellulose alone has a higher thermal stability than when combined with activated carbons. This may be the result of greater amount of hydrogen bonding between cellulose chains leading to more ordered and compact cellulose regions, which in turn can increase the decomposition temperature of cellulose. ³⁰ Results obtained by DTG showed that the thermal stability of the cellulosic material decreased after the addition of activated carbon where the decomposition temperature of cellulose decreased. This could be attributed to the interposition of activated carbon granules between the cellulose fibers. This is in agreement with the results of tensile tests.

3.5 Fourier-transform infrared spectroscopy

The spectra of the two activated carbons ACP and ACZn (Figures 4 and 5) are characteristic of activated carbons containing oxygenated functional groups such as carboxylic acids and lactones. The detailed interpretation of these spectra can be found in our previous works. ^{27 , 31}

Figure 4:

FTIR spectra of cellulose sheet (SP), the activated carbon ACP and its corresponding cellulose cell C-ACP.

Figure 5:

FTIR spectra of cellulose sheet (SP), the activated carbon ACZn and its corresponding cellulose cell C-ACZn.

On the other hand, the spectrum of the prepared cellulose sheet from sanitary paper is similar to that of cellulose as indicated in the literature. ^{32 , 33} The peak assignment is shown in Table 3.

The spectra of the two materials C-ACZn and C-ACP show the presence of the peaks of the respective activated carbons as well as those of the cellulose, which is predictable because these cells result from a physical mixture between the activated carbons and the sanitary paper. However, an interesting phenomenon should be highlighted. The latter concerns the disappearance of the peak at 1,640 cm⁻¹ in both C-ACZn and C-ACP spectra. This peak is that of the H–O–H bond of the absorbed water. According to the work of Fengel, ³⁴ an intense drying of the cellulose makes this peak disappear. Herein, and given that the samples were prepared and dried under the same conditions, the absence of this peak in the prepared cells indicates that the sites attracting water in the cellulose have probably formed hydrogen bonds with the activated carbon surface functional groups, particularly the carboxylic acid functions. This explains the results of tensile tests and the images obtained by optical microscope.

3.6 Adsorption tests

The adsorption ratio R for all tested materials is represented in Figure 6. Dispersing the activated carbons on the cellulose sheet leads to two kinetics. That is, one can observe the presence of two plateaus for both cells. This is resulting from a difference in the diffusion phenomenon which can be related to a delay in the contact between the activated carbon and the ions present in the solution caused by the presence of cellulose. However, after a certain time (30 and 40 min for C-ACZn and C-ACP, respectively), the kinetic is accelerated probably because of a better contact between the contaminated solution and the activated carbon. In the case of the activated carbons alone, there is also two adsorption kinetics however, the second starts later compared to the cells (at 60 min for both carbons). This is explained by the presence of smaller surface contact than in the case of the cells. This also explains the lower adsorption amount compared to the cells where the dispersion of carbon in the latter enhances the surface contact between the ions and the carbon grains hence a better adsorption amount. On the other hand, using the cells allows reaching higher adsorption ratios for both activated carbons.

Figure 6:

Co(II) adsorption ratio on the prepared activated carbons and their corresponding cellulose cells. Initial concentration: 50 ppm. Solution volume: 50 mL.

The pH of the prepared solution of Co(II) is around 6 which is higher than the pH_PZC of the used carbons which was determined by Joong et Schwarz method ³⁵ (5.5 and 2.8 for ACZn and ACP, respectively). This means that the adsorption of cations can take place. However, the predominant species of Co(II) at working pH are Co²⁺ and Co(OH)^{+ 36 , 37} consequently the retention of Co(II) on the surface of the prepared materials is favored.

Comparing now the efficiency of ACP, ACZn and their respective cells, one can notice that C-ACP gives the highest Co(II) retention ratio (25 %) because it combines all the favorable conditions for a better adsorption. Indeed, the activated carbon is well dispersed on the cellulose sheet allowing a better contact with the solution and a convenient pH_PZC allowing the adsorption of the present ions at the working pH.

These results show the effectiveness of this configuration in the adsorption of cobalt ions from stagnant waters.

Furthermore, when comparing Table 4 ^{10 , 36 , 37 , 38} the prepared cells with some usual materials used in the adsorption of cobalt ions, they show comparable efficiency. In fact, C-ACP gives interesting performances taking into account that the adsorption is carried out without any stirring knowing that this parameter has a big influence in the diffusion phenomena thus in the adsorption capacity.