Environmentally friendly low-cost graphene oxide-cellulose nanocomposite filter for dye removal from water

2.1 Materials

Graphite (99%), 10% v/v Sulphuric acid (H₂SO₄), Nitric acid (HNO₃), Potassium permanganate (KMnO₄), Copper Sulfate (CuSO₄), Sodium Hydroxide (NaOH), Ammonium Hydroxide (NH₄OH, 30%), Sodium Carbonate (Na₂CO₃), Methylene Blue (C16H18ClN3S) were obtained from Sigma Aldrich (India). All reagents were of AR grade ≥ 98.5% and used as received without further purification. PTFE filter paper was used. Standard cotton was used as a source of cellulose and a Peepal leaf was procured from a healthy tree. Double distilled water was used throughout the work.

2.2 Methods

2.2.4

2.2.4 Preparation of the peepal leaf filters

Fresh Peepal leaves were collected and cleaned with water to remove any dirt. A 10 wt% solution of washing soda (Na₂CO₃) and double distilled water was prepared and the leaves were boiled in it for 6 h. They were then taken out and washed with double distilled water, after which a soft brush was used to gently rub and remove the chlorophyll to leave only the leaf skeleton, which was later left to dry (see Fig. 2(a)).

Close

Fig. 2

(a) Peepal Leaf surface preparation (b) Dried peepal leaf (c) Dried peepal leaf cut to fit the size of silica crucible (c) Optical microscopy of porous prepared peepal leaf.

Export to PPT

2.2.6

2.2.6 Characterization

The crystalline properties of Graphite oxide, Graphene oxide, bare peepal leaf, synthesized GO-C NC Peepal leaf were recorded using a Rigaku Miniflex X-ray diffractometer. XRD analysis was carried out using CuKᾳ radiation (λ = 1.5406 nm) at 40 kV voltage and 30 mA current over a 2θ angle range of 5° to 55° at a scan rate of 2°/min. The microstructures of nanoparticles were recorded using a scanning electron microscope (FEI-APREO, Thermo Fisher). ⁵⁸ The chemical composition of Graphite oxide, Graphene oxide and C-GO NC were determined from the energy dispersive spectrum (EDS) recorded. A Perkin Elmer LAMBDA 650 Series spectrophotometer was used for analyzing the filtrate collected via UV–vis spectroscopy in the range of 190 nm to 900 nm. Filter capacity and capability testing was done via the column method, and the column design is constructed as shown in Fig. 3. The filter material of 0.1 g being investigated was packed into a column with an inner diameter of 15 mm and an overall height of 75 cm. The filter material was packed below and above with cotton (0.05 g) to prevent loss of material during the water flow.

Close

Fig. 3

Basic column design.

Export to PPT

3.1 Xrd

To investigate the crystalline structures, X-ray diffraction (XRD) was performed on the graphite oxide, graphene oxide, bare peepal leaf, synthesized graphene oxide cellulose nanocomposite, in the range of 2θ from 5° to 55° and results were shown in Fig. 4.

Close

Fig. 4

XRD patterns for Graphene, Graphite oxide, Graphene oxide, Bare peepal leaf, and Graphene oxide cellulose composite samples.

Export to PPT

The XRD data for Graphene shows a single significant diffraction peak at 2θ = 26.54° (d = 0.336 nm), which corresponds to the (002) plane. It can also be observed that the intensity of the prominent XRD peak at 10.66° sharply decreased in the case of graphite oxide (GrO). This is because after the oxidation of Graphite to Graphite oxide, the diffraction peak shifts from 2θ = 26° (d = 0.34 nm) ⁶⁰ to 2θ = 10.66° (d = 0.83 nm), reflecting a significant increase in the inter-planar distance due to the addition of oxygenated functional groups. The absence of excessive peak broadening of the characteristic peak of Graphite shows that the stacking was well ordered. Similar observations were reported in the literature.

After oxidation/exfoliation of graphite oxide to graphene oxide, these peaks are shifted to smaller angles; here, to a 2θ value of 10.2°, which corresponds to d = 0.87 nm using Bragg’s Law. This increased interplanar distance is due to a further introduction of additional oxygen moieties and oxygenated groups on the GO surface. Similar observations were reported in the literature. In addition to this, the disappearance of the sharp peak in graphene oxide can be attributed to the exfoliation of layered structures of graphene oxide. The broader peak may stem from the partial restacking of exfoliated graphene oxide layers, which effectively prevented the restacking of the graphene layers. The XRD patterns of peepal leaf were shown in Fig. 4. It was demonstrated that the prominent peak was observed at a 2θ angle of 26.8° and another peak at a 2θ angle of 28.1°. The peepal leaf was observed to be semi-crystalline in structure. It can be suggested that the lower crystal size structure tends to absorb more water than the higher crystal size structures. In all cases, more intense peaks indicate a more crystalline structure for the sample in question, implying that the GO-C composite was the most crystalline, followed by Graphene, Peepal leaf, Graphite oxide, and Graphene oxide.

3.2 Sem

The surface morphologies of graphite oxide and graphene oxide samples were characterized using a scanning electron microscope (SEM) and the figures were shown in Fig. 5(a and b) respectively.

Close

Fig. 5

SEM images of (a) Graphite oxide, (b) Graphene oxide, (c) Graphene oxide cellulose composite on the leaf at the center, (d) Graphene oxide cellulose composite on the leaf at the edge.

Export to PPT

In Fig. 5(a), the SEM image of GrO exhibits its standard layered structure, identical to graphite, which is due to the addition of a huge number of oxygen-containing groups, resulting in an increase in the interlayer spacing and additionally making the atomic-thick layers hydrophilic. In Fig. 5(b), the SEM image of GO, the creased overlapping exfoliated sheets containing only one or a few layers of carbon atoms like graphene can be observed. The particulate structure and an irregular, wrinkled surface of GO are observed as well. This may indicate a large number of “defective” adsorption sites favorable for capturing dye molecules. Similar observations were reported in the literature. (Feng et al., 2020) Furthermore, to analyze the surface morphology in the graphene oxide cellulose composite, SEM was performed. The GO nanofillers were uniformly distributed in the cotton fibers and confirmed the completion of preparation. The optical microscope image of the bare leaf shown in Fig. 5(c) was favorable results in that the surface was almost free of any leaf blade remnants and the porous leaf support structure was intact, and not degraded by the use of chemicals. This further established the usage of peepal leaf as appropriate support for the composite. The microstructure of the GO-cellulose composite deposited on the leaf at the center and the edge were characterized using SEM was shown in Fig. 5(c and d) respectively. In Fig. 5(c), the midrib and the porous structure of the treated leaf can be seen, and the composite is loaded on it uniformly. In Fig. 5(b), the layered nature of the leaf blade at the edge is seen, and the expectancy of uniform and proper embedding of the composite is achieved and further confirmed.

Fig. 5(c and d) reveals the interaction between graphene oxide and cellulose, and the network structure of GO cellulose composite loaded on the peepal leaf. This microstructure of GO cellulose composite loaded on the peepal leaf can help to exhibit better adsorption efficiency than its parental ingredients due to this system’s observed rough surface, increased porosity and enhanced surface area. The SEM images of GO cellulose composite-leaf samples were shown that the incorporation of GO did not create significant morphological changes in the surface roughness; composite film surfaces were similarly homogeneous and smooth. Nonetheless, smoothness observed in the SEM images indicates the GO sheets were uniformly dispersed, demonstrating films with high homogeneity without aggregation. Therefore, the GO oxygenated groups might be interacting with fibres of cellulose in the filter can provide good compatibility between the GO filler and the composite.

3.3 Column study and UV spectroscopy

For analysing the capabilities of water filtration of the various materials, dye removal tests were performed at set wavelengths. The column studies performed as batch experiments. On analyzing a pure MB sample to ascertain the reference value of absorbance to be subtracted from the data points observed for filtrates from different samples, the characteristic peak wavelengths values in order of decreasing absorbance were observed at 663.71 nm (1.3 A), 291.51 nm (0.8 A), and 245.45 nm (0.5 A). The λ_max, or the most significant peak, for MB, was observed at 663.71 nm, and dye removal tests were performed at this set wavelength to check the water filtration and dye removal capabilities of the various materials. The molecular structure and UV-Vis spectrum of MB are shown in Fig. 6(b).

Close

Fig. 6

(a) Calibration curve of Methylene Blue (MB) dye, (b) Molecular formula and UV–vis spectrum of methylene blue with λ_max at 663.71 nm, (c) Comparison of removal efficiencies of all the materials used for dye removal via column studies.

Export to PPT

In the pure MB sample, the characteristic peak wavelength values in order of decreasing absorbance were observed at 663.71 nm (1.3 A), 291.51 nm (0.8 A), and 245.45 nm (0.5 A). The λ_max, or the most significant peak, for MB, was observed at 663.71 nm. The molecular formula and UV–vis spectrum of methylene blue are shown in Fig. 6(b). Initially, the absorbance of methylene blue with concentrations 10 ppm and 20 ppm was investigated by passing 15 mL of solution (per cycle) through the column and subsequent UV–vis spectrophotometer measurement at λ_max = 663.71 nm for graphene oxide, graphite oxide, and graphene oxide-cellulose composites (1:1, 2:1, 4:1, 1:2, 1:4 wt%. graphene oxide-cellulose ratios); this was termed ‘Cycle-1′. This procedure was repeated till 12 cycles, or the system got choked, i.e. there was no filtrate- whichever occurred last.

Furthermore, MB solutions of different concentrations were treated through the columns in batches of 15 mL, the time taken for the solutions to filter out completely was noted. The resulting filtrates were analysed using UV-vis spectroscopy to obtain the absorbance, and consequently, the concentration, at 663.71 nm. Double distilled water was used as blank and the absorbance value of double distilled water was subtracted from all the recorded values for the samples. A calibration plot was constructed for MB to convert the absorbance into the concentration values for easier and more practical analysis using the Beer-Lambert Law as shown in Eq. (1) below: $$\alpha =k\ast c\ast l$$ where α is the absorbance, k is the proportionality constant, c is the concentration of the sample being analysed, and l (cm) is the length of the path (width of the cuvette between the two translucent sides).

The formula used to convert absorbance to concentration in this paper is: $$\alpha =\left(k\ast l\right)\ast c+R$$ $$c=\left(\alpha +0.28748\right)/0.15314$$

Since slope = k * l and R = y-intercept of the calibration curve’s trendline with an R² (regression coefficient) value of (Fig. 6(a)).

The regression coefficient R² was 0.9403 as shown in Fig. 6(a). To observe dye removal using bare peepal leaf, graphite oxide loaded peepal leaf, graphene oxide loaded peepal leaf, and graphite oxide cellulose nanocomposite loaded on peepal leaf filters, a silica crucible was placed on a conical flask, and the filter material was loaded. Subsequently, 15 mL of MB was transferred in regular intervals of 5 mL each to avoid overflow owing to the small size of the crucible. From Fig. 6(c), it can be understood that the filter choked out after 12 cycles (filtrations) (13th in series). During the last cycle (12th) for 2:1 graphene oxide-cellulose nanocomposite material, it was found that 10 ppm of MB was reduced by 80.76% based on filtrate absorption studies. Overall, the composite materials of graphene oxide, graphite oxide, graphene oxide and cellulose (1:1, 2:1, 4:1, 1:2, 1:4 wt ratios), bare peepal leaf), the absorbance decreased slowly. Graphite oxide was very effective with lower absorbance than graphene oxide till the eighth cycle. Since GrO and GO are negatively charged adsorbents (Yu et al., 2017), it exhibits an especially high affinity for cationic dyes like MB due to the strong electrostatic attraction observed between them. (Self-Assembly, 2021) Because of these advantages, GO (and GrO) has proven to be a promising adsorbent for the removal of cationic dyes like MB from water. Furthermore, there is no leaching of MB despite the number of cycles of filtration; only the choking of the bed was a hindrance which is to be expected while dealing with nanomaterials due to their small size, the tendency of aggregation, and the resulting lower porosity. This non-leaching of dye could prove to be a benefit of using bio-nano composites for water filtration.

During the last cycle (12th) for the synthesized 2:1 GO-C (Graphene oxide-cellulose) nanocomposite material, it was found that 10 ppm of MB was reduced by 87.82% based on the observed filtrate absorbance. For the 1:4 GO-C NC Peepal leaf composite, the removal efficiency was 80.17%, and comparable to standard adsorption materials like Graphene oxide and Graphite oxide. For all the materials analyzed (graphene oxide, graphite oxide, graphene oxide, and cellulose (1:1, 2:1, 4:1, 1:2, 1:4 wt ratios), bare peepal leaf), the adsorption removal efficiency decreased very gradually with subsequent cycles. This is because the number of adsorption sites left decreases as the number of MB molecules adsorbed on the material keeps increasing. According to the average values throughout the 12 cycles, the decreasing order of average efficiency across cycles and purification performance was:

GO-Peepal (80.93%) > Graphite oxide (80.82%) > 2:1 GO-C NC (80.76%) > 1:4 GO-C + Peepal (80.17%) > Graphene oxide (79.61%) > 4:1 GO-C + Peepal (73.85%) > 1:2 GO-C (68.48%) > Bare Peepal leaf (41.44%).

Fig. 6(c) shows that the filter choked out after 12 cycles (filtrations) (13th in series). Additionally, no leaching of MB was observed despite the number of cycles of filtration; only the choking of the bed was a hindrance- which is to be expected while dealing with nanomaterials due to their small size, the tendency of aggregation, and the resulting lower porosity. This non-leaching of dye could prove to be a benefit of using (bio-) nanocomposites for water filtration. Finally, noticed that the GO-C composites have comparable removal efficiencies to traditionally used and well-developed and commercialized materials like GrO and GO.