Study of the hydraulic correlation in the removal of pollutants from synthetic wastewater by means of a filter with Musa Paradisiaca

Introduction

The wastewater produced from anthropic origin produces alterations and environmental problems worldwide, these change the characteristics of ecosystems and encourage the spread of disease (Meneses Barroso et al., 2018). The textile industry is the main source of contamination of water sources, as its processes use chemicals, detergents, oils and others. For example, chemical dyes used by the textile industry for garment dyeing generate wastewater with intense colors, and disposal into the riverbeds produce changes in their ecosystem because they do not allow the passage of sunlight for the process of self-purification (Zhongming et al., 2021).

The textile, food, beverage and paper industries are the largest contributors to water resource pollution. Of the pollution that each of these productive sectors contributes 14.6% of industrial water pollution, worldwide, is associated with the textile industry. These figures refer to so-called low and middle income countries, located in Asia, Latin America, the Middle East, North Africa and the Caribbean (Grakova et al., 2021).

Waste water from the textile industries has been a subject of study for many years because of environmental problems, since it has a high biological oxygen demand (BOD) and chemical oxygen demand (COD) which are discharged at high temperatures and can present heavy metals and other complex chemicals threatening the environment in which they are discharged (Feuzer-Matos et al., 2022).

One of the most relevant indicators of water pollution is color, this parameter is mainly associated with the textile industry that uses color diversity (Montoya Alvarez, 2019). Dyes are generally organic compounds that have the ability to impart color to fibers of various origins, whether textile, as well as leather, paper, plastic and even food (Jabar et al., 2020). For a substance to function as a dye it must have an adequate color, ability to attach to the tissue and resistance to the action of water. There are more than 100000 types of dyes available commercially and it is estimated that around 7000 new dyes are produced annually (Saxena & Gupta, 2020).

The environmental impact generated by these toxic substances has led the scientific community to develop different methods for the treatment of industrial effluents such as: precipitation, oxy-reduction, ion exchange, filtration, electrochemical treatment, membrane technologies and evaporative recovery. However, these methods have been quite costly, in addition to the formation, disposal and storage of sludge and waste, originating during the processes, which becomes a major problem to be solved (Ramírez Calderón et al., 2020).

Because of this bioadsorption processes arise as an alternative for the treatment of effluents produced by the textile industry using as a sorbent different materials of biological origin that are low cost and are in great abundance in nature. Moreover, its biosorbent transformation is not a costly process by treating agricultural waste that previously had no use (Elgarahy et al., 2021). That is, this method mainly seeks the removal of heavy metals in wastewater from the industrial sector, using different biological sorbent materials such as: algae, fungi, bacteria, fruit peels, agricultural products and some types of biopolymers. These materials are inexpensive and are found in great abundance in nature, moreover, their biosorbent transformation is not a costly process (Corral-Bobadilla et al., 2019).

The bioadsorption process involves a solid phase (biomass) and a liquid phase (water) contains dissolved the contaminant to be treated (in this case, the ions of heavy metals). In order for the bioadsorption process to be carried out successfully, there must be a high affinity between the functional groups of the biomass and the pollutant, since the phenomenon of bioadsorption of metal ions, using biological materials as adsorbents, can be performed by various physico-chemical and metabolic mechanisms in which the process of capturing heavy metals may differ (Das et al., 2021).

In general, the extraction of metals by residual biomasses is attributed to their proteins, carbohydrates and phenolic components containing carboxyl, hydroxyl, sulphate, phosphate and amino groups, which have a high affinity for metal ions, facilitating their capture (Corral-Bobadilla et al., 2019).

Three classes of adsorption can be distinguished according to the type of attraction between solute and adsorbent. If the adsorption is given by the ion exchange in which the ions of a substance of interest are concentrated on a surface of the adsorbent material as a result of the electrostatic attraction between them, the adsorption is said to be of an electrical type. However, if the adsorbed molecule is not fixed at a specific place on the surface, but rather is free to move within the interface, adsorption is said to be due to van der Waals forces or also called physisorption. Incidentally, if the adsorbate forms strong bonds located in the active centers of the adsorbent, it can be said that adsorption has a chemical nature. It should be noted that in physisorption, the adsorbed species retains its chemical nature, while during the chemisorcion, the adsorbed species undergoes a transformation leading to a different species (Ratri et al., 2022).

There are variables that affect the phenomenon of bioadsorption such as temperature, pH, particle size and the presence of other ions. An elevated temperature may cause a change in sorbent texture and a deterioration of the material leading to a loss of sorption capacity (Corral-Bobadilla et al., 2019). The pH of the aqueous solution is an important parameter that controls the adsorption processes of metals in different adsorbents, due to the fact that hydrogen ions constitute a highly competitive adsorbate. The adsorption of metal ions depends both on the nature of the adsorbent surface and the distribution of the chemical species of the metal in the aqueous solution. The pH value of the aqueous phase is the most important factor in both cation and anion adsorption, the effect being different in both cases. Thus, while the adsorption of cations is usually favored for pH values higher than 4.5, the adsorption of anions prefers a low value of pH, between 1.5 and 4 (Sepulveda et al., 2022). Particle size effect adsorption takes place mainly inside the particles, on the pore walls at specific points. The amount of adsorbate (solute) that can be adsorbed is directly proportional to the volume, and it is well known that this volume is directly proportional to the external area and also that a small particle has greater surface area, greater area of the inner surface by the amount of pores per unit of mass. The presence of ions in the solution allows them to compete with metal in the interest of sorption zones (Corral-Bobadilla et al., 2019).

Among some of the wastewater treatment with vegetable fibers (Bioadsorbents) we can mention:

Evaluation of the biosorbent power of orange peel for the removal of heavy metals, Pb (II) and Zn (II). In the present work we investigated the biosorption of Pb (II) and Zn (II) from the biomass of dried orange peel, crushed, with and without crosslinking treatment (with CaCl2). Of the 8 experiments, it was found that for Pb (II) showed removal percentage of 99.5 % with removal capacity of 9.39 mg Pb (II)/g orange peel. The best percentage of Zn (II) removal was 99.5%, whose removal capacity was 9.95 mg of Zn (II)/g orange peel (Fola Akinhanmi et al., 2020).

Use of biosorbents for nickel and lead removal in binary systems. The adsorption of Pb (II) and Ni (II) on yam shells and palm bagasse was systematically studied in an individual and binary system, reaching for yam shells a maximum adsorption capacity of 362.45 for nickel and 68.14 mg/g for lead. For palm bagasse, an adsorption capacity of 162.64 mg/g was estimated for nickel and 90.28 mg/g for lead. In a binary system, an antagonistic effect was observed for the combined action of metals, although the removal of lead was significantly increased in yam shells when in solution accuses with nickel (Benettayeb et al., 2021).

Adsorption of Cr⁺⁶ by Cocos nucífera L. in residual of Fibrocemento in Santiago de Cuba. This work studied the adsorption of Cr⁺⁶ using the fruit shell of the Cocos nucífera L. plant as organic biomass. The optimal adsorption values of Cr⁺⁶ are: pH of 3 units; particle size less than 0.074 mm; adsorbent dose of 5 g.dm-3 and contact time of 1 hour. At low metal concentration values (1,0; 1.5 and 1.84 mg.dm-3) was obtained removal percentage greater than 90, however at high concentration values (2.5 and 3 mg.dm-3), values lower than 90 % are obtained (Itankar & Patil, 2022).

Within this line of research, the work aims to decontaminate synthetic water prepared from the BRL blue dye with characteristics similar to that of a textile industry, to determine the correlation between the flow rate, permeability, removal of heavy metals (hexavalent chromium and copper), chemical oxygen demand (COD) and colour using a filter bed with paradisiacal musa fibre, so that future studies can determine the feasibility of the actual implementation of fiber at the textile industry level based on the results obtained in this research.

Methods

The research work is experimental, and the procedure was divided into 5 parts: the characterization of the fiber musa paradisiaca, elaboration and characterization of the synthetic water, application of the natural fiber as filter bed and the characterization of the treated water. ANOVA analysis of two factors was performed to treat the results.

In the characterization of the natural fiber musa paradisiaca was used from the data reported in the work of Aucancela (2018), where the following were reported: pH, humidity percentage, ash percentage, organic carbon determination, lignin percentage, fiber density, porosity and permeability.

For the processing of synthetic wastewater, the color parameter was taken as the basis, considering a value of 4540 U Pt-Co for high concentrations because in the study "Analysis of the fiber obtained from the rachis of the musa paradisiaca plant, used as a filter bed in the adsorption of the color parameter" was observed that for high color concentrations (41305 U Pt-Co) the fiber is not effective in color adsorption (Aucancela, 2018, pag 20-25). Three wastewater types were made using the blue dye BRL in concentrations as indicated in Table 1. This in the drinking water base which has electrical conductivity values of 593 μS/cm, 7.1 pH, Cr⁺⁶, COD and copper with zero values.

The characterization of the synthetic wastewater was carried out, the pH and conductivity analysis with the multiparameter HACH HQ40d, for which the sample of 100 mL of synthetic water was placed in a beaker and the reading was done with the equipment. Color analysis, COD, hexavalent chromium (Cr⁺⁶) and copper (Cu) were performed using the HACH DR500 spectrophotometer following the different processes stipulated in the water analysis manual (Davidson et al., 2022; Hach Company, 2000):

Color: use UV/VIS DR5000 equipment with parameter code 120 and wavelength at 455 nm (nanometers), take a deionized water sample in the 25 mL cell and use as white and mark as ZERO in the equipment, the sample is shaken for 1 minute and a sample of 25 mL is taken in the cell and placed on the computer, READ is clicked, and the value is recorded.

COD: The sample is shaken for 1 minute, then a 2 mL aliquot is taken with a graduated pipette and poured into the high-range COD-AC vial (0-1500 mg/L COD), in the same way 2 mL of deionized water is taken and poured into the high-range COD-AC vial (0-1500 mg/L COD), the vials are shaken gently with the samples and placed in the thermal reactor at 150 °C for 2 hours, after which the vials are allowed to cool. Then search the code 430 in the UV/VIS DR5000, use the vial with deionized water is clicked on ZERO, then the vial is placed with the synthetic water sample and READ is clicked, the COD value in mg/L is recorded.

Cr⁺⁶ and Cromo were samples sent to be analyzed at the Environmental Services Laboratory of the National University of Chimborazo, whose methods correspond to those described in Standard Methods Ed 21 (Rice et al., 2017)

The flow cell used for treatability tests has the following internal dimensions: internal diameter D = 6.4 cm and height L = 15.3 cm. The flow rate was controlled by a peristaltic pump obtaining high speed (V3 = 0,86 ml/s), average speed (V2 = 0.45 ml/s), low speed (V1 = 0.22 ml/s). The permeability was controlled by cutting the length of the fiber, so the treatment will be with normal fiber (Fn), fiber cut to 1cm (F1), to 2cm (F2) and to 3cm (F3). Table 2 shows the experimental design used and the abbreviations given for each of the treatability tests.

Each test included a treatment lasting 180 minutes at continuous flow, pH, conductivity and color values were taken every 5 minutes giving a total of 36 samples, of COD, Cr⁺⁶ and Cu were analyzed at the beginning of treatment and in minutes 1, 5, 30, 60, 90, 120 and 180 giving a total of 8 samples. The same procedures and parameters described above were performed for the characterization of treated water.

Finally, the Statistical Analysis was obtained the average of the 3 values obtained in the laboratory over the 180 minutes of treatment, applied data smoothing, this in order to correct erroneous values. The method used is a simple exponential smoothing, this method contains a self-correcting mechanism that adjusts forecasts in the opposite direction to past errors. It is a particular case of weighted moving averages of current and previous values in which weights decrease. It is used for both smoothing and forecasting (Wei & Chen, 2022). The equation used is (1):

Where: ${\widehat{X}}_t$ = Average sales in units in the period, t, α = smoothing constant, ${\widehat{X}}_{t-1}$ = forecast sales in units of the period t-1,$X_{t-1}$ = Actual sales in units of the period t-1.

The data obtained in the experimentation were evaluated by the statistical analysis of the ANOVA variance of 2 factors with a significance level of 95%. The percentage of removal or adsorption of each of the analyzed parameters was obtained with the initial value and the value obtained at the end of the treatment. Finally, the coefficient correlation between the flow rate vs the percentage of removal of pollutants and the permeability vs the percentage of removal of pollutants was calculated to find whether or not there is a correlation between them.

Results and Discussion

Table 3 reports the results of the physico-chemical characterization of the fiber and Table 4 the hydraulic characteristics of the musa paradisiaca fiber obtained in the flow cell.

The values regarding the lignin percentage, ash percentage and density obtained in this research are similar to those reported by (Aucancela, 2018).

The permeability data allow us to observe the directly proportional relationship between the fiber cut and the permeability, the same has been observed with porosity whose data are close to the values reported in a study of hydraulic characterization of coconut fiber (similar to the paradisiacal muse) with a porosity of 81% (Luis Huertas Blanco et al., 2019).

Table 5 shows the results of the characterization of the three types of synthetic water. As can be seen, synthetic wastewaters have color, Cr⁺⁶ and Cu concentrations similar to the characteristics of effluents from textile industries. Regarding conductivity, pH and COD, the values obtained are low because these parameters are influenced in the textile industry mainly by the dyeing and finishing processes making contact with other chemicals

Table 6 shows the removal of contaminants in high concentration synthetic water after 3 hours of filtering water. The behavior for each parameter was different: Thus for the pH values are shown between 7.61 and 8.12 observing that the initial value of 8.16 is not altered staying within the range (5.5-9.5) for discharges to freshwater bodies (OFFICIAL RECORD, 2015). The conductivity presents mostly positive values whose results are better while increasing the flow velocity, reducing up to 22.33% in the F1CaV3 treatment. The color shows reduction values between 48.27 and 65.15% for each treatment. In the COD, better results were obtained while increasing the flow velocity, reducing up to 63.39% in the F1CaV3 treatment. Cr⁺⁶ reduction percentages are favorable for each treatment whose values are between 59.74 and 80.82% of removal. Finally, for Cu the percentages of reduction present fluctuations that, for example, in the treatment F3CaV2 is 18.03%, a small value relative to that obtained in the treatment F3CaV1 that was 83.54%.

In Table 7 pH remains under regulation. The conductivity shows positive removal values only for high flow rates by removing up to 7.73% in the F1CmV3 treatment. In color, reduction values are found between 52.71% and 69.67%. The COD presents almost mostly negative values meaning that, it was not possible to lower the concentration of this parameter but rather increase it by factors such as ripeness and decomposition, the presence of fine particles and even the presence of water-soluble salts presumed to possess the fiber. The percentages of reduction of Cr⁺⁶ are between 43.29 and 64.74% of removal and those of copper range between 45.92% and 88.69%.

Table 8 shows the results obtained for each low concentration treatability test after 3 hours of filtering the water. For synthetic concentration wastewater, pH remains under regulation. The conductivity and the COD have almost mostly negative values meaning that, this parameter was not lowered but rather increased by the explained above. The color is between 51.14% and 67.93%, Cr⁺⁶ between 30.18% and 53.95% and Cu between 40.21 and 87.31%. Because unusual behavior was observed at the start of treatment in the treatability tests with the three concentrations of synthetic water referring to conductivity parameters, color and COD, it was necessary to apply more treatability tests with the pre-wash of the fiber, since it provides values to these parameters.

The conductivity, color and COD values with which the fibre provides are presented in Table 9 below. Obtaining an average of 39187.75 μs/cm, 446.92 UPtCo, 5215.83 mg/l, for conductivity, color and COD respectively. Among the factors that influence the contribution of conductivity, color and COD by the fiber is its state of ripeness and decomposition, the presence of fine particles and even the presence of water-soluble salts that the fiber is presumed to have.

It should be noted that, although fiber provides concentration values for conductivity, color and COD, this problem was solved by washing the fiber favoring treatment since, the fiber after washing did not lose its adsorption characteristics.

Table 10 shows the results obtained for each treatability test after a 3-hour treatment by the filter bed for synthetic water of low concentrations with pre-wash of the fiber. The conductivity despite the washing of the fiber has almost mostly negative values that means that this parameter was not lowered, but it was reduced compared to the values obtained in Table 10. In color, the percentage of removal is between 69.86% and 77.92% and in the case of COD better results were obtained while the flow rate is increased and the fiber cut is lower. The pH presents percentages between 1.34% and 6.18%, the Cr⁺⁶ percentages of removal between 32.78% and 46.95%. Copper reported percentages between 32.35% and 79.60%.

The two-factor ANOVA is based on two hypothesis tests a null H0 and an alternative H1 that evaluate the two categorical variables. H0 states that all treatments are the same, meaning that there is no variation between the results in the treatability tests. H1 states that not all treatments are the same, meaning that there is variability between the results of one treatment and the other. The ANOVA variance analysis of two factors was performed using GNU PSPP open software and its statistical package Data Analysis for each parameter for all treatments according to the type of synthetic wastewater as follows with a confidence level of 95 %:

Table 11 shows the results of the ANOVA analysis of two factors for both the velocity variable and the permeability variable with respect to the removal of pollutants from synthetic wastewater of high, medium and low concentrations. The obtained values dictate that in the case of pH the hypothesis H0 was accepted for all concentrations in the variable permeability and the variable speed H0 was accepted for high and medium concentrations, that is, similar results are obtained in these treatments, however for Fsinlav. Cb and Flav. Cb treatments H1 was accepted which means that, in these treatments different results were obtained. In conductivity, H0 was accepted for all concentrations in the variable permeability, determining that all treatments are equal and H1 was accepted for all concentrations in the variable speed, determining that not all treatments are equal. Color hypothesis tests determined that in treatments where no fiber pre-wash was performed all treatments are the same in the two variables accepting H0, this is because, as already observed in the laboratory experiments the fiber treatment without prewash alter them, However, when prewashing is performed, conditions improve, which allows us to observe that there is variation in each treatment, accepting H1 in the permeability and velocity variable.

In COD H1 it was accepted for all concentrations in the variable speed. In the variable permeability only for low concentrations with pre-wash of the fiber H1 was accepted, while H0 was accepted for treatments of the different concentrations without pre-wash Cr⁺⁶ accepted H1 for each of the treatments in the two variables, which means that different removal results were obtained in each of the treatability tests performed for this parameter. Finally, for Cu, the flow velocities are adjusted to H1while in permeability H0 was accepted in each of the tests at different concentration.

To obtain the correlation coefficient, the removal percentage was correlated with the different velocities at the same permeability and the different permeabilities at the same velocity for each parameter at all concentrations. The correlation coefficient will allow us to know if the flow rate or permeability are related to the removal of pollutants where if the value approaches 1 the correlation will be strong or perfect and if it approaches 0 the correlation will be weak or zero. The sign determines whether it is directly proportional (positive sign) or inversely proportional (negative sign). Table 12 below shows the correlation coefficients obtained between the different permeabilities and velocities for each treatment according to its concentration.

Although the correlation coefficient was obtained for all treatments, the analysis of those tests in which H1 was accepted in the ANOVA analysis of two factors and which are highlighted in bold in Table 13. Thus, at pH the results indicate the velocity and permeability does not influence the removal of this parameter. In the conductivity a significant and perfect correlation was obtained, that is to say, the flow velocity if it influences in a directly proportional way since, at higher speed, greater percentage of conductivity removal. In color, the correlation of permeability is strongly inversely proportional, since the coefficient values are between -0.85 and – 0.95 and in the case of velocity the correlation is directly significant and strong since the coefficient values are between 0.70 and 0.95. Which means that color removal in the pre-wash treatment of the fiber is more effective at lower permeability and higher flow rate. In the COD the variable velocity, the correlation are between 6 and 1, i.e. between moderate and perfect which means that the higher speed is, the decrease in COD, while for the variable permeability correlation coefficients are between -8.13 to -8.88, that is, the correlation is significant and strong which means that the lower permeability the removal of COD is greater. In the Cr⁺⁶, the coefficients of correlation between significant and perfect that are greater than -0.70 so the lower the permeability and the flow velocity, the removal of hexavalent chromium will be more effective. Finally, in the case of copper, it was determined that there is no suitable correlation between the flow rate and the removal of contaminants.

The following (Table 13) explains the best treatments for each parameter in the different concentrations:

Conclusions

The statistical analysis determined that the flow rate and permeability variables are correlated with the removal of conductivity, color, COD and Cr⁺⁶ while pH and Cu are not correlated.

For the permeability variable the correlation is inversely proportional while for the velocity variable in the case of conductivity, color and COD the correlation is directly proportional and for the case of Cr⁺⁶ the correlation is inversely proportional.

The best treatment was to use 1 cm of fiber and a high flow rate at which 23.37% was removed; 77.92% and 70.01% for color conductivity and COD respectively. For the Cr⁺⁶ the best treatment was with fiber at 1 cm and a low flow rate removing up to 80% the concentration of this contaminant. While for Cu the best treatment was fiber at 3 cm and low flow rate removing between 79% and 89%. Thus, proving that the filter bed of musa paradisiaca is effective for the removal of these contaminants.

Despite obtaining good percentages of Cu removal, Cr⁺⁶ and color, in synthetic water of high concentration do not comply with the permissible limits established by Ecuadorian legislation so it is recommended to use the filter bed especially for wastewater of low concentrations or that have been treated in physical - chemical operations.