Improved defluoridation and energy production using dimethyl sulfoxide modified carbon cloth as bioanode in microbial desalination cell

Abstract

In the present study, carbon cloth (CC) was functionalized using dimethyl sulfoxide (DMSO) and employed as an excellent bioanode for improving defluoridation efficiency, wastewater treatment, and power output from a microbial desalination cell (MDC). The Raman spectroscopy and X-ray photoelectron spectroscopy (XPS) analysis of DMSO modified carbon cloth (CC_DMSO) confirmed the functionalization of CC_DMSO, and the water drop contact angle of 0° ensured its superior hydrophilicity. The presence of –COOH (carboxyl), S O (sulfoxide) and O–C O (carbonyl) functional groups on CC_DMSO aids in enhancing the performance of the MDC. Besides, cyclic voltametric and electrochemical impedance analysis revealed that CC_DMSO had an excellent electrochemical performance with low charge transfer resistance. Replacing CC with CC_DMSO as anode in MDC, the time required for 3,10 and 20 mg/L of initial fluoride (F⁻) concentrations in the middle chamber was reduced from 24 ± 0.75 to 17 ± 0.37, 72 ± 1 to 48 ± 0.70, and 120 ± 0.5 to 96 ± 0.53 h, respectively to meet the prescribed standards (1.5 mg/L). Furthermore, using CC_DMSO, the anode chamber of MDC exhibited a maximum of 83% substrate degradation, and simultaneously, the power output is increased by 2–2.8 times. CC_DMSO improved the power production from 0.009 ± 0.003, 1.394 ± 0.06 and 1.423 ± 0.15 mW/m² to 0.020 ± 0.07, 2.748 ± 0.22 and 3.245 ± 0.16 mW/m², respectively, for initial F⁻ concentrations of 3,10, and 20 mg/L. Modifying CC with DMSO thus proved to be an efficient and simple methodology for enhancing the overall performance of MDC.

Graphical abstract

Highlights

• •Dimethyl sulfoxide modified carbon cloth (CC_DMSO) is used as bioanode in MDC.
• •MDC is used for defluoridation of water along with bioelectricity production.
• •Oxygen functional groups on CC_DMSO improved the electrochemical performance of MDC.
• •CC_DMSO reduced the time required to achieve the permissible limit of fluoride in water.
• •The power output from MDC increased by 2–2.28 times using CC_DMSO.

1. Introduction

Geogenic-originated groundwater pollution is a topic of major concern, and fluoride (F⁻) stands first among them. F⁻ is introduced into the environment by both natural (rocks containing F⁻, volcanic ash) and anthropogenic sources (industrial effluents from glass, phosphate fertilizers, semiconductors etc.) [1]. Unlimited tapping of groundwater also accelerates the leaching of F⁻ into the groundwater [2]. Drinking water with F⁻ concentrations up to 1.5 mg/L is advantageous as it checks dental decay, but concentrations >1.5 mg/L leads to chronic diseases like dental fluorosis and skeletal fluorosis [3]. The Geographical Survey of India has branded several areas under the “fluoride red alert” category, for instance, a concentration as high as 21 ppm was reported in Andhra Pradesh, India [4]. In recent years, more incidences of dental caries, twisted limbs, and yellow teeth have been reported among people of all ages in India [5]. As fluorosis is a prevalent disease, defluoridation of drinking water is highly essential for arresting the endemic effects on human health and physiology.

Several efficient methodologies are in existence for the defluoridation of groundwater. Defluoridation can be accomplished by coagulation and precipitation [6], membrane separation processes [7], adsorption [[8], [9], [10]], electro-dialysis [11], and electro coagulation [12,13]. Although the traditional treatment techniques improve water quality, most of them are either energy-intensive or demands chemical addition, which eventually generate large amounts of sludge [14]. The produced sludge requires further treatment prior to disposal. For instance, in conventional Nalgonda technique 1 kg of alum and 0.5 g of lime produces approximately 0.37 kg and 0.4 kg of sludge respectively [6]. Hence, it necessitates switching to a viable energy-positive technology with a reduced burden of secondary pollutants.

Recently, microbial desalination cells (MDCs) have gained wide attention as they are capable of desalinating water while simultaneously treating wastewater with energy recovery [15,16]. The MDC is a three chambered reactor consisting of an anodic chamber followed by a desalination chamber. An anion exchange membrane (AEM) is used to separate the two chambers. A cathodic chamber is provided after the desalination chamber, and the chambers are parted by a cation exchange membrane (CEM). MDC works on the principle of oxidation of organic matter by microbes in the anolyte and reduction of oxygen in the catholyte, resulting in a maximum electric potential gradient of 1.1 V (acetate = 16.9 mM; pH = 7; oxygen = 0.2 atm. (partial pressure)) [17,18]. The generated electric field drifts the anions (Na⁺) and cations (Cl⁻) through AEM and CEM, respectively, to anolyte and catholyte, resulting in water desalination in the middle chamber [19]. Using MDC, desalination can be achieved without applying pressure, or use of draw solutions, and/or the consumption of electrical energy. MDCs are flexible in treating various pollutants as they are unaffected by the nature of ions to be removed from contaminated water. Hence, research has been directed to use of MDCs for water softening [20,21], heavy metals removal [[22], [23], [24], [25]], etc. However, its application in defluoridation has not been explored very much [23,24].

Although MDC is a promising technique for contaminant removal, its overall performance is severely affected by the properties of the anode, which are crucial for the development of exoelectrogenic microbes [26]. The major characteristics for the anode are biocompatibility, high porosity, specific surface area, conductivity, mechanical and chemical stability, and low cost [27]. In MDC, commonly used anodes are carbonaceous materials like carbon felt, carbon cloth, carbon brush, graphite rods, etc. [28]. Popular among them is carbon cloth (CC), as it is biocompatible and has good conductivity for efficient electron transfer [29,30]. Application of CC electrodes is hindered due to overpotential losses associated with limited active sites and poor hydrophilicity [31,32]. Recent studies have shown that these shortcomings might be resolved by appropriately activating the electrode surface of CC; as a result, more electrons can be transmitted with high power output. Numerous activation techniques, such as surface functionalization through the introduction of different functional groups, heteroatom doping, porous structure and morphology engineering, etc., have been reported in order to improve the physical and electrochemical properties of CC [33,34]. Among these, surface functionalization of CC is a viable option as the process is simple and it reduces the activation energy at the electrode-electrolyte interface [35,36]. Although the functionalization improves the overall performance of the electrode, the major challenge/constrains lies in the modification procedure. Most of the modification processes are either complicated, energy intensive, or require catalysts for efficient functioning [37], as given in Table 1.

Table 1.

Comparison of carbon cloth (CC) anode modification procedures adopted in different bioelectrochemical systems.

Sl. No	Material used for CC modification	Modification method	Ref
1	Iodine (I2) doped polythiophene nanoparticles (PTh –NP)	1.Doping of I2 onto PTh 2.Dip dry method coating of doped PTh and I2 onto CC	[38]
2	N-doped C/Fe3O4 nanotube	1.Hydrothermal deposition of nanowire template onto CC 2.Synthesis of precursor 3.Dipping in FeCl3 to form FeOOH nanotube arrays on CC 4.Dipping in glucose and heating of CC-FeOOH to form C/Fe3O4-nanotube on CC	[39]
3	Iron nanostructure was synthesized on CC by electroplating	Electroplating in 0.16 M FeCl3 solution at 5 V for 2 min using CC as cathode.	[40]
4	MnO2, Pd and Fe3O4 nanoparticles	1. Synthesis of nanoparticles 2. Loading onto CC using binder	[41]
5	Atmospheric-pressure plasma jet processed reduced graphene oxides	1.Preparation of reduced graphene oxide paste (rGO) 2.Screen printing of rGO onto CC 3.Annealing using Atmospheric-pressure plasma jet	[42]
6	Atmospheric-pressure plasma jets.	Using N2 at an operational voltage of 275 V	[43]

Introducing active functional groups like amine (-NH₂), hydroxyl (-OH), carboxyl (-COOH), carbonyl (-C O), etc., on to the surface of carbon electrodes has been performed by common methods like acid and thermal treatment, CO₂ activation, electrochemical methods, etc. [44,45]. Recently, incorporation of oxygen-containing functional groups on carbon-based electrodes has gained popularity due to its extraordinary ability to improve the catalytic activity, surface area and electrochemical properties of CC [46,47]. Dangling bonds and structural flaws are added to the carbon structure by oxygen functional groups. This leads to an increase in surface area and a decrease in activation energy between the electrode and electrolyte. Due to the binding of polar groups such as C–OH and –COOH, a strongly hydroxylated surface will be formed. This enhances the affinity for water [46]. The enhanced hydrophilicity of electrodes in polar solvents, as a result, improves the biocatalytic activity by increasing the electron transfer between microbes and anodes [48]. Oxygen functional groups can be incorporated by treating CC with compounds like nitric acid [46], potassium ferrate [45], benzoyl peroxide [47], etc. However, these processes require multiple levels of treatment that include strong acids and/or noble metals. Hence, selecting an appropriate compound that can modify CC in a facile way is imperative.

In this work, an attempt has been made to effectively functionalize the surface of CC in a simple step using dimethyl sulfoxide (DMSO). DMSO is one such compound that enables facile introduction of an O containing functional group onto the CC surface [49]. DMSO – a common laboratory and industrial reagent, is an outstanding solvent for both organic and inorganic materials and has excellent electrochemical properties [50]. Unlike the existing complex CC modification methods as given in Table 1, DMSO can be easily applied to CC in a single step without affecting the structure of CC. To the best of our knowledge the possibility of facile modification of CC using a common laboratory/industrial solvent for bioelectrochemical applications has not been reported. The objective of the present study is to explore the feasibility of using DMSO modified CC (CC_DMSO) as an anode in MDC and compare its performance with unmodified CC for defluoridation, wastewater treatment, and power output.

2. Materials and methods

2.1. Materials

CC was procured from Sainergy fuel cell India Pvt. Ltd. and ion exchange membranes (AEM, CEM) from Ralex ion exchange membranes, India. Analytical grade chemicals - NH₄Cl, NaCl, CaCl₂, MgSO₄, NaHCO₃, KH₂PO₄, K₂HPO₄, NaF, CaCl₂.2H₂O, MgCl₂.6H₂O, NaF, and sodium acetate were obtained from Sigma Aldrich, India, and DMSO from Merck, India. All the materials were used without any purification or treatment.

2.2. Coating dimethyl sulfoxide onto electrodes

Commercially available DMSO was coated onto CC using the drop cast method [51]. CC of size 3.5 × 3.5 cm² were rinsed in deionized water and heated at 100 °C until completely dry. 5 mL of DMSO was applied to the oven dried CC, drop by drop, until the electrode was fully immersed in the solvent. This was allowed to remain in contact for 30 min at room temperature, followed by oven drying at 100 °C for 1 h. The schematic representation of the principle of coating DMSO onto CC is shown in Fig. 1. The modified electrodes were stored in a desiccator before use.

Fig. 1.

Schematic representation of the principle of coating dimethyl sulfoxide (DMSO) onto carbon cloth (CC).

2.3. Reactor setup

A three chambered MDC reactor was fabricated using acrylic sheets. The reactor constitutes of three identical chambers, each of 100 mL capacity. The first one being the anode chamber, followed by a chamber for defluoridation in the middle, and finally the cathode chamber. The chambers were separated by an AEM (3.14 cm²) between the middle and the anode chambers, and a CEM (3.14 cm²) between the cathode and the middle chambers. The reactor was made water and airtight using O-rings and clamped with stainless steel bolts. Sampling ports were provided on the top of each chamber. CC_DMSO and CC (control) were used as anodes (3.2 × 3.2 cm²), and CC was used as the cathode (3.2 × 3.2 cm²) in both cases. The electrodes were coupled physically with copper wire across a 1 Ω resistor (high current mode).

2.4. Inoculation and operation

The anode chamber was filled with synthetic wastewater. The anolyte (per L of Milli-Q water, 18.2 MΩ) was prepared using: NH₄Cl, 0.11 g; KCl, 0.036 g; MgSO₄·7H₂O, 0.09 g; CaCl₂·2H₂O, 0.014 g; yeast extract, 0.001 g; and 0.3 mL of trace elements [52]. The reactor was inoculated with anaerobically digested sludge, which was obtained from an effluent treatment plant in Ahalya Medical College, Palakkad, India. Sodium acetate was used as the carbon source for the microbes. Nitrogen gas was purged continuously to maintain anaerobic conditions in the anode chamber. Aerated phosphate buffer prepared using KH₂PO₄ (0.53 g/L); and K₂HPO₄ (1.07 g/L) (per L of Milli-Q water) was used as the catholyte [21]. 3, 10, and 20 mg/L of F⁻ solutions were synthetically readied by dissolving reagent grade NaF in deionized water and were added to the middle desalination chamber [7].

2.5. Measurements and analyses

pH and F⁻ were measured in a pH/ion selective electrode (ISE) (Thermo Scientific Orion Star A214 pH/ISE). Conductivity was monitored using a benchtop conductivity meter (Thermo Scientific Orion Star A222, portable conductivity meter). The COD (substrate degradation efficiency) of anolyte was measured using the closed reflux method as described in the standard method (APHA, 2012). The substrate degradation and defluoridation efficiencies were determined using Eq. (1).

$$\mathrm{R}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y}(\%)=\frac{A-B}{A}\times 100$$ (1)

where, A and B are the initial and final concentration of contaminant, respectively.

Voltage (V) was monitored using a data logger (DAQ6510, Keithley). As per Ohm's law, current (I) V/fixed resistance. Corresponding to a particular external resistance, the current density (mA/m²) was estimated by dividing I and power density (mW/m²) by dividing VI by the effective area of anode (m²). Polarization study was conducted by manually varying the external resistance from 1 Ω to 10.2 MΩ.

2.6. Characterization of the modified electrodes

The macroscale distribution of DMSO on CC was examined using a Raman spectrophotometer (Horiba Labram HR Evo), and the surface chemistry was investigated by X-ray photoelectron spectroscopy (XPS) (ULVAC-PHI Inc. Model: PHI5000 Version Probe III). The water drop contact angle measurements were done by an optical angle measurement device (Holmarc contact angle meter, HO-IAD-CAM-01A). The surface image of the modified CC was obtained using field emission scanning electron microscopy (FESEM) (Carl Zeiss, Gemini SEM 300), and X-ray diffraction peaks were noted in the scanning window of 10°–60° by a diffractometer (XRD Smart Lab, Rigaku).

The electrochemical analysis of electrodes was performed as a three-electrode system using a Potentiostat (Multi Autolab M204, Metrohm) and NOVA 2.1 software. The working electrode used was CC/CC_DMSO, while platinum wire being the counter electrode, and the reference electrode chosen was a saturated calomel electrode in an electrolyte solution prepared using potassium ferricyanide (5 mM), phosphate buffer (5 mM), and KCl (0.1 M). Cyclic voltametric (CV) analysis was carried out from 0.1 V to 1 V, and the scan rate was fixed at 0.02 V/s. Electrochemical impedance spectroscopic (EIS) analyses were done in the frequency range of 0.1 Hz–10⁵ Hz with an amplitude of 0.1 V. The equivalent circuit of the EIS spectra was simulated to determine charge transfer resistance (R_ct) in the software ZSimpwin 3.21.

3. Results and discussion

3.1. Chemical characterization of carbon cloth and dimethyl sulfoxide modified carbon cloth

Raman spectra of CC_DMSO show an increase in the I_D/I_G of CC_DMSO (1.106) compared to CC (0.94), indicating a lower level of graphitization and a more disorganized structure. This process, is crucial in the introduction of functional groups like –COOH and –OH onto the electrode surface and causes more flaws and dangling bonds (Fig. 2a) [46]. The obtained results follow the same pattern as those obtained in the study by Ref. [53] which reported an increase in the I_D/I_G value when graphene was doped with nitrogen. The effectiveness of reduction and chemical composition (C 1s and O 1s) of CC and CC_DMSO were evaluated by XPS (Fig. 2b). The CC_DMSO showed an additional peak for S 2p, which can be derived from the S atom present in DMSO.

Fig. 2.

(a) Raman spectrum and (b) X-ray photoelectron spectroscopic survey spectra of carbon cloth (CC) and dimethyl sulfoxide modified carbon cloth (CC_DMSO).

The high resolution (HR) scan of C 1s of CC shows a sharp and minor peak at 284.6 eV and 284.9 eV, respectively, caused by the C–C/C C group [39,48]. A peak at 285.4 eV observed on CC can be attributed to the C O group, and a small peak at 286.9 eV could be related to the C–O–C group [49,50]. CC_DMSO also exhibited similar peaks, however, after modification, the C O and C–O–C peaks were found to be enhanced, while the intensity of the peak for C–C/C C was reduced. The HR scan of O 1s of CC (Fig. 3a) shows peaks at 532.5 eV and 533.7 eV that can be attributed to the –C-OH and –COOH groups, respectively [46]. Additionally, a small peak that corresponds to the presence of O–C is observed at 530.3 eV [54] (Fig. 3b). The HR scan of C 1s of CC_DMSO shows a similar peak of O–C at 530.3 with higher intensity, which confirms the attachment of oxygen during modification (Fig. 3c). Similarly, the intensity of –COOH peak is enhanced after the DMSO modification. An additional peak at 532.9 eV is observed after DMSO modification, which can be attributed to O–C O [54] (Fig. 3d). The enhancement in the peaks of O–C and –COOH along with the new peak for O–C O and S O confirms the functionalization after the treatment. The peak observed at 532 eV in the CC_DMSO originates from the S O of DMSO [55]. A peak at 534.6 eV is observed in both samples, which corresponds to adsorbed H₂O on the CC surface [56].

Fig. 3.

The HR scans for (a) C 1s and (b) O 1s for CC, (c) C 1s and (d) O 1s for dimethyl sulfoxide modified carbon cloth (CC_DMSO).

The surface hydrophilicity of CC and CC_DMSO was investigated by performing the waterdrop contact angle measurement (Fig. 4). The contact angle between deionized water and CC was 132.2° (Fig. 4a). DMSO lowered the contact angle to 123.4° within 15 min (Fig. 4b) of contact time. The angle further reduced to 69.7° after 30 min (Fig. 4c) of contact with DMSO. The contact angle was 0° when the time of dip of CC in DMSO was increased to 60 min (Fig. 4d), attributed to the superior wetting phenomenon of CC_DMSO. The hydrophilic property of CC_DMSO was contributed by the presence of –COOH and -C-OH functional groups [42].

Fig. 4.

(a) The waterdrop contact angle of bare carbon cloth (CC). The waterdrop contact angle of dimethyl sulfoxide modified carbon cloth after: (b) 15 min contact time (CC_DMSO – 15); (c) 30 min contact time (CC_DMSO – 30); and (d) 60 min contact time (CC_DMSO – 60).

The SEM analysis for structural characterization revealed that the application of DMSO did not cause any changes to the morphology of CC (Fig. S1a and S1b). Little change in the characteristic peak intensity and position in the XRD patterns of CC and CC_DMSO (Fig. S₂) highlights that DMSO did not alter the structural characteristics of CC. The two characteristic peaks at 25.8° and 43.7° can be assigned to the (002) and (101) diffraction peaks of graphitic carbon of CC [57].

3.2. Electrochemical characterization of carbon cloth and dimethyl sulfoxide modified carbon cloth

In the electrochemical analysis, CC_DMSO exhibited a higher anodic current density (∼1.45 mA) than CC (∼0.75 mA). The improved response of CC_DMSO can be correlated with the reduced overpotential loss owing to the enhanced electron transfer and superior hydrophilic property [35]. The CC did not exhibit distinct redox peaks (Fig. 5a), signifying the presence of electric double-layer capacitive characteristics and the non-availability of considerable redox mediators, whereas a single peak was evident in the case of CC_DMSO [58]. The higher area under the CV curve and peak current of CC_DMSO could be owing to an increase in the available electroactive sites that enhance the charge transfer capability [59,60]. This implies that coating DMSO onto CC increases the number of available attachment sites for microbes and the electrochemically active surface area, which is an important requisite for a high-performance anode in MDC [39].

Fig. 5.

(a) Cyclic Voltametric and (b) Nyquist curves.

The Nyquist curves of electrodes before and after modification with DMSO are given in Fig. 5b. It comprises three sections indicating: (i) ohmic resistance - an intercept at high frequency; (ii) charge transfer resistance (R_ct), – diameter of the semicircle part; and (iii) Warburg diffusion, - straight line portion at low frequencies. The R_ct value of CC_DMSO is noticeably lower (23 Ω) than that of CC (175 Ω), indicating a decline in the resistance for charge transfer at the electrode–electrolyte interface of CC_DMSO [61]. The charge transfer resistance of CC_DMSO was reduced due to the combined effects of good electrical conductivity, wettability, and the superior electrocatalytic properties of DMSO. All the mentioned factors help in improving the electrolyte - electrode interfacial characteristics and thus lowering the internal resistance and mass transfer losses [35]. Improving the charge transfer rate can result in reducing the energy of activation required for initiating the redox reactions [31].

3.2. Defluoridation and wastewater treatment

The CC and CC_DMSO were employed as anodes in MDCs, and their performance in defluoridation and wastewater treatment were investigated. The defluoridation efficiency was evaluated by monitoring the time required to reduce the F⁻ concentration to the prescribed standard limit of 1.5 mg/L. A significant enhancement in the performance of MDC for defluoridation was observed when the CC anode was replaced with CC_DMSO. Using CC as a bioanode, 3, 10, and 20 mg/L of F⁻ ions in the middle chamber took 24 ± 0.75, 72 ± 1, and 120 ± 0.5 h, respectively, to reach the permissible level (1.5 mg/L). However, with CC_DMSO, the same level of treatment was achieved within 17 ± 0.37, 48 ± 0.70, and 96 ± 0.53 h for 3,10 and 20 mg/L, respectively (Fig. 6a).

Fig. 6.

(a) Time taken to achieve permissible limit of 1.5 mg/L of fluoride. (b) Maximum substrate degradation obtained. (c) SEM image of biofilm on carbon cloth (CC) anode, and (d) SEM image of denser biofilm on dimethyl sulfoxide modified carbon cloth (CC_DMSO) anode.

Ionic separation at the middle chamber of MDC occurs due to the electric potential gradient generated by the anode and cathode. The presence of functional groups such as –COOH, O–C, O–C O, and S O groups on CC_DMSO indicates extensive functionalization of the CC surface. Eventually, the wettability and electrochemical activity of the electrode were improved, resulting in a high electric potential gradient across the electrode. Due to hydrogen bonding between the –COOH functional group and the membrane-bound peptide bonds in the bacterial cytochromes (that are linked to the intracellular electron transfer chain), the exoelectrogenic activity of the microbes was improved [42]. This eventually increases the potential gradient. As the potential gradient and ionic removal are correlated, the migration of F⁻ ions in the middle chamber was accelerated to the anode chamber with a CC_DMSO anode [46]. In addition to the potential gradient, the concentration gradient also plays a major role in the defluoridation. Initially, the concentration gradient would be sufficiently high to accelerate the rate of ion diffusion [62]. The concentration gradient decreases with time due to the movement of ions into the anolyte. This automatically reduces the rate of ion diffusion, and as a result, higher F⁻ concentrations take more time to reach 1.5 mg/L. Therefore, the rate of defluoridation in MDC can primarily be attributed to the potential and the concentration gradient between the middle and the adjacent chambers.

Simultaneously, the substrate degradation in the anode chamber of MDC was examined for CC and CC_DMSO. The maximum COD removal was found to increase from 16.7 ± 1.7 to 19.66 ± 3.5% with 3 mg/L of initial F⁻, from 42.6 ± 2.47 to 54.1 ± 1.92% with 10 mg/L of initial F⁻, and from to 70.2 ± 1.09 to 83 ± 1.68% with 20 mg/L of initial F⁻ using CC and CC_DMSO, respectively (Fig. 6b). The improved hydrophilicity and electrochemically active sites of CC_DMSO facilitate more bacterial colonization, and as a result, enhanced substrate degradation can be obtained [43]. This was further validated with the SEM analysis of electrodes used in MDC. The results revealed that the thickness and abundance of biofilm were found to be greater on CC_DMSO (Fig. 6d) than on CC (Fig. 6c).

The high defluoridation and COD removal efficiency of CC_DMSO could be due to (i) the presence of –COOH, O–C, -C-OH and O–C O functional groups on CC_DMSO, which reduced the overpotential losses associated with the oxidation of organic substrates [44]; (ii) the higher hydrophilicity of CC_DMSO improves the biocompatibility of anodes, which allows rapid attachment of microorganisms to the anode surface; and (iii) the enhanced wettability of DMSO, which increases the interaction between microbes, substrates, and electrode, as a result, the organic matter is degraded more effectively.

3.3. Power production

The performance of CC and CC_DMSO in MDC for power production was examined at initial F⁻ concentrations of 3, 10, and 20 mg/L. The optimal power density for CC and CC_DMSO was obtained by conducting polarization studies. The maximum power densities achieved for CC_DMSO at 3, 10, and 20 mg/L was 0.020 ± 0.07, 2.748 ± 0.22 and 3.245 ± 0.16 mW/m², respectively. However, using CC, lower power densities of 0.009 ± 0.003, 1.394 ± 0.06 and 1.423 ± 0.15 mW/m², were obtained respectively, with 3 (Fig. 7a), 10 (Figs. 7b), and 20 mg/L (Fig. 7c) of F⁻. The obtained results are in good agreement with the defluoridation efficiency and substrate degradation as described in the previous sections.

Fig. 7.

Polarization profiles of (a) 3 mg/L, (b) 10 mg/L, (c) 20 mg/L fluoride using carbon cloth (CC) and dimethyl sulfoxide modified carbon cloth (CC_DMSO) anode and (d) maximum power density comparison.

The presence of hydrophilic functional groups on the CC_DMSO anode reduced overpotential losses associated with the oxidation of organic substrates and improved electrode-microbe interaction, leading to higher power production than CC. The higher power output can also be correlated to the low R_ct and the higher redox peak intensity of CC_DMSO. Both the parameters resulted in a drastic improvement in the electron transfer from the microorganisms to the anode surface [39,40] which benefited the energy output from MDC.

In MDC, the power production is severely affected by: (i) the bioelectrochemical process driven by microbes in the anode chamber; and (ii) the concentration of ions in the middle chamber of MDC [20,35]. With the increase in F⁻ from 3 to 20 mg/L, the power output also improved from 0.009 ± 0.003 to 1.423 ± 0.15 mW/m² and from 0.020 ± 0.07 to 3.245 ± 0.16 mW/m² for CC and CC_DMSO, respectively. Similar results have been reported by Jafary et al. [63]. The ionic diffusion driven by the concentration gradient between the chambers could be a possible reason for this phenomenon [15,20]. With the increase in the concentration of ions in the middle chamber, the concentration gradient across the anode and middle chamber also increases. This would cause the diffusion of more ions at higher concentrations of F⁻ than at lower concentrations. High current generation resulted from a faster desalination rate, while low current generation from a slow desalination rate. Generally, it has been reported that the electrical gradient is the primary force at low salt concentrations, whereas the concentration gradient is dominant at high salt concentrations in the middle desalination chamber [63]. This phenomenon results in increasing the conductivity in the anode chamber, which lowers the overall ohmic resistance of MDC [64]. Hence, the synergetic effect of biofilm, DMSO in CC, and reduced ohmic resistance of MDC improved the power output by 2–2.28 times (Fig. 7d). The results thus proved the effectiveness of the single step facile modification of CC by DMSO in functionalizing the CC electrode.

4. Conclusions and outlook

In the present study, a novel electrode was fabricated by coating a solvent - DMSO onto CC via a facile modification method. The CC_DMSO was used as an anode in the MDC for improved defluoridation, wastewater treatment, and power production. The presence of O–C, –COOH, O–C O, C–OH, and S O groups on CC_DMSO improved the conductivity and hydrophilicity of CC and resulted in an enriched microbial growth on the anode surface, as indicated in Fig. 6c and d. This might have improved the redox activity of the anode. Additionally, the higher redox peak and lower charge transfer resistance of CC_DMSO reduced the internal resistance of the anode and facilitated high power output from the MDC. Moreover, the MDC equipped with CC_DMSO bioanode considerably reduced the time taken to bring the F⁻ concentrations to the permissible limit (1.5 mg/L). Simultaneously, 83% of substrate degradation was obtained in the anodic chamber of MDC. CC_DMSO exhibited a higher power density, of approximately 2–2.28-fold higher than the CC electrode.

The utilisation of a lab scale MDC for defluoridating water was effectively proven in the current study. Compared with conventional techniques, MDC does not require the addition of chemicals or energy (Table 2). It is a self-sufficient system that uses a bioelectrochemical process to remove fluoride from water, treat wastewater, and recover energy, concurrently. Though MDC exhibits various applications, its long-term performance is hindered due to various factors. The major operating concern was the consistent drop in the ion migration rate over the time. This might be explained by the gradual increase in the internal resistance, caused by the ion transport across the membranes and membrane fouling [65]. Improving the membrane properties with antifoulant coatings or materials will help in boosting up the MDC performance. In addition, concentration polarization due to the back diffusion of ions from the electrolytes (anolyte and catholyte) to the middle chamber can further reduce its performance [66]. A continuous recirculation of electrolytes will reduce the adverse effects of concentration polarization.

Table 2.

Comparison of different defluoridation techniques.

Sl. No	Technique	Energy	Chemical requirement	Performance	Reference
1	Enhanced Nalgonda technique	(−)* 230 V/240 V, 20 min	1 g/L alum + 0.15 g/L lime + 0.05 g/L Aluminium hydroxide + NaOH for pH adjustment	9.3 mg/L F− reduced to 2.8 mg/L in 5 h	[6]
2	Electrolytic defluoridation (EDF)	(−)* 16 V, 4 h	_______	7.9 mg/L F− reduced to 2.8 mg/L in 4 h	[6]
3	Adsorption	(−)*	1 g pulverized spent cation exchange resin + Zn (IV) powder	12.2 mg/L F− reduced to 1.5 mg/L in 24 h	[8]
4	Electrodialysis	(−)* 12 V, 21 h 30 min	_______	3.94 mg/L F− reduced to 0.283 mg/L in 21 h 30 min	[11]
5	Electrocoagulation	(−)* 4 mA/m2, 15 min	_______	2.5 mg/L F− reduced to 1.5 mg/L in 3 h	[13]
6	Microbial desalination cell (MDC)	(+)* 3.245 ± 0.16 mW/m2	_______	3 mg/L F−—17, 10 mg/L F−—48, 20 mg/L F−– 96 h to reach 1.5 mg/L	Present study

* Energy consumed (−), Energy produced (+).

Application of DMSO in MDC is a promising technique for defluoridation. Future research will be focused on the optimization of operational and design parameters to upscale MDC. A number of MDC configurations, such as stacked resin packed MDC [67], Up flow two chambered MDC [68], quadruple MDC [63], etc., have been investigated by various researchers, but the studies were limited to the desalination of brine or groundwater only. Future studies will be directed to analyse the sustainability and future impacts of upscaling MDC with electrode modification using DMSO and other nanoparticles (MXene [69], titanium dioxide nanosheets), conductive polymers (polypyrrole) [70] etc., for defluoridation. Tools like life cycle assessment (LCA), techno-economic analysis, emergy analysis (measures all the energy expended in producing a good or service in terms of a single type of energy), exergoeconomic analysis (involves both energy and the cost of the system) [71] will help in evaluating the overall sustainability and feasibility of the developed systems. Furthermore, DMSO based electrodes can be explored in redox flow batteries, wearable sensors, etc., owing to their superior electrochemical properties.

Credit author statement

Sabarija A Mohandas: methodology, experimental analysis, data interpretation, validation, writing -original draft. Sravan Janardhanan: experimental analysis, data interpretation. P Abdul Rasheed: supervision, conceptualization, methodology, writing - review and editing - original draft. Praveena Gangadharan: supervision, funding acquisition, conceptualization, methodology, data interpretation, writing - original draft.

Data availability statement

Data will be made available on request.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Praveena Gangadharan reports financial support was provided by Science and Engineering Research Board (SERB), India.

Appendix A. Supplementary data

The following is the Supplementary data to this article: