Somak Chatterjee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Somak Chatterjee has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
Sirshendu De
7 shared publications- Sirshendu De
Department of Chemical Engineering, Indian Institute of Technology Kharagpur, Kharagpur, India
6 shared publications Dharmesh Mishra
BITS Pilani, Dept Chem Engn, Pilani 333031, Rajasthan, India
3 shared publicationsAndrew Krause
MKS Vis Grp, Louisville, KY 40228 USA
2 shared publicationsAnkit Pal
Birla Institute of Technology and Science-Pilani
2 shared publicationsManu Sharma
Department of Chemical Engineering, BITS-Pilani, Pilani, Rajasthan-333031, India
2 shared publications
Latest Publications
- 23 August 2026
- Published by American Chemical Society (ACS) in Industrial & Engineering Chemistry Research
- Vol. 65 (37) , 19780-19800
Machine Learning Pipeline for Wastewater Effluent Quality Forecasting: Gradient Boosting-Based BOD5 Prediction with Feature Importance and Seasonal Pattern Analysis
Realtime BOD5 forecasting is essential for optimizing wastewater treatment operations and ensuring regulatory compliance. This study proposes a machine-learning-based predictive framework integrating temporal feature engineering, ensemble gradient-boosting models, and explainable artificial intelligence for multimonth BOD5 prediction. Monthly wastewater quality data from 2008 to 2023 (192 months), including BOD5, total suspended solids, total phosphorus, and flow rate, were used to construct 48 temporal features based on lagged variables, rolling statistics, and physicochemical ratios. Four ensemble models─Gradient Boosting, Random Forest, XGBoost, and LightGBM─were trained on 2008–2020 data and evaluated on independent data from 2021 to 2023. LightGBM achieved the highest accuracy (R2 = 0.9894, RMSE = 0.223 mg·L–1, MAPE = 2.26%), enabling reliable predictions up to 12 months ahead. Feature importance analysis identified six month lagged BOD5 and the BOD5-to-flow ratio as dominant predictors. The framework supports proactive effluent quality management and early warning decision support.
- 23 August 2026
- Published by American Chemical Society (ACS) in Industrial & Engineering Chemistry Research
- Vol. 65 (37) , 19780-19800
Machine Learning Pipeline for Wastewater Effluent Quality Forecasting: Gradient Boosting-Based BOD5 Prediction with Feature Importance and Seasonal Pattern Analysis
Realtime BOD5 forecasting is essential for optimizing wastewater treatment operations and ensuring regulatory compliance. This study proposes a machine-learning-based predictive framework integrating temporal feature engineering, ensemble gradient-boosting models, and explainable artificial intelligence for multimonth BOD5 prediction. Monthly wastewater quality data from 2008 to 2023 (192 months), including BOD5, total suspended solids, total phosphorus, and flow rate, were used to construct 48 temporal features based on lagged variables, rolling statistics, and physicochemical ratios. Four ensemble models─Gradient Boosting, Random Forest, XGBoost, and LightGBM─were trained on 2008–2020 data and evaluated on independent data from 2021 to 2023. LightGBM achieved the highest accuracy (R2 = 0.9894, RMSE = 0.223 mg·L–1, MAPE = 2.26%), enabling reliable predictions up to 12 months ahead. Feature importance analysis identified six month lagged BOD5 and the BOD5-to-flow ratio as dominant predictors. The framework supports proactive effluent quality management and early warning decision support.
- Open Access
- 24 November 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 41601
Graphene oxide-iron oxide based thin film nanocomposite membrane for congo red dye removal and simultaneous salt recovery from aqueous stream
Current work focusses on the synthesis of iron oxide impregnated-graphene oxide incorporated polyvinylidene fluoride based thin-film nanocomposite (TFN) membrane and its application in Congo Red dye removal. Interfacial polymerization with 0.02 wt% graphene oxide significantly improved performance of prepared TFN membranes, achieving over 94% rejection of dye with permeability of 3.6 l/m2·h·bar. Testing performance of the membrane was further validated in the presence of real-life feed wastewater, showing 72% dye rejection over 12 h, while an enhanced 94% rejection was achieved with synthetic dye solution. Superior performance was observed in removal of dissolved solids, conductivity, salinity, nitrate, sulfate and dyes compared to commercial nanofiltration membranes. Effect of different operating conditions on membrane performance were examined, confirming its robust nature. Comprehensive characterization of prepared membrane revealed its superiority in texture and chemical nature. Optimized TFN membrane demonstrated recovery efficiency of 70% for sodium chloride and magnesium sulfate solutions, indicating its potential for desalination approach. Thermal regeneration and long-term filtration studies were conducted to evaluate membrane’s lifespan, indicating long-term durability and effectiveness in industrial applications. Studies also indicated weight loss of 62% over two months, which corroborated biodegradable nature of the membrane samples after exhaustion.
- 24 November 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 41601
Open AccessGraphene oxide-iron oxide based thin film nanocomposite membrane for congo red dye removal and simultaneous salt recovery from aqueous stream
Current work focusses on the synthesis of iron oxide impregnated-graphene oxide incorporated polyvinylidene fluoride based thin-film nanocomposite (TFN) membrane and its application in Congo Red dye removal. Interfacial polymerization with 0.02 wt% graphene oxide significantly improved performance of prepared TFN membranes, achieving over 94% rejection of dye with permeability of 3.6 l/m2·h·bar. Testing performance of the membrane was further validated in the presence of real-life feed wastewater, showing 72% dye rejection over 12 h, while an enhanced 94% rejection was achieved with synthetic dye solution. Superior performance was observed in removal of dissolved solids, conductivity, salinity, nitrate, sulfate and dyes compared to commercial nanofiltration membranes. Effect of different operating conditions on membrane performance were examined, confirming its robust nature. Comprehensive characterization of prepared membrane revealed its superiority in texture and chemical nature. Optimized TFN membrane demonstrated recovery efficiency of 70% for sodium chloride and magnesium sulfate solutions, indicating its potential for desalination approach. Thermal regeneration and long-term filtration studies were conducted to evaluate membrane’s lifespan, indicating long-term durability and effectiveness in industrial applications. Studies also indicated weight loss of 62% over two months, which corroborated biodegradable nature of the membrane samples after exhaustion.
- Open Access
- 22 April 2025
- Published by Royal Society of Chemistry (RSC) in RSC Advances
- Vol. 15 (17) , 13337-13352
Layered double hydroxide based composite core–shell electrospun nanofibers for lead and fluoride filtration from contaminated streams
Coaxial electrospinning was used to synthesize polyacrylonitrile–polyethersulfone (PAN–PES) core–shell nanofibers with magnesium–aluminum layered double hydroxide (Mg–Al LDH) for filtration of lead and fluoride from contaminated streams. Fiber geometry was characterized at a 0.5 mL h−1 flow rate for the core polymer (PES/LDH) and 0.8 mL h−1 flow rate for the shell polymer (PAN), with a potential of 23 kV and a distance of 15–17 cm between the collector and the needle head. A homogeneous fiber shape was achieved using an optimal LDH concentration of 0.7%. The prepared nanofibers served as an ultrafiltration membrane with a permeability of 5 × 10−12 m s−1 Pa−1. The uptake capacity of the produced nanofibers for fluoride and lead was estimated to be 948 mg g−1 and 196 mg g−1, respectively at 298 K as per Langmuir's isotherm model. These fibers exhibited hydrophilic properties and possessed a significant level of porosity. XPS study revealed binding energies of 139.3 eV and 685.2 eV, indicating lead and fluoride uptake by the nanofibers. Ether, sulfone, hydroxyl and nitrile groups found in the nanofibers' shell and core most likely contributed to the lead and fluoride uptake. This facilitated the uptake of both ions on the surface of the nanofibers. In terms of the inhibition effect, fluoride had a stronger masking effect compared with lead in a multicomponent solution (consisting of lead and fluoride). Dynamic vacuum filtration was also investigated using the prepared nanofibers in artificial and real-life feed solutions.
- 22 April 2025
- Published by Royal Society of Chemistry (RSC) in RSC Advances
- Vol. 15 (17) , 13337-13352
Open AccessLayered double hydroxide based composite core–shell electrospun nanofibers for lead and fluoride filtration from contaminated streams
Coaxial electrospinning was used to synthesize polyacrylonitrile–polyethersulfone (PAN–PES) core–shell nanofibers with magnesium–aluminum layered double hydroxide (Mg–Al LDH) for filtration of lead and fluoride from contaminated streams. Fiber geometry was characterized at a 0.5 mL h−1 flow rate for the core polymer (PES/LDH) and 0.8 mL h−1 flow rate for the shell polymer (PAN), with a potential of 23 kV and a distance of 15–17 cm between the collector and the needle head. A homogeneous fiber shape was achieved using an optimal LDH concentration of 0.7%. The prepared nanofibers served as an ultrafiltration membrane with a permeability of 5 × 10−12 m s−1 Pa−1. The uptake capacity of the produced nanofibers for fluoride and lead was estimated to be 948 mg g−1 and 196 mg g−1, respectively at 298 K as per Langmuir's isotherm model. These fibers exhibited hydrophilic properties and possessed a significant level of porosity. XPS study revealed binding energies of 139.3 eV and 685.2 eV, indicating lead and fluoride uptake by the nanofibers. Ether, sulfone, hydroxyl and nitrile groups found in the nanofibers' shell and core most likely contributed to the lead and fluoride uptake. This facilitated the uptake of both ions on the surface of the nanofibers. In terms of the inhibition effect, fluoride had a stronger masking effect compared with lead in a multicomponent solution (consisting of lead and fluoride). Dynamic vacuum filtration was also investigated using the prepared nanofibers in artificial and real-life feed solutions.
- 5 February 2025
- Published by Springer Nature
- p. 225-256
Application of Composite Membrane-Based Technology in Treatment of Textile Industry Effluents
- 5 February 2025
- Published by Springer Nature
- p. 225-256
Application of Composite Membrane-Based Technology in Treatment of Textile Industry Effluents
- 7 December 2023
- Published by Elsevier in Materials Today: Proceedings
- Vol. 111, 137-146
Undoped polyaniline-modified sawdust as an adsorbent for lead removal
- 7 December 2023
- Published by Elsevier in Materials Today: Proceedings
- Vol. 111, 137-146
Undoped polyaniline-modified sawdust as an adsorbent for lead removal
- 6 December 2023
- Published by Elsevier in Chemosphere
- Vol. 349, 140837
Polyacrylonitrile and polyethersulfone based co-axial electrospun nanofibers for fluoride removal from contaminated stream
- 6 December 2023
- Published by Elsevier in Chemosphere
- Vol. 349, 140837
Polyacrylonitrile and polyethersulfone based co-axial electrospun nanofibers for fluoride removal from contaminated stream
- 4 July 2022
- Published by Wiley in Journal of Food Science
- Vol. 87 (8) , 3338-3354
A review on pectinase properties, application in juice clarification, and membranes as immobilization support
Pectic substances cause haziness and high viscosity of fruit juices. Pectinase enzymes are biological compounds that degrade pectic compounds. Nontoxicity and ecofriendly nature make pectinases excellent biocatalysts for juice clarification. However, the poor stability and nonreusability of pectinases trim down the effectiveness of the operation. The immobilization techniques have gained the attention of researchers as it augments the properties of the enzymes. Literature has reported the stability improvement of enzymes like lipase, laccase, hydrogen peroxidase, and cellulase upon immobilization on the membrane. However, only a few research articles divulge pectinase immobilization using a membrane. The catalysis‐separation synergy of membrane‐reactor has put indelible imprints in industrial applications. Immobilization of pectinase on the membrane can enhance its performance in juice processing. This review delineates the importance of physicochemical and kinematic properties of pectinases relating to the juice processing parameters. It also includes the influence of metal‐ion cofactors on enzymes’ activity. Considering the support and catalytic‐separation facets of the membrane, the prediction of the membrane as support for pectinase immobilization has also been carried out.
- 4 July 2022
- Published by Wiley in Journal of Food Science
- Vol. 87 (8) , 3338-3354
A review on pectinase properties, application in juice clarification, and membranes as immobilization support
Pectic substances cause haziness and high viscosity of fruit juices. Pectinase enzymes are biological compounds that degrade pectic compounds. Nontoxicity and ecofriendly nature make pectinases excellent biocatalysts for juice clarification. However, the poor stability and nonreusability of pectinases trim down the effectiveness of the operation. The immobilization techniques have gained the attention of researchers as it augments the properties of the enzymes. Literature has reported the stability improvement of enzymes like lipase, laccase, hydrogen peroxidase, and cellulase upon immobilization on the membrane. However, only a few research articles divulge pectinase immobilization using a membrane. The catalysis‐separation synergy of membrane‐reactor has put indelible imprints in industrial applications. Immobilization of pectinase on the membrane can enhance its performance in juice processing. This review delineates the importance of physicochemical and kinematic properties of pectinases relating to the juice processing parameters. It also includes the influence of metal‐ion cofactors on enzymes’ activity. Considering the support and catalytic‐separation facets of the membrane, the prediction of the membrane as support for pectinase immobilization has also been carried out.