Applied machine learning for predicting the properties and carbon and phosphorus fate of pristine and engineered hydrochar
Shiyu Xie,
Tao Zhang,
Siming You,
Santanu Mukherjee,
Mingjun Pu,
Qing Chen,
Yaosheng Wang,
Esmat F. Ali,
Hamada Abdelrahman,
Jörg Rinklebe,
Sang Soo Lee,
Sabry M. Shaheen
Affiliations
Shiyu Xie
State Key Laboratory of Nutrient Use and Management, Beijing Key Laboratory of Farmland Soil Pollution Prevention and Remediation, Key Laboratory of Plant-Soil Interactions of Ministry of Education, College of Resources and Environmental Sciences, China Agricultural University
Tao Zhang
State Key Laboratory of Nutrient Use and Management, Beijing Key Laboratory of Farmland Soil Pollution Prevention and Remediation, Key Laboratory of Plant-Soil Interactions of Ministry of Education, College of Resources and Environmental Sciences, China Agricultural University
Siming You
James Watt School of Engineering, University of Glasgow
Santanu Mukherjee
School of Agriculture Sciences, Shoolini University of Biotechnology and Management Sciences
Mingjun Pu
State Key Laboratory of Nutrient Use and Management, Beijing Key Laboratory of Farmland Soil Pollution Prevention and Remediation, Key Laboratory of Plant-Soil Interactions of Ministry of Education, College of Resources and Environmental Sciences, China Agricultural University
Qing Chen
State Key Laboratory of Nutrient Use and Management, Beijing Key Laboratory of Farmland Soil Pollution Prevention and Remediation, Key Laboratory of Plant-Soil Interactions of Ministry of Education, College of Resources and Environmental Sciences, China Agricultural University
Yaosheng Wang
Institute of Environment and Sustainable Development in Agriculture, Chinese Academy of Agricultural Sciences
Esmat F. Ali
Department of Biology, College of Science, Taif University
Hamada Abdelrahman
Faculty of Agriculture, Soil Science Department, Cairo University
Jörg Rinklebe
University of Wuppertal, School of Architecture and Civil Engineering, Institute of Foundation Engineering, Water and Waste Management, Laboratory of Soil and Groundwater Management
Sang Soo Lee
Department of Environmental and Energy Engineering, Yonsei University
Sabry M. Shaheen
University of Wuppertal, School of Architecture and Civil Engineering, Institute of Foundation Engineering, Water and Waste Management, Laboratory of Soil and Groundwater Management
Abstract Application of advanced techniques and machine learning (ML) for designing and predicting the properties of engineered hydrochar/biochar is of great agro-environmental concern. Carbon (C) stability and phosphorus (P) availability in hydrochar (HC) are among the key limitations as they cannot be accurately predicted by traditional one-factor tests and might be overcome by engineering the pristine HC. Therefore, the aims of this study were (1) to determine the optimal production conditions of engineered swine manure HC with high C stability and P availability, and (2) to develop the best ML models to predict the properties of HC derived from different feedstocks. Pristine- (HC) and FeCl3 impregnated swine manure-derived HC (HC-Fe) were produced by hydrothermal carbonization under different pH (4, 7, and 10), reaction temperature (180, 220, and 260 ℃), and residence time (60, 120, and 180 min) and characterized using thermo-gravimetric, microscopic, and spectroscopic analyses. Also, different ML algorithms were used to model and predict the hydrochar solid yield, properties, and nutrients content. FeCl3 impregnation increased Fe-phosphate content, while it reduced H/C and O/C ratios and hydroxyapatite P content, and therefore improved C stability and P availability in the HC-Fe as compared to HC, particularly under lower pH (4), temperature of 220 ℃, and at 120 min. The generalized additive ML model outperformed the other models for predicting the HC properties with a correlation coefficient of 0.86. The ML analysis showed that the most influential features on the hydrochar C stability were the H and O contents in the biomass, while P availability in HC was more dependent on the C, N and O contents in biomass. These results provided optimal production conditions for Fe-engineered manure hydrochar and identified the best performing ML model for predicting hydrochar properties. The main implication of this study is that it offers a high potential to improve the utilization of biowastes and produce biowaste-derived engineered hydrochar with high C stability and P availability on a large scale. Graphical Abstract
