Effects Of Biomass Species And Pyrolysis Temperature On Biochar Properties

Biomass is a kind of carbon neutral renewable energy.The production of biochar from biomass by the thermochemical method is regarded as an effective method to mitigate climate change.Biochar from biomass pyrolysis shows excellent carbon sequestration ability and has the potential to replace fossil fuels.Biochar is widely used in the field of environment and energy with rich C content.However,the dissolved organic matter derived from biochar（BDOM）is not stable and may affect the soil or water environment and further hinder the application of biochar in more fields.At present,there is no standardized and reliable method for the determination of biochar stability,and there are significant differences in the evaluation of biochar stability by different methods.The higher heating value（HHV）of biochar varies with different biomass sources and pyrolysis conditions,so it is very important to establish a time-saving and labor-saving prediction model for the combustion performance of biochar by using the mathematical statistics method.Therefore,in this study,BDOM,stability,and HHV of various biochar were studied using different analytical methods and mathematical statistical methods.The main conclusions of this research include the following three parts:（1）Fluorescence excitation-emission matrix spectra,Fourier transform infrared spectroscopy,and UV-visible spectrophotometer were used to characterize and analyze the BDOMs of sewage sludge biochar produced at 400～700℃.The results showed that BDOMs mainly were derived from biological or aquatic bacteria.The leaching ratios of BDOMs are in the range of 2.80%～3.74%.The pyrolysis temperature affects the properties and component distribution of BDOM.At low pyrolysis temperatures,the components of BDOM were evenly distributed and the molecular weight was larger.The content of aromatic functional groups and the degree of humification decreased with the increase of pyrolysis temperature.The relative content of humic-like substances increased with the increase of pyrolysis temperature.（2）18 biochars were prepared by pyrolysis of soybean straw,sawdust,and chlorella in the temperature range of 300～800℃.The stability of biochar was evaluated by elemental analysis,proximate analysis,the Edinburgh stability method,X-ray photoelectron spectroscopy（XPS）,¹³C solid nuclear magnetic resonance（NMR）,and thermogravimetric analysis.The results showed that the stability indicators obtained by most methods,excluding aromatic C from NMR and C–C/C=C/C–H from XPS,were closely related to the pyrolysis temperature（ANOVA analysis,p<0.05）.The VM/（FC+VM）（VM:volatile matter,FC:fixed carbon）had high correlations with other indicators（Pearson coefficient,|r|>0.36,p<0.05）except for C–C/C=C/C–H from XPS.The cluster analysis indicates the studied stability methods could be divided into three categories,and the proximate analysis may be developed as an alternative to replace O/C and H/C for the determination of biochar stability.（3）52 biochars were produced by pyrolysis of rice straw,pig manure,soybean straw,wood sawdust,sewage sludge,Chlorella Vulgaris,and their mixtures at the temperature ranging from 300 to 800°C.The multi-linear regression（MLR）and the machine learning（ML）models were developed to predict the measured HHV of biochar.The Pearson correlation（r）and relative importance analysis between HHV values and the indicators derived from the proximate and ultimate analysis were carried out.The measured HHV was used to train and test the MLR and the ML models.The results showed that HHV had strong correlations（|r|>0.9,p<0.05）with Ash,FC,and C.The co-pyrolysis of mixed biomass provided an alternative method for increasing biochar yield.The contents of Ash,FC,and C increased as the incremental pyrolysis temperature for most biochars.The MLR correlations based on either proximate or ultimate analysis showed acceptable prediction performance with test R²>0.90.The testing R² of the optimal ML models before and after adding extra data for model construction were 0.95（random forest model）and 0.98（random forest model）,respectively.Feature importance analysis of the ML models showed that Ash and C were the most important inputs to predict biochar HHV.