Anaerobic digestion of sugarcane trash and bagasse for biomethane production
File version
Author(s)
Primary Supervisor
Kaparaju, Laxmi Narasimha Prasad
Other Supervisors
Herat, Kusumsiri S
Editor(s)
Date
Size
File type(s)
Location
License
Abstract
Sugarcane cultivation is a major source and sinks of greenhouse gas emissions (GHGs). In 2019, approximately 30.04 x 106 t of sugarcane was harvested from 364,428 ha of land. Of the total cane harvested, sugarcane bagasse (SB) and sugarcane trash (ST) accounted for 30.1% and 14.9%, respectively. Further, fossil fuel consumption in the transport of cane to mills was 29.15 ML and is equivalent to 221.5 Mt CO2-equivalent GHG emissions. Anaerobic digestion (AD) of sugar industry wastes for biomethane production and use as vehicle fuel (bioCNG) would reduce the fossil fuel consumption and the associated GHG emissions in cane transportation to mill. The study aims to optimize biogas production and upgrade the produced biogas to vehicle fuel. For that the study is divided into different objectives. To determine the substrate characteristic and suitable AD parameters viz biodegradability (Experiment I), particle size (Experiment II), acid/base thermal pretreatment (Experiment III), C:N ratio (Experiment IV), and trace nutrient supplementation (Experiment V), to generate the maximum methane yield from ST with respect to SB was designed as objective I. At first, chemical composition and methane yields of ST and SB were determined through bench-scale biochemical methane potential (BMP) tests (Experiment I). Buswell’s equation predicted a theoretical methane yield of 291.0 and 349.4 mL CH4 g-1VSadded for ST (C107.8H179.1O101.4N1S0.08) and SB (C86.7H134.3O64.9N1S0.07), respectively. The corresponding methane yields with Modified Dulong’s equation were 266.3 and 298.7 mL CH4 g-1VSadded, respectively. The calculated energy value was 14.1 MJ for ST and 12.6 MJ for SB. However, experimental methane yields obtained were 161.8 and 187.9 mL CH4 g-1VSadded for ST and SB, respectively. First-order kinetic model revealed that experimental data fitted well (R2 = 0.99) with the modelling data and the hydraulic rate constant (khyd) values of 0.04 and 0.06 day-1 were obtained for ST and SB, respectively. However, modified Gompertz model had a lag phase () of 2.1 and 1.7 day, for ST and SB, respectively indicating hydrolysis was the rate limiting step for the studied lignocellulosic feedstocks. Thus, the effect of mechanical (Experiment II)., thermal and chemical pretreatments (Experiment III) on chemical composition and methane yields of ST and SB were evaluated. The effect of particle size of <0.25, 0.25-0.50, 0.50-1 and 1-2 mm on chemical composition and methane yields were determined. Results showed that particle size reduction had a profound effect on methane yields, especially for SB than for ST. For ST, particle size of 1-2 mm showed an improvement in methane yields by 19.1% over control (161.8 mL CH4 g-1 VSadded). For SB, the increase in methane yields over control (189.7 mL CH4 g-1 VSadded) were by 23.6%, 20.3%, 18.1% and 6.4% respectively at particle sizes of 1-2, 0.5-1, 0.25-0.5, 0.13-0.25 mm, respectively. These results suggest that the optimal particle size for anaerobic digestion of ST and SB will be 1-2 mm for maximum methane yield. Further, mechanical pretreatment through milling did not solubilise hemi-cellulose and/or improve delignfication but improved the surface area of the holocellulose. Therefore, the effect of chemical catalysts (dilute NaOH, H2SO4, HNO3) with and without steam explosion on chemical composition and methane yields was evaluated (Experiment III). Pretreatment conditions used for the steam explosion were 130 °C for 5 minutes at acid/base concentration of 2.5% catalyst loading. Results showed that the studied pretreatments had a profound effect on chemical composition and methane yields of ST. On comparison to control, dilute H2SO4, followed by NaOH and HNO3 addition with steam explosion improved the methane yields of ST by 63.5%, 52.1% and 45.6%, respectively. Steam explosion alone also improved the methane yields of ST by 40% over control. Biomass composition analysis showed that dilute H2SO4, HNO3, NaOH and steam explosion alone had improved the glucan content by 13.7%, 11.7%, 9.3% and 3% respectively than control. Dilute H2SO4 pretreatment improved the glucan availability by 45.2% and hemicellulose (xylan and arabinan) solubilisation by 63.7%-66.9%. Lignin depolymerisation in pretreated ST was improved (16.7%) over untreated ST. In Experiment I, chemical composition of ST and SB showed that the studied substrates were deficit in trace elements and contained high carbon to nitrogen (C/N) ratio of 92.4and 146.5 respectively. Therefore, the effect of C/N ratios of 15:1, 20:1, 25:1, 30:1, 35:1 and 40:1 with urea addition on methane yields of ST and SB was investigated (Experiment IV). Results showed that methane yields improved by 13.6% and 11.3% for ST when the C/N ratio was at 20:1 and 25:1, respectively. The corresponding values for SB were 14.2% and 14.3% at 20:1 to 25:1 C/N ratio, respectively. Both these results indicate that the optimal C/N was 20-25:1 for AD of lignocellulosic residues such as ST and SB. On the other hand, the effect of trace nutrients nickel (Ni), cobalt (Co), molybdenum (Mo), manganese (Mn), copper (Cu) and zinc (Zn) on methane yields during AD of ST and SB (Experiment V) showed that trace elements supplementation influenced the methane yields and both substrates responded differently. With ST, methane yields of 68.1% and 68.7% increase over control were noticed with addition of Co and Mo, respectively. For SB, methane yields increased by 48.6%, 63.9% and 4.8% with Co, Mo and Mn dosing at 2, 3, 90 mg kg-1 respectively. All other TE addition resulted in lower methane yields than control or inhibited the biogas production at different stages of incubation. All the batch BMP tests were conducted in triplicates at inoculum to substrate (ISR) ratio of 2 in serum glass bottles with a working volume of 100 mL and incubated statically at 37 °C. All the results were analysed for variance using LSD and Dunnett-t test giving methane yield as dependent variable (p<0.05). Second objective was designed to study the effect of organic loading on process performance and methane yields in four lab-scale stainless reactors (10 L working volume) and operated at an initial organic loading rate (OLR) of 1.5 gVS L-1 d-1 with hydraulic retention times (HRT) of 35 days for 225 days. Reactors were fed with untreated ST (ST), untreated SB (SB), pretreated SB (TB) and pretreated ST (TT). Dilute H2SO4 followed by steam explosion (Experiment II) was used for pretreatment of ST and SB. OLR was increased in a stepwise manner from the initial rate of 1.5 to 2.5 and 3.5 gVS L-1 d-1. OLR was changed upon achieving steady-state condition and/or operating for 2 consecutive HRTs. Methane production rates and yields responded with increase in OLR from 1.5 to 2.5 gVS L-1 d-1. Mean methane yields of 138, 173, 248 and 252 ml g-1VSfed were obtained at an OLR 1.5 gVS L-1 d-1 in ST, SB, TT and TB reactors, respectively. Increase in OLR to 2.5 gVS L-1 d-1 showed decrease in methane yields. Mean methane yields obtained for TB, TT, SB and ST were 121, 148, 226, 236 ml g-1VSfed with a VS removal rate of 48.5, 51.4, 51.5 and 52.4%, respectively. Process parameters such as pH, total ammoniacal nitrogen (TAN) and total volatile fatty acids (TVFA) were shown to be stable and were 7.3-7.5, 0.36-0.54 g L-1 and 0.79-0.98 g L-1 respectively during operational OLR’s. Further increase in OLR from 2.5 to 3.5 gVS L-1 d-1 resulted in further decrease in methane yields and unstable AD process. At OLR 3.5 gVS L-1 d-1, methane yields were 119, 139, 189 and 199 ml g-1VSfed for substrates ST, SB, TT and TB respectively. TVFA accumulation was noticed (1.55-2.49 g L-1) , pH was 7.4-7.5 and lower methane concentration (50.5-51.9%). Residual methane production (RMP) test after each OLR indicated the process efficiency. At OLR 2.5 gVS L-1 d-1, TT and TB reactors had the lowest RMP (32.1% and 30.2% respectively) with relatively high VS removal compared with SB and ST reactors. These results indicate that steam explosion with dilute sulphuric acid improved the biodegradability and methane yields of ST and SB. The results obtained from the lab-scale reactors were used to design and optimise process performance and methane yields from pretreated sugarcane trash in pilot-scale reactors (date not presented). Third objective was designed to evaluate detailed biogas composition and to develop and optimise high pressure water scrubbing technology (HPWS). For that, the biogas composition, energy content, siloxanes and trace volatile organic compounds in biogas generated from lab-scale biogas reactors were determined and compared with the pilot-scale. Laboratory biogas samples were collected during the steady-state condition when the reactors were operated at an OLR of 2.5 gVS L-1 d-1. Results showed that biogas collected from ST, SB, TT and TB reactor had methane concentration of 52.3, 52.2, 52.7, 52.7 %, respectively. The corresponding lower calorific values (LCV) were 18.4, 18.1, 18.9 and 19.2 MJ m-3 respectively. The wobbe index values in the biogases were 18.3, 18.2, 18.7 and 19.0 MJ m-3, respectively. Volatile organic compounds were noticed in the biogas samples. Organic silicon compounds (siloxanes) were in the range of 0-0.4 mg m-3. The reduced sulphur compounds and benzene and toluene content in the biogases were in the range of 0.7-1.3 mg m-3 and 0.2-0.7 mg m-3, respectively. Among the studied siloxanes, the proportion of cyclic siloxane (D3:D4:D5) viz., hexamethylcyclotrisiloxane (D3), octamethylcyclotetrasiloxane (D4) and decamethylcyclopentasiloxane (D5) were noticed. The ratio of D3:D4:D5 in biogases produced from lab-reactors were 68.5:5.7: 22.4 for ST, 33.9:4.9: 60.1 for SB, 25:8.6:69.3 for TT and 14.8:3.7:81.8 for TB. Trimethylsilanol, linear siloxanes and decamethylcyclohexasiloxane (D6) content in the biogases were below the detection limits. Volatile organic compounds, reduced sulphur compounds and siloxanes cause environmental impact and affect biomethane quality for vehicle fuel use. Biogas composition from pilot-scale biogas reactors (1.2 m3 reactor with 0.8 m3 working volume) fed with steam exploded ST at an OLR of 1.5 kgVS m-3 d-1 and HRT of 35 d was analysed to optimise the process parameters to achieve the desired biomethane quality and evaluate the energy requirements of pilot-scale biogas upgrading unit (10 m3 h-1) for biogas upgrading and bottling. Results showed that the biogas had 54.1% CH4 and 39.7% CO2 and the produced biogas was upgraded to 96.7% biomethane purity by using high pressure water scrubbing process. Experimental data from the biogas upgrading process was used to optimise biogas upgrading by using Aspen Plus software. The influence of process parameters such as absorber column pressure, water to gas flow rate, temperature on biomethane purity and percentage of H2S and CO2 removal were evaluated. Experimental results showed that at liquid flow rate of 3 m3/hr, fluid temperature of 20°C, at absorption column pressure of 8 bar with 4 m random packing material with redistributor at 2 m with 25 mm plastic pall ring packing material; biogas can be upgraded to biomethane of 96.8% CH4, 2.9% CO2, < 1 ppm H2S. These model results were validated with software simulation.
Journal Title
Conference Title
Book Title
Edition
Volume
Issue
Thesis Type
Thesis (PhD Doctorate)
Degree Program
Doctor of Philosophy (PhD)
School
School of Eng & Built Env
Publisher link
Patent number
Funder(s)
Grant identifier(s)
Rights Statement
Rights Statement
The author owns the copyright in this thesis, unless stated otherwise.
Item Access Status
Note
Access the data
Related item(s)
Subject
biogas production
vehicle fuel