POLITECNICO DI MILANO
School of Industrial Engineering
Department of Energy
SUPERCRITICAL WATER GASIFICATION OF BIOMASS:
KINETIC APPROACH & PROCESS SIMULATION
Relatore: Professor Emanuele Martelli
Corelatore: Professor Wiebren de Jong
(Delft University of Technology)
Master of Science Thesis of:
Guchan Yapar - 797694
Academic Year 2014-2015
To the coal miners of my country and the world; I hope one day you will not have to risk your lives to give us a light'
*to my beloved Mother and nephew Aslan.
ACKNOWLEDGEMENTS
I would like to first give my thanks to Prof. Emanule Martelli, supervisor of my thesis work in Politecnico di Milano, for his great help even from an other country.
I am very grateful to my supervisor in Delft University of Technology, Prof. Wiebren de Jong, for all his support and effort he spent for the success of this work.
Special thanks to my daily supervisor Onursal Yakaboylu who is a great science person and a friend. Wishing you the best of luck.
I would like to say thank you, to my friends in Italy, Turkey and new ones in Netherlands; you have made this journey more memorable.
ABSTRACT
Thermochemical biomass conversion is a method to transform stored energy to higher energy density solid, liquid or gas fuels by chemically decomposing the high molecular weight biopolymers. Biomass derived green gas has been worked extensively with the aim of taking over the place of fossil based combustible gas, for people's self-sufficiency in energy consumption along the earth and preserving a sustainable environment. Biomass gasification at high temperature - low pressure conditions targets to produce a combustible gas, abundant of H2 (high calorific gas) and CO (reactant of further shift reactions). Stated technology is able to scale up to 600 MW energy production capacities for various biomass or residues but with different thermal efficiencies due to the external heat requirement. Among all other physical and chemical properties, moisture content of the biomass may result in thermal efficiency issues and parallel to this power production decreases.
Supercritical Water Gasification (SCWG) allows the thermochemical conversion of wet biomass in hydrothermal media. High-pressure biomass slurry can be kept in aqueous media until the critical point of water. Right above near-critical temperature level (∼330 ℃) hydrogen bond number decreases and water molecule becomes more apolar, as a consequence water solubility increases allowing one-phase homogenous reactions; furthermore, ion concentration in the hydrothermal media increases, which favors the acid or base catalyzed reactions. Thermo dynamical benefits have aroused the curiosity of several science groups for the last 20 years, providing modeling and experiment studies on supercritical water gasification of biomass (SCWG). However, detailed kinetic modeling of the process for lignocellulosic biomass is missing, strongly needed for further technology improvements.
In this work, review and evaluation of the biomass kinetic approach has been performed with the aim of proposing reaction network and rate parameters for three main biomass structural compounds; cellulose, hemicellulose and lignin. AspenPlus 8.3 engineering software package has been aided for the simulation of the tubular reactors, integrated with the compiled kinetic data from literature and proposed network. Validation results show that reactor simulation has a good accuracy of predicting gasification efficiency, carbon efficiency and gas yield for a range of residence time (15-47 s), pressure (20-30 MPa) and biochemical composition (mostly agricultural residues). Reactor simulation validations showed the possibility of inserting consecutive reactors to a simplified SCWG process simulation. Process scheme including 3 tubular flow reactors and auxiliary units has been designed; according to which, carbonmonoxide low product gas and high gasification efficiency (4,45 % , 92,24%, respectively) have been achieved. The results of this work show that kinetic model developed for lignocellulosic biomass, based on the literature data is able to predict gasification efficiency, intermediate compound composition, gas composition and thermal energy output for different operating conditions and reactor lengths.
table of contents
1 Introduction 1
1.1 Background 1
1.2 Biomass Conversion technologies-Placement of Hydrothermal Biomass Gasification 2
1.3 Supercritical Water 5
1.4 SCWG of biomass 7
1.4.1 Main Research Groups and Experiments on SCWG 9
1.5 Reactor Systems 12
1.6 Scope and Objectives of the Study 14
2 Literature Review of Biomass Degradation in Hydrothermal Media: Reaction Mechanism and Kinetics 16
2.1 Cellulose Decomposition 17
2.1.1 Glucose Decomposition 20
2.2 Hemicellulose Decomposition 23
2.3 Lignin Decomposition 25
3 Proposed Reaction Pathway 31
3.1 Cellulose Pathway: 32
3.1.1 Decomposition of Cellobiose 32
3.1.2 Reactions of Glucose-Fructose decomposition products 33
3.2 Hemicellulose Pathway 35
3.3 Lignin Pathway: 37
3.4 Acid Decomposition Pathway: 39
3.5 Kinetics Data and Arrhenius Parameters 41
4 Integration of Kinetic Data Set to the Reactor Simulation- Test Run & Validation 52
4.1 Integration of Kinetic Data Set to the AspenPlusTM Simulation 52
4.2 Results of the Reactor Test-Run 54
4.3 Validation of the simulation 59
4.3.1 Effect of Residence Time 61
4.3.2 Effect of Pressure 64
4.3.3 Effect of Feedstock 66
5 Sensitivity Analysis 69
5.1 Sensitivity Analysis for Reactor Heating Rate Effect: 69
5.2 SCWG Process Scheme Build-Up & Sensitivity Analysis for Parameter Variations on SCWG Plant Simulation 77
5.2.1 Design Parameters and Unit Selections 77
5.2.2 SCWG of Cellulosic Biomass Plant Process Simulation 80
5.2.3 Effect of Maximum Temperature 81
5.2.4 Effect of Biomass Load 82
5.2.5 Effect of Biomass Type 85
Conclusions and Further Developments 88
References: 90
Annexes 96
A- Subcritical Region Reactions 96
B- Supercritical Region Reactions 97
C- RGibss reaction stoichiometries (for Lignin Pathway) 99
D- Component List 99
Figure 1.1 Application option scheme of SCWG process outlet [7] 4
Figure 1.2 Phase diagram of water [8] 5
Figure 1.3 Properties of supercritical water [10] 6
Figure 1.4 Simplified process scheme of the PDU for SCWG of biomass operated at University of Twente [7] 13
Figure 2.1 Cellobiose decomposition pathway: 300 ℃T<400 ℃, 25<P<40 MPa [51] 17
Figure 2.2 -ln(1 - X) versus residence time for cellobiose decomposition [51] 18
Figure 2.3 Arrhenius plot of the rate constant of conversion of microcrystalline cellulose in water (290 C – 400 oC, 25 MPa), [50]. 19
Figure 2.4 Mechanism of glucose decomposition in sub- and supercritical water: 300 ℃T<400 ℃,<P<30 MPa [53] 21
Figure 2.5 Glucose decomposition pathway in sub- and supercritical water: 300 ℃T<460 ℃, P=25 MPa [3]. 22
Figure 2.6 D-xylose decomposition in hot compressed water: 180℃T<220℃, P=10 MPa [56] 23
Figure 2.7 Kinetic model reaction mechanism for decomposition of xylose in supercritical water 450℃T<650℃, P=25 MPa [57] 24
Figure 2.8 D-xylose decomposition and gasification path in supercritical water: Kinetic model reaction mechanism for decomposition of xylose in supercritical water 450℃T<650℃, P=25 MPa [57] 24
Figure 2.9 GC-MS chromatogram for soluble products of guaiacol decomposition in supercritical water [59] 25
Figure 2.10 Simple reaction pathway for guaiacol (lignin model component) in supercritical water [59] 26
Figure 2.11 Simple reaction pathway of catechol decomposition in near- and supercritical water [60] 27
Figure 2.12 Catechol formation from guaiacol in sub- and supercritical water [61] 28
Figure 2.13 Phenol formation from guaiacol and catechol in sub- and supercritical water [61] 29
Figure 2.14 Reaction pathway of guaiacol under hydrothermal conditions: 300℃T<450℃, P=25 MPa [61] 30
Figure 2.15 Reaction scheme of phenol benzene decomposition in supercritical water: 300℃T<450℃, P=25 MPa [39] 30
Figure 3.1 Reaction scheme of cellobiose in near critical water: k1, k2, k3, kge.g, kgg.g from [51], kg.e, kg.a, kg.gly, kg.f, kg.a, kf.e, kf.gly from [53], kgly.dih, kdih.gly, kgly.p, kdih.p from [22], kg.5, kf.5, kf.fu , kfu.ch from [3] 33
Figure 3.2 Intermediate decomposition scheme kf.acid, kp.acid, ka.acid, ke.acid [54]; k5.lf, k5.ff [62] 34
Figure 3.3 Decomposition scheme of D-xylose in subcritical water [56] 35
Figure 3.4 D-xylose decomposition and gasification scheme in supercritical water kxy.fu,kxy.wshs,kfu.wshs,kaa.ga from [57] 36
Figure 3.5 Guaiacol decomposition scheme in subcritical water [61] 37
Figure 3.6 Guaiacol decomposition scheme in supercritical water kgu.ch, khu.ga, kgu.oc, kgu.c, kgu.t, kc.oc, kt.ch from [61]; kc.t, kp.c, kp.t, kp,ga, kp.ch, kt.b, kb.t, kb.p, kb.ga, kb.na, kna.ch, kb.ch from [39] 38
Figure 3.7 Organic acid gasification scheme in supercritical water. kfa.ga1, kfa.ga2 [63]; kaa.ga [64]; kwshs.ga, kpa.ga, kmf.aa [57]; lactic acid reactions [65] 39
Figure 3.8 Arrhenius relation of glucose to glyceraldehyde decomposition reaction 43
Figure 4.1 Simulation results of cellobiose decomposition products in isothermal subcritical reactor at 370 ℃ 55
Figure 4.2 Simulation results of glucose decomposition products in isothermal subcritical reactor at 370 ℃ 55
Figure 4.3 Simulation results of xylose decomposition products in isothermal subcritical reactor at 370 ℃ 56
Figure 4.4 Simulation results of guaiacol decomposition products in isothermal subcritical reactor at 370 ℃ 56
Figure 4.5 Simulation results of organic acid decomposition products in isothermal subcritical reactor at 370 ℃ 57
Figure 4.6 Simulation results of xylose decomposition products in isothermal supercritical reactor at 650 ℃ 58
TABLE OF FIGURES
Figure 4.7 Temperature profile in subcritical reactor 59
Figure 4.8 Temperature profile in supercritical reactor 60
Figure 4.9 Scheme of the experimental setup used in Lu’s work [9] 61
Figure 4.10 Effect of residence time on product gas molar flow 62
Figure 4.11 Comparison between simulation results GE, CE and experimental results GE*, CE* for residence time between 15 and 47 seconds 63
Figure 4.12 Comparison between product yield results of the simulation and experimental results (*), for residence time between 15 and 47 seconds 64
Figure 4.13 Comparison between simulation results GE, CE and experimental results GE*, CE* for residence reactor pressure between 200 and 300 bars 65
Figure 4.14 Comparison between product yield results of the simulation and experimental results (*), for reactor pressure between 200 and 300 bars 66
Figure 4.15 Comparison between simulation results GE, CE and experimental results GE*, CE* for different feedstock; rice straw, peanut shell, corn stalk, corn cob and wood sawdust. 67
Figure 5.1 Temperature profile in subcritical reactor for the 1st scenario 71
Figure 5.2 Temperature profile in supercritical reactor for the 1st scenario 71
Figure 5.3 Temperature profile in subcritical reactor for the 2nd scenario 72
Figure 5.4 Temperature profile in supercritical reactor for the 2nd scenario 72
Figure 5.5 Temperature profile in subcritical reactor for the 3rd scenario 73
Figure 5.6 Temperature profile in supercritical reactor for the 3rd scenario 73
Figure 5.7 GE and CE values of 1st 2nd and 3rd scenarios 74
Figure 5.8 Carbon mole ratio of unconverted liquid in the effluent, Carbon (unc.liq)% and molar flowrate of carbon contained in char for 1st 2nd and 3rd scenarios 75
Figure 5.9 Product gas composition values for first, second and third scenarios 76
Figure 5.10 Simplified heat exchange unit scheme 79
Figure 5.11 GE and CE results of the process simulation for Tmax= 650℃, 600 ℃, 550 ℃ 82
Figure 5.12 Biomass load effect on GE and CE 83
Figure 5.13 Biomass load effect on thermal energy supplied and net thermal energy production rate 84
Figure 5.14 Biomass load effect on specific thermal energy supplied and net thermal energy production rate 85
Figure 5.15 Biomass type effect on GE, CE and CGE 86
Figure 5.16 Biomass type effect on product gas compositions 87
Table 1.1 Proximate analysis and calorific values of pig manure, algae, cow manure and rice straw samples [5] 3
Table 1.2 Selected SCWG research groups [25] 9
Table 3.1 Cellobiose decomposition rate constants 41
Table 3.2 Glucose decomposition rate constants 42
Table 3.3Arrhenius Parameters of cellobiose and glucose decomposition 43
Table 3.4 Glucose decomposition rate constants continued [3] 44
Table 3.5 Arrhenius Parameters of glucose decomposition continued 44
Table 3.6 Xylose decomposition rate constants in subcritical water 45
Table 3.7 Arrhenius parameters of d-xylose decomposition subcritical water 45
Table 3.8 Arrhenius parameter of d-xylose decomposition supercritical water 45
Table 3.9 Lignin decomposition rate constants sub- and supercritical water [61] 46
Table 3.10 Arrhenius Parameters of guaiacol decomposition subcritical water 47
Table 3.11 Arrhenius Parameters of guaiacol decomposition supercritical water 47
Table 3.12 Phenol- benzene decomposition reaction rate constants supercritical [39] 48
Table 3.13 Arrhenius parameter of phenol-benzene decomposition in supercritical water 48
Table 3.14 Arrhenius parameters of formic acid decomposition [63] 49
Table 3.15 Arrhenius parameters of acetic and propionic acid decomposition 49
Table 3.16 Lactic acid decomposition reaction rate constants [65] 50
Table 3.17 Arrhenius parameters of lactic acid decomposition supercritical water 50
Table 4.1Simulation set parameters for preliminary run 54
Table 4.2 Simulation operating parameters; comparisons for varying residence time 61
Table 4.3 Simulation operating parameters; comparisons for varying pressure 64
Table 4.4 Biochemical and ash mass compositions of dry biomass samples used in the simulation 66
Table 5.1 Simulation set parameters 70
Table 5.2 Tubular reactor sizing parameters 79
Table 5.3 Base case simulation results 80
NOMENCLATURE
Roman / UnitsM / Moisture content / kgwater/kgbio
∆Hvapwater / Heat of vaporization of water / MJ.kg-1
Pc / Critical pressure / MPa
Tc / Critical temperature / 0C
V / Volume / m3
S / Surface area / m2
t / Time / s
k / Reaction rate constant / s-1
ks / Surface reaction rate constant / m.s-1
A / Pre-exponential factor / s-1
Ea / Activation energy / kJ.mol-1
Kw / Water ion product / mol2.dm-6
r / Reaction rate / mol.m-3s-1
R / Gas constant / J.mol-1K-1
T0 / Reference temperature / K
Ci / Molar concentration of i / mol.m-1
a, b / Volume parameter of SRK
c / Mathias-Copeman constants
mg,o / Mass flow rate of gas outlet / kg/hr
mbio,dry / Mass flow rate of dry biomass inlet / kg/hr
nc, g / Molar flow rate of carbon in gas outlet / mol/hr
nc,bio / Molar flow-rate of carbon in biomass inlet / mol/hr
ni / Molar flow rate of gas i / mol/hr
Yi / Molar yield of gas i / mol/kg
∆TLM / Log-mean temperature difference / K
U / Overall heat transfer coefficient / W.m-2K-1
hi / Convective heat transfer coefficient of fluid i / W.m-2K-1
kwall / Conductive heat transfer coefficient of wall / W.m-1K-1
r / Radius / m
Q / Heat transfer rate / MJ.s-1
Ltube / Tube length / m
Dtube / Diameter of tube / m
Qsupp, th / Externally supplied heat / MJ.s-1
Qth, net / Net thermal energy production rate / MJ.s-1
Q specific / Net specific thermal energy production / MJ.kg-1
Tr / Reduced temperature
Greek
ε / Dielectric constant
αi / Stoichiometric coefficient of i
v / Specific volume / m3.kg-1
ω / Accentric factor
Multiplication
Abbreviations
PFR / Plug flow reactor
CSTR / Continuous stirred tank reactor
TOC / Water soluble organic compound
DP / Decomposition product
WSHS / Water soluble humic solid
SCW / Supercritical water
SCWG / Supercritical water Gasification
SCWO / Supercritical water Oxidation
HHV / Higher (Gross) heating Value / MJ.kg-1
LHV / Lower (Net) heating value / MJ.kg-1
SOFC / Solid oxide fuel cell
PEM / Proton exchange membrane
CMC / Carboxy methyl cellulose
GE / Gasification efficiency / %
CE / Carbon gas efficiency / %
CGE / Cold gas efficiency / %
1 Introduction
1.1 Background
Pursue of continual global economic growth obliges mankind to guarantee a way to store, convert and amplify the energy. Huge difference between human energy exploitation rate and fossil fuel recovery (millions of years) do not leave any doubt about necessity of finding efficient and wide range applicable renewable energy production ways. Biomass is a dispersed, energy source contained in plants and animal wastes [1]. Energy from biomass is carbon neutral and environmentally benign thanks to the continuous loop of carbon between plants and atmosphere. Its exploitation is mainly done through biological or thermal conversion processes. Biological technologies mainly use microorganisms as energy converting agents from biomass to fuel. The oldest thermal biomass conversion, combustion have been used by generations and in developing countries still in general use. Biomass accounts for 14% of the world energy consumption while even though world annual biomass potential is estimated to be around 146 billion metric tons (almost half of the worlds annual energy demand) [2]. Fully exploitation of biomass is not realistic and reasonable, however it is clear that effective and widespread technologies are capable of decreasing the consumption of other fossil fuels. Higher technology processes, including biological technologies, have been worked in order to achieve high energy production capacity and wide scope of biomass use. Standard gasification has the biggest energy production share from biomass sources due to high-energy content gas product and technology maturity. Performance of the gasification process depend many operating conditions and biomass properties, including moisture. Thermal or electrical energy dissipated for the evaporation of water contained in wet biomass (grass, rice straw, sugar cane, algae, manure etc.) is still one of the biggest challenges of the current technology. However, degradation of biomass with the presence of water is also possible according to the early thermodynamic findings. Activity of water in the process can be multiplied by increasing the slurry temperature near the critical point under critical pressure (Tc, Pc). Physical changes in hot and compressed water are favorable for rapid hydrolysis, isomerization, dehydration and some other decomposition reactions [3]. Low carbonmonoxide (CO) compositions is a result of favored forward water gas shift reaction (WGSR) in SCWG.