Biochar production integration into Phoenix BioPower BTC technology

(1)

1

Biochar production integration into Phoenix BioPower BTC

technology

Àurea Solé Carulla

Master of Science Thesis

InnoEnergy SELECT Master Program

MJ232X Degree Project in Heat and Power Technology KTH - Department of Energy Technology EGI-2018

TRITA-ITM-EX 2018:667 SE-100 44 STOCKHOLM

(2)

2 Master of Science Thesis TRITA-ITM-EX 2018:667 Biokolproduktion integrerad med Phoenix BioPowers BTC teknik

Àurea Solé Carulla Godkänd

2018-09-12

Examinator Peter Hagström

Handledare Klas Engvall Uppdragsgivare

Phoenix Biopower

Kontaktperson Michael Bartlett

Sammanfattning

Phoenix BioPowers teknik BTC (Biomass fired TopCycle) integrerar förgasning av biomassa som ett försteg till en TopCycle högtrycksgasturbin. Dessutom injiceras ånga tillsammans med produktgasen i förbränningen, för att uppnå hög effektivitet samt minskar de totala kostnaderna samtidigt som tekniken erbjuder kontrollerbar och pålitlig förnybar kraft. Värmintegrationen med hjälp av ånga som värmebärare resulterar i hög värmeåtervinning genom processen och utgör en ytterligare effektivisering av BTC anläggningen.

Företaget har erhållit finansiering för ett projekt för CO2 negativ kraftproduktion, vilket syftar till att verifiera om BTC-teknik kan bli den första skalbara kommersiellt genomförbara tekniken som erbjuder CO2-negativ kraftproduktion genom att skapa en ny separat intäktsström från biokol. Genom att använda BTC-enheten under olika förhållanden kan biokol produceras tillsammans med det nödvändiga bränslet för att driva gasturbinen.

Uppsatsen presenterar en genomförbarhetsbedömning av integrationen av biokolproduktion i en BTC 30MWe-anläggning. Därför kommer integrationen av pyrolys och förgasningssteg före kraftenheten studeras och analyseras termodynamiskt och ytterligare optimeras för att uppnå de förhållanden under vilka högkvalitativ biokol produceras samtidigt som kraftproduktion och värmeåtervinning säkerställs. En beräkningsmodellen utvecklades för Aspen Plus® för att validera systemets genomförbarhet från teknisk synvinkel.

Resultaten av simuleringen visar en total anläggningsverkningsgrad på höga 88–91%, och elverkningsgrader på 52–55%. Medan modellen visar att systemet är i viss utsträckning är känsligt för förgasningstemperaturen, har temperaturen i pyrolyssteget en signifikant inverkan på den totala prestandan.

(3)

3 Master of Science Thesis TRITA-ITM-EX 2018:667 Biochar production integration into Phoenix BioPower BTC technology

Àurea Solé Carulla Approved

2018-09-12

Examiner Peter Hagström

Supervisor Klas Engvall Commissioner

Phoenix Biopower

Contact person Michael Bartlett

Abstract

Phoenix BioPower technology BTC (Biomass fired TopCycle) integrates biomass gasification as a prior step to a TopCycle high pressure gas turbine. Additionally, steam is injected together with the product gas into the combustor, achieving high efficiencies and reducing overall costs while offering controllable and reliable renewable power. Heat recovery is implemented within the system by means of steam as a heat carrier, making optimised heat integration an added asset of the plant.

The company has been funded to do a carbon negative power project, which aims to verify if the BTC technology can become the first scalable commercially viable technology offering carbon negative power, by generating a new separate revenue stream from biochar. By operating the BTC plant under different conditions, biochar can be produced along with the required fuel to run the gas turbine.

This thesis presents a feasibility assessment of the integration of biochar production into the BTC 30MWe plant. Therefore, the integration of pyrolysis and gasification steps before the power island will be studied and analysed thermodynamically and further optimised to obtain the conditions under which high quality biochar is produced while ensuring power generation and heat recovery. A computational model using Aspen Plus® was built to validate the feasibility of the system from the technical point of view.

The results of the simulation show an overall plant efficiency as high as 89-91%, and electrical efficiencies of 52-55%. While the model reveals that the system is slightly sensible to gasification temperature, pyrolysis temperature has a significant impact at the overall performance.

(4)

4

Acknowledgements

In the first place I would like to thank my supervisors: Michael Bartlett and professor Klas Engvall, who guided me through the development of the thesis and allowed me to fulfil this rewarding experience giving me the chance to work in cooperation with Phoenix BioPower. I really appreciate their dedication to me.

To my examiner Peter Hagström, for willing to help and making all the boring official procedures simple and effortless.

To Henrik Bage, who created an amazing atmosphere with his cheerfulness and optimism, in which one feels comfortable and willing to work.

To MSc SELECT family, whose friendship goes way beyond the studies, for sharing so many unforgettable moments during these two intense years. Coming from different continents and backgrounds, the willingness to make real impact to the energy transition and the power contained in every single person to accomplish their goals has been an incredible source of inspiration to me.

Finally, but not least, I acknowledge Innoenergy for proving the opportunity to make this journey possible, and to contribute towards a better, more sustainable and greener world.

Àurea Solé Carulla Stockholm, September 2018

(5)

5

Nomenclature

Variable/Parameter name Symbol

Biomass TopCycle BTC

Carbon Dioxide Capture and Sequestration CCS

Combined Cycle Gas Turbine CCGT

Combined Heat and Power CHP

Dry basis db

European Biochar Certificate EBC

Gas Turbine GT

Greenhouse Gases GHG

Heat Recovery Steam Generation HRSG

Heat Exchanger HX

Higher Heating Value HHV

High Pressure Compressor HPC

High Pressure Turbine HPT

Integrated Gasification Combined Cycle IGCC

Intercooler IC

Lower Heating Value LHV

Low Pressure Compressor LPC

Low Pressure Turbine LPT

Steam Injected Gas Turbine Cycle STIG

Volume basis vol

Weight basis wt.

Wet basis wb

(8)

8

Index of figures

Figure 1. Process flowchart of BTC plant. ... 12

Figure 2. CCGT generating set ... 19

Figure 3. Methodology approach of the master thesis ... 22

Figure 4. Flowsheet of the dryer simulation model ... 24

Figure 5. Flowsheet of the pyrolysis model ... 25

Figure 6. Flowsheet of gasification model based on Hannula et. al. [45] ... 27

Figure 7. Sequence of commands followed by gasification simulation model ... 28

Figure 8. Flowsheet of TopCycle Gas Turbine ... 28

Figure 9. Tar management simulation approach in the combustion reactor ... 29

Figure 10. BTC 30MWe plant, Aspen Plus® simulation sequence ... 31

Figure 11. Process streams Composite Curves. Baseline set-up conditions. ... 34

Figure 12. Optimal Heat Exchanger network design ... 35

Figure 13. BTC plant efficiency results for the study cases ... 36

Figure 14. Thermochemical conversion paths of biomass, operation conditions and main products ... 48

Figure 15. Net heat output for gasification of 7ton/h of biochar ... 51

Figure 16. Schematic of a TwoStage gasifier [55] ^, ... 53

Figure 17. BTC 30MW plant. Aspen Plus simulation flowsheet ... 55

(9)

9

Index of tables

Table 1. Operation conditions and typical char yield from thermochemical processes [13] ... 15

Table 2. Preferred conditions for biochar production with pyrolysis ... 16

Table 3. Effect of gasifying agent on producer gas composition [25] ... 17

Table 4. The major reactions in air/oxygen gasification, and respective heat of reaction ΔHº(KJ/mol) [28] ... 18

Table 5. Comparison of different power plants [34] ... 20

Table 6. Operating conditions of simulation models assessed ... 33

Table 7. BTC plant simulation system conditions results: Baseline and study cases ... 36

Table 8. BTC plant simulation energy balance results: Baseline and study cases ... 37

Table 9. Gantt chart of the master thesis development ... 49

Table 10. Air composition (%vol) at 1.013bar, 15 ºC ... 50

Table 11. Proximate and Ultimate analysis and HV of wet biomass and dry biomass ... 50

Table 12. Chemical characterization of wood chips and biochar residue from TwoStage gasification (%wt.) [47] ... 54

(10)

10

1 Background

Carbon-intense human activities have led to an alarming increase of greenhouse gas emissions in the atmosphere since mid-20th century. Extreme weather patterns, increase on global temperatures and arctic ice melting shields are some dreadful consequences emanated from climate change, affecting all regions around the world. Climate change has become one of the major concerns in our society, and its alleviation is an urgent need that involves the cooperation of all nations.

Paris climate accord within the United Nations Framework Convention on Climate Change came into force in 2017 to address global greenhouse gas emissions mitigation, aiming to pursue individual efforts to limit the temperature increase further to 1.5ºC above preindustrial levels [1]. To reduce emissions, economies must lower their carbon intensity, which implies a decisive shift towards a low carbon economy. Under a soft landing scenario, where the transition moving away from fossil fuels occurs gradually, the development of new technology and increased energy efficiency could stimulate innovation, create new jobs and lower production costs, resulting in a positive overall effect on the economy [2]. The growth of renewable energy technologies for power generation can also benefit national markets by reducing dependency on fossil fuels, opening the way for energy source diversification.

In contrast to fossil fuels, biomass releases carbon dioxide which is balanced with the carbon dioxide captured in its own growth. Biomass can derive to different types of products depending on the end uses: biofuel, biopower and bioproducts. While biofuel term indicates gas, solid or liquid fuels (synthetic gas, ethanol, biodiesel, biochar, etc.) derived from thermochemical or biological biomass conversion, biopower refers to a biomass or biofuel fuelled power CHP system.

Biopower is considered a renewable energy source, which contributes positively to energy security, offering a controllable power generation that can complement intermittent power produced by other renewables such as solar and wind. Additionally, biopower can be deemed a reliable energy source since the risk for plant shutdowns is small, as it uses well established technologies. From the large number of fuels that can be used in biopower plants, wood is still the largest biomass energy resource, yet other sources such as energy crops, grassy, municipal solid waste or algae can be used to produce bioenergy [3].

(11)

11 Recently, biochar has received increasing attention for its contribution to climate change mitigation. Its potential lies primarily on its highly recalcitrant nature, which slows down the fixed carbon return rate to the atmosphere [4].

The use of bioenergy in Sweden has increased from 40 TWh/year in 1970s to about 140 TWh in 2015, surpassing oil as the leading energy source by 2009. Biomass as energy source is particularly used in Swedish heat market for district heating purposes. According to Svebio, political support and strong incentives such as carbon tax (introduced in 1991), green electricity certificates (2003) and tax exemption for biofuels transport have given tremendous rise in bioenergy contribution, resulting in a decrease of GHG by 25% between 1990 and 2014 [5].

Sweden is currently at the leading edge internationally for wood supply systems and wood for process energy in the forest industry, district heating and CHP sector [6].

(12)

12

2 Introduction: BTC technology

The Swedish start-up Phoenix BioPower develops the Biomass-fired TopCycle gas turbine technology, namely BTC [7]. The plant integrates biomass gasification in combination with a top cycle gas turbine process running at high pressures and massive steam injection, achieving higher efficiencies of those compared to conventional Steam Cycle technology. Phoenix BioPower mission is to reach 60% electrical efficiency from biomass by year 2030 and aims entering the market with an electrical efficiency over 50% by 2023.

When converting from steam cycle to BTC, both marginal and lifecycle costs are significantly lowered and there is no need to invest in new infrastructure since the current infrastructure for fuel, power and heat can be utilised.

Figure 1. Process flowchart of BTC plant.

The plant is comprised of three main islands: biomass treatment island, power island, and heat recovery island represented in Figure 1 by yellow, red and blue, respectively.

As first step biomass needs to be converted to a gaseous fuel by means of a set of operations comprising the biomass treatment island: dryer, pyrolizer, gasifier, and gas clean-up. After lowering the moisture content of the wet biomass through a dryer, the resultant biomass is subjected to pyrolysis. The biochar product from pyrolysis is the gasification feedstock.

Gasification product gases together with pyrolysis gases and tars encompass the gaseous fuel

(13)

13 of the power island. The steam injected in both pyrolysis and gasification is also directed together with the fuel to the BTC system.

Secondly, gaseous fuel is fed to the the TopCycle gas turbine, where electricity is generated by operating at high pressures.

Finally, heat is recovered from the exhaust gases of the gas turbine. High temperature heat is first used to generate the steam needed for the system (i.e. pyrolizer, gasifier and cooling streams for the gas turbine) through a Heat Recovery Steam Generator (HRSG). The remaining low temperature heat left is further recovered by a condenser, connecting it to the district heating network.

Therefore, from the input biomass containing approximately 50%wt. of moisture content, the outputs of the BTC plant are the following:

• Unconverted biochar from gasification

• Electricity produced in the gas turbine

• Excess heat connected to the district heating

Another aspect of the plant worth to mention is that it is self-sustained in terms of heat requirements.

(14)

14

3 State of the art

This section provides a brief state of the art of the crucial operation blocks contained in the BTC system.

3.1 Pyrolysis

Even though biomass has a great potential as sustainable substitute of fossil fuels, its use presents a series of challenges as well. Its bulky and inconvenient form is a major barrier, as requires space for growing and expensive cost of transportation. This, with other disadvantages as high oxygen and water content, provides a major motivation for the conversion of solid biomass into solid, liquid and gaseous fuels [8]. The different existing thermochemical conversion methods are presented in Appendix A.

Pyrolysis is defined as the thermal degradation of biomass by applying heat either in the total absence of oxidizing agent or with a limited supply that does not permit to initiate gasification.

The resulting products include a range of useful products: the chemically altered solid particle (biochar), liquids (tars) and permanent gases including methane, hydrogen, carbon monoxide, and carbon dioxide.

Low temperature mild pyrolysis, the so-called torrefaction process, is a thermochemical conversion process that can be applied as thermal pre-treatment of both gasification and combustion of biomass [8]–[10], since it improves biomass properties by increasing energy density and decreasing both moisture and hydrophilicity [11]. Also, the resulting biochar presents lower O/C ratio than the original biomass, which enhances gasification efficiency.

Pyrolysis is commonly carried out in a relatively low temperature range of 200-600ºC, in comparison to gasification which achieves 700-1000ºC [8].

The overall reaction of pyrolysis could be approximated as follows:

𝐶𝐶_𝑛𝑛𝐻𝐻_𝑚𝑚𝑂𝑂_𝑃𝑃(𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵)^{ℎ𝑒𝑒𝑒𝑒𝑒𝑒}�⎯� 𝐶𝐶(𝑏𝑏𝐵𝐵𝐵𝐵𝑏𝑏ℎ𝐵𝐵𝑎𝑎) + � 𝐶𝐶_𝑥𝑥𝐻𝐻_𝑦𝑦𝑂𝑂_𝑧𝑧

𝑔𝑔𝑒𝑒𝑔𝑔 + � 𝐶𝐶_𝑥𝑥𝐻𝐻_𝑦𝑦𝑂𝑂_𝑧𝑧

𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 + 𝐻𝐻₂𝑂𝑂 (R1)

The pyrolytic decomposition of carbonaceous feedstock involves a complex series of reactions, yet the overall process is endothermic. Yields and properties of the products strongly rest on feedstock biomass composition, operating conditions (heating rate, final temperature, residence time, pressure, etc) and reactor configuration. Recari et. al studied the effect of pyrolysis

(15)

15 temperature and pressure on the product yields, concluding that formation of secondary char by cracking of tars becomes more important at high pressures [11].

3.1.1 Pyrolysis products

Biochar is defined as a solid material obtained from pyrolysis, generated through thermal degradation of lignin and hemicellulose which has a promising fuel characteristics [12].

Not only biochar can already be used in many applications with remarkable effects, but an appealing feature is that it represents a cheap, sustainable and easy-produced process [13], sometimes even as a by-product. Although some of the applications are still under research development the usage of biochar as soil enhancer has received increasing attention as it sequestrates carbon, while simultaneously provides energy and increases crop yields [4].

Biochar can alternatively be used as a gasification feedstock or to provide heat.

Biochar can be produced from domestic level to large industrial processes, produced in solid form by dry carbonization, pyrolysis or gasification of biomass. Qian et al [13] studied the conditions under which the different thermochemical processes are commonly operated and the biochar yields, summarised in Table 1.

Table 1. Operation conditions and typical char yield from thermochemical processes [13]

Process Temperature (ºC) Residence time Char yield (wt%)

Slow Pyrolysis 400-600 Min to days 20-40

Fast pyrolysis 400-600 1 s 10-20

Gasification 800-1000 5-20 s 10

Hydrothermal

carbonization 180-250 1-12 h 30-60

The elemental composition of the produced biochar varies depending on the feedstock. Also, H/C and O/C atomic ratio decrease when torrefaction/pyrolysis process is applied.

Consequently, the char trends towards pure carbon resulting in higher heating values than the biomass feedstock [14]–[16]. Energy densification from pyrolysis occurs only in the organic fraction of the feedstock [16].

Given that the present thesis focuses on the biochar production integrated in the BTC system, pyrolysis operation conditions are of crucial importance to maximize quality and yield of primary biochar. Table 2 presents the conditions that favour biochar’s yield.

(16)

16 Table 2. Preferred conditions for biochar production with pyrolysis

Preferred condition Ref.

Lignin content High [17]

Sample size Large [8]

Heating rate Slow (<0.01–2.0 °C/s) [8], [13]

Pyrolysis temperature Low (300-350ºC) [8], [11]

Pressure High [11]

Gas residence time Long [8], [13], [14]

Reactor type and shape, feedstock type, particle size, drying treatment, heating rate, pressure and inert gas flow are parameters that have considerable effect on both biochar yield and biochar physical characteristics [18]. For instance, the increment of pressure favours the formation of biochar, while pyrolysis temperature has the inverse effect as at high temperatures biomass components further decompose [19]. According to Basu [8], as larger particles offer higher resistance to escape of the primary pyrolysis product, secondary cracking takes place more easily.

As the biochar market for agricultural uses is expected to increase significantly in the next years, the European Biochar Foundation has created the European Biochar Certificate aiming to enable and guarantee sustainable biochar production [20], [21].

Pyrolysis liquid yield, also known as tar or bio-oil, is composed by complex hydrocarbon compounds and contains up to 20% of water. According to Piskorz et. al. [22], the main compounds found in tars are phenols, carboxylic acids, hydroxyaldehydes and sugar among others. It is produced by depolymerisation and fragmentation of the lignin, hemicellulose and cellulose, and its yield is enhanced by a rapid increase in temperature followed by a prompt quenching to prevent further degradation [8].

The non-condensable fraction of gases contains low molecular weight compounds such as CO2

and CO with some traces of CH4, C2H6 and C2H4 [8], [23].

(17)

17 3.2 Gasification

Gasification is a partial oxidation of solid fuels which aim is to generate gaseous fuels. By means of an oxidizing medium, the biomass is treated under sub-stoichiometric conditions yielding producer gas and lower amounts of by-products [9], [24]. Table 3 shows the producer gas composition utilising different gasifying agents. Once producer gas is purified, one of its potential applications is producing electricity in a gas turbine system.

Table 3. Effect of gasifying agent on producer gas composition [25]

Gas composition

(vol%, db) Air Steam Steam + oxygen

H2 11-16 35-40 23-28

CO 13-18 22-25 45-55

CO2 12-16 20-25 10-15

CH4 2-6 <9-11 <1

N2 45-60 <1 <5

LHV (MJ/Nm³) 4-6 12-14 10-12

Similarly, combustion also converts carbonaceous species into gases, although the principles are different. Combustion is an exothermic reaction taking place in an oxidizing environment producing flue gases, which do not have useful heating value. On the other hand, gasification produces a combustible energy gas with stored chemical energy, as the reaction takes place under a reducing environment requiring heat [8].

Gasification can be coupled with advanced turbines for electricity generation in an Integrated Gasification Combined Cycle (IGCC) offering better efficiencies than coal power plants, especially in a carbon dioxide capture and sequestration (CCS) scenario [26].

Air, oxygen, steam or a mix of the previous can be used as a gasification agent. Air is the most common oxidation medium due to its economic and operational advantage. However, the gas produced possesses low heating values (see Table 3). While air is applied to generate a producer gas utilised as an energy gas, oxygen is applied when the aim is to use the producer gas for further upgrading for synthetic fuel applications. Applying oxygen as an oxidation medium the gas produced reaches high heating values [27]. The principal reactions that take place under air/oxygen gasification are summarised in Table 4.

(18)

18 Table 4. The major reactions in air/oxygen gasification, and respective heat of reaction ΔHº(KJ/mol) [28]

𝐶𝐶 + 𝑂𝑂2↔ 𝐶𝐶𝑂𝑂2 −394 (R2)

𝐶𝐶 + 0.5𝑂𝑂2↔ 𝐶𝐶𝑂𝑂 −111 (R3)

𝐶𝐶 + 2𝐻𝐻2↔ 𝐶𝐶𝐻𝐻4 −74 (R4)

𝐶𝐶 + 𝐶𝐶𝑂𝑂₂ ↔ 2𝐶𝐶𝑂𝑂 +172 (R5)

𝐶𝐶𝐻𝐻₄+ 𝐻𝐻₂𝑂𝑂 ↔ 𝐶𝐶𝑂𝑂 + 3𝐻𝐻₂ +206 (R6)

𝐶𝐶𝑂𝑂 + 𝐻𝐻₂𝑂𝑂 ↔ 𝐶𝐶𝑂𝑂₂+ 𝐻𝐻₂ −41.1 (R7)

Gasification with steam operates under slightly different conditions, as an external heat source is required. As water-gas shift reaction occurs (R7), hydrogen content increases. Steam gasification is particularly interesting when gasification is coupled in a system with heat recovery integration, as the high temperature enhances the devolatilization process of biomass and at the same time steam improves heating value.

It is worth to differentiate between gasification, partial gasification and mild gasification processes. In gasification the feedstock goes through complete devolatilization and thermal cracking into light compounds as CO, H2, CO2, H2 and CH4. Alternatively, only part of the char is reacted in partial gasification by controlling the contact surface and time between CO2/H2O with the char after de-volatilization. Mild gasification does not deal with gasification level, but to the level of thermal cracking of volatiles. It emphasizes preserving the heavy volatiles without further gasifying fixed carbon by controlling the temperature and residence time [29].

The yield and composition of producer gas particularly depend on the performance of the solid particle. In fact, several studies prove the improvement of yield and quality content of the product gas by adding a previous pyrolysis step [15], [30]. Other parameters as gasification conditions (temperature, equivalent ratio, and pressure) and reactor design have high influence on the producer gas quality as well. Nonetheless, the bibliographic revision denotes that pressure has not been thoroughly studied [15].

(19)

19 3.3 A review on Gas Turbines

Gas turbines are the favoured prime mover in large-scale cogeneration system. The technology is based on heat engines that use air as their working fluid, operating under the so-called Brayton thermodynamic cycle. The temperature of exhaust gases exiting the turbine reach 400-620ºC.

With such high temperature gases, heat can be recovered and be used thereafter to provide heating, hot water or to generate steam.

There are essentially three types of gas turbine cycles depending on how the exhaust gases are processed: simple cycle, regenerative cycle and combined cycle. Simple gas turbines are basically composed of three main elements: compressor, combustor and power turbine. They are a completely self-contained unit that requires only fuel and air in order to generate mechanical, rotational power [31]. The gas exiting the turbine is exhausted directly to the atmosphere. The electricity produced by the turbine is sufficient to drive both the compressor and the generator, resulting with efficiencies from 20% to 42% [32], [33]. In regenerative cycles, the exhaust gas is used in an exchanger (regenerator) to preheat the compressor discharge prior to the combustor, improving the efficiencies to 30-45% [32].

Figure 2. CCGT generating set¹

A more complex configuration is the Combined Cycle Gas Turbine (CCGT), where the exhaust gas from the gas turbine produces high-pressure steam which then powers a second generator.

CCGT can reach efficiencies of 45-61% at very large scale. The steam turbine thermodynamic performance is based on Rankine cycle, which converts heat into mechanical work while

1 Source: Energy Solution Center, Distributed Generation Consortium

(20)

20 undergoing phase change. Figure 2 depicts the configuration of a CCGT generation set.

Additionally, the resulting steam from the steam turbine can be used to provide heat load.

Gas turbines can utilize a variety of fuels such as natural gas, fuel oils and synthetic fuels. In contrast to internal combustion engines in which combustion occurs intermittently, gas turbines require continuous combustion. Given the continuous operation, carbon based solid fuels such as coal or biomass need a gasification step to convert them to pressurised gaseous fuel. Also, waste fuels such as biogas or landfill gas can be used to feed the combustion chamber. As the combusted fuel passes through the turbine, clean gases are required to prevent blade erosion [33].

The fuel choice will determine the atmospheric emissions from the gas turbine. Among the most common pollutants produced during gas turbine operation NOx, unburnt carbons, sulphur oxides and particulates are especially necessary to be considered and control applications need to be implemented to ensure regulation levels are not surpassed.

Among several methods applied to decrease NOx emissions, the steam injected gas turbine cycle (STIG) results particularly interesting. STIG consist of injecting high-pressure steam generated from heat recovery at the combustion chamber increasing the flow through the turbine. By this working principle, the peaks in flame temperature are decreased and thus resulting in thermal NOx generation reduction. Additionally, it also influences in the overall cycle by improving the thermodynamic performance increasing the power output and the electrical efficiency. Steam injection gas turbines are very suitable for process industries given their flexible operation conditions.

Table 5 compares the most common values of power output, electrical efficiency and investment cost between simple, combined cycle and steam injection gas turbines.

Table 5. Comparison of different power plants [34]

Simple GT CCGT STIG

Power output (MW) 29.1 41.3 37.3

Electrical efficiency (%) 36 51.1 46

Investment cost (M€) 11.6 31 20.9

Investment cost (€/kW) 398.6 750.6 560.3

(21)

21

4 Objectives and scope

The company is currently working on a project, whose aim is to verify if BTC technology can become the first scalable commercially viable technology offering carbon negative power, by generating a new separate revenue stream from biochar.

The scope of the master thesis proposal falls under the carbon negative power project objectives, covering the technical approach with a thermodynamic analysis of the system. The overall potential is compared between alternatives, assessing the concept feasibility.

The carbon negative power project features a cooperation between Phoenix BioPower, KTH Gasification research group and TU/Berlin combustion research group for developing the requirements of the system, with detailed customer input from Stockholm Exergi AB.

The master thesis is subdivided into the following work tasks:

 Literature research

 Low Temperature Pyrolysis model

 Fluid Bed Gasification model

 Establishment of the baseline model: Integration of power island into pyrolysis and gasification model

 Heat recovery optimization

 Sensitivity analyses on the operation conditions of the baseline model

 Modelling of study cases: high temperature gasification (A) and high temperature pyrolysis (B)

As it can be seen from the previous tasks list, firstly the focus is on the biochar and gaseous fuel production from thermochemical conversions, namely biomass treatment island. Later, the power island, comprising the TopCycle gas turbine, is integrated into the gasifier island, and the heat recovery of the whole system is optimised.

A simulation of the baseline system and study cases A and B models, accompanied by the list of parameters, assumptions and empirical data used for its development will be summarised in the present report. A preceding literature review of the different parts of the system, and the follow up technical analysis of simulation results are also included.

The Gantt chart of the project, which contains the timeline of the abovementioned tasks implementation, is included in Appendix B.

(22)

22

5 Methodology

The method of attack of the thesis is depicted in Figure 3, where the interconnection of different tasks is presented. The inputs considered for the specific achievements are represented by yellow boxes, while red arrows stand for flow of information.

Figure 3. Methodology approach of the master thesis

Aspen Plus® software has been chosen to design, simulate and optimise the system integration thermodynamics [35], [36]. It is also adequate to estimate compound properties and perform sensitivity analyses of different process variables, among other features.

Phoenix BioPower, in cooperation with KTH Gasification research group, has undertaken experiments to assess the pyrolizer operation under high pressures at the existing laboratory rig located at Finspång (Sweden). The experimental results have been considered as data input for the simulation of the pyrolizer.

A constant supervision from both Phoenix BioPower and KTH has been supporting the development of the project and has ensured its realization throughout the duration of the thesis.

(23)

23

6 Simulation analysis

Building on the set-up of the BTC plant in chapter 2, this section deals with the description of the approach followed for building the thermodynamic simulation model.

The possibility to study the performance of the BTC plant by using simulation tools as Aspen Plus® is valuable for a better understanding of the overall process. A validated simulation model could give an approximate answer to many factors affecting the efficiency of the plant.

In this sense, the simulation approach of the different blocks has been studied thoroughly, so that the integration of all the blocks following a bottom-up approach gives a prediction of the plant performance.

The process-oriented software Aspen Plus V9 is used to create a computational model for the mass and energy balances of the BTC plant. The heat exchanger integration system is designed using Aspen Energy Analyzer, a module from the Aspen One Engineering suite, which can be used to find and achieve the energy savings opportunities within the simulation flowsheet in Aspen Plus.

The global property method used is Redlich-Kwong Soave equation of state. However, some components taking part in the system are not present in Aspen Plus data banks, i.e. biomass, biochar, tars and ash. Therefore, they are defined as Nonconventional components and do not participate in chemical or phase equilibrium. The property methods used for the abovementioned components are HCOALGEN and DCHARIGT, which use the proximate, ultimate and sulphur analyses to calculate the enthalpy and density of Nonconventional components.

Given that pyrolysis and gasification reactor blocks need to deal with reactions that involve Nonconventional components, FORTRAN external subroutines are added and dynamically linked to the simulation when it runs.

The simulation was completed for the baseline case, as well as for the different study cases varying some thermodynamic conditions of the system, i.e. temperature of pyrolysis and gasification. Appendix C presents the boundary conditions of the BTC system.

(24)

24 6.1 Dryer

Most of solid fuels especially biomass contain high concentrations of moisture and a drying step may be of crucial importance for saving both time and energy in the pyrolysis or gasification reaction.

Following the specifications of Phoenix BioPower, the belt conveyor is the drying technology identified to be suitable for the BTC plant. In particular, the simulation has been based following the technical details of SWISS COMBI belt dryer [37]. As the belt drier is designated for the thermal treatment of the biomass, the simulation model considers the principles of a heat exchanger. In the first place, the heat required to vaporize the moisture from 50%wt to 15%wt.

at the operating pressure is calculated.

Applying the heat exchanger efficiency, the mass of air needed to dry the biomass particle to the given target is calculated, as well as the power needed to compress the corresponding air flow. The flowsheet of the drying simulation block is depicted in Figure 4.

Q_wat= Q_air· η (Eq. 1)

mwat[c_wat(Tsat− T0) + λwat] = [mairc_air(Tout− T0)]·η (Eq. 2)

It is worth mentioning that given the biomass definition as non-conventional component, additional blocks are considered in the Aspen Plus simulation to overcome this limitation. In fact, the input wet biomass and the dry biomass obtained as a product after the drying process are defined as two different components which differ only by the moisture content. Therefore, in order to tackle the abovementioned components, a calculator block is added to convert the wet biomass to dry biomass and water, ensuring the mass balance by a FORTRAN subroutine.

Figure 4. Flowsheet of the dryer simulation model 50% moist 15% moist ^S

15% moist Air, H2O

Air Biomass Water

(25)

25 6.2 Pyrolysis

Several studies of thermodynamic equilibrium simulation models with Aspen Plus for biomass pyrolysis have been performed [18], [38], [39]. Although the use of an equilibrium approach gives the opportunity to estimate yields and composition of pyrolysis products depending on reactor conditions, the main conclusion drawn from the authors is that equilibrium approach is not appropriate as the products are not predicted correctly. In fact, Sinha et al. [40] identified that the complication of wood pyrolysis models is mainly due to complexities of wood constituents composition, structural effects, heating rate effects and residence time effects.

This simulation presents an empirical model of a pyrolysis reactor based on experimental data extracted from tests performed by Phoenix BioPower in collaboration with KTH Gasification group. The reaction tests were taken at low temperature (320 – 407ºC) and high pressure (20 – 42 bar), injecting high pressure steam. The experimental data applied to the simulation model is the following:

• Pyrolysis product mass yields (biochar, tars, pyrolysis gases)

• Permanent gases composition

• Ultimate and proximate analysis of biochar

It is worth mentioning that little to no experimental data about high pressure pyrolysis with steam has been found in literature.

The pyrolysis model takes the dried biomass and steam as input, as Figure 5 shows. While the steam injected is not acting as a reactant in the pyrolysis reaction, it has been assumed that the moisture present in the biomass is involved in the reaction. Given the complexity on the composition of some compounds, biomass, tars and biochar are defined as non-conventional components in Aspen Plus. Therefore, a FORTRAN subroutine is added to ensure the mass balance of the block.

While biochar’s ultimate analysis and the permanent gases composition are taken from the experimental analysis, the tars composition becomes more complex as it contains hundreds of

Figure 5. Flowsheet of the pyrolysis model Tars CO, CO2

Biochar

15% moist^S Biomass

Steam

(26)

26 organic oxygenate compounds such as phenol, cresols, organic acids, etc. Therefore, the tars composition is calculated by atom mass balance of the pyrolysis system. Additionally, reaction water formed is assumed to account for 20% of total mass of condensate [41].

6.3 Fluidised Bed Gasification

An exhaustive literature search about gasification thermodynamic models shows that the authors tend to use either equilibrium or kinetic approaches to predict product gas composition.

Dynamic kinetic model gives a better interpretation of the real case. For instance, a study developed by Pryianka et. al. [42] showed that the model is capable of predicting gasifier performance in light with experimental data, being validated under various operating conditions. However, an Aspen Plus model considering dynamic parameters involved in the reaction becomes excessively complex for the scope of this master thesis.

Else ways, if gasification reaction is assumed to occur fast enough to reach equilibrium and thus disregarding kinetics, the model becomes significantly simpler. The Gibbs reactor uses Gibbs free energy minimization with phase splitting to calculate equilibrium [35]. However, the equilibrium model results lack accuracy in the product gas composition. Following these lines, Guilnaz et. al. [43] compared different approaches proposed by several authors to overcome the barriers and complexities involved in developing an accurate model for the gasification of biomass [44]–[46].

The gasification model as part of the BTC system has been based on Hannula et. al. study [45], where a semiempirical model is proposed by adding a correction factor for altering the equilibrium constant of some components which would otherwise be estimated erroneously.

The correction factors are based on experimental data of air-blown fluidised-bed gasification of sawdust biomass at 4-5bars and 856-955ºC. The correlations perceived depend on the air ratio E defined in Eq. 3, where 𝐵𝐵𝑂𝑂 is the weight of used oxygen and 𝐵𝐵𝑔𝑔𝑒𝑒,𝑜𝑜 is the weight of oxygen required to reach full combustion.

𝐸𝐸 = 𝐵𝐵_𝑂𝑂

𝐵𝐵_{𝑔𝑔𝑒𝑒,𝑜𝑜} (Eq. 3)

The adjustments obtained from the study correct carbon conversion and light hydrocarbon formation reactions. The correction factors are presented in Eq. 4- Eq. 8.

(27)

27

𝐶𝐶_{𝑐𝑐𝑜𝑜𝑛𝑛𝑐𝑐}= 25.7𝐸𝐸 + 88.5 (Eq. 4)

𝐶𝐶𝐻𝐻4= 0.5166 − 0.8621𝐸𝐸 (Eq. 5)

𝐶𝐶₂𝐻𝐻₂= 0.0046 (Eq. 6)

𝐶𝐶₂𝐻𝐻₄= 0.138 − 0.311𝐸𝐸 (Eq. 7)

𝐶𝐶2𝐻𝐻6= 0.02 − 0.038𝐸𝐸 (Eq. 8)

The inputs of the gasification system considered are the biochar, air and steam; product gas and unconverted biochar are the outputs. In order to build the model in Aspen Plus, different unit blocks are required, as schematised in the structure of the model at Figure 6.

In the first place the biochar is decomposed into its elemental composition following its ultimate analysis by means of a yield reactor supported by a FORTRAN subroutine. Then the unconverted biochar is separated accordingly to the carbon conversion correlation (Eq. 4). The composition of the unconverted biochar is based on a study made by V. Hansen et. al [47], where wood gasification biochar residue is chemically and physically characterised.

Specifications on the chemical characterisation of unconverted biochar provided by the study are presented in Appendix F.

The third block of the gasification flowsheet is a stoichiometric reactor, where methane, ethane, ethylene and acetylene are formed based on the hydrocarbon formation correlations applied.

Both the separator and the hydrocarbon formation stoichiometric reactor are backed by FORTRAN subroutines.

Decomposition HC Formation

Equilibrium

Q

^losses

Unconverted biochar Biochar

S

S Air

Steam

Figure 6. Flowsheet of gasification model based on Hannula et. al. [45]

(28)

28 Finally, the elements are mixed with air and steam and subjected to equilibrium in a Gibbs reactor, following the reactions presented at Table 4. The hydrocarbons formed at the previous step are assigned as inert in the equilibrium reactor.

As the gasification conditions (pressure and temperature) are predefined, it is important to control the heat output of the system. This is done by connecting all the heat streams to a heat sum block and adjusting the net total heat output from the block by varying the amount of air injected to the gasifier; a sensitivity analysis of gasification variables is attached in Appendix E. In this way assuming a certain percentage of gasification losses, the air ratio is calculated.

Taking into consideration that the gasification system encompasses several operation units and calculator blocks, Figure 7 summarizes the sequences order stipulated in the Aspen Plus model.

6.4 Power Train

The TopCycle Gas Turbine modelled comprises two stages of pressure with the aim to generate electricity from the combustion of the gaseous fuel at high pressures. The structure of the model is depicted at Figure 8.

Between the first and second compressor an intercooler is placed to lower the temperature at the high-pressure compressor. Note that the mass flow of air being compressed encompasses the amount required for both gasification and fuel combustion. Combustion of the gaseous fuel occurs under equilibrium at 55 bar, with 10% oxygen excess.

Figure 8. Flowsheet of TopCycle Gas Turbine

Figure 7. Sequence of commands followed by gasification simulation model

Set net heat losses Calculate air ratio Apply correlation factors Run simulation environment

G G

20bar

55bar Combustor

to Gasifier

Fuel Quench

Air Water Fuel

300ºC

55bar

(29)

29 Considering the fuel conformation, while gasification product gases and pyrolysis gases are conventional components, tars are defined as non-conventional components. Therefore, if no action is taken tars would bypass combustion as inert. To address this limitation, tars need to be decomposed by its ultimate analysis elements (similarly to the biomass decomposition in pyrolysis and the biochar decomposition in gasification), as Figure 9 shows. The heat requirement for the decomposition is provided by the combustor, so that the overall net heat balance for the combustion block is zero. The balances of both mass and energy are specified by FORTRAN subroutines.

Quench water injection ensures that the inlet stream temperature to the high-pressure turbine do not surpass the target combustion temperature. The pressure drops from compressors, diffusers, duct and turbines are considered, as well as the cooling flows injected at the turbines for safety reasons. All the above-mentioned information, together with the compressors and turbines efficiencies assumed are provided by Phoenix BioPower.

6.5 Heat Recovery

As there are heat requirements as well as cooling requirements within the BTC plant, the heat integration is strategically allocated in the system in order to have a self-sustainable process which maximizes the heat sent to the district heating network.

High mass flow and temperature of the flue gas can provide the necessary heat to produce steam at high pressures and hot water required for the system. Steam is needed in pyrolysis and gasification, while hot water is used as quench agent and as cooling stream in the power train.

HRSG is the process where the high temperature heat from the hot flue gas is captured and utilised to generate steam. The air passing through the intercooler is another high temperature

Figure 9. Tar management simulation approach in the combustion reactor Air

Steam

Pyrolysis gases Gasification product TAR

Q = 0

C, H2, O2

Exhaust Gases

S

(30)

30 heat source available in the system. On the other hand, the heat source of the dryer is the low temperature water from the flue gas condenser.

The heat exchanger integration system is designed using Aspen Energy Analyzer, a module from the Aspen Tech process design suite, which analyses all the heat requirements and finds an optimal heat exchanger network design using pinch point analysis method.

As the high temperature heat available in the system is limited, there is a maximum amount of steam that can be self-generated. This parameter is referred as steam-to-carbon ratio. It is defined as the amount of steam injected in the system divided by the amount of carbon coming from the biomass, on weight basis.

𝑆𝑆/𝐶𝐶𝑔𝑔𝑙𝑙𝑜𝑜𝑔𝑔𝑒𝑒𝑙𝑙 =𝑆𝑆𝑆𝑆𝑆𝑆𝐵𝐵𝐵𝐵_{𝑝𝑝𝑦𝑦𝑝𝑝𝑜𝑜}+ 𝑆𝑆𝑆𝑆𝑆𝑆𝐵𝐵𝐵𝐵_{𝑔𝑔𝑒𝑒𝑔𝑔𝑙𝑙} 𝐶𝐶𝑔𝑔𝑙𝑙𝑜𝑜𝑚𝑚𝑒𝑒𝑔𝑔𝑔𝑔

(Eq. 9)

The efficiency of the system is being influenced positively as the steam-to-carbon ratio increases. Additionally, the low temperature excess heat is likewise recovered, and connected to the district heating network.

6.6 Model integration

Once all the different simulation blocks are built and validated independently, their integration to simulate the plant as a unit is crucial. The Aspen Plus flowsheet of the BTC 30MW power plant model comprises 5 non-conventional components (i.e. ash, wet biomass, dry biomass, biochar and tars), 116 streams, 60 different blocks, 16 calculator blocks, 4 convergence design specifications and 10 transfer flows. The whole BTC plant flowsheet is attached at Appendix H.

In Aspen Plus software, each unit functioning block is solved as per certain sequences.

Considering the complexity of the simulation system and the interconnexion of dependent variables between the different processes, a sequence order needs to be defined by the user.

Figure 10 shows the sequence implemented every run. The boxes represent the FORTRAN calculator blocks and design specifications.

(31)

31 Figure 10. BTC 30MWe plant, Aspen Plus® simulation sequence

(32)

32 6.7 Assumptions

Some assumptions have been required in order to keep the development of the simulation model built with Aspen Plus as much realistic as possible within the scope of the master thesis.

• The compounds considered at the ultimate analysis of biochar and biomass are C, H and O. Therefore, N, Cl and S have not been contemplated, as they are not essential for the evaluation of the BTC system performance (i.e. power generation).

• As stated in section 6.2, reaction water formed in pyrolysis accounts for 20%wt of the liquid fraction [8], [41].

• Oxygen content present in the ultimate analysis is decomposed as O2 in biochar decomposition reactor (see Figure 6). When calculating the air ratio, this amount of oxygen is subtracted from the stoichiometric air needed. Although it is a big assumption and it is not what happens in reality, it was the only way to advance with the Aspen Plus simulation without achieving a too high oxidation level for the products. Tar decomposition (see Figure 9) is addressed in the same way when calculating the excess air combustion.

(33)

33

7 Results

The simulation of 30MWe BTC plant has been run under different conditions. Firstly, the baseline model reflects the scenario where the biochar yield is maximised in pyrolysis (high pressure, low temperature) and the biochar is subjected to moderate gasification temperature (850 ºC) to produce gaseous fuel. Study cases analyse the system performances by changing the temperature of both gasification temperature (Case A) or pyrolysis temperature (Case B) separately. Table 6 summarises the operating conditions of each simulation case.

Table 6. Operating conditions of simulation models assessed

Pyrolysis temperature (ºC) Gasification temperature (ºC)

Baseline model 326 850

Case A 326 950

Case B 396 850

As mentioned in chapter 6, two important input parameters are dependent variables defined by the simulation approach: steam-to-carbon ratio and air ratio. The other operating conditions remain unchanged, so that a more realistic comparison between scenarios can be made.

A more detailed study has been performed by the baseline model, where a heat exchanger network has been suggested.

7.1 Baseline model

Under the baseline conditions, 24.4 ton/h of wet biomass (50% wt. moisture content) are required to generate 30 MWe. In this scenario, 410 kg/h of unconverted gasification biochar are produced, being equivalent to 3.3 MWth; at the same time 18 MWth of heat excess are connected to the district heating network. Overall, the system efficiency is as high as 88.5%, reaching an electrical efficiency of 51.8%. The results of the BTC baseline Aspen Plus simulation are summarised in Table 7.

Regarding heat integration, the maximum amount of steam that is possible to self-generate within the system is 25.3 ton/h, corresponding to a global steam-to-carbon ratio of 4.27 (weight basis), i.e. total steam available to total carbon contained in the raw biomass.

(34)

34 The composite curves of the hot and cold process streams are plotted in Figure 11. From the hot streams side, there are the flue gases exiting at 531ºC and the pressurised air entering the intercooler at 447ºC, being cooled down to 300ºC. On the other hand, the cold streams to be heated are the following: hot air required to dry the biomass, quench water (injected at 75ºC), and feedwater to generate the steam at 300ºC, 75bar. Note that while the high temperature excess heat is all recovered by producing steam (right side of the plot), there is still some low temperature excess heat left, which is connected to the district heating network. However, since the district heating is considered a utility stream, it is not included in the composite curve.

Figure 11. Process streams Composite Curves. Baseline set-up conditions.

An optimal heat exchanger network design proposed by Aspen Energy Analyzer using pinch point analysis method comprises 12 Heat Exchangers (see Figure 12). The minimum temperature difference, which is defined as the minimum permissible temperature difference between hot and cold streams entering and exiting an individual heat exchanger between process streams, is set to 10ºC. Note that two heat exchangers are allocated for supplying the excess heat to the district heating network.

It is necessary to mention that the sizing and rating of the heat exchangers has been considered outside the scope of the master thesis. Therefore, there is a need to validate how realistic the area of the proposed heat exchangers is, which has been identified as a key issue for continued work.

District Heating

Air dryer + Feed water

Flue gas condenser Air dryer

(35)

35 According to Stockholm Exergi AB [48] the returning temperature to the district heating distribution plant in Stockholm is approximately 41ºC. In this simulation, the outlet temperature to the distribution is assumed 5 Celsius degrees below the flue gases dew point temperature.

Considering that the flue gases conditions change depending on the set-up of the plant, a FORTRAN calculator block has been added to calculate such temperature. In the baseline scenario case, this temperature is 71.5 ºC.

District Heating

Flue gas

Intercooler air

Steam boiler

Feed water heated

Air dryer Feed water preheated

Figure 12. Optimal Heat Exchanger network design

(36)

36 7.2 Study cases

After having analysed the results of the system running under the baseline set-up conditions, it is worthwhile to assess the BTC plant performance varying the temperature of pyrolysis and gasification. The outcomes obtained are plotted in Figure 13, and detailed results are gathered in Table 7 and Table 8.

Figure 13. BTC plant efficiency results for the study cases

Table 7. BTC plant simulation system conditions results: Baseline and study cases Baseline Case A Case B System conditions results

Fuel input (ton/h) 24.3 24.2 23.0

Global Steam-to-Carbon ratio (wt %) 4.3 4.3 4.5

Global carbon conversion (vol %) 94.4 94.9 97.9

Gasification air ratio² E (wb) 0.21 0.23 0.27

Gasification Steam-to-Carbon ratio (wt %) 0.27 0.27 0.69 Gasification carbon conversion (vol %) 93.9 94.4 95.3

Biochar production (ton/h) 0.41 0.38 0.16

Biochar LHV (MJ/kg) 28.6 28.6 28.6

Fuel temperature (ºC) 487 528 480

Air flow Low Pressure Compressor (kg/s) 22.3 22.3 21.9

Flue gas temperature (ºC) 531 531 532

2 Defined at Eq. 3

51,8% 52,0% 54,9%

31,1% 31,3% 34,3%

5,6% 5,2% 2,3%

B A S E L I N E C A S E A C A S E B

Electrical efficiency Useful heat efficiency Biochar production

(37)

37 Table 8. BTC plant simulation energy balance results: Baseline and study cases

The efficiencies of the system refer to the amount of heat assossiated with the ouput divided by the energy contained in the raw biomass (based on LHV).

7.2.1 Study Case A

In the first study case, where the pyrolysis temperature remains at 326ºC and gasification temperature increases from 850ºC to 950ºC, the overall simulation plant performance does not vary significantly in comparison to the baseline scenario.

Given that the product yields of pyrolysis applied are identical to the baseline case, the biochar composition entering the gasification is the same. Following the gasification model approach explained at section 6.3 more air is required to reach higher gasification temperatures. At the same time, carbon conversion in gasification is higher as the air ratio increases (following Eq.

4), resulting in slightly less biochar output. Additionally, the higher the air ratio is the more advanced oxidation level is reached, thus generating more CO2 instead of CO.

While the heating value of the gasification product gas is a bit higher in the baseline case (TGasif=850ºC), the fuel has more sensible heat in the study case A (TGasif=950ºC). This situation results into a very similar performance in the power train and heat recovery blocks dynamics, reaching very analogous electrical and useful heat efficiencies.

From the results obtained, it is deemed that overall performance is not significantly sensitive to gasification temperature. This would mean that, without penalizing the heat balance, the gasification could be run at lower temperatures, which would have the following advantages:

• More flexibility for using other biomass feedstock (i.e. higher in ash content)

• Cheaper equipment materials, which could not be used in higher operation temperatures

• Longer lifetime of the system equipment

Energy balance Baseline Case A Case B

Input: Biomass LHV (MWth) 57.9 57.6 54.7

Output: Gas turbine (MWe) 30 30 30

Output: District heating (MWth) 18.0 18.0 18.7

Output: Biochar LHV (MWth) 3.3 3.0 1.3

Heat-to-power ratio 1.7 1.7 1.6

Efficiencies

Electrical efficiency (%) 52 52 55

Useful heat efficiency (%) 31 31 34

Biochar production (%) 6 5 2

Overall efficiency (%) 89 89 91

(38)

38 Nevertheless, it is necessary to bear in mind that in practice lower temperature in gasification implies slower reaction rates, and thus a need for longer residence time in the reactor.

7.2.2 Study Case B

The second study case analyses how the change in pyrolysis temperature from 326ºC to 396ºC affect the dynamics of the system, keeping constant the gasification temperature at 850ºC.

The pyrolysis product yields applied in this case are different, based as well on a set of experimental tests. The mass yields have been provided by Phoenix Biopower. The amount of pyrolysis biochar formed is almost halved, while the permanent gases and tars produced are much higher; consequently, less biochar undergoes gasification. On the top of that, the fixed carbon of the biochar produced is increased at the expense of decreasing hydrogen and oxygen content.

The aim was to keep the same steam-to-carbon ratio in the gasification as in baseline and study case A. However, due to unknown Aspen Plus limitations, a higher steam-to-carbon ratio was necessary to make the equilibrium reactor run. This behaviour is probably due to the different composition of the biochar undergoing gasification. A possible reason could be that since there is more carbon available and less hydrogen, additional rate of oxidizer is needed. However, a clear explanation could not have been formulated.

Higher air ratio is needed to heat up the additional steam injected from 300ºC to 850ºC, resulting in an air ratio as high as 0.27. This fact in turn induces to a carbon conversion as high as 95%, generating very little biochar output. Nevertheless, considering the amount of fuel gases and tars formed from both, pyrolysis and gasification, the electrical efficiency is considerably higher, as it is the useful excess heat sent to the district heating network.

The global steam-to-carbon ratio that can be sustained by the system is greater than in the previous cases. The reasoning of this behaviour is attributed to the fact that more biomass heat is available in form of fuel to the combustor and in the flue gas exhaust.

(39)

39

8 Discussion

The primary objective of the thermodynamic model is to provide a diagnostic tool for predicting the performance of the BTC system, evaluating the relevance of different parameters and compare the overall performance by varying system characteristics.

The first focus has been on developing an equilibrium model for biomass pre-treatment block, so that products could be predicted specifying running conditions only. However, the chemical reactions involved in the thermochemical conversions of biomass are complex and do not occur under equilibrium, so results lack accuracy in product composition. To address such limitation, undertaking pyrolysis and gasification with a kinetic model would have been optimal if not for the inconvenience and complexity of using Aspen Plus software to do so. Therefore, it has been opted to go for empirical and semi-empirical approach for pyrolysis and gasification, respectively. Given the lack of empirical data of such processes operating at high pressures, the experimental correlations applied have been as close as possible to the working conditions.

The BTC plant simulation model is found to be useful, however a calibration of pyrolysis and gasification is required to get a more realistic and practical picture of the system. Therefore, more experimental data is needed to adjust the model to the simulation operating conditions.

Firstly, the empirical data applied to pyrolysis model comes from tests taken at 30 bar, instead of 60 bar. Although the results are expected to follow the same trend, experiments at 60bar would ensure a more reliable prediction. Similarly, the semiempirical model applied for gasification presents correlations operating at 4-5 bar instead of 60 bar. Adjusting the correction factors to the given set-up would be required.

On the top of that, gasification model presents carbon conversion and hydrocarbon formation adjustments based on woody biomass instead of biochar gasification, which has been found to cause a great impact on the overall plant performance. This fact leads to achieving relatively the same results between cases with different gasification temperature. If biomass and biochar are compared, the last one has more carbon content (75% vs. 49%) and less hydrogen and oxygen content. On relative basis, more air needs to be injected per kg of gasification feedstock to reach the same oxidation degree and consequently the carbon conversion is lower following Eq. 4. Hence, it can be concluded that applying the given correlations, carbon conversion is underpredicted by the used gasification model. This unprediction implies that a higher efficiency is possible.

Biochar production integration into Phoenix BioPower BTC technology