Characterization of Biomass Waste as Fuels-- Effect of Pre Treatments and Nitrogen Chemistry

Funding

EU commission through the INECSE programme (contract nr. MEST-CT-2005-021018)

Background

Biomass fuels represent already today a big share of the Renewable Energy Sources used in Europe. Furthermore the European Commission is trying to stimulate the member countries to increase the generation of electricity and heat from biomass sources, as well as transportation fuels and chemicals through the concept of Biorefinery. Truth is that in the transition towards a renewables-based world, biomass will surely contribute in a big percentage.

Moreover, biomass fuels generally contains a high percentage of volatiles (> 70% mass) and the devolatilization phase, thus, plays a major role in the whole process of combustion/gasification in terms of conversion, yields and stability of the flame. So the need for characterization of the behaviour of these fuels in thermochemical conversion is evident and fundamental in order to provide the market with sustainable, reliable, efficient and cost-competitive technologies.

According to the previously stated paradigm, the materials we chose to study are mainly agricultural residues (wheat straw and residues from the food industry) and other biomass waste (Chicken Litter, Palm Kernel Cake and Dry Distillers Grain with Solubles (DDGS) from corn ethanol production) in order to upgrade waste streams into valuable energy sources. Among these, straw is already one of the most abundant, 300 Mton are produced every year only in Europe, and it is already employed for energy purposes, mainly co-fired with coal to produce power and heat. But we’ll also study residues from the food industry, like residues from the olive oil production and peach stones; both are abundant waste streams in Mediterranean countries.

But, like the benefits, the issues with these fuels are well known: the high content of ashes and theis composition rich in alkali (mainly K and Na) together with Chlorine and Silica, gives rise to many issues on the long-time scale, like slagging, corrosion and fouling.

Fluidized bed reactors (Combustors and Gasifiers) are very flexible regarding fuel feed; this makes them one of the most suitable technologies for biomass conversion, considering the heterogeneous nature of such fuels, but, at the same time, one of the most affected by the low-temperature melting ashes that badly affect the fluidization itself.

So, in order to improve this behaviour and also the efficiency of conversion, several pre-treatments have been studied and some of them have proven effective. Among these, water leaching, fractionation (maintaining only the coarse part) and the combination of the two seem to give the best results.

These treatments are relatively easy and cheap to perform considering that the leaching process consists only in soaking the fuel in water for a certain time (changing the mass-to-water ratio and the residence time affects the efficiency of the leaching) followed by drying. The mechanical fractionation is basically a sieveing process where the fine part (dimensions less than 1 mm) is discharged and the coarse part results improved in its ash composition.

Objectives

The main goal of this project is to compare the initial materials with the treated ones in terms of global reactivities, volatiles released (mainly NOx precursors like NH3, HCN and HNCO) and composition of the inorganic matter.

Abundant literature is available for coal characterization in terms of Nitrogen functional groups and Nitrogen emissions, but a clear and extensive analysis as we are planning on doing it’s not certainly available for biomass fuels. Infact, identifying both the gaseous compounds released during pyrolysis process (through FTIR measurements) and the solid-bound functional groups of Nitrogen in the original sample and in the char left (through XPS and 15N-NMR measurements), will give fundamental information in order to try and correlate the released volatiles with their source.

Validation of the kinetic data and the volatile partitioning obtained in slow pyrolysis conditions (with a Thermogravimetric analyzer) will be the goal of flash pyrolysis tests on a heated grid reactor connected with an FTIR spectrometer.

Moreover inorganic matter in the samples will be analyzed and its catalytic effect will be investigated.

Modelling of the devolatilization process will also be a goal of our project since all the previous measurements will give as "by-product" abundant data that can be used for modelling purposes.

Experimental Work

Experiments will be carried out at slow pyrolysis conditions on a Thermogravimetric Analyzer (TA instruments) in order to be able to retrieve proper kinetic parameters. We will explore different heating rates (5-10-20-30 and 100°C/min) and two final temperatures of 800°C and 900°C under a flow of Helium.

Figure 1: DTG and TG curves for Untreated Wheat straw

sample (-.-) and Leached sample (-) at 10ºC/min in He flow

Kinetic of such measurements will be analyzed using the software KINETICS05, created by Mr.A.K.Burnham and Mr.R.L.Braun and licensed by Lawrence Livermore National Laboratory, California, USA. Through this software, data at different heating rates will be simultaneously fitted with a Gaussian Distribution of activation energies in order to identify the pre-exponential factor, mean activation energy and standard deviation.

Flash pyrolysis measurements at heating rates of 1000ºK/min will be performed on a heated grid reactor.

Volatiles released, both from TGA and Heated grid, will be analyzed on-line with an FT-IR spectrometer.

As previously mentioned, several analyses will also be performed on the solid initial samples and on the chars. We’ll perform XPS and 15N-NMR measurements on the chars produced at different heating rates and from different samples at different final temperatures and SEM-EDX measurements will illustrate the composition of the inorganic matter in the initial samples and in the chars.

References:

J.Giuntoli, W. de Jong, S. Arvelakis, H. Spliethoff and A.H.M Verkooijen, "Influence of pre-treatments on thermal conversion of Agricultural Residues: Effects on Nitrogen chemistry during Pyrolysis", Proceedings of the 15th European Biomass Conference & Exhibition held in Berlin 7-11 May 2007.

Moreover, biomass fuels generally contains a high percentage of volatiles (> 70% mass) and the devolatilization phase, thus, plays a major role in the whole process of combustion/gasification in terms of conversion, yields and stability of the flame. So the need for characterization of the behaviour of these fuels in thermochemical conversion is evident and fundamental in order to provide the market with sustainable, reliable, efficient and cost-competitive technologies.

According to the previously stated paradigm, the materials we chose to study are mainly agricultural residues (wheat straw and residues from the food industry) and other biomass waste (Chicken Litter, Palm Kernel Cake and Dry Distillers Grain with Solubles (DDGS) from corn ethanol production) in order to upgrade waste streams into valuable energy sources. Among these, straw is already one of the most abundant, 300 Mton are produced every year only in Europe, and it is already employed for energy purposes, mainly co-fired with coal to produce power and heat. But we’ll also study residues from the food industry, like residues from the olive oil production and peach stones; both are abundant waste streams in Mediterranean countries.

But, like the benefits, the issues with these fuels are well known: the high content of ashes and theis composition rich in alkali (mainly K and Na) together with Chlorine and Silica, gives rise to many issues on the long-time scale, like slagging, corrosion and fouling.

Fluidized bed reactors (Combustors and Gasifiers) are very flexible regarding fuel feed; this makes them one of the most suitable technologies for biomass conversion, considering the heterogeneous nature of such fuels, but, at the same time, one of the most affected by the low-temperature melting ashes that badly affect the fluidization itself.

So, in order to improve this behaviour and also the efficiency of conversion, several pre-treatments have been studied and some of them have proven effective. Among these, water leaching, fractionation (maintaining only the coarse part) and the combination of the two seem to give the best results.

These treatments are relatively easy and cheap to perform considering that the leaching process consists only in soaking the fuel in water for a certain time (changing the mass-to-water ratio and the residence time affects the efficiency of the leaching) followed by drying. The mechanical fractionation is basically a sieveing process where the fine part (dimensions less than 1 mm) is discharged and the coarse part results improved in its ash composition.

Objectives

The main goal of this project is to compare the initial materials with the treated ones in terms of global reactivities, volatiles released (mainly NOx precursors like NH3, HCN and HNCO) and composition of the inorganic matter.

Abundant literature is available for coal characterization in terms of Nitrogen functional groups and Nitrogen emissions, but a clear and extensive analysis as we are planning on doing it’s not certainly available for biomass fuels. Infact, identifying both the gaseous compounds released during pyrolysis process (through FTIR measurements) and the solid-bound functional groups of Nitrogen in the original sample and in the char left (through XPS and 15N-NMR measurements), will give fundamental information in order to try and correlate the released volatiles with their source.

Validation of the kinetic data and the volatile partitioning obtained in slow pyrolysis conditions (with a Thermogravimetric analyzer) will be the goal of flash pyrolysis tests on a heated grid reactor connected with an FTIR spectrometer.

Moreover inorganic matter in the samples will be analyzed and its catalytic effect will be investigated.

Modelling of the devolatilization process will also be a goal of our project since all the previous measurements will give as "by-product" abundant data that can be used for modelling purposes.

Experimental Work

Experiments will be carried out at slow pyrolysis conditions on a Thermogravimetric Analyzer (TA instruments) in order to be able to retrieve proper kinetic parameters. We will explore different heating rates (5-10-20-30 and 100°C/min) and two final temperatures of 800°C and 900°C under a flow of Helium.

Figure 1: DTG and TG curves for Untreated Wheat straw

sample (-.-) and Leached sample (-) at 10ºC/min in He flow

Kinetic of such measurements will be analyzed using the software KINETICS05, created by Mr.A.K.Burnham and Mr.R.L.Braun and licensed by Lawrence Livermore National Laboratory, California, USA. Through this software, data at different heating rates will be simultaneously fitted with a Gaussian Distribution of activation energies in order to identify the pre-exponential factor, mean activation energy and standard deviation.

Flash pyrolysis measurements at heating rates of 1000ºK/min will be performed on a heated grid reactor.

Volatiles released, both from TGA and Heated grid, will be analyzed on-line with an FT-IR spectrometer.

As previously mentioned, several analyses will also be performed on the solid initial samples and on the chars. We’ll perform XPS and 15N-NMR measurements on the chars produced at different heating rates and from different samples at different final temperatures and SEM-EDX measurements will illustrate the composition of the inorganic matter in the initial samples and in the chars.

References:

J.Giuntoli, W. de Jong, S. Arvelakis, H. Spliethoff and A.H.M Verkooijen, "Influence of pre-treatments on thermal conversion of Agricultural Residues: Effects on Nitrogen chemistry during Pyrolysis", Proceedings of the 15th European Biomass Conference & Exhibition held in Berlin 7-11 May 2007.