Resumen de Impact milling of biomass: process and product characterization
This Ph.D. Thesis aims to contribute to the new energy production concept in which the renewables energies must play a relevant role in order to mitigate the greenhouse effect. As it is well-known, this huge problem is mainly caused by the greenhouse gases emissions (mainly CO2), as a consequence of the massive consumption of fossil fuels.
Biomass is one of the renewable energies with the highest potential of fossil fuel substitution. Several physical pre-treatments and thermo-chemical technologies transform the raw biomass resources in solid, liquid and gaseous bio-fuels. However, nowadays, several barriers must still be overtook in order to be competitive in terms of technical feasibility, economical profit and environmental impact contribution. One of them is the milling pre-treatment process.
Milling pre-treatment reduces the particle size, making possible biomass energy valorization by co-combustion with coal in current power plant, by the production of pellets and briquettes as well as obtaining syngas and bio-ethanol. However, the milling is a high energy consumption process with computable economic and environmental impacts within the value chain of the biomass. Efficiency, control and characterization of the process are required to minimize these impacts.
The current doctoral research began with the characterization of a hammer-milling pilot plant. This analysis comprises the evaluation of several milling strategies as combination of mills and classifiers as well as other operational aspects as instrumentation, monitoring, control and sampling. This characterization was described in the first publication of the Thesis:
Publication nº 1: Gil, M.; Ramos, I.; Arauzo, I.; Román, J. Characterization of a Biomass Milling Pilot Plant. Proceedings 17th European Biomass Conference & Exhibition. Hamburg, 2009.
Within BIOCARD project (FPVI), the value chain of the cardoon (Cynara Cardunculus L.) was studied, including the milling pre-treatment. Three different forms of cardoon (whole cardoon fractions, only stem and in pellet form) were milled, focusing on the influences of the final particle size (once milled) on the energy consumption process, the moisture content reduction (drying effect) and bulk density increase (densification process). Experimental results were published in Biomass & Bioenergy:
Publication nº 2: Gil, M.; Arauzo, I.; Teruel, E.; Bartolomé Rubio, C. Milling and handling Cynara Cardunculus L. for use as solid biofuel: Experimental tests. Biomass & Bioenergy, 41, 145-156, 2012.
Further studies were conducted to take into consideration more variables, enclosed on two groups of factors that affect the milling performance: 1) the physical characteristics of the resource and 2) the operational variables. Within the former, it is obvious that the essential constitution of each type of biomass determines its milling behavior. In addition, other physical conditions as the moisture content and the input particle size play also a relevant role. Among the operational variables, the angular speed of the hammer and the opening size of the screen surrounding the mill chamber determine the particle fracture by impact (hammer vs particle) and the particle classification by size (by crossing the openings) inside the mill chamber, respectively.
Within IDSCALE project, an intensive experimental campaign was performed with two biomass resources: short-rotation poplar (Populus spp.) and corn stover (Zea Mays L.). Their differences come from their constitution (woody vs herbaceous), final use (energy crop vs residual biomass), physical conditions (chips vs steams/leaves/cob), chemical composition as well as other multiples considerations regarding to their cultivation, harvest and handling.
As a result of this analysis, two papers were published. The first one focused on the effect of these input variables (material and process) on the energy consumption, moisture content reduction, bulk density increase and handling behavior (by the angle of repose). The second one analyzed these effects on the particle size of the milled particles:
• Publication nº 3: Gil M.; Arauzo I.; Teruel. E. Influence of input biomass conditions and operational parameters on comminution of short-rotation forestry poplar and corn stover using neural networks. Energy & Fuels, 27, 2649-59, 2013.
• Publication nº 4: Gil, M.; Arauzo I. Hammer mill operating and biomass physical conditions effects on particle size distribution of solid pulverized biofuels. Fuel Processing Technology, 127, 80-87, 2014.
Up to now, the milling studies were focused on experimental tests at semi-industrial scale. These tests analyzed the effects of input variables, related to material and process, on output variables. From this point of view, the mill is observed as a black box. Input-output relations and derived conclusions allowed to understand to some extent the underlying mechanisms inside the mill chamber, but no direct evidences were found.
Therefore, further studies were performed in order to find new proofs that elucidate these underlying mechanisms. For that, single impact tests were carried out in a hammer mill at laboratory scale in which the impact conditions between the hammer and particle were varied with the aim to obtain the breakage probability (S) and the breakage function (B) of a biomass particle. The first describes the fracture probability of the particle in the impact and the second one describes the particle size distribution of the new particle progeny generated after the mother particle fracture. Briefly, the breakage probability (S) and breakage function (B) describe whether a biomass particle is broken and how.
Breakage probability was described by the mastercurve, following the methodology of Vogel & Peukert and a new formulation was proposed by the breakage function. The results were published in Fuel Processing Technology:
Publication nº 5: Gil, M.; Luciano, E.; Arauzo, I. Approach to the breakage behavior of comminuted poplar and corn stover under single impact. Fuel Processing Technology, 131, 142-149, 2015.
The most elemental events inside the mill chamber are: 1) the particle fracture and 2) the particle classification. The former is full defined by the breakage probability and the breakage function. The latter express the probability of the particle to leave the mill chamber by crossing the openings of the metal screen. Both events can be incorporated in a Population Balance Model (PBM), modelling the milling process. Model outputs show the particle size distribution of the product as well as the number of impacts that the particle suffers inside the mill.
At the same time, an experimental campaign in the laboratory hammer mill under different operation conditions was performed using poplar and corn stover. The results were utilized to validate the results of the Population Balance Model and checking that the model is sensitive to variation related to the physical conditions and operational mill parameters. Model and validation were described and published in Powder Technology:
Publication nº 6: Gil, M.; Luciano, E.; Arauzo, I. Population balance model for biomass milling. Powder Technology, 276, 34-44, 2015.
This study completes the characterization of the milling behavior of biomass resources. Experimental and theoretical concepts were developed at laboratory and semi-industrial scales. Results and conclusions at different scales were compared in this document in order to find new findings.
During the milling experiments and product characterization, several limitations were observed in relation to the particle size characterization of the milled biomass. Furthermore, the particle size is one of the most important parameters of product quality in milling processes. The goal of milling is to fulfill the size requirements of the milled product imposed by the final user.
International standards establish the procedure to obtain the size distribution of milled biomass particles by means of vibrational screening. Taking a cascade of standards sieves, the particle size is defined by the sieve in which the particle is retained. The results are expressed by the particle size distribution, considering one representative particle diameter. Therefore, the particle size is full characterized for spherical shapes but not for irregular ones, unknowing the particle dimensions obtained or the relationships between these dimensions.
Within the product characterization activities, we developed a study about the morphology characterization of biomass particles by analysis of 2D images. Multiples results were found and reported in the seventh publication of this Thesis:
Publication nº 7: Gil, M.; Arauzo, I.; Teruel, E. Analysis of standard sieving methods for milled biomass through image processing. Effects of particle shape and size for poplar and corn stover. FUEL. 116, 328-340. 2014.
However, these results may help to understand some aspects of the standard particle size characterization, but the 2D image procedure is hardly extended at scientific and industrial level. Current standard procedure is cheap, quickly and easy to implement. We also developed deep studies about the characterization of standard particle size distribution in paper 2 and 4, previously mentioned.
Finally, the handling behavior of any bulk solids strongly depends on its physical characteristics in terms of material properties, moisture content, particle size and shape. Therefore, the handling behavior of biomass resources is determined by the milling process and, obviously, of the particle size and shape. The underlying mechanisms inside the mill chamber (fracture and classification of the particles) determine the particle size and shape and, hence, also the handling behavior of the milled product.
Therefore, the handling behavior of cardoon, poplar and corn stover was analyzed at semi-industrial and laboratory scales. The first study comprised experimental discharge tests of three form of cardoon in three different hoppers, analyzing the advantage and drawbacks in terms of continuous discharge, ratholes, preferential discharge channels, bridges, etc. These experiences were reported in the publication nº 2, previously cited.
The second work was focused on obtaining the main handling parameters for milled biomass and analyzing the influences of the type of biomass, the moisture content, the particle size and shape on the handling behavior. Since CIRCE has no experience and laboratories on this issue, a research stay was carried out in the Technische Universiteit of Delft, under the supervision of Dr. Dingena Schott. The results were reported in Fuel Processing Technology:
Publication nº 8: Gil, M.; Schott, D.; Arauzo, I.; Teruel, E., Handling behavior of two milled biomass: SRF poplar and corn stover. Fuel Processing Technology. 112, 76-85. 2013.
To summarize, this Ph.D. Thesis comprises the characterization of impact milling of biomass; the process and the product. The milling process were analyzed at semi-industrial and laboratory scales. At semi-industrial level, the influences of the main variables, related to the biomass characteristics and milling operational parameters, on the specific energy consumption and quality product were determined. At laboratory scale, the underlying mechanisms, that govern the elemental events of fracture and classification of the particle inside of the mill chamber, were also characterized. Finally, a Population Balance Model was developed, being sensible to the effects of the main variable at play, and it was further validated with experimental tests at laboratory scale.
The characteristics of the milled product were studied in terms of moisture content reduction and bulk density increase during the milling. In addition, special efforts were made to obtain the characterization of the particle morphology and handling behavior for biomass resources. Techniques for 2D image processing were applied to obtain the particle morphology and two experimental campaigns were carried out for the handling behavior characterization at semi-industrial and laboratory scale.
