NC State
BioResources
Stelte, W., Sanadi, A. R., Shang, L., Holm, J. K., Ahrenfeldt, J., and Henriksen, U. B. (2012). "Recent developments in biomass pelletization - A review," BioRes. 7(3), 4451-4490.

Abstract

The depletion of fossil fuels and the need to reduce greenhouse gas emissions has resulted in a strong growth of biomass utilization for heat and power production. Attempts to overcome the poor handling properties of biomass, i.e. its low bulk density and inhomogeneous structure, have resulted in an increasing interest in biomass densification technologies, such as pelletization and briquetting. The global pellet market has developed quickly, and strong growth is expected for the coming years. Due to an increase in demand for biomass, the traditionally used wood residues from sawmills and pulp and paper industry are not sufficient to meet future needs. An extended raw material base consisting of a broad variety of fibrous residues from agriculture and food industries, as well as thermal pre-treatment processes, provides new challenges for the pellet industry. Pellet production has been an established process for several decades, but only in the past five years has there been significant progress made to understand the key factors affecting pelletizing processes. A good understanding about the pelletizing process, especially the processing parameters and their effect on pellet formation and bonding are important for process and product optimization. The present review provides a comprehensive overview of the latest insights into the biomass pelletization processes, such as the forces involved in the pelletizing processes, modeling, bonding, and adhesive mechanisms. Furthermore, thermal pretreatment of the biomass, i.e. torrefaction and other thermal treatment to enhance the fuel properties of biomass pellets are discussed.


Download PDF

Full Article

RECENT DEVELOPMENT IN BIOMASS PELLETIZATION – REVIEW

Wolfgang Stelte,* Anand R. Sanadi,b Lei Shang,c Jens K. Holm,d Jesper Ahrenfeldt,c and Ulrik B. Henriksen c

The depletion of fossil fuels and the need to reduce greenhouse gas emissions has resulted in a strong growth of biomass utilization for heat and power production. Attempts to overcome the poor handling properties of biomass, i.e. its low bulk density and inhomogeneous structure, have resulted in an increasing interest in biomass densification technologies, such as pelletization and briquetting. The global pellet market has developed quickly, and strong growth is expected for the coming years. Due to an increase in demand for biomass, the traditionally used wood residues from sawmills and pulp and paper industry are not sufficient to meet future needs. An extended raw material base consisting of a broad variety of fibrous residues from agriculture and food industries, as well as thermal pre-treatment processes, provides new challenges for the pellet industry. Pellet production has been an established process for several decades, but only in the past five years has there been significant progress made to understand the key factors affecting pelletizing processes. A good understanding about the pelletizing process, especially the processing parameters and their effect on pellet formation and bonding are important for process and product optimization. The present review provides a comprehensive overview of the latest insights into the biomass pelletization processes, such as the forces involved in the pelletizing processes, modeling, bonding, and adhesive mechanisms. Furthermore, thermal pretreatment of the biomass, i.e. torrefaction and other thermal treatment to enhance the fuel properties of biomass pellets are discussed.

Keywords: Biomass; Pelletization; Bonding; Torrefaction; Pre-treatment; Wood pellets

Contact information: a: Centre for Renewable Energy and Transport, Division for Energy and Climate, Danish Technological Institute, Gregersensvej, DK-2630 Taastrup, Denmark.; b: Biomass and Ecosystem Science, Faculty of Life Sciences, University of Copenhagen, 1958 Frederiksberg, Denmark; c: Department of Chemical and Biochemical Engineering, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark; d: Chemical Engineering, DONG Energy Power A/S, 2820, Gentofte, Denmark; *Corresponding author: stelte@gmail.com

INTRODUCTION

The depletion of fossil fuels (i.e. natural gas and oil) and the discussion about CO2-induced climate change (Berner 2003) have resulted in political decisions triggering the use of biomass for heat and power production in Europe and elsewhere (Tolon-Becerra et al. 2011). The utilization of biomass is an attractive option for power producers to reduce their CO2 emissions due to relatively easy implementation in existing infrastructures, especially when co-fired in coal-based power plants (Neville 2011). In contrast to wind and solar power that depend on weather and seasonal change, biomass can be stored and utilized when needed, thereby providing a continuous source for heat and power production (Dunnett and Shah 2007; Rosillo-Calle 2007; Schubert et al. 2009).

One of the major factors limiting the utilization of biomass for heat and power production is its low bulk density, resulting in inefficient and cost-intensive handling properties. The distances between biomass production sites, such as forest and agricultural land to industrial and residential areas, where the energy is needed, are often long and require significant logistics for transportation and storage (Rentizelas et al. 2009). The bulk density of biomass is about 40 to 150 kg/m3 for grasses (Adapa et al. 2002; Larsson et al. 2008) and about 150 to 200 kg/m3 for commercial woodchips (Robbins 1982). Pelletization of biomass increases the biomass bulk density to about 700 kg/m(Sokhansanj and Turhollow 2004). Apart from density increase, pelletization offers several other benefits, such as a homogeneous shape and structure that is advantageous for automated feeding into boiler systems.

The pellet marked consists of global trade of biomass, and the European and North American pellet markets have been studied in great detail by Sikkema et al. (2011) (European Union) and Spelter and Toth (2009) (USA/Canada). Wood pellet production in the European Union was estimated to be about 10 million tons (Sikkema et al. 2011) and 6.2 million tons for North America (Spelter and Toth 2009) for the year 2009, which sums up to about 16.2 million tons total. Another study by Cocchi et al. (2011) estimated a slightly lower value for the global pellet production of about 14.3 million tons for the year 2010. The globally installed pellet production capacity for 2011 was estimated to be at about 30 million tons. All studies suggest a strong growth for both the European and North American pellet markets. The Finnish Pöyry Industry consulting company has predicted growth of the global pellet production capacity to 46 million tons by 2020 (Pöyry 2011). The market price for pellets is rather volatile and has been fluctuating between 112 and 142 Euro per ton within the years 2007 to 2011; the price seems to follow the price for crude oil and economic development. The price as of May 2012 was, according to the APX-ENDEX (2012), about 127 Euro per ton. Sikkema et al. (2011) have made a study of the European pellet market and conclude that the current pellet price is under pressure and that future demand of pellets is highly uncertain, while the European market is subsidy driven. High feed in tariffs and combined sulfur and carbon tax are named as major factors affecting the pellet market. The future development of the pellet market is seen to be highly dependent on the abolishment of existing and/or the establishment of new public support schemes and fossil fuel price development as well as new renewable energy obligations.

Increase in demand for pellets and the limited availability of wood resources has resulted in efforts to broaden the raw material base used for pellet production; fibrous residues from agriculture and food processing industry, such as straws, husks, and pulps are being considered. Co-firing of biomass in existing combined heat and power (CHP) plants is a relatively inexpensive and easy way to reduce the CO2 emissions of coal-fired power plants. There has been an increasing interest of power producers to obtain a “coal-like” biofuel, such as pellets made from thermally pretreated (torrefied) biomass. Torre-fied pellets have just been commercialized (van der Stelt et al. 2011).

Earlier pellet reviews (García-Maravera et al. 2011; Kaliyan and Morey 2009d; Tumuluru et al. 2011) have focused on different aspects of biomass pelletization. However, a lot of new developments have taken place in the pellet sector lately, e.g. the introduction of new raw materials and pre-treatment technologies, such as torrefaction. Furthermore, newer studies have focused on the fundamental understanding of the pelletizing processes, such as key process parameters, bonding mechanisms, and pre-treatment technologies. Parts of the present review have also been published in a Ph.D. thesis at the Technical University of Denmark (Stelte 2011e).

PELLETIZING PROCESS AND MODELLING

The compaction of biomass into briquettes and pellets is an old process that has been known for more than 130 years. William H. Smith registered the first patent for biomass densification in 1880 (Smith 1880) in Chicago, Illinois. It describes a process where sawdust was heated up to 150°C, put in a strong mold, and compressed using a steam hammer.

The different types of biomass densification are bailing, briquetting, extrusion, and pelletization (Tumuluru et al. 2010). With respect to the production of high density, solid energy carriers from biomass, briquetting, and pelletization are the two standard methods, and that is why the present review is limited to these two techniques. Both processes occur at high pressures and are two very closely related techniques. The basic processing difference is the use of a piston or screw press for briquetting, while biomass pellets are produced in a pellet mill. The size of the final products is different; pellets have a cylindrical shape and are about 6 to 25 mm in diameter and 3 to 50 mm in length (Alakangas 2010). There are European standards relevant for biomass pellets and raw material classification (EN 14961-1, EN 14961-2 and EN 14961-6), and corresponding international ISO standards are under development (ISO/CD 17225-1, ISO/CD 17225-2 and ISO/CD 17225-6). These standards set clear values for pellet dimensions, pellet quality, and composition (i.e. raw materials, ash content, and heavy metals) and are based on earlier-developed national standards that have been reviewed and compared in detail by García-Maraver et al. (2011).

Briquettes can have different shapes and are greater in their dimensions. The terms briquettes and pellets are often confused in the literature since they only recently have been defined more exactly with the introduction of international standards for solid biofuels. The specifications for biomass briquettes are defined accordingly in the European standard EN 14961-3 and the international standard ISO/CD 17225-3.

Biomass densification first became a commercial, large-scale process in the second half of the last century, and was used to increase the handling properties of biomass both for energy production and for animal feedstock. In North America, wood pellets came into existence in the 1970s with the primary purpose to resolve the energy crisis. They were mainly used by industrial, commercial, and institutional sectors for heating. Residential consumers followed in 1983 when the first pellet stoves were introduced to the market. The European markets started later, with Sweden running at the forefront beginning about 1980; afterwards, the market expanded all over Europe (Peksa-Blanchard et al. 2007). This development was initially driven by an increase in prices for fossil fuels and good availability of residues from sawmills and the pulp and paper industry. Environmental regulations aiming to reduce carbon dioxide emissions and an increase in environmental consciousness of our society became important factors triggering the utilization of biomass for energy production (Bauen et al. 2009; Peksa-Blanchard et al. 2007). Political decision makers have set clear aims to reduce carbon dioxide emissions within the European Union within a short time, better known as the European 2020 goal. It targets to cut down the emission of greenhouse gases by 20% (based on emissions in 1990) and to have a share of 20% renewables in energy production by the year 2020 (Tolon-Becerra, 2011). The utilization of biomass is seen as one of the major contributors to reach this target in time and the reason for the strong growth of the European pellet market.

Process Overview

Pelletizing processes consist of multiple steps (Fig. 1), which include raw material pre-treatment, pelletization, and post-treatment. Pre-treatment steps depend a lot on the raw material characteristics, and generally consist of size reduction, drying, and conditioning. After pelletization, the pellets are transferred into a pellet cooler and screened to remove small particles.

Fig. 1. A typical overview of the biomass densification process

Biomass feedstock for solid biofuels can be categorized into forestry, agriculture, and waste-based materials. These categories can be sub-divided into primary sources (directly produced materials) or secondary sources (derived from other processes) (Panoutsou 2011). Several studies have dealt with the optimization of biomass compaction processes. Some of the raw materials studied are listed in Table 1.

Wood is by far the most widely used raw material for pellet production today, while agricultural residues and grasses are more commonly used for briquetting processes. With respect to combustion properties (e.g. ash formation, slagging, and corrosion), wood has more favorable properties compared to agricultural residues and grasses. Nevertheless due to a limited availability of wood resources and a growing demand for pellets, alternative raw materials will gain further importance in the near future. In particular, industrial consumers are increasingly interested in firing agricultural biomass residues (Skøt 2011).

Table 1. Biomass Resources in Densification Studies

Pellets are produced in a mill that generally consist of a die with cylindrical press channels and rollers that force the biomass to flow into and through the channels. Due to the friction between the steel surface and the biomass in the press channel, a high back pressure is built-up and heat is generated. The physical forces that build up in the press channel of a pellet mill are crucial for understanding and optimizing the pelletizing process, and they have been the subject of multiple studies (Holm et al. 2006, 2007, 2011; Stelte et al. 2011b). A die with press channels and roller(s) are the basic parts of a pellet mill. The die can either be in the shape of a ring or a flat plate, as shown in Fig. 2. Either the die or the rollers can be rotating, and due to that movement, the biomass particles are squeezed into the openings of the press channel.

Fig. 2. Typical pellet mill design a) ring die and b) flat die (Graphic adapted from Alakangas and Paju 2002).

The pelletization process takes place in the press channel of a pellet mill; this process has been the subject of several studies and modeling approaches (Holm et al. 2006, 2007, 2011; Larsson 2010; Nielsen et al. 2009b; Stelte et al. 2011b). During pelletization, biomass particles are fed into the mill. The basic design of a typical ring die pellet mill consists of a die with press channels and (usually two) eccentrically installed rollers. The rollers are located in close proximity to the die. The biomass in the mill is squeezed between roller and die and forced to flow into the press channels. Every time the roller passes the channel, the raw biomass is pressed into the channel, as illustrated in Fig. 3, and a layered structure of the pellet is produced.

Fig. 3. Assembly of a pellet in the press channel of a pellet mill (Graphic modified based on Alakangas and Paju 2002)

Studies have shown that the aspect ratio ‘c’ (length/diameter) of the press channel is one of the most influential parameters determining the magnitude of the pressure (Px) generated in the press channel of a pellet mill (Faborode and Ocallaghan 1987; Faborode 1989; Holm et al. 2006, 2007; Stelte et al. 2011b).

The pressure exerted by the rollers of the pellet press (PR) is oppositely directed to the pressure built up in the press channel (Px) (Fig. 3). Under the assumption of steady state conditions both pressures (Px and PR) are in equilibrium. PR is limited within a certain range, as set by the size and motor power of the mill. When Px exceeds this range the press channels of the pellet mill will be blocked, since the rollers will not be able to provide the necessary pressure to push the materials through the channels. The optimal magnitude of Px is therefore a trade-off between the necessary pressure to produce stable pellets and the energy uptake by the pellet mill. High Px increases the risk of fires due to excessive heat development caused by friction, as well as energy uptake of the pellet mill.

Forces and Modeling

The mechanical forces acting on the biomass in the press channel of a pellet mill have recently been studied in great detail by Holm et al. (2011; 2006; 2007), and Nielsen et al. (2009a).

According to Holm et al. (2006), the major force acting on the biomass in the press channel is the friction force between the press channel wall and the biomass. Assuming that the biomass is an orthotropic material, where the fibers align perpendic-ular to the direction of the press channel, and that the biomass also is an elastic material, the pressure build-up in the press channel of a pellet mill (Px) can be described as a function of the friction coefficient (µ), the Poisson ratio (υ), press channel length (L), and its radius (R). The fibers are elastically deformed when exposed to a radial pressure, and this pressure gives rise to a longitudinal elongation of the fibers (Poisson effect). Since the channel walls are fixed, it is converted into a transverse pressure on the channel walls, resulting in friction. The friction gives rise to back pressure that is under steady-state conditions in equilibrium with the oppositely directed pressure exerted by the rollers and has been expressed by Holm et al. (2006), as described in Equation 1.

 (1)

This model was later verified by experiments using a single pellet press (Holm et al. 2007) and further optimized (Holm et al. 2011) so that it can be used to facilitate fast testing of pelletization behavior of new types of biomass. The latest study also addresses the dependency of the forces observed in the press channel on the temperature. All together, these studies present a method that allows pellet producers to estimate the pelletizing behavior of new types of raw materials by conducting a few tests in a single pellet press unit. The data from these tests in combination with the presented model can be used to estimate key process parameters for a production-size pellet mill, i.e. optimal press channel length and moisture content. This can reduce both time and costs for process development and optimization. The schematic diagram in Fig. 4 shows how a few tests in a single pellet press at low aspect ratios suffice to predict the force in a production size pellet mill. It has to be noted that high aspect ratios obtained in production scale pellet presses are difficult to study in single pellet presses due to the extremely high pressures that set a physical limit to what can be achieved in a single pellet press unit. This also explains the gap between single pellet press and industrial scale pellet press data points in Fig. 4.

Fig. 4. Schematic diagram showing how testing with a single pellet press (SPP) unit in combination with modeling can predict the performance of an industrial pellet mill. Px is the pressure build-up in the press channel and c is the aspect ratio (length/diameter) of the pellet. Single pellet press trials were done as aspect ratios between 0.1 and 5, and the model was used to estimate pressures that will likely occur in production size pellet mills with higher aspect ratios of about 8 to 10. Graphic reprinted with permission from Holm et al. 2011 © American Chemical Society.

Nielsen et al. (2009a) studied the pelletizing process in great detail and subdivided it into three different components that sum up to the overall energy consumption of the process. These components were compression, flow, and friction, as shown in Fig. 5. Every time the roller approaches the surface of the die, the biomass is compressed and forms a temporary layer on the die surface (compression component). The flow component represents the energy required to force the compressed layer into the press channels, and the friction component stands for the energy required to press the compressed biomass in the channels. The work required for the three components can be determined experimentally for different raw materials and process parameters. The resulting data provide an estimate for the energy consumption of the pelletizing process that can be used for process design and optimization.

Other studies (Adapa et al. 2002; Mani et al. 2004) have focused on the compaction behavior of different biomass types by conducting a pelletizing test in a single pellet unit and fitting the obtained data to different mathematical models found in the literature. The models having the best fit to the experimental data were identified and used to explain the compaction mechanism of biomass.

Fig. 5. Illustration of the pelletizing process (Nielsen et al. 2009a). A section of a press channel row is used to illustrate the die/roll/sawdust system. The components of the pelletizing process are allocated to the positions marked by the white dashed lines (see text). The lower part of the press channels with larger diameter is not part of the pelletizing. Graphic reprinted with permission from Nielsen et al. 2009a © Society of Wood Science and Technology.

Process Energy Consumption

The energy requirement for the pelletization of biomass has been the subject of different studies (Nielsen et al. 2009a; Odogherty and Wheeler 1984; Reed and Bryant 1979). Reed and Bryant (1979) studied the energy required for commercial pelletizing processes. They showed that both production rate and electrical energy used are strongly correlated to the raw material type and processing conditions, such as feed size and moisture content. The tested materials included saw dust, aspen wood, Douglas fir, and municipal solid waste. The fraction of product energy consumed during pelletization was about 1 to 3.1%, while the average electrical energy required to pelletize biomass was roughly between 16 and 49 kWh/t. Odogherty and Wheeler (1984) studied the energy required to press wafers out of wheat straw and found a linear relationship between the pellet density and specific energy required for pressing. Nielsen et al. (2009a) studied the contribution of the different components of a pelletizing process (Fig. 5) to the total energy consumption and determined the work required for the flow, compression, and friction contributions for the pelletization of beech and pine at various temperatures and moisture contents. They concluded from their studies that a large fraction of the process energy is used to make the biomass flow into the inlets of the press channels, what they defined as the “flow component”.

Key Process Parameters

Several studies have been made to identify the key process parameters of biomass densification, affecting both process and product:

Moisture content

The effect of raw material moisture content on the pelletizing properties and product quality has been the subject of several studies (Andreiko and Grochowicz 2007; Arshadi et al. 2008; Carone et al. 2011; Filbakk et al. 2011; Kaliyan and Morey 2009b; Mani et al. 2006; Nielsen et al. 2009b, 2010; Odogherty and Wheeler 1984; Rhen et al. 2005; Ryu et al. 2008; Serrano et al. 2011; Smith et al. 1977; Stelte et al. 2011b). In these studies, biomass was pelletized at different levels of moisture content, and its impact on the pellet quality (durability or compression stability) was analyzed. In general, the optimum moisture content for wood species was found to be between 5 to 10% (wt.), while it was slightly higher for agricultural grasses (between 10 to 20% (wt.)).

The optimum moisture content for the densification of beech was 6 to 10% (wt.) (Nielsen et al. 2009a; Stelte et al. 2011b), for spruce it was about 10% (wt.) (Stelte et al. 2011b), for olive pulp it was 5% (wt.) (Carone et al. 2011), and for pine it was about 6 to 8% (wt.) moisture content (Nielsen et al. 2009a). The optimum moisture contents for the pelletization of grasses are significantly higher. For unspecified straw, in general, the optimum moisture content was found to be between 10 to 15 % (wt.) (Odogherty and Wheeler 1984), for barley it was straw 19 to 23 % (wt) (Serrano et al. 2011), for wheat straw it was about 15 % (wt) (Smith et al. 1977; Stelte et al. 2011b), and for corn stover it was about 10 % (wt) (Kaliyan and Morey 2009b). Increase in moisture contents above the optimum have been shown to have a negative influence on the pellet’s mechanical properties (Carone et al. 2011; Kaliyan and Morey 2009d; Nielsen et al. 2009a; Serrano et al. 2011; Stelte et al. 2011b) and to reduce the pellet’s density (Kaliyan and Morey 2009d; Odogherty and Wheeler 1984).

Apart from the quality of the densified product, the densification process itself is influenced by the moisture content. Andrejko and Grochwitz (2007) concluded from their studies that the energy consumption to compact ground lupine seeds into pellets was dependent on moisture content. They found that the energy input necessary to compact the ground seeds to a constant volume decreased with an increase in moisture content within the range of 9.5 to 15.0% (wt.). Nielsen et al. (2009a) have shown that an increase in moisture content for pine and beech results in a decrease of the energy requirement for different components of the pelletizing process. The influence of moisture content on the behavior the amorphous polymers in the biomass and the properties of the pellets (Stelte 2011c) will be discussed later.

Temperature

The effect of temperature on biomass densification has been studied to a significant extent (Filbakk et al. 2011; Gilbert et al. 2009; Kaliyan and Morey 2009b; Nielsen et al. 2009a; Rhen et al. 2005; Serrano et al. 2011; Stelte et al. 2011b). Heat is generated during pelletization due to the friction between biomass and the press channels of the mill. Serrano et al. (2011) studied the heat distribution by taking thermographic images of a pellet press, and found that the temperature of a die under operation at stable conditions is about 90°C, while the temperature of a pellet leaving the press channel is just at about 70°C and cooling rapidly once it has left the press channel (Fig. 6). This difference could be due to a limited heat transfer from the die to the pellet due to a short retention time of the biomass in the press channel and/or a poor heat transfer between the metal surface and the biomass.

The first studies of the temperature effects on biomass densification were conducted by Smith et al. (1977) where they investigated the dependency of straw briquette density on the applied temperature within a range of 60 to 140°C. Their study showed that the density increased as the temperature increased until it reached 90°C, while any further temperature increase did not result in higher densification. They also observed that the stability of pellets were enhanced and suggested that thin layers of waxes around the stem (cuticula) melt and solidify during this process. The waxes can serve as an adhesive between individual straw fibers. Other studies suggest the opposite, that is a negative effect of plant waxes (such as in wheat straw) on interparticle bonding due to the formation of a weak boundary layer (Bikerman 1967; Stelte et al. 2011d, 2012a).

Fig. 6. Thermographic image of a pellet press die. Graphic reprinted with permission from Serrano et al. 2011 © Elsevier.

A recent study conducted by Stelte et al. (2011c) has shown that wheat straw waxes undergo a glass transition at 40 to 50°C (below the temperature where density improvements were observed), while wheat straw lignins showed a strong glass transition at 65 to 75°C (8% moisture content), which is much closer to the observed improve-ments. Indeed, several studies (Gilbert et al. 2009; Kaliyan and Morey 2010b; Stelte et al. 2011d) have suggested that lignin glass transition and subsequent flow and hardening result in improved wetting and enhanced contact area, followed by the inter-penetration of polymer chains from adjacent particles. This leads to higher mechanical properties of a pellet.

The increase in mechanical properties of the pellets with an increase in temperature was reported for spruce (Rhen et al. 2005), corn stover (Kaliyan and Morey 2009b), switch grass (Gilbert et al. 2009), pine (Nielsen et al. 2009a), olive (Carone et al. 2011), beech (Nielsen et al. 2009a), and wheat straw (Stelte et al. 2012a). Furthermore, it was reported that an increase in temperature reduces the friction in the press channel of the mill (Stelte et al. 2011b) (Fig. 7) and lowers the energy requirement for different components of the pelletizing process (Nielsen et al. 2009a).

Considering the effect of die temperature on the pelletizing process, it has been investigated that the friction decreases with an increase in die temperature. Stelte et al. (2011b) have measured the backpressure built up in the press channel during the pelletization of different biomass resources (beech, spruce, and wheat straw) and found a decrease with an increase in temperature. They suggested that wood extractives migrating to the pellet surface at elevated temperatures and polymer softening lower the friction in the press channel. In case of wood, it has been shown by means of infrared spectroscopy of the pellet surface that spruce contains lipophilic extractives, most likely tall oils, which lowers the friction in the press channel. Tall oil (also known as pine oil or rosin oil) is a mixture of fatty acids, resin acids, and sterols. It can migrate to the pellet surface at elevated temperatures. Such substances are substantially absent at ambient temperature or in case of pellets made from wood species containing little amounts of tall oil, such as beech (Stelte et al. 2011b). The migration of tall oil to the pellet surface is enhanced at higher temperatures. Its lubricating properties are likely the reason for the sharp drop of Px at about 70°C (Fig 7).

Fig. 7. Pressure build-up in the press channel of a pellet press (Px) related to temperature. The shown data is based on the pelletization of beech, spruce and straw (8% wt. moisture content) in a single pellet press test unit. Graphic reprinted with permission from Stelte et al. 2011b © Elsevier.

Particle size

Different studies have been conducted on the impact of particle size on the compaction properties of biomass (Filbakk et al. 2011; Jensen et al. 2011; Kaliyan and Morey 2009b; Mani et al. 2006; Serrano et al. 2011; Stelte et al. 2011b). Stelte et al. (2011b) have shown that the friction in the press channel of a pellet mill increases with decreasing particle size for beech particles due to an increase in surface area contact between the particles and the channel wall. Regarding pellet quality, Kaliyan and Morey (2009b) have found out that decreasing particle size for corn stover grinds results in an increased briquette density. Similar results were observed by Mani et al. (2006), who found that particle size significantly affects the density for pellets made from barley straw, corn stover, and switch grass, but not in the case of wheat straw. A study made by Serrano et al. (2011) indicated opposite results, suggesting smaller particles resulted in less dense pellets. Their difference compared to other studies was explained by the use of an industrial pellet mill instead of laboratory scale single pellet press units.

The optimum particle size depends on the densification process, and briquetting processes can in general tolerate larger particles then pelletizing processes, for pellet production particles are usually below 5 mm in diameter. And in general, a broad varia-tion of particle size is best with respect to pellet quality. However an amount of fines (particles smaller than 0.5 mm in diameter) that is too high in the raw material has a negative impact both on friction and pellet quality. As a rule of thumb, the amount of fines should not exceed 10 to 20% unless a binding agent is added.

Jensen et al. (2011) described a method that can be used to investigate the particle size distribution within a biomass pellet, which is different from the raw material (before pelletization), since particle size decreases during pelletization. They suggest that wet disintegration combined with mechanical impact is the most suitable method to determine the internal particle size distribution of a biomass pellet.

Press channel dimensions

The diameter of a press channel varies according to the desired product diameter, usually between 6 to 25 mm for a biomass pellet (Alakangas 2010). The press channel length and the ratio between length and diameter, also known as the compression ratio (c) or aspect ratio, is the most important factor influencing the pressure built up in the press channel of a pellet press (Holm et al. 2006). It has therefore been subject to multiple studies (Faborode and Ocallaghan 1987; Faborode 1989; Holm et al. 2006, 2007, 2011; Odogherty and Wheeler 1984; Stelte et al. 2011b; Čolović et al. 2010). Faborode and O’Callaghan (1987) were the first to study the relationship between the aspect ratio and the compaction pressure. They tested the impact of the aspect ratio on the pressure that is built-up in the press channel during pelletization. Their data clearly shows an exponential correlation between the aspect ratio and the built-up pelletizing pressure. The physical forces acting on a pellet in the press channel of a mill were further studied by Holm et al. (2006; 2007; 2011). Their studies resulted in a pellet model that can be used to estimate the pressure build up in a pellet mill (Fig. 4).

Čolović et al. (2010) studied the impact of the press channel length on the physical quality of cattle feed pellets, and they concluded that an increase in press channel length resulted in higher mechanical properties of the pellets (e.g., harder pellets).

For wood pellet production, the aspect ratio is usually around 6, while it can be up to 11 to 12 in the case of pellet production from wheat straw. The optimum length depends to great extent on the chosen raw material and processing conditions (i.e. temperature, moisture content, particle size).

Pelletizing pressure

The pressure the biomass is exposed to during pelletizing and briquetting has a significant impact on the product density and durability, as well as on the process energy consumption. This process parameter has therefore been subject of various studies (Adapa et al. 2009; Butler and McColly 1959; Carone et al. 2011; Gilbert et al. 2009; Kaliyan and Morey 2009b; Mani et al. 2006; Odogherty and Wheeler 1984; Smith et al. 1977; Stelte et al. 2011b). Studies have been conducted for straws, wheat (Adapa et al. 2009; Gilbert et al. 2009; Mani et al. 2006; Smith et al. 1977), barley (Adapa et al. 2009; Mani et al. 2006; Odogherty and Wheeler 1984), canola (Adapa et al. 2009), alfalfa (Butler and McColly 1959), and oat (Adapa et al. 2009). Other biomass residues tested where corn stover (Kaliyan and Morey 2009b; Mani et al. 2006), switchgrass (Gilbert et al. 2009; Kaliyan and Morey 2009b; Mani et al. 2006), olive residues (Carone et al. 2011), and wood (Stelte et al. 2011b). The product density of biomass compacted at different pressures has been studied extensively (Adapa et al. 2009; Faborode and Ocallaghan 1987; Gilbert et al. 2009; Kaliyan and Morey 2009b; Mani et al. 2006; Odogherty and Wheeler 1984; Smith et al. 1977; Stelte et al. 2011b). Furthermore, the mechanical properties (Carone et al. 2011; Gilbert et al. 2009; Kaliyan and Morey 2009b) and energy content (Odogherty and Wheeler 1984) were measured and compared for samples pressed at different pressures. In all the studies, there is a very clear agreement that concludes that pellet and briquette density increases with an increase in pressure. The correlation between pressure and density follows a saturation curve with the plant cell wall density as an upper limit.

Maximum applied pressures ranged from 50 MPa (Odogherty and Wheeler 1984) to 600 MPa (Stelte et al. 2011b); the pressure typically used in most studies was above 50 MPa (Adapa et al. 2009; Mani et al. 2006; Stelte et al. 2011b). The mechanical properties, compressive strength (Gilbert et al. 2009), and durability (Carone et al. 2011; Kaliyan and Morey 2009a) improved with an increase in pressure and follow a saturation curve, as already observed for the pellet density (Stelte et al. 2011b). Building up the pressure in either a pellet mill or briquetting press requires energy. From previous studies, it can be concluded that there is a certain threshold of pressure and that above this pressure, additional energy put into the process is mainly converted into excess heat, rather than contributing to better pellet quality.

BONDING MECHANISMS

The understanding of the bonding mechanisms between biomass particles within a fuel pellet or briquette is crucial for the production of high quality fuels. The bonding mechanisms have been the subject of only a few studies so far (Kaliyan and Morey 2010b; Lam et al. 2011; Stelte et al. 2011d). Nevertheless, a lot of knowledge can be transferred from related fields, such as pharmaceutical tableting (Hiestand 1997; Leuenberger and Rohera 1986), powder and agglomeration technology (Pietsch 2002), fiber board manufacturing (Bouajila et al. 2006), wood-thermoplastic panel boards (Sanadi and Caulfield, 2008), wood welding (Delmotte et al. 2008; Delmotte et al. 2009; Mansouri et al. 2010; Pizzi et al. 2006), and materials science in general.

According to Rumpf (1962) and Pietsch (2002), bonding forces between particles in compacted bodies can be classified into: 1) solid bridges, 2) attraction forces between solid particles, 3) mechanical interlocking, 4) adhesion and cohesion, and 5) interfacial forces and capillary pressure. These bonding mechanisms have been identified and assumed also to be valid for densified forage and wood residues (Mohsenin and Zaske 1976; Tabil 1996).One area of science that has been studied intensively during the last century is the production of pharmaceutical tablets. Tablet pressing of powder materials is a well-known process that has been studied intensively and has been the subject of extensive reviews (Hiestand 1997; Leuenberger and Rohera 1986). In general, there are two important aspects to consider when pelletizing granular materials and these are: 1) the behavior of the particle under pressure, and 2) the interactions between the particles (Steward 1950). The pressure at a constant compression rate increases with time during densification processes as shown in the compression curve in Fig. 8 (Pietsch 2002).

Fig. 8. Powder compression curve (Graphic modified from Pietsch 2002). The densification process can be separated into different stages, i.e. particle rearrangement, elastic and plastic deformation and hardening.

Mani et al. (2004) have analyzed compression curves of various grasses and interpreted them. The densification process can be separated into different stages, as shown in Fig. 8. Initially, the pressure builds up slowly because particles rearrange in a way that they fill less space, and air located in the pores between the particles is removed when pressing. With further compression of the materials, particles are in very close proximity to each other and short range bonding forces, i.e. van der Waals forces and electrostatic forces make them adhere to one another. After a certain point, no further packing can be obtained, and particles are pressed against each other, undergoing elastic and plastic deformation and fiber interlocking (Pietsch 2002). In case of plant cells that contain a large inner volume (vacuole) filled with air (dried biomass), the cell structure breaks up and the vacuole is compressed. At the same time, cell wall compounds (i.e. lignin and hemicelluloses) are expected to be released from the cell and to interact with surrounding particles (Odogherty 1989). Due to the high temperature and pressure, lignin softens and flows, resulting in improved wetting along with molecular inter-diffusion and entanglement of polymer chains between adjacent fibers. This phenomenon has earlier been described as “solid bridge” formation (Kaliyan and Morey 2010b) and is important for the pellet strength (Stelte et al. 2011d). The density increases with pressure until it reaches a maximum which, in the case of plant biomass, can be expected to be close to the density of the plant cell wall (Stelte et al. 2011b).

Bond formation between wood particles has been intensively addressed in wood science and wood technology, and knowledge can be transferred to pelletizing processes. Wood welding, like pelletization and briquetting, is a process where heat and pressure is applied to biomass, resulting in bond formation. The mechanism behind friction welding of wood has been studied in great detail (Delmotte et al. 2008, 2009; Mansouri et al. 2010; Pizzi et al. 2006). According to these studies, the amorphous wood polymers lignin and hemicelluloses “melt” and flow in a process termed the “un-gluing” of wood cells. This process results in the formation of an “entanglement network of molten polymers” that solidifies subsequently into a “wood fiber entanglement network composite” (Gfeller et al. 2003). Kaliyan and Morey (2009d) have recently reviewed the densification mechanisms in solid biofuels in great detail. Stelte et al.(2011d) have studied the bonding mechanism in fuels pellets made from different biomass resources in great detail. The quality of the bonding can be studied by means of compression testing and fracture surface analysis of the surfaces created at break. Both Stelte et al. (2011d) and Kaliyan and Morey (2010b) have used this technique to study the inter-particle bonding within a biomass pellet. To study the fracture surface, the pellets are coated with a conductive layer and viewed in a scanning electron microscope as shown in Fig. 9 (Stelte et al. 2011d). A rough surface (Fig. 9a) showing fiber pullouts or tear-out and polymeric flow are indications for stronger adhesion due to interpenetration of polymer chains from adjacent biomass particles. Flat fracture surfaces (Fig. 9b), on the contrary, are indicators for brittle failure mechanisms. This can be due to the presence of a weak boundary layer, i.e. hydrophobic extractives or a lack of polymeric flow as a result of too low temperature and/or moisture content (Stelte et al. 2011d). Weak boundary layers have been identified as one possible reason why straw pellets have lower mechanical properties compared to wood pellets. Straws have a high content of extractives concentrated on their outer surface (the plant cuticle), which can lead to the creation of a layer that hinders/prevents the formation of stable bonds between adjacent biomass particles during pelletizing processes (Stelte 2012a).

Fig. 9. a) Fracture surface of a pellet showing signs of strong adhesion (i.e. rough surface polymeric flow, fiber pullouts) and b) fracture surface of a pellet showing signs of poor adhesion (i.e. brittle failure, flat surface, lack of fiber pullouts, and polymeric flow). Graphics reprinted with permission from Stelte et al. 2011d © Elsevier

Kaliyan and Morey (2010b) have also studied fracture surfaces of biomass briquettes using light microscopy (Fig. 10). The figures support the suggested bonding mechanisms and show indications for both polymer melting (Fig. 10a) and interlocking of fibers (Fig. 10b).