Torrefaction: Pre- or Post-Pelletization

Oakridge National Laboratory tests identify differing physical and chemical characteristics and energy use of two torrefied pellet pathways.
By Shahab Sokhansanj | October 07, 2013

Recent developments in densification technologies, including pelletization, have substantially improved the economics of moving biomass around the globe. This is one of the reasons that the wood pellet industry has become an inevitable part of the bioeconomy in the U.S., Canada and elsewhere.
To be an effective alternative to coal, increasing the energy density of pellets is as important as increasing bulk density and durability. The high heating value of currently marketed pellets is about 19  gigajoule (GJ) per metric ton, which limits the proportion that can be used in cofiring with coal, with a heating value as high as 28 to 30 GJ per metric ton. In this context, removing the low-heat content volatiles from biomass is a promising strategy to increase the overall heat value of pellets and the ability to store them outside brings storage and handling costs on par with coal.

Torrefaction, also known as mild pyrolysis/carbonization, is a thermal pretreatment process to upgrade lignocellulosic biomass into a higher-quality energy and carbon carrier, to augment coal. The process subjects the biomass to temperatures ranging from 200 to 300 degrees Celsius in the absence of oxygen for up to 30 minutes. The slow heating process roasts biomass, releasing volatile compounds and breaking down hemicelluloses. Torrefaction has been proposed as a robust strategy to overcome the heterogeneity among different types of cellulosic feedstocks, thus producing a uniform-quality energy commodity with improved energy/carbon content and grindability. The bulk density of the torrefied material, however, is generally lower than that of the raw biomass, making transport and storage economically challenging. Therefore, combining torrefaction and pelletization has great potential to upgrade raw biomass to a universal energy commodity.

Pellets to the Test

In a research project for the Environmental Sciences Division of Oak Ridge National Laboratory, two process pathways were investigated to make torrefied wood pellets from Douglas fir wood chips. Both pathways started with dry wood chips as the raw material and ended with torrefied wood pellets as the final product. For Pathway I, the chips were ground, pelletized and pellets were torrefied. For Pathway II, the chips were initially torrefied and then ground and pelletized. Since there are pros and cons associated with each of these process schemes, a detailed laboratory experiment was planned to evaluate the characteristics of each operation, along with the compositional and physical properties as well as energy and mass balance.

For the experiments, Douglas fir wood chips with an average size of 30 to 50 millimeter and initial moisture content of 45 to 50 percent were collected from Fibreco Co. located in North Vancouver, British Columbia. The material was first dried to 15 percent moisture content for grinding or almost bone-dry condition for torrefaction prior to grind. Both untreated wood chips and torrefied wood chips were ground with a hammer mill under the same experimental conditions (the same feed rate and screen mesh size). Drying, followed by torrefaction, was performed in a Carbolite oven, which operates at temperatures ranging from room temperature to 600 C. Torrefaction was carried out in a stainless steel box installed inside the oven with nitrogen introduced to provide an oxygen-free environment. A lab-scale California pellet mill was used for pelletizing both torrefied and untreated materials at 13 percent moisture content.

Comparing Systems

Since biomass is generally flexible and tenacious, it is energy intensive to reduce the particle size prior to the pelletization process or to use them in pulverized combustion systems. From the comparison of processes, it is clear that loss of moisture and some of the volatiles during  torrefaction process makes biomass more brittle and easier to grind. The torrefied material was found to consume nearly 20 times less energy to grind than untreated wood chips. The torrefied material had to be reconditioned, however, to bring it back up to 13 percent moisture from the 2 percent level that followed torrefaction, before pelletization. Pelletization of the untreated materials was possible without using any binding agents, while the addition of a wheat flour binder, at 5 percent mass basis, was necessary to enable effective pelletization of torrefied material. Pelletization of the torrefied material with the use of binder resulted in pellets with relatively higher density, durability and lower moisture absorption than torrefied pellets.

In order to understand the stability of the pellets in water, both types of pellets were immersed. Compared to the raw material pellets, which readily disintegrated when immersed in water, both types of torrefied material exhibited higher stability in water. However, it was found that the pellets made from torrefied wood chips fell apart relatively easily, whereas the pelletized-and-torrefied material stayed intact for more than two hours. Similar results were seen when the three samples were subjected to high humidity. It appears that penetration of the water vapor and its subsequent condensation is critically dependent on the external surface area of the pellets. In the case of pelletized and subsequently torrefied material, it appears that torrefaction substantially increased the external porosity of the pellets and more water vapor can potentially get condensed in the small pores present in the outer surface of the pellets. In the case of untreated pellets and torrefied and subsequently pelletized material, pelletizing could pack the materials tighter with less external surface area and less porosity available for vapors condensation.

Pathway I, which involved drying, grinding, pelletization and torrefaction, consumed slightly higher amount of energy compared to Pathway II, which consisted of the direct torrefaction of the wet wood chips, grinding and pelletization. However, this lower energy consumption was at the expense of using binders during pelletization. Additionally, the quality of the wood pellets resulting from the pelletization of the torrefied material was still poor, including lower heat values and higher moisture content.
Therefore, in order to get the benefit from the energy savings of Pathway II, we need to develop effective process strategies to improve the binding and pelletability of the torrefied material to make it more durable, denser and stronger. Torrefaction following pelletization currently appears to be a promising strategy to obtain torrefied wood pellets which are transportable with improved durability, reduced moisture content and higher energy value.

Author: Shahab Sokhansanj
Bioenergy Resource and Engineering Systems Group Environmental Sciences Division, Oak Ridge National Laboratory

University of British Columbia researchers involved in the torrefaction project  include: Bahman Ghiasi, Takaaki Furubayashi, Linoj Kumar, Tony Bi, Anthony Lau, Jim Lim, Chang Soo Kim and Jack Saddler.