By Harry Short
Bioenergy plants today are defined by supply chains that use truck-based transport. Such transport has high fuel use per biomass transported as compared to bulk transport modes such as rail or ship. One can easily imagine a future in which a significant percentage of biofuel production is cannibalized to feed the fuel demands of its truck fleet (Hill et al., 2006). This investigative review explores two alternative transport options, rail and pipelines, and attempts to update previous analyses by using higher fuel diesel costs.
Transportation Costs for Biomass
The transportation of bioenergy feedstocks differs from fossil fuels in two important ways. The energy density of fossil fuels is double (for coal) to triple (for petroleum) that of biomass (Ashton & Cassidy, 2007). Therefore, a smaller mass of fossil fuel needs to be transported to generate a unit of electricity. Secondly, fossil fuels have concentrated extraction points (mines or wells) and thus the infrastructure used to move these fuels can be streamlined and dedicated to the fuel (pipelines and coal trains). Dependence on truck transport is small because of the concentration of extraction, and pipelines and dedicated unit trains can achieve economies of scale at medium to long distances. However, biomass is produced over large areas requiring
significant truck transport to collect the feedstock (typically in loads of 20-40t) for delivery to a bioenergy plant. This reliance on truck transport, which is expensive relative to those options used by fossil fuels, limits the size of bioenergy plants as escalating transportation costs reduce the profitability of large-scale biorefineries (Yemshanov & McKenney, 2008).
Alternatives to Truck Transport
If alternatives to truck transport could be found that lower costs, this might affect the viability of large scale biorefineries (small scale plants would be unable to benefit from economies of scale). Recently, several researchers have considered alternatives to truck transport for biomass delivery. Two alternatives have been explored: transport by rail and transport by pipeline. Though some truck transport will be required to move biomass from field in either of these alternative scenarios, a large portion of the total distance from source to biorefinery would be carried by rail or pipeline.
Transport by Pipeline
Kumar et al. (2004) explore the transport of mixed hardwood and softwood chips from western Canada in pipelines, in which the carrier fluid is either water or heavy gas oil. They evaluate two flow scenarios: a one-way system in which carrier fluid is released at end of pipe and a two-way system in which the fluid is pumped back to its origin for reuse (a hydrocarbon-based carrier could only use the two-way system, thought the authors do not state why). They find that for large flows (2 million dry t/yr) and at medium to long distances (100 to 500 km), pipelines cost less than truck transport.
One problem that arises immediately when using a carrier liquid and wood chips is the uptake of the liquid by the chips. The water content of the chips rises by 13% when water is used, and incredibly, the chips absorb upwards of 50% of their weight in oil when a hydrocarbon carrier is employed. The heating value of wood chips, which have absorbed water, is roughly 2.5MJ/kg less than green chips, an reduction the authors argue is prohibitive for use of the wood chips in direct-combustion plants (though they don’t include an economic analysis of this). This problem disappears if the wood chips are intended for cellulosic ethanol production in which wetting of the biomass is necessary. Unfortunately, they do not pursue this option in their analysis, but I will address this option towards the end of this post.
Instead, the authors pursue the scenario where wood chips, carried by heavy gas oil, are used for direct combustion. In this scenario, the chips laden with oil would be burned for electricity generation with the unabsorbed oil being re-circulated. Two thirds of the energy released would be from the absorbed oil and one third from the biomass. As it is more economical to use heavy gas oil as a precursor to transportation fuels (via cracking), the Kumar et al. (2004) conclude that co-firing wood chips with the absorbed oil would involve opportunity costs that are too high to be practical.
Even though pipelines may be more cost-effective at large distances compared to trucking, the costs of moving biomass for such long distances is still quite expensive regardless of method (on average $15 per dry tonne by pipeline for a 100km distance to nearly $40/dry tonne for 500km). With the price of a tonne of woodchips between $22 and $33 (Ashton & Cassidy, 2007), transportation costs at medium to large distances, rival if not exceed fuel costs. Thus, even if the problems associated with using a liquid carrier discussed above, were not present, the economics of transporting biomass long distances would still not compute. Therefore, transporting biomass by pipeline is not a viable alternative to truck transport in a direct combustion bioenergy plant.
Transportation by Rail
A second alterative discussed in the literature is the use of dedicated unit trains to move biomass from collection points to a central plant. Trucks would be used to move the biomass from field to the collection points, after which the train would finish delivery. Mahmudi and Flynn (2006) break transportation costs (see Figure 1) into distance fixed costs (DFCs) i.e. costs that are independent of the distance a unit is transported, and distance variable costs (DVCs) i.e. the costs that do depend on distance a unit is shipped. For trucking, DFCs would only include the cost of loading/unloading the truck as much of the infrastructure costs are subsidized by the government and other taxpayers. For rail, DVC would represent shipping costs such as the loading equipment, but also the siding, i.e. the biomass loading points, and sometimes the rail cars themselves plus infrastructure costs since the carrier owns the rails. Therefore, DFC for trucks is modest ($5/dry t) compared to rail ($27-28/dry t depending on feedstock). The DVC for trucks because of their reduced fuel efficiency per ton of cargo is high ($0.11- 0.13/dry t-km) compared to ultra-efficient rail ($0.03/dry t-km). DVC is linear as the wages, fuel, and capital recovery costs inherent to it vary directly with distance traveled.
The authors then generate cost graphs based on 2004 diesel fuel prices. From Graph 1, one can see that rail is only more cost effective for shipment of straw at distances greater than 200km.
I modified the DVC in their paper to see how the minimum cost-effective distance changed as diesel prices increased. The average price of diesel in 2008 was roughly three times the cost of diesel when the paper was written in 2004 (EIA, 2008). Since the DVC includes other costs besides fuel, I multiplied the DVC by a factor of 2.4 instead of 3. I could not find out the relative contributions of wages, fuel, and capital recovery costs to DVC, and instead assumed that 80% of the DVC came from fuel costs. As one can see in Graph 2, the increased costs of diesel fuel decreased the distance at which rail becomes cost competitive (now roughly 100km or approximately 60 miles).
I repeated these calculations for wood chips from forest harvest residues (FHR) and found a similar result (see Graphs 3 and 4): the distance to be cost competitive fell from 300km to 125km when diesel prices increased to 2008 levels.
Can Rail Replace Truck Transport?
Rail lines in much of North America are limited in their geographic reach. Roads, however, have high coverage in many, even rural locales. Mahmudi and Flynn (2006) show that despite the advantages rail has at medium distances, rail lines are often not located in adequate proximity to sources of biomass. In Alberta, a ‘central’ plant must be placed miles from the biomass sources at a point in the rail network where three separate rail lines finally meet. This increases the distance traveled so much that overall transportation costs eliminate profitability. Of course, this effect depends on the specific biomass feedstock shed and its attendant rail network, but it suggests that idealized models are limited without these specifications.
The question remains whether a plant that draws from medium to large distances should be built. Increasing distance means that the amount of biomass pooled will be much larger necessitating a bioenergy plant with a large capacity. Though previous research indicates that the power costs for larger bioenergy plants is relatively flat as a function of distance shipped (economies of scale from large size are offset by increased transportation costs) (Kumar et al. 2003), these models are based on cheap diesel. This suggests that the power costs associated with larger distances (and larger plants) would increase with distance in an expensive diesel market rather than stay flat as previously found. Therefore, even though rail transport may become cost competitive with truck transport because the latter is more affected by fuel price, the resulting increase in shipping costs for both modes limits profitability to small plants with a limited draw radius.
Other Transportation and Centralization Considerations
It has been proposed that rail might return to using coal-fired steam locomotion if liquid fuel prices continue to rise (Valentine 2009), which would allow for lower rail transportation costs as coal is expected to be cheaper per Btu than diesel for the long term. Holdouts might argue that coal-fueled trains could then become competitive. However, rail-based systems will still require large recruitment areas and thus large plants to process this biomass. Yet, centralized electricity production from biomass poses other challenges besides transportation costs. In order to maintain service in the case of a break in the supply change, surplus biomass would need to be stored. Dunnett et al. (2008) argue that a three day buffer storage at a cellulosic biorefinery from a relatively small draw area of 50km2 would require over 46,000 cubic meters of storage space. Finding space for buffers for even larger plants would be difficult.
Even if buffer space could be found, the fact remains that biomass has a low energy density compared to fossil fuels and maintaining continuous operation at a large plant will require continuous shipments of biomass. Estimates range from 50 truck deliveries per hour for a medium-sized biorefinery to a delivery every 4 minutes for a 450MW straw-based, direct combustion plant (Dennet et al. 2008, Kumar et al. 2006). These delivery schedules and their attendant traffic will almost certainly receive political backlash from affected locales. Since those transport alternatives that avoid traffic congestion (pipelines and rail) are not cost-competitive, the inevitable solution will be many, decentralized bioenergy plants.
Cellulosic biorefineries could utilize biomass piped in with water as a carrier. A pipeline obviates the need for buffer storage and would not affect traffic congestion once its construction is complete. Additionally, pre-digestion of the biomass could occur within the pipeline rather than at the biorefinery (Kumar et al. 2005). Pipelines involving flows of larger than 1.4 M dry tonnes/yr would be cheaper than truck transport even without considering economies of scale. Pipelines, by solving the problems elaborated above, open up the opportunity to create very large biorefineries where significant costs savings could be reaped and where large quantities of valuable byproducts (agroceuticals and other extracts) might be generated.
On the other hand, if you are afraid that cellulosic ethanol technology with its purported ability to turn any plant matter into biofuel could turn the world into a large ‘lawn’ for our mowing, then maybe pipelines and gargantuan biorefineries are not the kind of ‘progress’ we need…
Ashton, S and P. Cassidy. 2007, Fact Sheet 5.8 in Hubbard et al. (Eds.) Sustainable Forestry for Bioenergy and Bio-based Products. Trainers Curriculum Notebook. Southern Forest Research Partnership, Inc.
Dunnett, A., Adjiman, C., & Shah, N. (2008). A spatially explicit whole-system model of lignocellulosic bioethanol supply chain: an assessment of decentralized processing potential. Biotechnology for Biofuels 1:13-30.
Energy Information Administration. (2008). Diesel fuel prices: What consumers should know. Brochure #:DOE/EIA-X045. Department of Energy: Washington, D.C. Accessed at http://www.eia.doe.gov/bookshelf/brochures/diesel/
Hill, J., Nelson, E., Tilman, D., Polasky, S., & Tiffany, D. (2006). Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proceedings of the National Academy of the Sciences 103:11026-11210.
Kumar, A., Cameron, J., & Flynn, P. (2003). Biomass power cost and optimum plant size in western Canada. Biomass and Bioenergy 24:445-464.
Kumar, A., Cameron, J., & Flynn, P. (2004). Pipeline transport of biomass. Applied Biochemistry and Biotechnology 113:27-39.
Kumar, A., Cameron, J., & Flynn, P. (2005). Pipeline transport and simultaneous saccharification of corn stover. Bioresource Technology 96:819-829.
Valentine, H. (2009). Brief case for an investigation to be made into the suitability of modern steam locomotive production in areas with heavy production capability. Accessed at http://www.internationalsteam.co.uk/trains/newsteam/modern12.htm.
Yemshanov, D., & McKenney, D. (2008). Fast-growing poplar plantations as a bioenergy supply source for Canada. Biomass and Bioenergy 32:185-197.