Which sawdust route pays? Technoeconomic comparison of CHP, pellets and pretreated pellets for a Saskatchewan sawmill site

Introduction

Sawdust and planer shavings are the most visible by-products of a sawmill, and many mills still treat them as a low-value stream that is burned, sold cheaply or sent to pellet makers. A research group at the University of Saskatchewan and Agriculture and Agri-Food Canada (Saskatoon Research and Development Centre) asked a simple question with large consequences for mill economics: if a sawmill-adjacent plant had 72,000 tonnes of sawdust and shavings per year, which bioenergy route would earn the most money? The paper, published in the journal Energies on 15 February 2023, compares four process scenarios in a technoeconomic analysis (TEA).

The study site is hypothetical but located in a real setting: a plant at Meadow Lake, Saskatchewan, beside a sawmill and a bioenergy centre. According to the authors, earlier technoeconomic studies had not compared several bioenergy scenarios built on wood sawdust, which is the gap the paper addresses. The authors are Peyman Alizadeh, Lope G. Tabil, Edmund Mupondwa, Xue Li and Duncan Cree. Funding came from the Biofuel Network (ASC-16) and NSERC (RGPIN-2017-05287).

This is not a sawing study, but sawmill residues are central to the economics of a sawmill, so the paper belongs in the residues and economy category of the portal.

Research methodology

Scenarios. Four plant designs were simulated:

1. Combined heat and power (CHP) from raw sawdust.
2. Conventional pellets without CHP.
3. Steam-exploded pellets, with CHP and co-products.
4. Torrefied pellets, with CHP and co-products.

Tools and assumptions. Process modelling was done in SuperPro Designer V10.0. The plant processes 72 kt per year of sawdust and shavings and runs 7,920 hours a year (330 days). Utilisation was set at 80 percent in year 2, 90 percent in year 3 and 100 percent afterwards, with no operation in year 1. Financial assumptions included 2020 US dollars, a CAD/USD rate of 1.34, a 20-year project life, a 12-month construction period, 10-year depreciation, salvage value of 5 percent of direct fixed cost, a 40 percent revenue tax, electricity at 0.10 USD/kWh, water at 0.97 USD/m3, wastewater at 0.78 USD/m3 and a 10-year loan at 9 percent interest. The sawdust composition used was cellulose 46.5 percent, lignin 23.2 percent, hemicellulose 13.3 percent, water 7.74 percent and other solids 8.9 percent.

Feedstock cost. The feedstock price was excluded, effectively treating sawdust as free at the plant gate, and transport costs were excluded because the feedstock is on site.

Process conditions. Steam explosion was run at 180 C for 9 minutes (1000 kPa), which the paper says is lower than a literature optimum of 215 C. Torrefaction was run at 230 C for 45 minutes in a fluidised bed. Co-product prices assumed were furfural 800 USD/t, HMF 1,000 USD/t and acetic acid 1,200 USD/t; steam was valued at 20 USD/t (high pressure) and 12 USD/t (low pressure).

Economic indicators. Net present value (NPV) at a discount rate of 7 percent, internal rate of return (IRR), payback time (PBT) and return on investment (ROI). A sensitivity analysis varied the discount rate (7, 10 and 15 percent) and the main-product price (plus or minus 20 percent). Optimal plant size was also explored.

Key findings

Scenario 1: CHP from raw sawdust. The text reports about 1,161 kW of electrical power and roughly 4,896,453 kWh of electricity per year, with a total capital investment of about USD 14.5 million. The NPV was negative and stayed negative across all tested discount rates and a 20 percent steam-price increase. It was the weakest option on the economic indices.

Scenario 2: conventional pellets. Production of 72 kt per year of pellets at a stated cost of 44.33 USD/t against a selling price of 120 USD/t. Equipment cost was about USD 1.7 million and total capital investment about USD 10.5 million. The NPV at 7 percent was USD 30.9 million. The abstract reports payback of 2.49 years, IRR of 51.33 percent and ROI of 40.1 percent. The conclusions give the IRR as 49.6 percent, an inconsistency the portal could not resolve from the page.

Scenario 3: steam-exploded pellets with CHP and co-products. Output was 53 kt of pellets per year at a cost of 131.61 USD/t against a price of 144 USD/t. Equipment cost about USD 3.3 million and total capital investment about USD 20.8 million. Revenue was USD 9.17 million against operating costs of USD 7.05 million, a gross margin the paper gives as 25 percent. NPV at 7 percent was about USD 9.1 million, but it turned negative (minus USD 1.08 million) at a 15 percent discount rate and when the pellet price fell by 20 percent.

Scenario 4: torrefied pellets with CHP and co-products. Output was 46 kt of pellets per year at a cost of 149.03 USD/t against a price of 140 USD/t, so the pellets themselves were sold below production cost and the scenario relied on co-products. Equipment cost about USD 3.6 million and total capital investment about USD 22.5 million. Despite this, it had the highest NPV at 7 percent, USD 38.29 million, and the highest total revenue (about USD 14 million in the results, 14.174 million in the conclusions).

Which is best depends on the metric. The authors conclude that the torrefied-pellet route with co-products yields the highest revenue and NPV, while conventional pellet production is the simpler option with the higher IRR. They add that by-product valorisation can offset heavy capital investment.

Cost structure. Facility-dependent costs made up 56 to 61 percent of operating costs in all scenarios. Labour was 20, 29 and 29 percent in scenarios 1, 3 and 4 respectively, and utilities were 22 percent in scenario 2.

Plant size. The optimal CHP plant size was given as 250 to 300 kt of main product. Pellet plants, with or without pretreatment and CHP, were optimal at 150 to 200 kt per year. The authors compare these with earlier work that reported 150 to 190 kt for conventional pellets and 230 to 270 kt for steam pretreatment.

Payback. Cash-flow charts indicate payback after year 4 for most cases and after year 3 for scenario 2; the text also gives about 2.5 years for scenario 2 (the difference may reflect how years are counted, which the portal did not verify).

Practical applications for sawmills

This section is the portal's interpretation, not a statement by the authors.

• A relative ranking, not a business case. The study suggests that for a Saskatchewan-type setting with free sawdust, a plain pellet line is the low-complexity route and raw-sawdust CHP is the weak one. A mill should not copy these returns: feedstock was priced at zero, transport was excluded and the pellet price of 120 USD/t is an assumption of the paper.
• Scale matters. The optimal pellet plant (150 to 200 kt/year) is larger than the 72 kt/year of sawdust modelled. A single mid-sized sawmill may therefore lack the volume for an optimally sized plant and would need to pool residues with neighbours, or consider a partnership with a pellet producer.
• Ask what the co-product market looks like. Scenarios 3 and 4 depend on selling furfural, HMF and acetic acid. The pellet margins alone are thin or negative, so a mill assessing such routes would first need confirmed buyers for the chemicals.
• Use the paper as a checklist. The cost categories (facility-dependent costs, labour, utilities) and the sensitivity ranges give a practical template for a mill-specific spreadsheet.
• Residue-use decision. The comparison implies that using sawdust only for on-site electricity is the least attractive of the four options tested, which may be relevant to mills that currently invest in boiler or CHP capacity; local energy prices will change this.

Limitations

• Only text values were readable on the page. Tables 1 to 4 (equipment costs, production rates, cost and profitability summaries) showed captions only, Figures 1 to 3 were images without readable data, and Supplementary Tables S1 to S9 and Figures S1 to S4 were not accessed.
• The page shows internal inconsistencies: IRR for scenario 2 (51.33 percent in the abstract, 49.6 percent in the conclusions) and total revenue for scenario 4 (about USD 14 million versus 14.174 million). The portal reports both and recommends checking the PDF.
• Feedstock cost and transport cost were excluded, which makes every scenario look better than a mill buying or hauling sawdust would experience.
• The scenarios use fixed price assumptions and the authors say that pretreated-pellet margins are thin and depend on co-product sales; market dependency and effluent treatment are flagged as obstacles.
• The authors say the torrefied-pellet scenario needs an environmental analysis before adoption; none was done here.
• The plant is a simulation. No operating plant data were presented.
• The steam explosion condition used (180 C, 9 minutes) is not the literature optimum cited by the authors (215 C), which may affect yields.
• The portal read the article through a summarised fetch; the full discussion section was not separately visible and the reference list was truncated.

Conclusions

For 72 kt per year of sawdust and shavings at a Saskatchewan site, the technoeconomic simulation ranks four bioenergy routes. Raw-sawdust CHP performs worst, with a negative NPV. Conventional pellets are the simplest and have a high IRR and short payback by the paper's numbers. Torrefied pellets with CHP and co-products give the largest NPV (USD 38.29 million at 7 percent), though the pellets themselves are priced below their production cost and the result leans on co-product sales and a large capital outlay. Plant scale in the range of 150 to 200 kt per year was found to be optimal for the pellet routes. The results are sensitive to discount rate and product price, ignore feedstock cost and transport, and should be read as a screening-level comparison.

Original research

• Authors: Peyman Alizadeh, Lope G. Tabil, Edmund Mupondwa, Xue Li, Duncan Cree
• Title: Technoeconomic Feasibility of Bioenergy Production from Wood Sawdust
• Journal: Energies, Volume 16, Issue 4, Article 1914
• Year: 2023 (published 15 February 2023)
• DOI: 10.3390/en16041914
• Link: https://www.mdpi.com/1996-1073/16/4/1914
• Affiliations: Department of Chemical and Biological Engineering and Department of Mechanical Engineering, University of Saskatchewan; Saskatoon Research and Development Centre, Agriculture and Agri-Food Canada
• Funding: Biofuel Network (ASC-16) and NSERC (RGPIN-2017-05287)
• Licence: Open access; page metadata lists CC BY 3.0

This article is an independent summary prepared by the portal; the original paper should be consulted for full methods, tables and data.