Industries / Food, farming and forestry / Forestry and logging / Timber and sawmills / Lumber treating
Disposal of treated wood
Deciding what happens to treated wood when it comes out of service: separating it from clean wood, and landfilling, burning or reusing it without spreading its chemicals.
5
Milestones achieved, from 2002 to 2021.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
2002
Arsenic concentrates in ash
Helena Solo-Gabriele of the University of Miami, Timothy Townsend of the University of Florida and colleagues burned CCA-treated wood, mixtures and recycled construction and demolition (C&D) wood in an industrial furnace. Copper, chromium and arsenic together made up as much as 36 percent of the ash by weight for heavily treated wood, and ash from mixtures containing only 5 percent CCA-treated wood leached enough arsenic to be classed as hazardous waste under US rules.
LessonBurning shrinks treated wood but concentrates its metals, so a few treated pieces decide where all the ash can go.
Credit
Helena Solo-Gabriele, Timothy Townsend and colleagues
2002
Germany sorts waste wood
Germany's Waste Wood Ordinance (Altholzverordnung) of 15 August 2002 puts used wood into four classes, from A I, untreated or only cut, to A IV, wood treated with preservatives "such as railway sleepers, utility poles, hop poles, vine stakes". Wood in class A IV may not be chipped for particleboard; for material recovery the ordinance allows it only for making synthesis gas, activated carbon and industrial charcoal in permitted plants, and wood that is not recovered must go to an approved thermal treatment plant.
Credit
the German federal government
2003
Treated wood in garden mulch
Townsend, Solo-Gabriele and colleagues found that 18 of 22 samples of wood from C&D recycling plants in Florida leached arsenic above the state's groundwater cleanup target of 50 µg/L (micrograms per liter), and calculated that less than 0.1 percent CCA-treated wood would put a mulch over Florida's residential soil guideline for arsenic. A statewide follow-up published in 2007 found that most colored mulch made from recycled C&D wood held 1 to 5 percent CCA-treated wood, with up to 230 mg/kg of arsenic, and that dye or plywood chips in a bag flagged the suspect mulch.
LessonA trace of treated wood spoils a whole stream of recycled wood, so sorting has to come before chipping.
Credit
Timothy Townsend, Helena Solo-Gabriele, G. Jacobi and colleagues
2016
Old sleepers become fuel
On 8 February 2016 EPA added processed creosote-treated railroad ties to its list of "categorical non-waste fuels": once metal is removed and the ties are shredded or ground, plants built to burn both biomass and fuel oil may burn them without a case-by-case review as solid waste. A 2018 rule added ties treated with creosote-borate and copper naphthenate. The 2016 rule cites forest industry figures that about 17 million crossties come out of US track each year, about a third go to landscaping and most of the rest to energy recovery.
Credit
US EPA
2021
California's disposal rule lapses
For nearly 15 years California's Alternative Management Standards let treated wood waste go to the composite-lined part of approved solid waste landfills. They expired at the end of 2020, and for the first three months of 2021 the only lawful destination was a hazardous waste landfill; Ventura County's public works department had warned that stockpiling might become more common. The Department of Toxic Substances Control (DTSC) issued variances from March, and Assembly Bill 332, passed unanimously and signed on 31 August 2021, wrote the old standards into law with immediate effect.
LessonA disposal rule with an end date can strand a whole waste stream overnight.
Credit
California Legislature; DTSC
Watch for
Hyperspectral cameras find copper
A hyperspectral camera records hundreds of narrow bands of visible and infrared light for each pixel. Researchers at Ghent University's Woodlab and Unilin Panels, a Belgian maker of wood panels, built 1,320 samples of six species, untreated or treated with three copper-based and two organic preservatives at several strengths. Two machine-learning models placed samples in the right copper concentration range more than 90 percent of the time. When a species or preservative was left out of training, results depended strongly on what the model had been trained on.
Achieved whenanother group tests the method on recovered wood from a working recycling stream, with each piece's copper measured by chemical analysis, and publishes the accuracy.
Claimed by
Maxime Dekegeleer and colleagues, Ghent University and Unilin Panels, in Waste Management, April 2026
Open
01
Real waste wood, labeled
An open reference set of real recovered wood from recycling yards in several countries: each piece photographed, scanned with a near-infrared or hyperspectral sensor, and then analyzed by a laboratory for copper, chromium, arsenic and boron. Add a baseline classifier and test it on yards it never saw in training.
The briefCredit · unclaimed02
Mulch retested, twenty years on
A survey of dyed and recycled-wood mulch sold in stores across many states and countries, with arsenic, chromium and copper measured by a laboratory, repeating the method of the Florida studies of 2003 and 2007.
The briefCredit · unclaimed
Open 01 · the brief
Real waste wood, labeled
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open reference set of real recovered wood from recycling yards in several countries: each piece photographed, scanned with a near-infrared or hyperspectral sensor, and then analyzed by a laboratory for copper, chromium, arsenic and boron. Add a baseline classifier and test it on yards it never saw in training.
Why it is still open
Published sorting models are trained on wood treated in the laboratory, such as the 1,320 samples in the 2026 Ghent study, which found that accuracy depends on which species and preservatives are in the training set. Manuela Mancini and Åsmund Rinnan classified real waste wood from Italy and Denmark by near-infrared spectroscopy in 2023, 86.7 percent correctly for treated wood, but their Zenodo record holds the paper, not the spectra. A University of Miami and University of Florida group showed in 2004 that X-ray fluorescence (XRF), which identifies elements by the X-rays they give back, detects CCA through paint and in wet wood, and a full-scale automated XRF unit it tested in 2011 diverted 98 percent of the arsenic in the wood; no report of such sorting in routine commercial use was found. Searches of Zenodo and Hugging Face in November 2026 found no open dataset of real waste wood with measured preservative content.
Built with what exists
A phone camera; a handheld near-infrared spectrometer or a hyperspectral camera; metals analysis by an accredited laboratory; Python with scikit-learn and PyTorch; Zenodo to host the data; gloves and a dust mask for handling.
First step
Ask one wood recycler for 100 pieces from its incoming stream, and get a laboratory quote for copper, chromium, arsenic and boron in each.
Done looks like
At least 1,000 pieces from at least 5 yards in at least 2 countries, each with images, spectra and laboratory metal results, released openly, and a published share of treated pieces found at yards held out of training, set beside the 98 percent of arsenic that the 2011 XRF test diverted.
Open 02 · the brief
Mulch retested, twenty years on
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A survey of dyed and recycled-wood mulch sold in stores across many states and countries, with arsenic, chromium and copper measured by a laboratory, repeating the method of the Florida studies of 2003 and 2007.
Why it is still open
The store surveys that found CCA-treated wood in mulch were made in Florida in 2003 and 2007. A. S. Jones and colleagues (2019) estimate that waterborne-treated wood thrown away in the United States may pass 16 million m³ a year by 2030, with the copper in it rising to 20,900 metric tons, and that treated wood must be separated out for recycled C&D wood to meet guidelines. Searches of Europe PMC for 2009 to 2026 and of the web in November 2026 found no later survey of mulch in stores.
Built with what exists
Bags of store-bought mulch; a commercial environmental laboratory for total metals and the synthetic precipitation leaching procedure (SPLP), the standard US leaching test the Florida studies used; the visual screen from the 2007 study (dye, plywood chips); a spreadsheet and a public map.
First step
Buy 10 bags of dyed mulch from five stores in one state and send them to a laboratory for arsenic, chromium and copper.
Done looks like
At least 100 bags of at least 30 products from at least 10 states, each with its lot number, a photo and laboratory results, published with the share above the 2007 study's marks of 50 µg/L of leachable metals and 10 mg/kg of recoverable metals.