Tracking Material Provenance for a Circular Economy

15 concepts8 papers1 proposal

When a structural steel beam, an aluminum car chassis, or an electric vehicle battery reaches the end of its operational life, it does not suddenly lose its intrinsic material value. However, in our current industrial paradigm, these high-value engineered assets are routinely downcycled or discarded.

The barrier to true circularity is rarely a lack of recycling technology; rather, it is an information deficit. As highlighted by Raabe et al., the absence of precise chemical composition and historical data for post-consumer scrap prevents the high-fidelity recovery of specific alloy systems. Without knowing exactly what a material is made of, or what mechanical stresses it has endured, engineers cannot safely verify its reuse potential, forcing industries to rely on carbon-intensive virgin production.

To bridge this lifecycle traceability gap, a new research direction has emerged focused on establishing material provenance. This framework operates via two complementary strategies: proactive digital tracking (building a continuous digital thread) and reactive physical inference (using advanced sensing and machine learning to decode a material's hidden history).


The Proactive Strategy: Digital Material Passports

The most direct way to ensure material traceability is to carry structural and chemical data alongside the physical asset throughout its entire lifecycle. This digital ledger is frequently termed a "material passport."

[Physical Asset + Sensors] ---> [Standardized Ontology] ---> [Digital Twin / Passport]
       (RFID / Stress)               (Semantic Web)              (BIM / Circular Loop)

To implement this, Ness et al. proposed integrating Radio Frequency Identification (RFID) tags and real-time stress sensors directly into structural steel components. By linking this physical hardware to Building Information Modeling (BIM) software, the authors demonstrated how real-time structural health data can be preserved, transforming passive metal into "smart steel" with a verifiable service history.

However, a material passport is only useful if different stakeholders across the supply chain can read and trust the data decades later. Addressing this infrastructure gap, Elshani et al. developed a standardized semantic web ontology designed to track the "life story" of prefabricated building components. This formal data structure ensures that safety logs, environmental impacts, and material properties remain interoperable across multiple lifecycles.

As Davis et al. emphasize, this continuous in-service monitoring provides the foundational data infrastructure required for safe remanufacturing and reuse, shifting the recycling paradigm from blind processing to informed asset management.

Retrofitting the Past with AI

While digital passports work well for newly manufactured goods, the vast majority of our existing infrastructure was built without a digital thread. To prevent these legacy materials from being lost to the downcycling loop, Lauble et al. introduced an artificial intelligence workflow that uses computer vision and natural language processing to automatically extract material specifications from legacy 2D floor plans and PDF reports. This allows industries to retroactively generate material passports for existing building stocks, unlocking circularity for older assets.


The Reactive Strategy: Non-Destructive Inference

In reality, physical labels tear off, digital records are lost, and heterogeneous scrap streams inevitably become mixed. When the digital thread breaks, recyclers face what Haizhou Liu et al. describe as a complete loss of physical identification. In these scenarios, the material's provenance must be inferred through non-destructive sensing and machine learning.

[Unknown Scrap / Battery] ---> [Non-Destructive Sensing] ---> [Machine Learning Model] ---> [Inferred Provenance]
                                (PGNAA / Vision / Cycles)       (ResNet / Density / FL)       (Exact Grade/Chemistry)

High-Throughput Elemental Fingerprinting

To sort complex metal scrap at industrial scale, researchers are moving beyond slow, destructive chemical assays. Raabe et al. point to high-throughput automated sorting technologies, such as Laser-Induced Breakdown Spectroscopy (LIBS), as critical tools for identifying exact chemical compositions on sorting lines.

For even deeper elemental analysis, Shayan et al. utilized Prompt Gamma Neutron Activation Analysis (PGNAA). By treating the resulting gamma-ray spectra as probability distributions and applying kernel density estimation with maximum log-likelihood, the authors achieved near-perfect classification of aluminum alloys in less than a quarter of a second. This speed allows heterogeneous scrap streams to be sorted accurately in real time, preventing the mixing of incompatible alloys.

Low-Cost Machine Vision

Because advanced spectroscopic equipment like LIBS or PGNAA can be prohibitively expensive for smaller recycling operations, alternative methods leverage surface characteristics. Jian‐Hong Liu et al. demonstrated that deep learning architectures (specifically ResNet18) can classify aluminum scrap grades with remarkable accuracy by analyzing surface features and color spaces alone, offering a highly accessible, non-destructive alternative to nuclear or laser-based sensing.

Decoding Electrochemical History

The challenge of missing provenance is perhaps most acute in battery recycling, where the internal cathode chemistry of a retired electric vehicle battery is completely hidden from view. Dismantling a battery to check its chemistry is hazardous and expensive.

To solve this, Haizhou Liu et al. developed a federated machine learning framework capable of identifying a battery's internal cathode chemistry using data from just a single non-destructive charge-discharge cycle. The subtle thermodynamic and kinetic variations in the voltage curve act as a physical fingerprint of the battery's internal state. By utilizing federated learning, the authors enabled multiple manufacturing stakeholders to train this diagnostic model collaboratively without sharing proprietary, confidential design data.


Why It Works: The Intuition

The success of this research direction relies on a simple realization: a material's physical state and structure always bear the indelible marks of its composition and history.

Whether it is the specific gamma-ray emission spectrum produced when neutrons strike an aluminum alloy (Shayan et al.), the visual texture left by industrial processing (Jian‐Hong Liu et al.), or the voltage response of an aging electrochemical cell (Haizhou Liu et al.), the data is physically present within the asset.

By pairing these physical signatures with modern machine learning, or by preserving them proactively through standardized digital ontologies (Elshani et al.), this field closes the lifecycle information gap. Providing verifiable provenance data gives recyclers and structural engineers the confidence required to treat post-consumer scrap not as waste, but as a reliable, high-fidelity resource.

Key Foundational Work

Smart steel: new paradigms for the reuse of steel enabled by digital tracking and modelling (2014) — contributed 3 concept elements to this direction.
Strategies for improving the sustainability of structural metals (2019) — contributed 2 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_material_passports_bcbc0361,
  title = {material passports},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/material-passports/rdr_bcbc0361b93a43c688f4649af183c501},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_material_passports_bcbc0361_v1,
  title = {material passports},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/material-passports/rdr_bcbc0361b93a43c688f4649af183c501/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake