I recently noticed a few new mining studies that relied on toll milling as their base case operation. This made me wonder if this is a trend and if there are more examples of where this is happening.
If only there was a way to find out.
I have been tinkering with Ai (mainly ChatGPT and Gemini) to see what it can do, but I never really had a “project” for it to work on.
So I thought, lets see what Ai can tell us about toll milling. This would be an opportunity to test out Claude (AI assistant built by Anthropic) to see what it will deliver. I could ask it a few questions about the concept of toll milling to find out what it would say.
The following blog is the unedited responses that were given to my user prompts. The results, albeit lengthy, are interesting and informative. Let’s dive in.
By the way, this is the first blog post ever where I have applied Ai.
CLAUDE Ai CONVERSATION
USER PROMPT: Can you prepare a list of recent mining studies that used toll milling as the base case.
RESPONSE: Here is a summary of recent mining studies (primarily PEAs and PFS-level studies) that used toll milling as the base case or a key component, based on publicly available information from 2024-2026:
1. RPX GOLD – WAWA GOLD PROJECT (ONTARIO, CANADA) PEA – February 2026
The base case is built entirely around toll milling, with crushed ROM ore trucked to an off-site facility within 150 km for processing into gold dore. No on-site mill or tailings facility is planned, significantly reducing capital requirements. The project sits in the Michipicoten Greenstone Belt near Wesdome, Alamos, and Hemlo operations, all of which offer potential excess capacity.
2. AMEX EXPLORATION – PERRON GOLD PROJECT (QUEBEC, CANADA) Updated PEA – September 2025
Features a two-stage development strategy: a 4-year, 1,000 tpd contract mining and toll-milling operation in Phase 1, followed by construction of an on-site processing plant in Phase 2. The Abitibi region’s numerous operating mills make this approach practical. Initial capex is C$146M, partially offset by C$68.6M in pre-production revenues during the toll-milling phase.
3. CHALLENGER GOLD – HUALILAN PROJECT (ARGENTINA) PFS – 2024/2025
The PFS is structured around toll milling as the path to near-term production, with the study outlining forecast EBITDA of approximately A$221M over a 3-year tolling period at current metal prices. Challenger also executed a binding term sheet with Dolphin Real Assets/Elsztain Group to facilitate processing through a tolling agreement, with the arrangement expected to fund mining, transportation, and processing costs until initial gold revenue is received.
4. SCOTTIE RESOURCES – SCOTTIE GOLD MINE PROJECT (BRITISH COLUMBIA, CANADA) PEA – October 2025
The primary base case is a Direct-Ship Ore (DSO) scenario, but the study prominently presents toll milling at the nearby Premier mill as a significant upside scenario. Under the toll-milling arrangement, the after-tax NPV(5%) rises from C$215.8M to C$380.1M at US$2,600/oz gold – a meaningful enhancement. No formal toll-milling agreement was in place at the time of the study.
5. MAGNA MINING – CREAN HILL PROJECT (ONTARIO, CANADA) PEA – 2024
The PEA incorporates a definitive off-take agreement with Vale Base Metals to process Contact mineralization through Vale’s Clarabelle Mill in Sudbury. This is a firm, contracted toll-milling arrangement rather than a conceptual one, giving the study’s economics a higher degree of confidence in processing costs.
6. AURIC MINING – JEFFREYS FIND GOLD MINE (WESTERN AUSTRALIA) Operational toll-milling program – 2024
While not a formal feasibility study, Auric and joint venture partner BML Ventures executed a contract to toll mill a minimum of 300,000 tonnes of ore through the Greenfields Mill in Coolgardie during 2024. The first stage produced ~9,741 oz of gold at 1.86 g/t, demonstrating the model’s viability for small, high-grade open pit operations lacking on-site processing infrastructure.
GENERAL OBSERVATIONS:
Toll milling as a base case is most common in:
– Early-stage PEAs for junior developers in established mining districts (Abitibi, Michipicoten, Sudbury) where nearby mill capacity exists.
– High-grade, lower-tonnage projects where the capital cost of a dedicated mill is hard to justify at the outset.
– Projects pursuing a phased development path to generate near-term cash flow before committing to full plant construction.
USER PROMPT: What are the risks of toll milling plan.
RESPONSE: Toll milling carries several meaningful risks that study authors and investors should consider carefully.
OPERATIONAL & LOGISTICS RISKS
The most fundamental risk is mill availability. A toll miller may prioritize its own ore or other clients during peak periods, leaving your project without processing capacity exactly when you need it. Scheduling conflicts can delay production and erode project economics significantly. Transportation of ore or concentrate over long distances also introduces costs that are sensitive to fuel prices, road conditions, and seasonal access – particularly in remote northern jurisdictions.
COMMERCIAL & CONTRACTUAL RISKS
Toll milling agreements are typically not long-term ironclad contracts, and terms can be renegotiated or terminated. The toll miller holds significant leverage, especially if there are few alternative processors nearby. Processing rates, payable metal terms, and penalty clauses for deleterious elements can all shift unfavorably over time. If no binding agreement exists at the study stage – as was the case with Scottie Resources – the entire economic base case rests on an assumption rather than a firm commitment.
METALLURGICAL RISKS
The toll miller’s circuit is optimized for their own ore, not yours. Differences in mineralogy, grind size requirements, or the presence of penalty elements like arsenic, bismuth, or mercury can result in lower recoveries than modeled, higher processing charges, or outright refusal to accept the ore. Blending your material with the host mill’s feed can also complicate settlement and metal accounting.
FINANCIAL & ROYALTY RISKS
Without ownership of the processing facility, you have limited control over costs. Any increase in the toll miller’s operating costs – energy, reagents, labor – can be passed through to you. Additionally, some royalty structures are calculated on net smelter returns, meaning toll milling charges directly reduce the royalty base and affect third-party obligations.
REGULATORY & PERMITTING RISKS
Transportation of ore across jurisdictions may require additional permits, environmental assessments, or spill containment plans. If the toll mill’s own permits are challenged or its tailings facility reaches capacity, your project could be indirectly affected by regulatory actions aimed at a third party.
STRATEGIC RISKS
Relying on toll milling can delay the development of on-site infrastructure and institutional processing knowledge, which matters if the long-term plan is to build a dedicated mill. It can also make the project harder to finance or sell, since acquirers typically prefer integrated operations with controlled processing costs. Finally, if the toll miller is acquired, restructured, or shuts down, continuity of the arrangement is immediately in question.
In short, toll milling is an elegant solution for reducing upfront capital, but it trades capex risk for a different set of operational, commercial, and strategic risks that need to be carefully managed and clearly disclosed in any technical study.
USER PROMPT: What responsibility does the Qualified Person have in signing off on a study with toll milling.
RESPONSE: The Qualified Person (QP) signing off on a technical study that uses toll milling as the base case carries significant professional and regulatory responsibility, particularly under NI 43-101 in Canada or JORC in Australia.
CORE RESPONSIBILITY UNDER REGULATION
Under NI 43-101, the QP must take responsibility for the technical report in its entirety, or for clearly delineated sections where multiple QPs are involved. The QP must have sufficient relevant experience not just in geology or engineering broadly, but specifically in the type of deposit and the technical aspects being certified – which in a toll milling scenario includes process engineering, metallurgy, and mine economics. A geologist signing off on processing assumptions they are not qualified to assess would be in breach of their obligations.
REASONABLENESS OF ASSUMPTIONS
The QP is required to certify that the study’s assumptions are reasonable and reflect industry best practice. For toll milling specifically, this means the QP must be satisfied that the processing costs used in the economic model are realistic and supportable. If the toll milling rate is based on a verbal understanding or a letter of intent rather than a binding agreement, the QP must disclose this clearly and assess whether the assumed rate is achievable in the current market. Using a stale or unsupported toll rate without disclosure would represent a failure of duty.
DISCLOSURE OF RISKS
The QP has an affirmative obligation to disclose material risks that could affect the project’s viability. A rigorous technical report should explicitly identify the risks associated with toll milling – mill availability, contract uncertainty, metallurgical compatibility, transportation costs – rather than burying them in boilerplate language. Superficial or generic risk disclosure that does not reflect the specific circumstances of the toll milling arrangement would fall short of professional standards.
METALLURGICAL COMPETENCE
If the study relies on recoveries and processing costs derived from testwork conducted on the toll miller’s circuit, the QP must evaluate whether that testwork is adequate, recent, and representative of the ore to be processed. If no testwork has been done on the specific mill to be used, the QP must justify why analogous data is sufficient, and must disclose the uncertainty this introduces. Blindly accepting a recovery assumption without interrogating its basis is a serious professional risk.
INDEPENDENCE AND CONFLICTS OF INTEREST
NI 43-101 requires the QP to be independent in certain circumstances, and to disclose any relationships with the issuer. If the toll miller is a related party or has a financial interest in the project, the QP must consider whether this creates a conflict that needs to be disclosed or whether it affects the reliability of commercial terms assumed in the study.
SITE VISIT REQUIREMENTS
The QP is generally required to have conducted a site visit. In a toll milling scenario, a thorough QP would arguably need to be familiar not just with the mine site but with the processing facility to be used – its condition, spare capacity, metallurgical compatibility, and operational track record. Signing off on processing assumptions for a facility that has never been visited or independently assessed is a professional vulnerability.
PRACTICAL AND PROFESSIONAL CONSEQUENCES
If a study proves materially misleading due to unsupported toll milling assumptions, the QP faces regulatory action from securities regulators, professional discipline from their engineering or geoscience association, civil liability to investors, and reputational damage. Canadian securities regulators have historically taken a dim view of QPs who rubber-stamp economic assumptions without genuine independent verification.
In summary, the QP’s responsibility in a toll milling study is not merely to sign a form – it is to genuinely interrogate the commercial, metallurgical, and logistical assumptions underpinning the arrangement, ensure they are adequately disclosed, and take personal professional accountability for their reasonableness. A QP who treats toll milling as a convenient shortcut to a positive economic outcome without doing this work is exposing themselves to considerable risk.
I asked Claude to create a QP checklist for factors that should be reviewed for a toll milling study. You can download that file at this link DOWNLOAD FILE.
CONCLUSION
The results given by Claude are quite thorough and insightful. It’s hard to argue with its observations and conclusions. This research took all of 30 seconds, so I can see it is no longer difficult to become a blog writer. Writing isn’t the challenge; finding interesting topics is.
Syncrude had four large walking draglines, each with a 80 cubic metre bucket and 110 metre operating radius. These were very big machines; you could sit one in the end zone of a football field and the bucket would be digging (or dumping) in the other end zone. Two draglines were on the East side of the mine and two were on the West, mining the oilsand in 25 m wide strips.
There were numerous instances of East mine block slides, where large portions of the upper slope would fail as large blocks, 50 metres long and up to 30 metres back from the crest. The fear was that if a dragline happened to be sitting on one of these failing blocks, the entire machine would slide along into the pit. Many block slides did occur over the years, but only a few came close to jeopardizing a machine. The geotechnical monitoring programs in place were successful (described later).
The insitu clay structures were identified using oil and gas borehole logging technology, with tadpole dipmeter plots (see image) used to analyse the bedding (the tail on the tadpole shows the dip direction). The vertical axis is depth from surface or elevation. The geotech engineers would use this information, combined with structural mapping of previously mined faces, to forecast potentially unstable areas.
The main geotechnical issue on the West side were basal slope failures, termed this due to sliding along weak clays and muds at the base of the highwall. This photo shows a typical basal failure. Basal failures also occured on the East side.
Once our engineer-in-training rotation program was complete, we were to be assigned to a more permanent position. For me, that was going to be as an East side geotechnical engineer – ugh!. It’s at that time I decided to look for greener pastures. Three years was long enough from 1980 to 1983; given the amount of learning and responsibility I had undertaken. Other colleagues left the same time, while many other friends stayed in Ft McMurray for their entire careers.
In Part 1 of this two part blog post I would like to share some stories from the early days of my career working in Fort McMurray.
At the time Syncrude had an excellent engineer-in-training program for new graduates. Every six months they would rotate engineers into different technical areas.
Next we sampled that depth carefully, revealing that frozen muskeg layers were present. When we installed standpipe piezometers in these holes, we saw water flowing out of the top of the pipes. This means the foundation pore pressure is high, way too high.
For example, one project I had was to monitor the performance of different brands and styles of conveyor idlers. We would track about 2,000 individual idlers; when they were installed on the conveyors; when they were removed, why they were removed (bearing failure, cover failure, something else).
The mining industry is implementing more and more technology in the mining cycle.
Mine reconciliation requires information such as initial predictions from exploration data and geological models, actual measurement: data from mining sources, such as blast holes, stockpile samples, or mill feed. As well it will need data on the final product being shipped off site. Do the metal quantities balance out throughout the mining operation?
Each mine site may be unique with respect to; ore sources; terminology; ore types; mining methods; stockpiling philosophy; processing methods; technology availability; and personnel capability. So often the easiest approach for mine reconciliation is based on the Excel spreadsheet. (Reconciliation is generally not an easy undertaking).
An example of a satellite being used is the Sentinel-1, launched in mid-2015 by the European Space Agency. This satellite information is open-source data. It will have a 6 to 12 day revisit cycle in many locations.
On LinkedIn, one can see numerous posts where independent experts are examining historical InSAR data for recent failures to see whether early movement should have been detected. The results seem to be quite positive in that areas that have failed might have been red-flagged prior to failure.
A mining site consists of numerous constructed embankments and slopes of all types and heights. Many of these slopes may be creeping and moving all the time – it’s a living beast.
Two dilution approaches are common. One can either construct a diluted block model; or one can apply dilution afterwards in the production schedule. I have used both approaches at different times.
Sometimes lower grade stockpiles are built up by the mine each year but only processed at the end of the mine life. Periodically the ore mining rate may exceed the processing rate and other times it may be less. This is where the stockpile provides its service, smoothing the ore delivery to the plant.
Once the schedules are finalized, they are normally reviewed by the client for approval. The strip ratio and ore grade profile by date are of interest. One may then be asked to look to at different stockpiling approaches to see if an NPV (i.e. head grade) improvement is possible.
The last task for the mine engineer in Chapter 16 is estimating the open pit equipment fleet and manpower needs. The capital and operating costs for the mining operation will also be calculated as part of this work, but the costs are only presented in Chapter 21.
The support equipment needs (dozers, graders, pickups, mechanics trucks, etc.) are typically fixed. For example, 2 graders per year regardless if the annual tonnages mined fluctuate.
These two blog posts hopefully give an overview of some of the things that mining engineers do as part of their jobs. Hopefully the posts also shed light on the amount of work that goes into Chapter 16 of a 43-101 report. While that chapter may not seem that long compared to some of the others, a lot of the effort is behind the scenes.
So I thought what better way to explain the mining engineer role than by describing the anatomy of a typical Chapter 16 (MINING) in a 43-101 Technical Report. That chapter is a good example of the range of tasks typically undertaken by mining engineers.
There is always a mineral resource estimate available before doing a PEA. The way the resource is reported will indicate what type of mine this likely is. The geologists have already done some of the mining engineer’s work.
Before starting pit optimization, we require economic inputs from several people. The base case metal prices must be selected (normally with input from the client). The mining operating cost per tonne must be estimated (by the mining engineer). The processing engineers will provide the processing cost and recovery for each ore type.
Once the optimization is run, a series of nested pit shells are created, each with its own tonnes and grade. These shells are compared for incremental strip ratio, incremental head grade, total tonnes, and contained metal.
The mining engineer is now ready to undertake the pit design. The pit design step introduces a benched slope profile, smooths out the pit shape, and adds haulroads. Hence a couple of key input parameters are required at this time. The mining engineer will need to know the geotechnical pit slope criteria and the truck size & haul road widths. Let’s look at both of these.
Ramps: Next the mining engineer needs to select the truck size, even though the production schedule has not yet been created.
This ends Part 1. In Part 2 we will discuss the mining engineer’s next tasks; production scheduling; waste dump design; and equipment selection. The mining engineer QP will sign off and take responsibility for all the mine design work done so far. You can read Part 2 at this link “
This article is about the benefit of preparing (cutting) more geological cross-sections and the value they bring.
Long sections are aligned along the long axis of the deposit. They can be vertically oriented, although sometimes they may be tilted to follow the dip angle of an ore zone.
When looking at cross-sections, it is always important to look at multiple cross-sections across the orebody. Too often in reports one may be presented with the widest and juiciest ore zone, as if that was typical for the entire orebody. It likely is not typical.
Bench plans (or level plans) are horizontal slices across the ore body at various elevations. In these sections one is looking down on the orebody from above.
3D PDF files can be created by some of the geological software packages. They can export specific data of interest; for example topography, ore zone wireframes, underground workings, and block model information. These 3D files allows anyone to rotate an image, zoom in as needed and turn layers off and on.
The different types of geological sections all provide useful information. Don’t focus only on cross-sections, and don’t focus only on one typical section. Create more sections at different orientations to help everyone understand better.
When I am undertaking a due diligence review or working on a study, very early on I like to have a look at the grade-tonnage information. This could be for the entire deposit resource, within a resource constraining shell, or in the pit design.
However, if the tonnage curve profile resembled the light blue line in this image, with a concave shape, the ore tonnage is decreasing rapidly with increasing cutoff grade. This is generally not a favorable situation.
Regarding mineral resources, one should be required to disclose the waste tonnage and strip ratio when reporting resources inside a constraining shell. The constraining shell and cutoff grade are both based on defined economic factors such as unit mining costs, processing cost, process recoveries, and metal prices. With respect to the mining cost component, the strip ratio is a key aspect of the total mining cost, yet it normally isn’t disclosed.
In 43-101 technical reports, the financial Chapter 22 normally presents the project sensitivities expressed in a spider diagram or a table format.
The primary question to be answered is whether one can mine safely and economically without creating significant impacts on the environment.
Lake Turbidity: Dike construction will need to be done through the water column. Works such as dredging or dumping rock fill will create sediment plumes that can extend far beyond the dike. Is the area particularly sensitive to such turbidity disturbances, is there water current flow to carry away sediments?
Pit wall setback: Given the size and depth of the open pit, how far must the dike be from the pit crest? Its nice to have 200 metre setback distance, but that may push the dike out into deeper water.
Once the approximate location of the dike has been identified, the next step is to examine the design of the dike itself. Most of the issues to be considered relate to the geotechnical site conditions.
Each mine site is different, and that is what makes mining into water bodies a unique challenge. However many mine operators have done this successfully using various approaches to tackle the challenge.