Written by a member of the QIMC investor community

Mapping commercial flow

Note: The following is meant to be speculative and not confirmed by the company nor any of its representatives.

The following is a prediction of what the commercial flow will be at Bennett Hill, given all the news from QIMC and news from other companies in the industry. This prediction is backed up by research, but I’m not a scientist (nor claim to be) so just keep this in mind. There are a couple of papers linked.

Before asking how much hydrogen, we need to ask ourselves: how is hydrogen generated at Bennett Hill?

There are 3 proposed hydrogen-generation pathways and 3 potential accumulation/production mechanisms that could explain the hydrogen observed at Bennett Hill. The 3 pathways are proposed hydrogen-generation processes, whereas the three mechanisms describe how that generated hydrogen may exist, accumulate, migrate, and ultimately be produced. These pathways are under the term “serpentinization”, which is the umbrella term for the whole category of water-rock reactions that generate hydrogen. Each of pathways will be described and supplemented with Layman's terms:

Pathway 1. Magnetite oxidation — magnetite (Fe₃O₄) reacts with water under heat and pressure, releasing hydrogen as the iron oxidizes from Fe²⁺ to Fe³⁺. This is suggested at Bennett Hill through magnetic susceptibility signatures and explicitly named as one of the three simultaneous pathways. This is the dominant generation pathway and the primary contributor to Peak and repeated high-range readings in DDH-26-04.

Layman's terms: Chemical reaction with water under heat and make lots of hydrogen.

Pathway 2. Amphibole oxidation — iron-bearing amphibole minerals (hornblende, actinolite) undergo similar water-rock oxidation reactions. Also suggested at Bennett Hill through core description — amphibolite-rich intervals are documented in the geological logs from both DDH-26-04 and DDH-26-05. Contributes to the same free-gas generation pool as magnetite, running concurrently in the same rock package.

Layman's terms: More chemical react with more water to make lots of hydrogen.

Pathway 3. Biotite oxidation — iron-bearing biotite mica undergoes hydrothermal oxidation releasing hydrogen. Also hypothesized through core description. Biotite is specifically one of the minerals Richer-Laflèche cited in the IOCG-style alteration documentation — the hydrothermal breccia zones at 176-251m in DDH-26-05 and the equivalent zone in DDH-26-04 both contain biotite-bearing rock.

Layman's terms: Loss of electrons makes lots of Hydrogen.

These 3 pathways help generate the hydrogen, and are all distinct, although related. There are now 3 different mechanisms which come from these pathways to create the flow:

Mechanism 1 — Free gas from the vuggy/fractured zone:
All three generation pathways (magnetite + amphibole + biotite) contribute hydrogen to this mechanism — they're three taps feeding the same pipe. This is gas that has migrated upward from the generation zone, accumulated in the vuggy and brecciated intervals of the crystalline basement rock below the cap rock, and is trapped there by the 390m syenogranite seal above it. The host rock is strongly altered, hematitized, fault-brecciated crystalline rock with increasing vugginess toward depth — the classic IOCG-style hydrothermal alteration setting. The evidence: DDH-26-04's 24.3% peak at 707m, 104 of 284 samples ≥1%, increasing vugginess toward final depth, and the 776-779m core loss/void.

Layman's terms: The 3 pathways feed into fractured zone making free gas. Lots of it. This is the deep tank.

Working production scenario: 350–700 Mcf/d.
This is a speculative range rather than a reservoir-engineering forecast created using a CSIRO analogue study on comparable Precambrian granite as the baseline generation model (found at https://www.sciencedirect.com/science/article/pii/S0009254123003984). The current drilling data demonstrate repeated high-concentration H₂ and, in DDH-26-05, observations consistent with mobile/free gas entering the borehole, but they do not yet provide the permeability, reservoir pressure, effective producing thickness or pressure-drawdown data required to calculate sustainable well deliverability. The range is therefore intended as a scenario informed by geological evidence and industry analogues—not as a demonstrated flow rate.

Mechanism 2 — Dissolved gas exsolution from the artesian brine:
One possible interpretation is that the 617 m artesian water zone contains H₂ dissolved in formation brine, potentially generated through water-rock reactions involving the same iron-bearing minerals. As the pressurized brine rises toward surface and pressure drops, dissolved hydrogen exsolves. Hanley 2020's Type B CaCl₂ brine data (20-31 wt% salinity, 61 bar pressure, 60-80°C) gives 1,000-3,000 mL dissolved H₂ per litre of formation water. This mechanism was suggested qualitatively by the July 20 artesian overflow — water overflowing at the wellhead under its own pressure.
(paper used is Fluid Inclusion Systematics Associated with Epithermal Gold Mineralization, Eastern Cobequid Highlands, Nova Scotia by Hanley, accessible from https://novascotia.ca/natr/meb/pdf/20re02.asp)

Layman's terms: Old water has brine that reacts with iron. overpressured formation allows the brine can flow toward the surface; decreasing pressure allows hydrogen to exsolve from solution.

Prediction: Working production scenario: 200-450 Mcf/d.
This is a speculative range derived using the same Hanley paper. The current drilling data demonstrate the artesian water overflow at 617m, consistent with an overpressured brine system, but do not yet provide direct measurement of dissolved hydrogen concentration in the formation water, water flow rate, or the fraction of dissolved gas that would exsolve under production conditions

Mechanism 3 — Shallow free gas from the 164m shattered fault corridor:
This is gas intercepted in a completely different rock type — competent, thinly-bedded siltstone — at a dramatically shallower depth, in a different structural setting. The siltstone package is described by the QP as having "potential trap geometry," suggesting the gas is trapped locally within the siltstone rather than migrating through it. The cap is the 108m thick graphitic fault breccia above (35-143m), not the syenogranite. The evidence: DDH-26-05's free gas at 164-167m with no water return, ambient air detection, 23.5% peak at 170m, and the distinct "potential trap geometry" language.

Layman's terms: Free gas corridor is good. A shallower tank discovered above Mechansim 1.

Prediction: It’s hard to tell how much this will add. There are 2 scenarios:

Scenario A — Connected to the deep system (same pool, different entry points):
If the 164m zone and the 665-818m zone are draining the same connected reservoir, then Mechanism 3 adds essentially nothing to the aggregate flow estimate — it's the same gas, accessible from two different depths. You just have more options for which depth to complete a pilot well at. This is actually still good news (shallower = cheaper wells = better economics) but it doesn't increase the volume estimate.

Scenario B — Independent, bounded accumulation (separate pool):
If the siltstone package at 143-197m is a genuinely discrete trapped accumulation — sealed above by the graphitic fault breccia (35-143m), and below by something we haven't yet identified — then it's an independent reservoir contributing its own flow rate on top of the deep system. In this case, Mechanism 3 is genuinely additive. Estimating how much it adds is difficult without permeability data, but working from the concentration profile: the shallow zone has 23.5% peak H₂ in a 54m corridor — comparable concentration to the deep zone, but in a thinner, potentially less porous/permeable siltstone package. A reasonable, conservative estimate for an independent shallow zone contribution would be 200-300 Mcf/d.

Overall: If all three mechanisms prove to represent independently productive volumes, the combined scenario would be approximately 750–1,450 Mcf/d per well. For illustrative purposes, let's use 1,000 Mcf/d as a round-number production scenario. These are all meant to be approximate:

1 Mcf of H₂ ≈ 2.4 kg.
So: 1,000 Mcf/d × 2.4 kg ≈ 2,400 kg/day ≈ 2.4 tonnes/day of H₂.
2,400 kg/day × 365 days = 876,000 kg/year
So 1,000 Mcf/d of natural hydrogen ≈ 876,000 kg/year, or ≈ 876 tonnes/year of H₂.

All for one well. This can also scale up.

For reference, Mali generates 5-50 tonnes of hydrogen per year. This comparison is intended to illustrate the potential scale difference between a confirmed geological hydrogen system and the world's only current producing site, not to imply Bennett Hill will achieve this rate.

Are these numbers justified?
We will take a look at Pulsar Helium for this comparison. Note that Pulsar is looking for Helium, so a bit different, but similar settings.

Pulsar Helium's planned liquefaction plant is designed for ~4,000 Mcf/d of total field output with four wells (as far as I know), implying a target of ~1,000 Mcf/d per well. This is my inference from the plant design, not a demonstrated 1,000 Mcf/d per-well flow rate.

The geological similarities with Bennett Hill are similiar: both are rift-hosted, radiogenic Precambrian basement systems; both have a real confirmed cap rock; both show overpressured free gas and both have multiple confirmed gas-bearing wells. The key difference: QIMC has a couple of methods to generate Hydrogen: dry free gas, a separate artesian aqueous phase, and a shallow free gas phase, which are independent production mechanisms in the same structural corridor.

The Pulsar comparison suggests that a ~1,000+ Mcf/d per-well target is not inherently unreasonable as an industry analogue.

What would confirm these estimates?
The key missing data are reservoir pressure, permeability/transmissibility, effective producing thickness, pressure drawdown, and sustained flow testing. A pilot production test should provide the first direct measurement of Bennett Hill's actual deliverability. Until then, the numbers above should be viewed as scenario ranges rather than reserves, resources, or demonstrated production forecasts.

The biggest takeaway:
QIMC's Bennett Hill data support a model involving three proposed generation pathways and three potential production/accumulation mechanisms, with a free-gas phase. This is different from other companies in the industry, such as Gold Hydrogen and HyTerra who have been running into issues with their systems being primarily aqueous (look up flow rates at Ramsay-1, McCoy-1). Pilot testing will ultimately be required to determine the actual flow rates, but for now, investors should be pretty excited.

Bonus:
There is a chance that QIMC finds Helium-3. There was a paper from Dottin et al in 2025 (https://www.science.org/doi/10.1126/sciadv.adr2917)

Which specifically analyzed noble gas isotopes in rocks associated with the same mantle plume system that underlies the CCFZ and found evidence of mantle-derived helium — elevated ³He/⁴He ratios consistent with a mantle source contributing to the fault system's fluid budget. This is precisely the fingerprint that distinguishes mantle/primordial contributions from purely crustal/serpentinization ones. If the CCFZ has genuine mantle-fluid input (which Dottin's paper suggests), then the hydrogen at Bennett Hill could have both a serpentinization component (dominant, near-surface, the three Fe²⁺ oxidation pathways) AND a smaller mantle-degassing component (the primordial contribution, manifesting as anomalously high ³He/⁴He ratios and potentially contributing additional hydrogen from depth).

[Liquid helium density is ~0.125 kg/L. 940 L/hr × 0.125 kg/L = 117.5 kg/hr → 2,820 kg/day of liquid helium. Helium's molar volume at standard conditions gives roughly 5.6 m³ of gas per kg [At 0°C and 1 atm, 1 kg of helium occupies approximately 5.6 m³ as a gas] 2,820 kg/day × 5.6 m³/kg ≈ 15,790 m³/day of pure gaseous He ≈ 557 Mcf/ of pure helium. Using Pulsar's confirmed 14.5% He4 concentration: 557 Mcf/d ÷ 0.145 ≈ ~3,840 Mcf/d of total raw gas needed to feed this plant at its designed output. [This assumes ~100% helium recovery efficiency, which real plants don't achieve. Actual required raw gas intake is likely somewhat higher than 3,840 Mcf/d to account, say around 4,000 Mcf/d. Also no DST to confirm]

This was shared by a member of the investor community. This is theoretical and does not indicate any factual evidence QIMC is involved with companies like Microsoft.

The Opportunity: Data Centres Are Becoming an Energy Problem

Yesterday, the Government of Ontario released its Data Centre Playbook . One of the ideas it emphasizes is BYOP (Bring Your Own Power) essentially requiring a data centre operator to demonstrate that it has secured a dedicated power supply that does not strain the existing grid.

According to my research, Microsoft plans to build data centres in the GTA region of Ontario and the Quebec City region of Quebec. With QIMC’s properties concentrated near the Temiskaming region, the physical distance between these regions is obviously significant.
I also don't imagine Microsoft will build data centres in Northern Ontario toward the Temiskaming region, although that is obviously not confirmed (Microsoft is planning to build a large number of data centres)

[But there is an important possibility here: Ontario’s data centre approval process (including the September 12 consultation deadline) matters because if Ontario’s framework creates incentives for data centres to locate closer to clean power sources rather than simply contracting for power from distant sources, that could organically pull data centre development northward toward where QIMC’s hydrogen actually is.]

QIMC Does Not Necessarily Need to Sell Hydrogen to Microsoft:
If QIMC were to supply Microsoft directly with hydrogen, it would obviously be difficult. It could take years or decades to build the necessary pipeline infrastructure, while trucking hydrogen over long distances would be dangerous, inefficient, and expensive.

So I think the most efficient near-term solution is different:

The grid itself — with QIMC as a power seller rather than a hydrogen seller.
QIMC could sell electricity, converting its geological hydrogen into power on-site through fuel cells or turbines — including GE Vernova’s hydrogen turbines, which John shared on March 5. That electricity could then be fed into Ontario’s or Quebec’s existing grid, allowing the grid to physically deliver the power to Microsoft’s Toronto or Quebec City data centres.

The infrastructure is already there.
Under Ontario’s BYOP framework, Microsoft could theoretically contract for that generation as a dedicated clean energy supply, purchasing the output of specific QIMC generation units through a long-term Power Purchase Agreement (PPA), even if the actual electrons travel through the shared grid. This potentially turns QIMC’s geographic disadvantage into a different kind of advantage: QIMC does not need to physically move hydrogen to the data centre. It needs to convert the hydrogen into electricity where the hydrogen is produced and sell the resulting power into the grid.

What the QIMC-to-Microsoft Model could look like:
The process could be relatively straightforward.

1. QIMC drills and produces geological hydrogen:
   Rather than compressing, purifying, and shipping that hydrogen to a distant buyer, QIMC installs on-site generation — such as GE Vernova hydrogen turbines or fuel cells — at or near the wellhead.
   Those units convert the hydrogen into electricity locally.

2. The electricity enters the existing grid:
   [In Ontario] The electricity feeds into Ontario’s existing high-voltage grid at a nearby transmission interconnection point.
   This is where the existing infrastructure becomes particularly important.
   The key line is the Hydro One Northeast transmission system — a 230kV and 115kV network that connects Northern Ontario’s generation assets, historically hydro and now wind, southward toward the GTA load centre.

Kirkland Lake (adjacent to the Timiskaming corridor where QIMC is drilling) already has an existing grid connection point because it historically hosted gold mines with significant power demand. Ontario’s grid has significant spare transmission capacity on the north-south corridor specifically. Northern Ontario’s industrial and mining demand has declined over decades as mines closed or mechanized, while much of the transmission infrastructure built to serve those operations remains.
That means adding a new generation source in the Timiskaming region wouldn't necessarily require building entirely new transmission lines. It might be possible to use existing spare capacity. If so, that dramatically simplifies and accelerates the interconnection process compared with a greenfield transmission project.

3. Microsoft purchases the power through a PPA:
   Microsoft contracts to buy a specific quantity of the clean, Canadian-sourced power through a PPA, which is a long-term contract specifying price, volume, and term. The grid physically moves the electricity from north to south, while Microsoft books the associated clean-energy attributes (provenance certificates confirming that the power came from geological hydrogen and is zero-carbon at source) against its sustainability commitments. The important distinction is that QIMC does not need to deliver molecules of hydrogen to Microsoft. It can deliver electrons, which move at the speed of light.

Moving on, the economics could become significant. At Ontario’s current industrial electricity rate (roughly $80–100/MWh for contracted clean power, potentially higher under a premium Microsoft PPA for certified clean hydrogen power) the economics begin to become meaningful.
For example:
25 MW continuous × 8,760 hours/year × $90/MWh ≈ ~$19.7 million/year in electricity revenue.
And that is only 25 MW.
The real opportunity is the ability to scale generation as QIMC proves out its hydrogen resource and develops additional wells.

Two Technologies:
There are two primary ways QIMC could convert its hydrogen into electricity:

1. Fuel Cells:
   Fuel cells use electrochemical conversion.
   Hydrogen and oxygen combine directly to produce electricity, water, and heat, with no combustion at all. There are no moving parts in the core reaction.
   Bloom’s specification sheet already confirmed 52% electrical efficiency (LHV net AC) — meaning approximately 52% of the hydrogen’s energy content comes out as usable electricity, with the remainder available as heat. That heat can potentially be captured through waste-heat recovery, consistent with the Ontario Playbook’s own stated preference. There is also no CO₂ produced because there is no carbon in the fuel chain. The process does consume some parasitic power to operate pumps and controls, but this is already factored into the 52% net figure.
   A possible technology and commercialization pipeline could therefore look like:

Lambton → Altair Nanotechnologies, Ceres Power, Ballard Power Systems → Bloom Energy [for example]

2. Gas Turbines:
   Gas turbines use thermodynamic combustion. Hydrogen burns with oxygen to produce heat, which drives a turbine, which spins a generator.
   Electrical efficiency is somewhat lower than fuel cells at smaller scales — approximately 35–45%, depending on turbine size and whether it is combined-cycle — but the power output per unit is dramatically higher. That makes turbines much better suited to grid-scale injection. They also produce zero CO₂ because there is no carbon in the hydrogen. The only combustion byproduct is water vapour. If the waste heat is captured through combined heat and power (CHP), overall energy utilization can reach 70–90%, making it one of the most efficient power-generation methods available.
   A possible technology and commercialization pipeline could therefore look like:

Lambton → NovaLT → 7HA (full commercial development, utility-scale, 10–20 wells, Microsoft-scale demand) [for example]

So the turbine route is simpler at larger scale, while the fuel-cell route requires more surface processing but generates cleaner, higher-efficiency electricity at smaller scale.

The Lambton College H2-RE DCPS system fits naturally into this strategy as a smaller-scale version of the same logic.
The Lambton system (modular, hydrogen-fuelled, 15–25 kW per unit) is explicitly designed for edge and remote deployment.
That makes it potentially valuable as the first step in demonstrating the broader QIMC model:
Produce geological hydrogen → convert it to electricity locally → deploy that electricity where it is needed.
The scale simply changes.

The same idea also works in Nova Scotia.
The Maritime Electricity Cooperation MOU was designed specifically to enable cross-border clean-power flows between Nova Scotia, New Brunswick, Prince Edward Island, and eventually New England. That means a QIMC hydrogen-to-power facility at Bennett Hill could potentially sell into a market that extends well beyond Nova Scotia’s own modest grid. And this is where the GE Vernova connection becomes particularly interesting.

GE Vernova’s global headquarters is in Cambridge, Massachusetts — right next to Boston, directly in the Dracut/New England demand corridor that the M&NP pipeline terminates at and that the June 15 NR explicitly named as a target market, “encompassing markets in New England, New York, and beyond.”
In other words, the same geography that QIMC has identified as a potential hydrogen market is also home to GE Vernova’s global headquarters.
That is not necessarily a coincidence, but it is certainly a relationship worth exploring.

The Bigger Picture:
If GE Vernova is evaluating geological hydrogen as a future fuel for its turbine fleet, QIMC’s Bennett Hill project — described as the closest confirmed geological hydrogen source to New England and producing the cleanest gas in the sector becomes a potentially compelling candidate for a reference project.

QIMC does not have to become a hydrogen transportation company.
It does not necessarily need to build pipelines across provinces or truck hydrogen hundreds of kilometres to distant customers.
Instead, QIMC could become a clean-power producer located at the source of geological hydrogen.
The model is simple:
Geological hydrogen → local generation → electricity → existing grid → hyperscale demand.
Microsoft and other hyperscalers could ultimately become the customers.
And Bennett Hill could potentially open a second market through the Maritime grid and the Northeast U.S.
The strategic opportunity, therefore, may not be about figuring out how to get QIMC’s hydrogen to Microsoft.
It may be about making QIMC’s hydrogen into the power that Microsoft needs.

Obviously all speculative, formatted using AI because I'm too lazy to write it all out