From Doré to 99.99% Gold: What the Refining Process Costs

Gold leaves the ground looking nothing like the metal a central bank will accept. It arrives at a refinery as doré, a rough alloy of gold and silver streaked with base-metal contamination, closer to industrial waste than treasure. What it must become is a bar of 99.99% purity, and that gap is not just metallurgical.

It is the gap between ore and a tradeable financial instrument.

For anyone holding shares in gold producers, this matters because the value a miner captures per ounce is shaped directly by the purity grade it achieves and the refining terms it negotiates. Refining is not backstage plumbing. It sits inside the return calculation.

Here is what this covers: the specific steps that turn raw material into marketable metal, what each step costs, and why purity grade runs a direct line to the revenue a gold producer actually books. By the end, you will know where in the chain the real financial risk hides, and it is not where most investors look.

What raw gold actually looks like before refining begins

Doré is the starting point, and it is a problem before it is an asset. Gold pulled from a mine arrives at the refinery as this crude alloy of gold and silver, carrying variable amounts of copper, zinc, lead, and other base metals depending on the ore body it came from.

That variability is the whole issue. Doré from one mine looks nothing like doré from another, which is precisely why standardised refining is a precondition for entering the market rather than an optional polish. No two batches can be trusted on sight, so the industry built a purity gateway that every producer must pass through.

The history makes the scale of that gap clear. Standard trading bars were once produced simply by melting coins, yielding fineness ranges of roughly 899-901 and 915.5-917. That was considered acceptable in its era.

Compare that to where the market sits now:

• Coin-melt era: fineness of roughly 899 to 917, produced by melting existing coinage.
• LBMA Good Delivery minimum: 995.0 (99.5%), the wholesale trading floor set by the London Bullion Market Association.
• Shanghai Gold Exchange mandate: under the SGEB1-2019 standard, 99.99% is the only grade permitted for standard bars.

The commercial turning point arrived in 1979.

The LBMA Good Delivery Rules codify the minimum 995.0 fineness, physical dimension tolerances, and assay requirements that a bar must meet before any LBMA-accredited refiner can present it into the wholesale market.

The four-nines milestone The Royal Canadian Mint pioneered the four-nines standard commercially with its Gold Maple Leaf in 1979, setting the benchmark that institutional and investment buyers would come to expect as routine.

These tightening standards were not arbitrary thresholds handed down by regulators. They are the outcome of market demand hardening over decades, and today’s four-nines expectation reflects what institutional buyers now require. For you as an investor, that carries a direct implication: a miner that cannot meet the standard cannot reach institutional buyers at full spot price. Refining is a quality gateway, not a cost to shave.

The two-stage process that turns doré into 99.99% pure gold

Getting from doré to four-nines gold happens in two acts, and they are not interchangeable. The first stage does the heavy lifting chemically, fast and at scale. The second stage achieves precision electrically, slower and more exact. You need both because neither can do the other’s job.

Stage 1: Miller chlorination, the rough refining step

The Miller process is chemistry doing bulk work. Here is the sequence:

1. Impure doré is melted in a refractory crucible until fully molten.
2. Dry chlorine gas is injected below the surface of the molten metal.
3. Base metals such as copper, zinc, and lead, along with silver, preferentially form chloride compounds.
4. Those chlorides either volatilise off or rise as a skimmable slag that is removed.
5. The remaining metal is upgraded to approximately 99.5% purity.

The advantage here is speed and throughput. Chlorination is rapid and cheap at scale, which is why it is the industry-standard first stage. But it hits a ceiling. It cannot reach four-nines alone, and that limitation is what makes the second stage necessary.

Stage 2: Wohlwill electrolysis, the precision finishing step

Electrolysis handles the final purity increment. The mechanism runs like this:

1. The chlorination output is cast into anodes, the gold bars that will dissolve.
2. These anodes are suspended in an electrolyte bath of gold chloride and hydrochloric acid, alongside pure gold sheets acting as cathodes.
3. An electrical current is applied, causing gold to oxidise and dissolve into the solution.
4. The dissolved gold is selectively deposited onto the cathode as a dense, fine-grained layer of 99.99% pure gold.
5. Remaining base metals stay in solution or settle out as anode slime, a byproduct that is itself processed for recovery.

This stage is slower, but it delivers the four-nines standard that institutional markets demand.

The sequence tells you something practical. Reaching four-nines is an engineering choice with cost implications at each step, not a single operation. Producers and refiners negotiate over exactly these process terms when they set treatment and refining charges, and the cost profile changes sharply between the 99.5% and 99.99% thresholds. If you are evaluating a mining company’s offtake or refining agreement, that step-change is where refining economics meet revenue realisation.

What refining actually costs, and how those costs sit against the spot price

The cost data reframes the whole conversation. Refining fees are often assumed to be a drag on producer margins, but the numbers tell a different story once you separate the upstream bulk processing from the downstream per-ounce refining.

Note the split. Ore processing is measured per tonne of rock, running US$15-40 for standard oxidised ores and up to US$40-60 per tonne for refractory ores that need pre-oxidation. Final refining is measured per ounce of metal, a completely different unit, and conflating the two distorts the picture.

The cost split between per-tonne ore processing and per-ounce refining reflects a broader principle in mineral processing economics: value extraction happens in stages, and each stage carries its own cost structure, recovery rate, and margin exposure before refined metal ever reaches a buyer.

The headline figure is the last one. Combined smelting and refining costs of US$10-25 per ounce work out to roughly 0.5-0.7% of the gold’s market value at current spot prices. Mining companies often model refining at around US$2.65 per ounce in their project economics. Smaller-volume or non-institutional feed carries a commercial premium, with retail and semi-industrial rates running US$30-150 per troy ounce, but for institutional-scale producers, the refining process itself is not where margin disappears.

So where does the real exposure sit? In the places that do not show up in an investor presentation.

The least-known risk in gold refining Combined sampling and sample-preparation error can run to the order of plus or minus 5%. A 1 kg sample drawn from a 500 kg lot typically shows a 2-6% relative standard deviation under standard grab sampling practices.

Sampling and assaying is where payable-metal disputes originate, often with no malicious intent at all. On top of that, standard industry smelting and refining terms build in assumed metal losses of 0.05-0.2%. These contractual assumptions quietly shave the ounces a producer gets paid for.

The read you should take is this. The 0.5-0.7% refining cost is not the number to worry about. When you scrutinise a gold producer’s cost structure, the less-visible sampling risk and contractual loss terms are where payable-metal calculations actually move, and they rarely make it into the disclosures investors read most closely.

How purity, weight, and spot price combine to set a refined gold bar’s value

Now the pieces come together into a valuation model. The worth of any refined gold bar is not a single mysterious figure. It is the product of three variables working in sequence, and understanding how they stack lets you see where value is genuinely set and where it merely gets modified.

The three variables in the valuation formula

The valuation rests on three inputs:

• Purity-adjusted weight (fine ounces): the gross weight multiplied by the decimal fineness. A 1 kg bar at 99.99% fineness equals roughly 32.147 troy ounces of fine gold.
• Spot price: quoted per troy ounce of fine gold, which is what makes purity the bridge between physical weight and financial value.
• Purity premium: the modifier that four-nines gold carries over the 995.0 wholesale minimum.

The spot price is the dominant variable by a wide margin. It is set in wholesale markets driven by macroeconomic and investment forces, not by refining quality, and the purity premium sits on top of that baseline as a modifier rather than a driver.

LBMA spot price formation operates through a wholesale mechanism that sits entirely outside the refinery, aggregating bids and offers from institutional participants whose demand is driven by macro variables rather than by the physical purity of the bars being settled.

That premium is also less quantifiable than many assume. No explicit average premium for four-nines bars over 995.0 bars is widely published in current market data. The real advantage of four-nines is institutional acceptance and liquidity, not a fixed price differential. Buyers pay for the certainty that the bar clears any gate, not for a published surcharge.

Why macro demand now dominates spot price

The floor under gold prices has become structural, and the numbers behind that shift are substantial.

The scale of the market Total gold demand hit a record 4,974 tonnes in 2024, carrying a value of US$382 billion. Prices rose 27% across the year, driven by inflation concerns, geopolitical uncertainty, and falling real interest rates.

Central banks are now embedded in price formation. They have purchased over 1,000 tonnes of gold per year for three consecutive years to 2024, a persistent official-sector bid that puts a floor under the market. Sitting above that floor is the investment layer, with global gold ETF holdings growing by 801 tonnes in 2025.

Central bank gold demand has shifted from cyclical to structural over the past three years, with the persistent official-sector bid now functioning as a price floor that reshapes the revenue environment for every accredited producer, regardless of the refining quality it achieves.

World Gold Council demand data for 2025 shows central bank purchases reached 863 tonnes for the year, confirming that the persistent official-sector bid documented across prior years has remained a structural feature of the market rather than a cyclical episode.

What this tells you is important for how you frame gold price risk in a portfolio. The value a producer captures for its refined output is being set by institutional forces well beyond the refinery gate. Purity grade matters at the margin. The macro demand structure matters far more.

Where quality failures and ethical risks can erase the value of refined gold entirely

Compliance is not box-ticking in this market. It is an existential commercial variable, and a refinery’s LBMA Good Delivery accreditation is itself a financial asset that can be destroyed.

Accreditation is the key that opens central-bank flows and major institutional buyers. Lose it, and the consequence is not a warning letter.

The compliance fact that matters most A refinery removed from the LBMA Good Delivery List is barred from reapplying for a minimum of five years. This is not a temporary inconvenience. It is medium-term exclusion from the institutional market.

The recent suspension cases show what that exclusion looks like in practice:

1. 2021: Kyrgyzaltyn JSC of Kyrgyzstan was suspended over failures to meet responsible-sourcing standards amid fraud concerns.
2. 2022: Six Russian precious-metals refineries were suspended following geopolitical sanctions.
3. 2026: Shandong Gold Smelting Co. was suspended after its addition to the U.S. Uyghur Forced Labor Prevention Act entity list over supply chain concerns.

A December 2022 lawsuit regarding gold from the North Mara mine in Tanzania further illustrates the continuous scrutiny facing accredited refiners on human rights and supply chain transparency. The pressure is not easing.

It helps to separate the two categories of risk a refinery carries:

• Technical risks: sampling and assaying error of up to plus or minus 5%, and contractual metal loss assumptions of 0.05-0.2% built into standard terms.
• Ethical and governance risks: responsible-sourcing failures, sanctions exposure, and forced-labour entity listings that trigger suspension.

Here is why this reaches you as an investor. The five-year reapplication bar means accreditation is a permanent commercial asset, not a renewable certificate, and its loss is medium-term market exclusion. A gold producer whose metal is processed by a suspended refiner faces immediate disruption to its offtake chain, and by extension to its revenue. Refinery accreditation status belongs on your due-diligence list.

What the refining chain tells you about a gold producer’s financial position

Pull the chain together and it becomes an investor framework, not a chemistry lesson. From doré to four-nines bar, each link carries a specific financial signal you can actually use.

Run through the checklist:

• Purity grade and market access: four-nines is the threshold that opens institutional buyers at full spot price. A producer that cannot meet it is locked out of premium demand.
• Refining cost and contractual terms: the refining fee itself is small, at 0.5-0.7% of spot value, but the sampling error and assumed metal loss terms are where payable ounces genuinely move.
• Refiner LBMA accreditation status: with a minimum five-year bar on reapplication after suspension, the accreditation of a producer’s chosen refiner is a strategic variable, not an afterthought.

The spot price, the dominant value driver, is shaped by macro forces no single producer controls. That is exactly why the controllable variables deserve your attention: refining quality, refiner accreditation, and contractual metal loss terms are the levers that separate producers within the same price environment.

There is a forward-looking angle here too. With central banks buying over 1,000 tonnes a year for three consecutive years to 2024, structural demand keeps supporting spot prices, and the purity threshold to access that demand is likely to stay at four-nines or above. Access itself is gated: only 66 LBMA-accredited refineries handle approved institutional flows globally, so a producer’s relationship with one of them is a strategic asset worth scrutinising.

Knowing where to look in a company’s disclosures is worth more than knowing the chemistry.

For investors ready to apply the refining framework to stock selection decisions, our dedicated guide to evaluating gold mining stocks covers margin analysis, cost-structure benchmarking, and the producer-specific variables that separate strong performers from the field when spot prices are elevated.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors.