Gold, Silver, and Copper in Data Centers: The Real Metal Ratio Behind the Digital World

By

 If you’ve ever walked into a data center (or even just stood near one), you know the vibe.

It’s not “server room.” It’s more like “industrial-grade aquarium,” except the fish are fans, the water is chilled air, and the bubbling sound is your electric meter hyperventilating.

And here’s the part most people don’t think about: a data center isn’t just software in a building. It’s a giant, physical metal-moving project.

Not rare earths. Not some sci-fi mineral nobody can pronounce.

The big three that show up in the conversation again and again are:

  • Copper (the heavyweight)

  • Silver (the specialist)

  • Gold (the tiny but essential diva)

So let’s talk about the “ratio” question—how much gold, silver, and copper go into building a data center—and then zoom out: production volumes, why mining is hard, and why prices act the way they do.

No hype. No “buy this now.” Just the mechanics.


Infographic showing the approximate material ratio used in data center construction: copper dominates by mass for power and wiring, while silver and gold appear in much smaller amounts mainly in electronic contacts and connectors, with a comparison of global production volumes and mining difficulty.

First: What counts as “in a data center”?

When people ask about metals “used in data center construction,” they often mix three different buckets:

  1. The building + power delivery
    Think switchgear, transformers, busbars, power cables, grounding, backup systems, etc.

  2. Cooling + mechanical
    Lots of steel, aluminum, fluids, and—yes—more copper in motors, wiring, and heat exchange hardware.

  3. IT hardware
    Servers, GPUs, storage, networking gear, racks, and all the connectors inside them.

Here’s the key: Copper lives mostly in bucket #1 and #2.
Gold and silver live mostly in bucket #3 (and they show up there in tiny amounts per device).

That’s why any clean “ratio” is slippery. The answer changes depending on whether you mean:

  • just the IT equipment, or

  • the entire facility (everything from utility feed to rack)

Most people mean the whole thing—so I’ll aim there.

Copper: the real boss of the room

Copper is the main event because data centers are basically power distribution factories that happen to compute.

How much copper per MW?

You’ll see wildly different numbers online, because people accidentally compare apples to… busbars.

  • Nvidia (in a data center power discussion) referenced 200 kg copper busbars to feed a single rack, in the context of shifting to higher-voltage DC designs. That’s one component in a rack power path, not the whole building.

  • Other discussions point to facility-level copper per MW being far higher—measured in tens of tonnes per MW in some cases (cabling, power distribution, transformers, etc.).

So what’s a reasonable “whole facility” working range?

A practical way to phrase it (without pretending we have a single perfect number) is:

Copper in a data center is measured in tonnes per MW, not kilograms per MW, once you count the full electrical and mechanical buildout.

Even if you argue about the exact coefficient, the directional truth is solid: copper dominates by mass.

Why copper adds up so fast

Copper shows up everywhere that electricity needs to move reliably:

  • utility interconnect and onsite distribution

  • switchgear and transformers

  • busways, busbars, grounding

  • power to cooling equipment and pumps

  • rack power delivery

  • and sometimes even design choices like redundancy levels (N+1, 2N, etc.) push copper usage up

And because data centers are dense loads, a lot of copper gets concentrated in a small footprint.

That’s why serious research groups are calling copper a critical enabler for the AI/data-center buildout.

Gold and silver: tiny by weight, huge by importance

If copper is the highway system, gold and silver are the little precision parts inside the engine.

What gold is doing in there

Gold is used because it:

  • conducts well

  • doesn’t corrode

  • holds up in high-reliability contacts and connectors

It’s not used as “bulk metal.” It’s used as thin plating—sometimes measured in microinches on connector fingers.

So the mass is tiny, but it’s in the exact place you don’t want oxidation or intermittent contact.

Rough gold amounts in electronics (for intuition)

Estimates vary a lot by device class and quality tier, but for a motherboard-scale reference:

  • “Typical” motherboard gold content often gets cited around ~0.2 to 0.5 grams, with high-end server boards potentially higher.

  • Another recycler-oriented reference pegs a standard consumer motherboard around ~0.3 grams.

Are these perfect? No. They’re estimates, and the real number depends on connector count, plating thickness, and design.

But they’re good enough to understand the order of magnitude: fractions of a gram per board, not grams and grams.

What silver is doing in there

Silver is used where you want:

  • excellent conductivity

  • reliable contact surfaces

  • certain electronic components and coatings

Industry groups are explicitly tying silver demand growth to technology sectors, including data centers.

Silver usage per device is also usually measured in fractions of a gram to grams, depending on category. For example, smartphone estimates commonly land in the sub-gram range.

Servers and networking gear can have higher loadings than phones simply because they’re bigger, more complex, and built for reliability—but again, we’re not talking kilograms.

So what’s the “ratio” of copper vs silver vs gold?

Here’s the most honest answer:

By mass, copper overwhelms everything. Gold and silver are “spice,” not “protein.”

A clean way to express the ratio

Instead of pretending we can give one universal recipe, I like giving a “rule of thumb” ratio range for the whole facility:

  • Copper: tonnes per MW (often tens of tonnes per MW when counting full buildout)

  • Silver: typically kilograms per MW or less when aggregated across IT gear (depends heavily on server count, type, and refresh cycle)

  • Gold: typically hundreds of grams per MW or less when aggregated across IT gear (again depends heavily on hardware mix)

So if you force me to translate that into a “ratio vibe”:

  • Copper : Silver : Gold ≈ 1,000,000 : 10–100 : 1 (order-of-magnitude thinking)

Meaning: copper is often millions of times heavier than gold, and tens of thousands to hundreds of thousands of times heavier than silver, depending on how you count the facility and the gear.

Not because gold/silver aren’t used—because they’re used as thin, precise material in small locations.

Why you see bad numbers online

Two main reasons:

  1. People quote a number about one component (like busbars feeding a rack) and label it as “data center copper per MW.”

  2. People confuse IT power (MW of compute load) with total facility buildout materials, which includes big electrical infrastructure that scales with redundancy and cooling choices.

Production volumes: how much the world makes

Now zoom out. If demand grows, the next question is always: “Okay, can supply keep up?”

Let’s ground that with global mine production estimates from USGS Mineral Commodity Summaries (2024 edition, reporting 2023 estimates):

  • Copper (world mine production, 2023e): ~22,000 thousand metric tons (≈ 22 million tonnes)

  • Gold (world mine production, 2023e): ~3,000 metric tons

  • Silver (world mine production, 2023e): ~26,000 metric tons

Already you can see the scale mismatch:

  • Copper is produced in tens of millions of tonnes

  • Silver in tens of thousands of tonnes

  • Gold in a few thousand tonnes

That’s why copper feels “industrial” and gold feels “scarce,” even before prices enter the chat.

Why mining is hard (and different for each metal)

Copper mining: big, slow, capital-heavy

Copper is abundant-ish in the Earth’s crust, but the challenge is grade and scale.

A lot of copper comes from huge deposits where the rock contains a small percentage of copper. That means:

  • gigantic material moved

  • big energy inputs

  • long permitting timelines

  • huge capex for mines, mills, smelters, refineries

Even the USGS points out that copper resources are massive, but translating that into supply is a long, real-world process.

Copper also has a helpful pressure-release valve: recycling. Scrap contributes meaningfully to supply in many markets.

But recycling doesn’t magically solve everything, because demand growth can outrun scrap availability—especially when the “in use” copper is sitting inside buildings and infrastructure for decades.

Gold mining: high value, but not “easy”

Gold has insane value density, but it’s not like you just scoop it up.

Challenges:

  • ore grades vary widely

  • chemistry and recovery processes matter

  • geopolitical and permitting risk is real

  • and a lot of “new supply” is really existing mines extending life via more drilling and development

USGS notes global gold mine production was roughly flat around the 2022–2023 window, which is a classic signal that adding meaningful supply is not trivial.

Also interesting: USGS breaks out gold demand categories and lists electronics as about 5% of global gold consumption (excluding ETFs and similar).
So data centers matter for gold—but gold demand is still dominated by other uses.

Silver mining: the “byproduct” problem

Silver has a special twist: a lot of silver is not mined because a company wakes up and says “Let’s mine silver.”

USGS explicitly notes that silver is often obtained as a byproduct, especially from lead-zinc, copper, and gold mines.

This matters because it means:

  • silver supply can be tied to economics of other metals

  • even if silver demand rises, supply doesn’t always respond cleanly

  • mines don’t flip a switch and become “more silver productive” unless the host metal mining expands or processing changes

That’s why silver markets can feel… moody.

How prices get formed (without the mystical vibes)

Prices are basically the end result of three messy forces:

1) Marginal cost vs marginal need

  • For copper, the price tends to be disciplined by the cost of bringing on new supply (new mines, expansions, smelting capacity) and the cost curve of existing producers.

  • For gold, “cost” matters, but demand is heavily influenced by jewelry, central bank activity, and investor behavior. Electronics is real but not the majority slice.

  • For silver, because supply is often byproduct-driven, price can detach from a simple “silver mine cost” narrative.

2) Time (the most underrated ingredient)

Even if high prices scream “more supply,” mines take years to permit and build. That time delay makes commodities prone to cycles.

Copper is famous for this: demand moves faster than the mine pipeline.

3) Narrative compression

Markets love a headline that compresses complexity into one scary number.

Example: claims about gigantic copper tonnage per AI data center can spread fast—sometimes with math errors—because the narrative is clean and emotional.

The truth is more boring and more useful:

  • copper demand is growing from electrification and data centers

  • but the exact tonnage depends on design choices

  • and the world copper system is huge, so you have to compare “one project” to annual production to keep perspective

A practical “back-of-the-envelope” you can quote in a blog

If you want a simple line that won’t get you roasted too badly:

  • Copper dominates by mass. It’s the wiring, power delivery, and heavy electrical backbone. Expect tonnes per MW once you count full facility buildout.

  • Silver and gold are tiny by weight, used mainly in electronics contacts, coatings, and connectors, often as thin plating layers.

  • Global annual mine production is roughly ~22 million tonnes copper, ~26,000 tonnes silver, ~3,000 tonnes gold (2023 estimates).

If you want a one-liner ratio vibe for the whole facility:

Think “mountains of copper, handfuls of silver, and a few coffee spoons of gold,” per MW-scale buildout.

That’s not poetry. That’s just the physics of where each metal is actually used.

The fun twist: design changes can change the copper math

One reason copper estimates bounce around is that electrical designs are evolving.

Higher-voltage approaches (including Nvidia’s commentary on moving toward 800V DC in the rack ecosystem) are explicitly framed as ways to transmit more power with less copper in certain parts of the chain.

So yes—data center copper is big.
But also yes—engineers are actively trying to not use more copper than they have to.

Because copper is expensive, heavy, and annoying to route. (If you’ve ever wrestled thick cable through a tray, you know.)

Wrap-up: the ratio question, answered like a human

If your goal is to talk about this in a way that feels grounded:

  1. Copper is the dominant metal by mass in data center buildouts—especially once you count the facility-level electrical system.

  2. Gold and silver matter because they sit at failure points (contacts and connectors), not because they show up in bulk.

  3. Supply realities differ:

    • copper is huge-scale, long lead-time, capex-heavy

    • gold is high-value and multi-demand-driven; electronics is a smaller slice

    • silver is often byproduct supply, so it can be “sticky”


Popular posts from this blog

Bond ETF Structures Explained: Government Bonds vs Corporate Bonds vs High-Yield Bonds

By
Image
Introduction  For long-term investors, bond ETFs are often misunderstood. They are frequently described using simplified labels— safe , boring , income-focused , or defensive . But these labels hide the most important reality: bond ETFs are not a single asset class . They are a collection of structurally different instruments that behave very differently across market cycles. Government bond ETFs, corporate bond ETFs, and high-yield bond ETFs may all contain “bonds,” but they respond to interest rates, economic growth, inflation, and risk sentiment in fundamentally different ways. Treating them as interchangeable can lead to unintended portfolio behavior, especially over long investment horizons. This matters because, for long-term investors, bonds are rarely about maximizing returns. Instead, they serve structural roles : Stabilizing portfolio volatility Offsetting equity drawdowns Providing liquidity during stress periods Managing duration and interest rate expo...
Read more

Tesla’s Weight in Major ETFs: What Most Investors Don’t Realize

By
Image
  Introduction: Tesla Is Already in Your Portfolio — Whether You Like It or Not Most long-term investors believe they have made a clear choice. Some say they invest in ETFs to avoid the risks of individual stocks. Others say they avoid Tesla because it is too volatile, too narrative-driven, or too controversial. Yet for a large portion of U.S. equity investors, this distinction is mostly theoretical. Without buying a single share of Tesla , many investors already hold meaningful Tesla exposure—embedded quietly inside the most popular ETFs in the world. This article explains where Tesla sits inside major ETFs, how its weight actually works, and why most investors misunderstand the nature of their exposure . Not to judge it. Not to recommend action. But to clarify the structure. 1. Tesla Is Not a “Side Holding” Inside ETFs Tesla is often described as just one company among hundreds inside large ETFs. Technically, that is true. Structurally, it is misleading. In capita...
Read more

Why U.S. Long-Term Investors Eventually Look Beyond the U.S.: A Structural Case for International ETFs

By
Image
  Introduction: When “U.S.-Only” Stops Being Enough For decades, U.S. equities have dominated global markets. The depth of the American capital market, the strength of the dollar, and the global reach of U.S. corporations have allowed many long-term investors to build portfolios that are almost entirely domestic. However, as portfolios mature and asset sizes grow, many long-term U.S. investors eventually begin to look abroad—not because the U.S. market has “failed,” but because concentration itself becomes a structural risk . This shift is not driven by short-term performance chasing or macro forecasts. Instead, it emerges from long-term portfolio construction logic: currency exposure, valuation dispersion, sector balance, and geopolitical diversification . This article explores why international ETFs often enter long-term portfolios—not as a replacement for U.S. equities, but as a structural complement. 1. The Hidden Concentration Inside “U.S.-Only” Portfolios At first gl...
Read more