The proposed Power of Siberia 2 pipeline is a roughly 2,600-kilometer corridor designed to carry West Siberian gas through eastern Mongolia into northern China, at a capacity of up to 50 billion cubic meters per year. Negotiations between Gazprom and China National Petroleum Corporation have produced binding memoranda, then further uncertainty, then more memoranda. The pipeline does not yet exist and may not for years.

And yet, in Beijing’s 15th five-year plan, between provisions for new-energy bases and power transmission corridors, the state has authorized “preliminary work” on what officials dub the China-Russia Central Line. “Preliminary work,” in the language of Chinese planning, is a technology of commitment, authorizing feasibility studies, coordinating interagency expectations, and, critically, creating the anticipation of sunk costs.

Cold War history provides an analogy.

The pipeline has a connection with semiconductor fabrication, although its mechanism is diffuse and ecological. A chip is made inside a system that runs on electricity, nitrogen, hydrogen, ultra-pure water, and climate control so exacting that a brief power disruption can scrap in-process wafers worth millions of dollars. Taiwan Semiconductor Manufacturing Company consumed 27,456 gigawatt-hours of electricity in 2024, roughly the annual power consumption of Connecticut. Natural gas accounted for less than 7% of that total. Electricity was everything, and electricity in northern China is produced partly by gas-fired plants that require continuous fuel supply.

The more interesting pathway runs through industrial gases. A modern fab consumes nitrogen and hydrogen at a scale that strains the imagination: tens of thousands of standard cubic meters of nitrogen per hour, used for inerting, purging, and deposition, and hundreds of standard cubic meters of hydrogen for annealing and epitaxial processes. Much of this hydrogen is produced from natural gas via steam methane reforming. Any shift in the economics or security of natural gas supply therefore propagates into the economics of hydrogen, and from there into the supply chains that sit beneath the clean-room floor. The pipeline is an upstream condition for chip-making, which explains what the official planning documents are actually doing.

Beijing understands the relationship. The 15th five-year plan is notable for placing natural gas pipeline networks and integrated circuits in the same national blueprint. The plan calls for improvement of mature fabrication nodes, advanced process capability, key equipment, and what it describes as “full-chain breakthroughs” achieved through “unconventional measures.” The phrase “unconventional measures” has the quality of bureaucratic candor: it acknowledges that the ordinary levers are insufficient. The “full-chain” framing treats the chip problem as a system vulnerability, where weakness anywhere in the chain, including in the mundane substrate industries that supply gases and chemicals and ultra-pure water, becomes a strategic exposure.

Back to the future

Cold War history provides an analogy. A declassified CIA intelligence estimate from 1982 examined the Soviet Siberia-to-Western Europe pipeline with the dry alarm that characterized Cold War strategic assessment. It noted that large pipeline projects tie together technology transfer, credit, markets, and long-run dependence in ways that create political dilemmas for everyone involved. The buyer gains energy security and loses leverage. The seller gains hard currency and loses flexibility. The pipeline, once built, becomes what analysts call a frozen option: a capital commitment so large that it biases future policy — abandoning sunk costs is politically difficult, and constituencies form around infrastructure.

RELATED: Russia’s and China’s superweapons are stunning the world. The US is struggling to catch up.

GREG BAKER/AFP/Getty Images

Nord Stream 2 carried 55 billion cubic meters per year when it was operating. Power of Siberia 2, at 50 billion cubic meters, is built to similar scale. The comparison is not reassuring to anyone, including China’s planners, who understand that a second large Russian pipeline would increase import concentration even as it reduces seaborne vulnerability. This is the paradox embedded in the corridor logic: The project that insulates itself from one chokepoint exposes itself to another.

An extended energy shock around the Strait of Hormuz, of the kind that analysts are tracking in 2026, makes overland pipelines look like strategic wisdom. A geopolitical rupture or rivalry with Russia would make the same pipeline look like a trap. China’s negotiators have read this history. Their unusual patience in signing on, their expansion of LNG capacity in parallel, their insistence on pricing terms that Russia finds inadequate, all reflect the recognition that the pipeline’s value as an unbuilt corridor may exceed its value as a built one. China wants optionality as well as leverage.

More energy, more chips

The binding constraint on China’s most advanced semiconductor fabrication is not electricity or nitrogen or hydrogen but extreme ultraviolet lithography and the specialized manufacturing equipment and intellectual property that surrounds it, as well as the export controls that the United States has used since 2022 to restrict Chinese access to the frontier tooling. A stable gas supply does not yield an EUV machine. The pipeline’s effects are on the ecology of scaling, not on the cutting edge, where the competition is most intense and the gap remains most visible.

What the pipeline can do is lower the infrastructure risk premium that makes certain chipmaking clusters too fragile to sustain. Imagine a provincial government courting a 28-nanometer foundry, a packaging campus, and several industrial-gas suppliers. The limiting questions in that negotiation are often quiet ones: Can the local grid guarantee continuous power? Can industrial gases be delivered without interruption? Can the region meet environmental compliance requirements without shutting down plants during winter pollution campaigns? A new trunkline does not answer these questions but shifts the feasible responses. It allows planners to make commitments that would otherwise require hedges, and hedges in industrial policy tend to become failures.

The plan to advance “preliminary work” on the Central Line is a political commitment embedded in security thinking, industrial strategy, and the institutional planning routines of a state that treats external dependence as a vulnerability to be managed by building redundancy and domestic capacity simultaneously. Chips increase the value of energy security. Energy security increases the feasibility of chip scaling. The state that grasps this feedback loop before its competitors will have done something more durable than winning a trade dispute. It will have changed the conditions under which the next dispute is conducted. Such change may take decades to become visible, and “preliminary work” is how it begins.

​Tech, Return, China, Russia, Power of siberia pipeline 2, Semiconductor