Read our COVID-19 research and news.

The Homestake gold mine, which produced ore from 1876 until 2002, was the largest and deepest gold mine in North America.

Matthew Kapust

Excavation starts for U.S. particle physicists’ next giant experiment

Today, physicists and politicians gathered at a former mine in South Dakota to break ground for the United States’s next great particle physics experiment. Known as the Long-Baseline Neutrino Facility (LBNF), the experiment will fire a beam of elusive particles called neutrinos from Fermi National Accelerator Laboratory (Fermilab) in Batavia, Illinois, to a gargantuan particle detector 1300 kilometers away in an abandoned gold mine in Lead, South Dakota.

To build the modular detector, workers have to carve out massive caverns 1480 meters underground, haul out stone that weighs as much as a dozen aircraft carries, and truck in millions of liters of frigid liquid argon. This afternoon officials gathered deep underground to turn the first few shovels of stone.

“We couldn’t be more excited to be actually starting construction,” says Mike Headley, head of the South Dakota Science and Technology Authority in Lead. “We’re absolutely thrilled that [the project] is moving forward and about what it’s going to do for the U.S. scientifically,” says Headley, who is also director of the Sanford Underground Research Facility (SURF), the small laboratory the state started at Homestake in 2006 with a $70 million donation from philanthropist T. Denny Sanford.

Perhaps the most mysterious of subatomic particles, neutrinos outnumber every other type of matter particle, yet interact so weakly that every second trillions of them pass unnoticed through each of us. They come in three types—electron, muon, and tau—that can morph, or oscillate, into one another. By firing muon neutrinos from particle accelerators to distant underground detectors, physicists have sketched out the basics of such “neutrino oscillations.” The LBNF aims to nail down all the details and put physicists’ theory of the phenomenon to the acid test. It will also look for a slight asymmetry between neutrinos and antineutrinos that could be key to explaining how the infant universe generated so much more matter than antimatter and search for signs of even weirder new physics.

The four caverns housing the neutrino detector will lie on either side of a shallower cavern housing utilities.

LBNF project

Scientists proposed building such an experiment as early as 2001 as part of a bigger, multipurpose lab at Homestake to be funded by the National Science Foundation (NSF). However, in 2010 the National Science Board, which sets policy for NSF, balked at that idea, leaving it to the Department of Energy (DOE) to build the neutrino experiment. The size and scope of the effort then oscillated up and down, as physicists and DOE officials haggled over what the department could afford. In 2014 they agreed to restore the experiment to its original scope and make it an international project. DOE now anticipates covering $1.5 billion of the total cost. The detector itself—now known independently as the Deep Underground Neutrino Experiment (DUNE)—will comprise four massive tanks of ultrapure liquid argon.

Big dig

But before scientists can build the experiment, engineers and workers must greatly expand the SURF. They will have to blast out four chambers measuring roughly 70 meters long, 20 meters wide, and 29 meters high, as well as a longer, lower service hall. They will extract about 790,000 metric tons of rock, says Tracy Lundin, a civil engineer at Fermilab and LBNF project manager for conventional facilities. Batch by batch, the material will go up the mine’s existing rock-hauling elevator or “skip,” Lundin says, and will then be carried by a 1200-meter-long conveyor belt and deposited in the hole left from a prior surface mining operation. The excavation should take about 3 years, he says.

In spite of the mind-numbing numbers, such excavation isn’t unprecedented, Lundin says. “In the underground construction business, it’s a medium to medium-large project," he says. Engineers and workers also have caught a break, Lundin says, in that the residual stresses in the rock are modest, reducing the risk that it will fracture or shift during excavation. “This mine has been known as a very stress-friendly rock mass,” he says.

Once workers have excavated the caverns, they will have to build the steel tanks that will hold the liquid argon. Engineers are borrowing a technology called membrane cryostat technology that is now used in tanker ships that carry liquid natural gas, says David Montanari, an engineer at Fermilab and LBNF project manager for cryogenic infrastructure. The inner liner of the tank will consist of a thin layer of corrugated steel. That will be surrounded by a thick layer of insulation and an outer steel support structure, Montanari says. “We are trying not to reinvent the wheel,” he says. “We are trying to use as much as possible conventional technology.”

The location of the tanks deep underground may make them somewhat harder to build than a tanker ship. “There’s a maximum size for a piece of steel that can go down the mine shaft,” says Eric James, a physicist at Fermilab and DUNE technical coordinator. “So you have to design things so that they can be bolted together piece by piece and still withstand the pressure applied by all the liquid argon.”

Thousands of trucks of argon

Getting the liquid argon to the lab will also require some thought. “There’s a big process to figure out how we get that much liquid argon out to South Dakota when all of the suppliers are on the Gulf Coast and the East Coast,” says Troy Lark, a mechanical engineer at Fermilab and procurement manager for LBNF. Argon is shipped as a liquid by truck, one of which typically hauls about 20 tons. So filling the detector will require 3500 deliveries.

Ironically, once the liquid reaches the SURF, workers will convert it to a gas before sending it underground. Piping the liquid down isn’t practical because of the colossal pressure that would build up at the bottom of the 1480 meter pipe, Lark says. So the argon must go down as a gas and be recondensed—at a chilly –185.8°C—once it reaches the lab. Filling each of the four modules will take between 7 months and a year, Montanari estimates.

The civil construction and infrastructure work at Homestake are pricy. The DOE budget request for fiscal year 2018 pegs the cost of those two things at a total of $398 million.

Physicists do not yet have approval from DOE to build the high-tech guts of the DUNE detector. Researchers at Fermilab and CERN, the European particle physics laboratory near Geneva, Switzerland, are still working with prototype detectors and ironing out their designs, James says. They hope to get final approval to start building detector hardware itself in 2019, in hopes of completing construction of the detectors in 2024, he says. The United States will cover about 25% of the costs of the detector, James says.

But first, physicists must have the caverns and facilities in which to place those high-tech devices. Today, the engineers finally started building them.

Correction, 31 July, 10:23 a.m.: The story has been changed: The amount of rock to be excavated has been corrected.