Knocking on Heaven's Door: How Physics and Scientific Thinking Illuminate the Universe and the Modern World - Lisa Randall (2011)



I am not one prone to overstatement, since I usually find that great events or achievements speak for themselves. This reluctance to embellish can get me into trouble in America, where people overuse superlatives so much that mere praise without an “est” at the end is sometimes misinterpreted as slander by faint praise. I’m frequently encouraged to add a few buzzwords or adverbs to my statements of support to avoid any misunderstanding. But in the case of the LHC I’ll go out on a limb and say there is no question that it’s a stupendous achievement. The LHC has an uncanny authority and beauty. The technology overwhelms.

In this chapter, we’ll embark on our exploration of this incredible machine. In the chapter that follows, we’ll enter the roller coaster construction adventure and a few chapters later, the world of the experiments that record what the LHC creates. But for the time being, we’ll focus on the machine itself, which isolates, accelerates, and collides together the energetic protons that we hope will reveal new inner worlds.


The first time I visited the LHC, I was surprised at the sense of awe it inspired—this in spite of my having visited particle colliders and detectors many times before. Its scale was simply different. We entered, put on our helmets, walked down into and through the LHC tunnel, stopped at an enormous pit into which the ATLAS (A Toroidal LHC ApparatuS) detector would ultimately be lowered, and finally arrived at the experimental apparatus itself. It was still under construction, which meant ATLAS was not yet covered up as it would be when running—but was instead on display in full view.

Although the scientist in me recoils at first in thinking of this incredibly precise technological miracle as an art project—even a major one—I couldn’t help taking out my camera and snapping away. The complexity, coherence, and magnitude, as well as the crisscrossing lines and colors, are hard to convey in words. The impression is simply awe-inspiring.

People from the art world have had similar reactions. When the art collector Francesca von Habsburg toured the site, she took along a professional photographer whose pictures were so beautiful they were published in the magazine Vanity Fair. When the filmmaker Jesse Dylan, who grew up in a world of culture, first visited the LHC, he viewed it as a remarkable art project—a “culminating achievement” whose beauty he wanted to share. Jesse embarked on a video to convey his impressions of the grandeur of the experiments and the machine.

The actor and science enthusiast Alan Alda, when moderating a panel about the LHC, likened it to one of the wonders of the ancient world. The physicist David Gross compared it to the pyramids. The engineer and entrepreneur Elon Musk—who cofounded PayPal, runs Tesla (the company that makes electric cars), and developed and operates SpaceX (which constructs rockets that will deliver machinery and products to the International Space Station)—said about the LHC, “Definitely one of humanity’s greatest achievements.”

I’ve heard such statements from people in all walks of life. The Internet, fast cars, green energy, and space travel are among the most exciting and active areas of applied research today. But going out and trying to understand the fundamental laws of the universe is in a category by itself that astounds and impresses. Art lovers and scientists alike want to understand the world and decipher its origins. You might debate the nature of humanity’s greatest achievement, but I don’t think anyone would question that one of the most remarkable things we do is to contemplate and investigate what lies beyond the easily accessible. Humans alone take on this challenge.

The collisions we’ll study at the LHC are akin to those that took place in the first trillionth of a millisecond after the Big Bang. They will teach us about small distances and about the nature of matter and forces at this very early time. You might think of the Large Hadron Collider as a super-microscope that allows us to study particles and forces at incredibly small sizes—on the order of a tenth of a thousandth of a trillionth of a millimeter.

The LHC achieves these tiny probes by creating higher energy particle collisions than ever before achieved on Earth—up to seven times the energy of the highest existing collider, the Tevatron in Batavia, Illinois. As explained in Chapter 6, quantum mechanics and its use of waves tells us these energies are essential for investigating such small distances. And—along with the increase in energy—the intensity will be 50 times higher than at the Tevatron, making discovering the rare events that could reveal nature’s inner workings that much more likely.

Despite my resistance to hyperbole, the LHC belongs to a world that can only be described with superlatives. It is not merely large: the LHC is the biggest machine ever built. It is not merely cold: the 1.9 kelvin (1.9 degrees Celsius above absolute zero) temperature necessary for the LHC’s superconducting magnets to operate is the coldest extended region that we know of in the universe—even colder than outer space. The magnetic field is not merely big: the superconducting dipole magnets generating a magnetic field more than 100,000 times stronger than the Earth’s are the strongest magnets in industrial production ever made.

And the extremes don’t end there. The vacuum inside the proton-containing tubes, a 10 trillionth of an atmosphere, is the most complete vacuum over the largest region ever produced. The energy of the collisions are the highest ever generated on Earth, allowing us to study the interactions that occurred in the early universe the furthest back in time.

The LHC also stores huge amounts of energy. The magnetic field itself stores an amount equivalent to a couple of tons of TNT, while the beams store about a tenth of that. That energy is stored in one-billionth of a gram of matter, a mere submicroscopic speck of material under ordinary circumstances. When the machine is done with the beam, this enormously concentrated energy is dumped into a cylinder of graphite composite eight meters long and one meter in diameter, which is encased in 1,000 tons of concrete.

The extremes achieved at the LHC push technology to its limits. They don’t come cheaply and the superlatives extend to cost. The LHC’s $9 billion price tag also makes it the most expensive machine ever built. CERN paid about two-thirds of the cost of the machine, with CERN’s 20 member countries contributing to the CERN budget according to their means, ranging from 20 percent from Germany to 0.2 percent from Bulgaria. The remainder was paid for by nonmember states, including the United States, Japan, and Canada. CERN contributes 20 percent to the experiments themselves, which are funded by international collaborations. As of 2008, when the machine was essentially built, the United States had more than 1,000 scientists working on CMS and ATLAS and had contributed $531 million toward the LHC enterprise.


CERN, which houses the LHC, is a research facility, with many programs operating simultaneously. However, CERN’s resources are generally concentrated in a single flagship program. In the 1980s, that program was the SpbarpScollider,38 which found the force carriers essential to the Standard Model of particle physics. The stellar experiments that took place there in 1983 discovered the weak gauge bosons—the two charged W bosons and the neutral Zboson, which communicate the weak force. Those were the key missing Standard Model ingredients at the time, and the discovery earned the accelerator project leaders a Nobel Prize.

Even so, while the SpbarpS was operating, scientists and engineers were already planning a collider known as LEP, which would collide together electrons and their antiparticles known as positrons to study the weak interactions and the Standard Model in exquisite detail. This dream came to fruition in the 1990s, when through its very accurate measurements, LEP studied millions of weak gauge bosons that taught physicists a great deal about Standard Model physics interactions.

LEP was a circular collider with a 27 kilometer circumference. Electrons and positrons were repeatedly boosted in this ring as they orbited around. As we saw in Chapter 6, circular colliders can be inefficient when accelerating light particles such as electrons, since such particles radiate when accelerated on a circular path. The electron beams at the LEP energy of about 100 GeV lost about three percent of their energy each time they went around. This wasn’t too great a loss, but if anyone had wanted to accelerate electrons around this tunnel at any higher energy, the loss during each rotation would have been a deal breaker. Increasing the energy by a factor of 10 would have increased energy loss by a factor of 10,000, which would have made the accelerator far too inefficient to be acceptable.

For this reason, while LEP was being envisioned, people were already thinking about CERN’s next flagship project—which would presumably run at even higher energy. Because of the electron’s unacceptable energy losses, if CERN was to ever build a higher-energy machine, it would require proton beams, which are much heavier and therefore radiate much less. The physicists and engineers who developed LEP were aware of this more desirable possibility so they built the LEP tunnel sufficiently wide to accommodate a possible proton collider in the future, after the electron-positron machine would be dismantled.

Finally, some 25 years later, proton beams now race through the tunnel originally excavated for LEP. (See Figure 24.) The Large Hadron Collider is a couple of years behind schedule and about 20 percent over budget. That’s a pity, but perhaps not so unreasonable given that the LHC is the biggest, most international, most expensive, most energetic, most ambitious experiment ever built. As the screenwriter and director James L. Brooks jokingly said when hearing about the LHC’s setbacks and recovery, “I know people who take approximately the same amount of time to get their wallpaper just so. Understanding the universe just might have a better kick to it. Then again there’s some pretty great wallpaper out there.”


FIGURE 24 ] The setting for the Large Hadron Collider, with the underground tunnel illustrated in white, and Lake Geneva and mountains in the background. (Photo courtesy of CERN)


Protons are everywhere around and within us. However, they are generally bound into nuclei surrounded by electrons inside atoms. They aren’t isolated from those electrons and they aren’t collimated (aligned into columns) inside beams. The LHC first separates and accelerates protons and then steers them to their ultimate destiny. In doing so, they utilize the LHC’s many extremes.

The first step in preparing proton beams is to heat hydrogen atoms, which strips off their electrons and leaves the isolated protons that are their nuclei. Magnetic fields divert these protons so that they are channeled into beams. The LHC then accelerates the beams in several stages in distinct regions, with the protons traveling from one accelerator to another, each time increasing their energy before they are diverted from one of the two parallel beams so that they can collide.

The initial acceleration phase takes place in CERN’s linac, which is a linear stretch of tunnel along which radio waves accelerate protons. When the radio wave is peaked, the associated electric field accelerates the protons. The protons are then made to drift away from the field so they don’t decelerate when the field goes down. They subsequently return to the field when it peaks again so that they repeatedly accelerate from one peak to the next. Essentially the radio waves pulse the protons in the way you push a child on a swing. The waves thereby boost the protons, increasing their energy, but only a tiny amount in this first acceleration stage.

In the next stage, the protons are kicked via magnets into a series of rings where they are further accelerated. Each of these accelerators functions similarly to the linear accelerator described above. However, because these next accelerators are ring shaped, they can repeatedly boost the protons’ energies as they circle around thousands of times. These circular accelerators thereby transfer quite a bit of energy.

This “fellowship of the rings” that accelerates protons before they enter the large LHC ring consists of the proton synchrotron booster (PSB) that accelerates protons to 1.4 GeV, the proton synchrotron (PS) that brings them up to 26 GeV in energy, and then the super proton synchrotron (SPS) that raises their energy to the so-called injection energy of 450 GeV. (See Figure 25 to see a proton’s journey.) This is the energy the protons carry when they enter the last acceleration stage in the large 27 kilometer tunnel.

A couple of these accelerating rings are relics of previous CERN projects. The proton synchrotron, which is the oldest, celebrated its golden anniversary in November 2009, and the proton synchrotron booster was critical to the operation of CERN’s last major project—namely, LEP—in the 1980s.

After protons leave the SPS, their 20 minute long injection phase begins. At this point the 450 GeV protons that emerged from the SPS are boosted to their full energy inside the large LHC tunnel. The protons in the tunnel travel along two separate beams going in opposite directions through narrow three-inch pipes that extend on the 27 kilometers of the underground LHC ring.


FIGURE 25 ] The path a proton travels on when accelerated by the LHC.

The 3.8 meter (12 ft.) wide tunnel that was built in the 1980s but that now houses the proton beams in their final acceleration stage is well lit and air conditioned and large enough to comfortably walk around in, as I had the opportunity to do while the LHC was still in the construction phase. I took only a short stroll inside the tunnel on my LHC tour, but it still took me far longer to traverse my few steps than the 89 millionths of a second it takes for the accelerated highly energetic protons traveling at 99.9999991 percent of the speed of light to make it around.

The tunnel sits about 100 meters underground, with the precise depth varying from 50 to 175 meters. This shields the surface from radiation and also means CERN didn’t have to buy up (and destroy) all the farmland lying over the tunnel’s location during the construction phase. Property rights did, however, delay tunnel excavation back in the 1980s when it was originally constructed for LEP. The problem was that in France, landowners are entitled to the entire region to the Earth’s center—not just the farmland they plow. The tunnel could be dug only after the French authorities blessed the operation by signing a “Déclaration d’Utilité Publique,” thereby making the underlying rock—and in principle the magma underneath too—public property.

Physicists debate whether the reason for the tilt in the tunnel’s depth was geology or if it was done to further defl ect radiation, but the fact is the tilt helps with both. The uneven terrain was in fact an interesting constraint on the tunnel’s depth and location. The region lying under the CERN site is mostly a type of compact rock known as molasses, but underneath the fluvial and marine deposits lie gravel, sand, and loam containing groundwater, and this would not be a good place for a tunnel. The slope keeps the tunnel in the good rock. It also meant that one section of the tunnel at the foot of the beautiful Jura Mountains lying at the edge of CERN could be a little less deep so that getting stuff in and out of vertical shafts in this location was a bit easier (and cheaper).

The final accelerating electric fields in this tunnel are not arranged in a precisely circular fashion. The LHC has eight large arcs alternating with eight 700-meter-long straight sections. Each of these eight sectors can be independently heated up and cooled down, which is important for repairs and instrumentation. After entering the tunnel, protons are accelerated in each of the short straight sections by radio waves, much as they were in the previous acceleration stages that brought them up to injection energy. The acceleration occurs in radio-frequency (RF) cavities that contain a 400 MHz radio signal, which is the same frequency you use when you remotely unlock your car door. When this field accelerates a proton bunch that enters such a cavity, it increases the energy of the protons by a mere 485 billionths of a TeV. This doesn’t sound like much, but the protons orbit the LHC ring 11,000 times a second. Therefore, it takes only 20 minutes to accelerate the proton beam from its injection energy of 450 GeV to its target energy of 7 TeV, about 15 times higher. Some protons are lost during collisions or stray loose, but most of those protons will continue to circulate for about half a day before the beam is depleted and needs to be dumped into the ground and replaced by fresh newly injected protons.

By design, the protons that circulate in the LHC ring aren’t uniformly distributed. They are sent around the ring in bunches—2,808 of them—each containing 115 billion protons. Each bunch starts off 10 centimeters long and one millimeter wide and is separated from the next bunch by about 10 meters. This helps with the acceleration since each bunch is accelerated separately. As a bonus, bundling the protons in this way guarantees that proton bunches interact at intervals of at least 25–75 nanoseconds, which is long enough apart that each bunch collision gets recorded separately. Since so many fewer protons are in a bunch than in a beam, the number of collisions that happen at the same time is under much better control because it is bunches, rather than the full quota of protons in the beam, that will collide at any one time.


Accelerating the protons to high energy is indeed an impressive achievement. But the real technological tour de force in building the LHC was designing and creating the high-field dipole magnets necessary to keep the protons properly circulating around the ring. Without the dipoles, the protons would go along a straight line. Keeping energetic protons circulating in a ring requires an enormous magnetic field.

Because of the existing tunnel size, the major technical engineering hurdle LHC engineers had to contend with was building magnets as strong as possible on an industrial scale—that is, they could be mass produced. The strong field is required to keep high-energy protons on track inside the hand-me-down tunnel that LEP had bequeathed. Keeping more energetic protons circulating requires either stronger magnets or a bigger tunnel so that the proton paths curve sufficiently to stay on track. With the LHC, the tunnel size was predetermined, so the target energy was governed by the maximum attainable magnetic field.

The American Superconducting Supercollider, had it been completed, would have resided in a much bigger tunnel (which in fact was partially excavated), 87 kilometer in circumference, and was planned with the goal of achieving 40 TeV—almost three times the LHC’s target energy. This vastly greater energy would have been possible because the machine was being designed from scratch, without the constraint in size of an existing tunnel and the consequent requirement of unrealistically large magnetic fields. However, the proposed European plan had the practical advantage that the tunnel and the CERN infrastructure of science, engineering, and logistics already existed.

One of the most impressive objects I saw when I visited CERN was a prototype of LHC’s gigantic cylindrical dipole magnets. (See Figure 26 for a cross section.) Even with 1,232 such magnets, each of them is an impressive 15 meters long and weighs 30 tons. The length was determined not by physics considerations but by the relatively narrow LHC tunnel—as well as the imperative of trucking the magnets around on European roads. Each of these magnets cost €700,000, making the net cost of the LHC magnets alone more than a billion dollars.

The narrow pipes that hold the proton beams extend inside the dipoles, which are strung together end to end so that they wind through the extent of the LHC tunnel’s interior. They produce a magnetic field that can be as strong as 8.3 tesla, about a thousand times the field of the average refrigerator magnet. As the energy of the proton beams increases from 450 GeV to 7 TeV, the magnetic field increases from 0.54 to 8.3 teslas, in order to keep guiding the increasingly energetic protons around.

The field these magnets produce is so enormous that it would displace the magnets themselves if no restraints were in place. This force is alleviated through the geometry of the coils, but the magnets are ultimately kept in place through specially constructed collars made of four-centimeter-thick steel.


FIGURE 26 ] Schematic of a cryodipole magnet. Protons are kept circulating around the LHC ring by 1232 such superconducting magnets.

Superconducting technology is responsible for the LHC’s powerful magnets. LHC engineers benefited from the superconducting technology that had been developed for the SSC, as well as for the American Tevatron collider at the Fermilab accelerator center near Chicago, Illinois, and for the German electron-positron collider at the DESY accelerator center in Hamburg.

Ordinary wires such as the copper wires in your home have resistance. This means energy is lost as the current passes through. Superconducting wires, on the other hand, don’t dissipate energy. Electrical current passes through unimpeded. Coils of superconducting wire can carry enormous magnetic fields, and, once in place, the field will be maintained.

Each LHC dipole contains coils of niobium-titanium superconducting cables, each of which contains stranded filaments a mere six microns thick—much smaller than a human hair. The LHC contains 1,200 tons of these remarkable filaments. If you unwrapped them, they would be long enough to encircle the orbit of Mars.

When operating, the dipoles need to be extremely cold, since they work only when the temperature is sufficiently low. The superconducting wires are maintained at 1.9 degrees above absolute zero, which is 271 degrees Celsius below the freezing temperature of water. This temperature is even lower than the 2.7-degree cosmic microwave background radiation in outer space. The LHC tunnel houses the coldest extended region in the universe—at least that we know of. The magnets are known as cryodipoles to take into account their special refrigerated nature.

In addition to the impressive filament technology used for the magnets, the refrigeration (cryogenic) system is also an imposing accomplishment meriting its own superlatives. The system is in fact the world’s largest. Flowing helium maintains the extremely low temperature. A casing of approximately 97 metric tons of liquid helium surrounds the magnets to cool the cables. It is not ordinary helium gas, but helium with the necessary pressure to keep it in a superfluid phase. Superfluid helium is not subject to the viscosity of ordinary materials, so it can dissipate any heat produced in the dipole system with great efficiency: 10,000 metric tons of liquid nitrogen are first cooled, and this in turn cools the 130 metric tons of helium that circulate in the dipoles.

Not everything at the LHC is beneath the ground. Surface buildings hold equipment, electronics, and refrigeration plants. A conventional refrigerator cools down the helium to 4.5 kelvin and then the final cooling takes place with the pressure reduced. This process (as well as warming up) takes about a month, which means that each time the machine is turned on and off, or any repair is attempted, a good deal of additional time is required to cool.

If something went wrong—for example a tiny amount of heat capable of raising the temperature—the system would quench, meaning that superconductivity would be destroyed. Such a quenching would be disastrous if the energy were not properly dissipated, since all the energy stored in the magnets would suddenly be released. Therefore, a special system for detecting quenches and spreading the energy release are in place. The system looks for differences in voltage inconsistent with superconductivity. If detected, the energy is released everywhere, within less than a second, so that the dipole will no longer be superconducting.

Even with superconducting technology, huge currents are needed to achieve the 8.3 tesla magnetic field. The current goes up to almost 12,000 amperes, which is about 40,000 times the current flowing through the lightbulb on your desk.

With the current and the refrigeration, the LHC when running uses an enormous amount of electricity—about the amount required for a small city such as nearby Geneva. To avoid excessive energy expenditures, the accelerator runs only until the cold Swiss winter months when electricity prices go up (with an exception made for the turn-on in 2009). This policy has the extra advantage that it gives the LHC engineers and scientists a nice long Christmas vacation.


The final LHC superlative applies to the vacuum inside the pipes where the protons circulate. The system needs to be kept as free as possible of excess matter in order to maintain the cold helium because any stray molecules could transport away heat and energy. Most critically, the proton beam regions have to be as free of gas as possible. If gas were present, protons could collide with it and destroy the nice circulation of the proton beam. The pressure inside the beams is therefore extremely tiny, 10 trillion times smaller than atmospheric pressure—the pressure one million meters above the Earth’s surface where the air is extremely rarified. At the LHC, 9,000 cubic meters of air was evacuated to achieve the welcoming space for the proton beam.

Even at this ridiculously low pressure, about three million molecules of gas still reside in every cubic centimeter region in the pipe, so protons do occasionally hit the gas and get deflected. Were enough of these protons to hit a superconducting magnet, they would quench it and destroy the superconductivity. Carbon collimators line the LHC beam in order to remove any stray beam particles that lie outside a three-millimeter aperture, which is plenty large enough to permit the approximately millimeter-wide beam to pass through.

Still, organizing the protons in a millimeter-wide bunch is a tricky task. It is accomplished by other magnets, known as quadrupole magnets, that effectively focus and squeeze the beam. The LHC contains 392 such magnets. Quadrupole magnets also divert the proton beams from their independent paths so that they can actually collide.

The beams don’t collide precisely or completely head-on, but rather at the infinitesimal angle of about a thousandth of a radian. This is to ensure that only one bunch from each beam collides at a time so that the data are less confusing and the beam stays intact.

When the two bunches from the two circulating beams collide, one hundred billion protons are up against another bunch of 100 billion protons. Quadrupole magnets are also responsible for the especially daunting task of focusing the beams at the regions along the beam where collisions occur and experiments that record the events are situated. At these locations, the magnets squeeze the beams to the tiny size of 16 microns. The beams have to be extremely small and dense so that the hundred billion protons in a bunch are more likely to find one of the hundred billion protons in the other bunch when they pass through.

Most of the protons in a bunch won’t find the protons in the other bunch, even when they are directed toward each other so as to collide. Individual protons are only about a millionth of a nanometer in diameter. This means that even though all these protons are kept in bunches of 16 microns, only about 20 protons collide head-on each time the bunches cross.

This is in fact a very good thing. If too many collisions occurred simultaneously, the data would simply be confusing. It would be impossible to tell which particles emerged from which collision. And of course if no collisions occurred, that would be a bad thing as well. By focusing just this number of protons into just this size, the LHC ensures the optimal number of events each time bunches cross.

The individual proton collisions, when they occur, do so almost instantaneously—in a time about 25 orders of magnitude less than a second. This means the time between the sets of proton collisions is set entirely by how frequently the bunches cross, which at full capacity is about every 25 nanoseconds. The beams are crossing more than 10 million times a second. With such frequent collisions, the LHC produces a huge amount of data—about a billion collisions per second. Fortunately, the time between bunch crossing is long enough to let the computers keep track of the interesting individual collisions without confusing collisions that originated in different bunches.

So in the end, the extremes at the LHC are necessary to guarantee both the highest possible energy collisions and the largest number of events that the experiments can handle. Most of the energy just stays in circulation with only the rare proton collision worthy of attention. Despite the massive energy in the beams, the energy of individual bunch collisions involves little more than the kinetic energy of a few mosquitoes in flight. These are protons colliding—not football players or cars. The LHC’s extremes concentrate energy in an extremely tiny region, and in elementary particle collisions that experimenters can follow. We’ll soon consider some of the hidden ingredients that they might find and the insights into the nature of matter and space that physicists hope those discoveries will provide.