Quantum Gravity

Physicists Eye Quantum-Gravity Interface

Gravity curves space and time around massive objects. What happens when such objects are put in quantum superpositions, causing space-time to curve in two different ways?

Courtesy of Dirk Bouwmeester

Gravity curves space and time around massive objects. What happens when such objects are put in quantum superpositions, causing space-time to curve in two different ways?

Comments (16)


It starts like a textbook physics experiment, with a ball attached to a spring. If a photon strikes the ball, the impact sets it oscillating very gently. But there’s a catch. Before reaching the ball, the photon encounters a half-silvered mirror, which reflects half of the light that strikes it and allows the other half to pass through.

What happens next depends on which of two extremely well-tested but conflicting theories is correct: quantum mechanics or Einstein’s theory of general relativity; these describe the small- and large-scale properties of the universe, respectively.

In a strange quantum mechanical effect called “superposition,” the photon simultaneously passes through and reflects backward off the mirror; it then both strikes and doesn’t strike the ball. If quantum mechanics works at the macroscopic level, then the ball will both begin oscillating and stay still, entering a superposition of the two states. Because the ball has mass, its gravitational field will also split into a superposition.

But according to general relativity, gravity warps space and time around the ball. The theory cannot tolerate space and time warping in two different ways, which could destabilize the superposition, forcing the ball to adopt one state or the other.

Knowing what happens to the ball could help physicists resolve the conflict between quantum mechanics and general relativity. But such experiments have long been considered infeasible: Only photon-size entities can be put in quantum superpositions, and only ball-size objects have detectable gravitational fields. Quantum mechanics and general relativity dominate in disparate domains, and they seem to converge only in enormously dense, quantum-size black holes. In the laboratory, as the physicist Freeman Dyson wrote in 2004, “any differences between their predictions are physically undetectable.”

In the past two years, that widely held view has begun to change. With the help of new precision instruments and clever approaches for indirectly probing imperceptible effects, experimentalists are now taking steps toward investigating the interface between quantum mechanics and general relativity in tests like the one with the photon and the ball. The new experimental possibilities are revitalizing the 80-year-old quest for a theory of quantum gravity.

“In the final showdown between quantum mechanics and gravity, our understanding of space and time will be completely changed.”

“The biggest single problem of all of physics is how to reconcile gravity and quantum mechanics,” said Philip Stamp, a theoretical physicist at the University of British Columbia. “All of a sudden, it’s clear there is a target.”

Theorists are thinking through how the experiments might play out, and what each outcome would mean for a more complete theory merging quantum mechanics and general relativity. “Neither of them has ever failed,” Stamp said. “They’re incompatible. If experiments can get to grips with that conflict, that’s a big deal.”

Quantum Nature

At the quantum scale, rather than being “here” or “there” as balls tend to be, elementary particles have a certain probability of existing in each of the locations. These probabilities are like the peaks of a wave that often extends through space. When a photon encounters two adjacent slits on a screen, for example, it has a 50-50 chance of passing through either of them. The probability peaks associated with its two paths meet on the far side of the screen, creating interference fringes of light and dark. These fringes prove that the photon existed in a superposition of both trajectories.

But quantum superpositions are delicate. The moment a particle in a superposition interacts with the environment, it appears to collapse into a definite state of “here” or “there.” Modern theory and experiments suggest that this effect, called environmental decoherence, occurs because the superposition leaks out and envelops whatever the particle encountered. Once leaked, the superposition quickly expands to include the physicist trying to study it, or the engineer attempting to harness it to build a quantum computer. From the inside, only one of the many superimposed versions of reality is perceptible.

A single photon is easy to keep in a superposition. Massive objects like a ball on a spring, however, “become exponentially sensitive to environmental disturbances,” explained Gerard Milburn, director of the Center for Engineered Quantum Systems at the University of Queensland in Australia. “The chances of any one of their particles getting disturbed by a random kick from the environment is extremely high.”

Because of environmental decoherence, the idea of probing quantum superpositions of massive objects in tabletop experiments seemed for decades to be dead in the water. “The problem is getting the isolation, making sure no disturbances come along other than gravity,” Milburn said. But the prospects have dramatically improved.

Dirk Bouwmeester, an experimental physicist who splits his time between the University of California, Santa Barbara, and Leiden University in the Netherlands, has developed a setup much like the photon-and-ball experiment, but replacing the ball on its spring with an object called an optomechanical oscillator — essentially a tiny mirror on a springboard. The goal is to put the oscillator in a quantum superposition of two vibration modes, and then see whether gravity destabilizes the superposition.

Ten years ago, the best optomechanical oscillators of the kind required for Bouwmeester’s experiment could wiggle back and forth 100,000 times without stopping. But that wasn’t long enough for the effects of gravity to kick in. Now, improved oscillators can wiggle one million times, which Bouwmeester calculates is close to what he needs in order to see, or rule out, decoherence caused by gravity. “Within three to five years, we will prove quantum superpositions of this mirror,” he said. After that, he and his team must reduce the environmental disturbances on the oscillator until it is sensitive to the impact of a single photon. “It’s going to work,” he insists.

Photo of Markus Aspelmeyer

Courtesy of Markus Aspelmeyer

Markus Aspelmeyer, a quantum physicist at the University of Vienna, is developing three experiments aimed at probing the interface between quantum mechanics and gravity.

Markus Aspelmeyer, a professor of physics at the University of Vienna, is equally optimistic. His group is developing three separate experiments at the quantum-gravity interface — two for the lab and one for an orbiting satellite. In the space-based experiment, a nanosphere will be cooled to its lowest energy state of motion, and a laser pulse will put the nanosphere in a quantum superposition of two locations, setting up a situation much like a double-slit experiment. The nanosphere will behave like a wave with two interfering peaks as it moves toward a detector. Each nanosphere can be detected in only a single location, but after multiple repetitions of the experiment, interference fringes will appear in the distribution of the nanospheres’ locations. If gravity destroys superpositions, the fringes won’t appear for nanospheres that are too massive.

The group is designing a similar experiment for Earth’s surface, but it will have to wait. At present, the nanospheres cannot be cooled enough, and they fall too quickly under Earth’s gravity, for the test to work. But “it turns out that optical platforms on satellites actually already meet the requirements that we need for our experiments,” said Aspelmeyer, who is collaborating with the European Aeronautic Defense and Space Company in Germany. His team recently demonstrated a key technical step required for the experiment. If it gets off the ground and goes as planned, it will reveal the relationship between the mass of the nanospheres and decoherence, pitting gravity against quantum mechanics.

The researchers laid out another terrestrial experiment last spring in Nature Physics. Many proposed quantum gravity theories involve modifications to Heisenberg’s uncertainty principle, a cornerstone of quantum mechanics that says it isn’t possible to precisely measure both the position and momentum of an object at the same time. Any deviations to Heisenberg’s formula should show up in the position-momentum uncertainty of an optomechanical oscillator, because it is affected by gravity. The uncertainty itself is immeasurably small — a blurriness just 100-million-trillionth the width of a proton — but Igor Pikovski, a theorist in Aspelmeyer’s group, has discovered a backdoor route to detecting it. When a light pulse strikes the oscillator, Pikovski claims that its phase (the position of its peaks and troughs) will undergo a discernible shift that depends on the uncertainty. Deviations from the predictions of traditional quantum mechanics could be experimental evidence of quantum gravity.

Aspelmeyer’s group has started to realize the first experimental steps. Pikovski’s idea “provides us with a quite, I have to admit, unexpected improvement in performance,” Aspelmeyer said. “We are all a little surprised, actually.”

The Showdown

Many physicists expect quantum theory to prevail. They believe the ball on a spring should, in principle, be able to exist in two places at once, just as a photon can. The ball’s gravitational field should be able to interfere with itself in a quantum superposition, just as the photon’s electromagnetic field does. “I don’t see why these concepts of quantum theory that have proven to be right for the case of light should fail for the case of gravity,” Aspelmeyer said.

But the incompatibility of general relativity and quantum mechanics itself suggests that gravity might behave differently. One compelling idea is that gravity could act as a sort of inescapable background noise that collapses superpositions.

“While you can get rid of air molecules and electromagnetic radiation, you can’t screen out gravity,” said Miles Blencowe, a professor of physics at Dartmouth College. “My view is that gravity is sort of like the fundamental, unavoidable, last-resort environment.”

Rendering of an optomechanical oscillator.

Christopher Baker and Ivan Favero at Université Paris Diderot-CNRS

In an optomechanical oscillator, the light confined between two mirrors causes one of the mirrors to oscillate on a spring. Experimentalists plan to use such devices to pit quantum mechanics against general relativity.

The background-noise idea was conceived in the 1980s and 1990s by Lajos Diósi of the Wigner Research Center for Physics in Hungary and, separately, by Roger Penrose of Oxford University. According to Penrose’s model, a discrepancy in the curvature of space and time could accumulate during a superposition, eventually destroying it. The more massive or energetic the object involved and, thus, the larger its gravitational field, the more quickly “gravitational decoherence” would happen. The space-time discrepancy ultimately results in an irreducible level of noise in the position and momentum of particles, consistent with the uncertainty principle.

“That would be a wonderful result if the ultimate reason for the uncertainty principle and the puzzling features of quantum physics are due to some quantum effects of space and time,” Milburn said.

Inspired by the possibility of experimental tests, Milburn and other theorists are expanding on Diósi and Penrose’s basic idea. In a July paper in Physical Review Letters, Blencowe derived an equation for the rate of gravitational decoherence by modeling gravity as a kind of ambient radiation. His equation contains a quantity called the Planck energy, which equals the mass of the smallest possible black hole. “When we see the Planck energy we think quantum gravity,” he said. “So it may be that this calculation is touching on elements of this undiscovered theory of quantum gravity, and if we had one, it would show us that gravity is fundamentally different than other forms of decoherence.”

Stamp is developing what he calls a “correlated path theory” of quantum gravity that pinpoints a possible mathematical mechanism for gravitational decoherence. In traditional quantum mechanics, probabilities of future outcomes are calculated by independently summing the various paths a particle can take, such as its simultaneous trajectories through both slits on a screen. Stamp found that when gravity is included in the calculations, the paths connect. “Gravity basically is the interaction that allows communication between the different paths,” he said. The correlation between paths results once more in decoherence. “No adjustable parameters,” he said. “No wiggle room. These predictions are absolutely definite.”

At meetings and workshops, theorists and experimentalists are working closely to coordinate the various proposals and plans for testing them. They say it’s a mutually motivating situation.

“In the final showdown between quantum mechanics and gravity, our understanding of space and time will be completely changed,” Milburn said. “We’re hoping these experiments will lead the way.”

This article was reprinted on ScientificAmerican.com.

Add a Comment

View Comments (16)

Comments for this entry

  • I would suggest that the greatest problem in physics is not the reconciliation of quantum theory and general relativity. Even if such a reconciliation were to happen it would shed light on only five percent of the matter and energy of the universe. Ninety-five percent of the universe would remain a mystery. Dark energy and dark matter seem to be far greater problems for physics.

  • principle: question everything…
    Has anyone ever actually demonstrated that fundamental entities such as an electron and a proton, or a pair of electrons, actually experience gravity, i.e. show a gravitational attraction for one another? It would be incredibly hard to demonstrate this, and swamped by the much larger electric forces. Okay, we know that large collections of neutral /uncharged particles aggregate under gravity – we think – as in a neutron star. But two neutrons ?

  • Sub-atomic particles such as neutrons or electrons experience gravity because they have mass. Their mass is very small, but it certifiably exists and we can therefore calculate exactly how much gravitational force a single neutron exerts upon any other particle in the universe through simple math. If it has mass, it experiences gravity.

  • Why should Quantum theory not be expect to follow its own premise (in that it can be right and wrong at the same time.)?

  • Hello Jordan,

    You are actually incorrect – mass and gravity aren’t the same thing – photons experience gravity, yet are massless particles. Just look up Einstein’s cross and black hole event horizons for proof. Mass arises from the Higgs interaction and essentially acts as ‘drag’ on massive particles, while massless particles travel at the speed of light. Both are affected by gravity.

  • I am a HUGE fan of Einstein. But, the worst thing that he ever did was give in to his early classical training and allow the theory of a universal “Ether” to shape his theory of
    Gravity. The worst thing that the scientific community has done, since then, has been to blindly accept it. The physical evidence against Gravity, as the “warping of space” is everywhere. One need only consider the shapes of solar systems or galaxies. Or contemplate the differing behaviors of photons and neutrinos, when passing near a star. Only the one carrying a charge is somehow affected by the “shape” of space in the star’s vicinity. A response eerily similar to that of any charged particle passing through or near the field of other charged particles. While not yet experimentally proven, the scientific community needs to seriously reconsider Gravity as an “attractive force acting at the Speed of Light” as posited by Newton.

  • @Nils Irland,
    In the case of photons, gravity is a distortion of spacetime caused by the momentum – which is something that photons indisputably have.
    @Dick Hamilton,
    Since photons experience gravity (and this has been shown, through the success of things like gravitational lensing being used to observe the light from distant galaxies), I feel like it’s not too much of a stretch to say that, at the least, other bosons (force-carriers, in the same class as photons) experience gravity. As for leptons (“matter” particles – electrons, quarks, protons, neutrons, etc.) – well, we can simply extend our knowledge of gravity downwards. If a large clump of stuff interacts gravitationally, why would a small clump of stuff not interact gravitationally? How small is too small? At what point is the cut-off, and why is it at that point rather than any other? Physics in general hates “fine-tuned” values for different phenomena – and this explanation causes more problems than it solves.

  • Neutrons feel gravity, but there are other kinds of forces much more stronger at the subatomic level:

    strong nuclear (scale 1), range 10^-15m, between nucleons
    electric (scale 10^-2), range 1/r^2, keeps the core together
    weak nuclear (scale 10^-12), range 10^-17m, beta disintegration within particles
    gravity (scale 10^-38), range 1/r^2, macroscopic

    So as you can see, gravity is insignificant at the subatomic level, yet we notice it in the macroscopic world.

    I believe if this experiment succeeds, it won’t contradict any of the two theories but come with a surprising outcome such as the ball rotating, warming or disintegrating particles instead of oscillating.

  • It is not true that Heisenberg’s uncertainty principle says “it isn’t possible to precisely measure both the position and momentum of an object at the same time. ”

    It is well known that any wave theory has an associated uncertainty principle, so this is a feature of wave mechanics itself. It is impossible to make an QM predictions about a single quantum measurement except when measuring a prepared state. So QM has nothing to say about an individual ‘object’. Indeed, there is a well known thought experiment to show this. Prepare a beam of electrons with precise momentum. This spreads the beam out to a width commensurate with the precision of the momentum as given by Heisenberg. We now have a prepared state and know the momentum precisely for any electron in the beam. But for an arbitrary electron (and we cannot predict which) we can precisely measure its position. It is easy to beat Heisenberg this way.

    If you read any introductory QM text you will see that Heisenberg is not a relation between position and momentum measurements but between position and momentum expectation values. Therefore this can only apply to multiple measurements.

    Is this a pedantic quibble? No, because many philosophers of physics believe in the de Broglie-Bohm interpretation which says that QM is a theory of what we can know about an underlying classical world. As the Bohm interpretation reproduces all of QM including the Bell inequalities it is hard to see how to construct an experiment to decide between interpretations. Maybe these QM vs Gravity experiments will do just that. If so, I’m putting my money on Gravity – I’d rather give up time than space.

  • @John Schlesinger:

    In my opinion, your thought experiment will not work out. You make the mistake that you implicitly assume to have an underlying classical reality. By that, I mean that you assume the momentum and the position to be realized at one moment, like in classical mechanics. However, the message of QM is, that this is not true. (Also in Bohmian Mechanics it is not true, as the only property that can be assigned to the particle is the position. Therefore in BM the particle does not have a momentum anyway, because the momentum is assigned to the guiding wave. This can be illustrated if one thinks of an eigenstate of the harmonic oscillator. As the eigenstates are real, the phase of the guiding wave is zero and therefore the particle does not move. However, the momentum will in general be different from zero).

    To be more concrete: In fact, your experiment has already been done (however slightly altered). Substitute momentum and position by the spin in the x and z direction respectively. Then, one of the first and probably one of the most famous experiments in the context of QM, namely the Stern-Gerlach, show exactly that. Using first a SG apparatus in the z direction, filtering out only the spin up particles, we have our perfect beam. Now install a second SG apparatus in the x direction and we measure the spin in the x direction and one might say “Yuhey, we know x and z spin of our particle, we beat Heisenberg” exactly as you said. But that is of course wrong. Measuring the z component again, we find the same splitting as in the first apparatus although we have already killed all the z-spin down particles. So in fact, maybe we new about the x direction (at least before the third splitting), but we could not really talk about the z direction.

    The same holds for your thought experiment. As soon as you measure the position of your particle, it won’t have a specific momentum anymore (indeed, it will be in a superposition of all momentum eigenstates).
    What is the message of this: you can not talk about specific values of complementary observables at the same time. They don’t have an ontology as in classical physics.

  • I didn’t know about investigations into “quantum gravity” but interestingly, in December I had postulated to my son that it was possible to create a situation analogous to the “double slit” experiment in which the gravity fields of the two possible locations of the quantum particle would be unaffected, but had no idea that it could ever be possible to test it.

  • If quantum mechanics allow for superposition (1 particle being in 2 places at the same time), then couldn’t it follow that gravity would exist between those 2 positions…as though they were 2 particles? If true, wouldn’t that reconcile quantum mechanics to general relativity?

  • I would put my money on the experiment counting as an act of observation, therefore the beam would experience decoherence. I don’t think it would be fair to say that it would be gravity that caused this decoherence, but it could be more correct to say that it was due to interaction with a massive object. Massive objects cannot exist in a state of superposition. I think it would be naive to conclude otherwise.

    If, say for instance, quantum gravity was found to be due to overlapping virtual particle waves, then mass would be the result and limit of this type of interaction. Then to attempt to put a massive body in superposition would be synergistic to putting these virtual particles beyond that limit.

Leave a Comment

Your email address will not be published. Your name will appear near your comment. Required *