The Fabric of the Cosmos: Space, Time, and the Texture of Reality (56 page)

In 1997, a group of physicists led by Anton Zeilinger, then at the University of Innsbruck, and another group led by A. Francesco De Martini at the University of Rome,
²
each carried out the first successful teleportation of a single photon. In both experiments, an initial photon in a particular quantum state was teleported a short distance across a laboratory, but there is every reason to expect that the procedures would have worked equally well over any distance. Each group used a technique based on theoretical insights reported in 1993 by a team of physicists—Charles Bennett of IBM's Watson Research Center; Gilles Brassard, Claude Crepeau, and Richard Josza of the University of Montreal; the Israeli physicist Asher Peres; and William Wootters of Williams College—that rely on quantum entanglement (Chapter 4).

Remember, two entangled particles, say two photons, have a strange and intimate relationship. While each has only a certain probability of spinning one way or another, and while each, when measured, seems to "choose" randomly between the various possibilities, whatever "choice" one makes the other immediately makes too, regardless of their spatial separation. In Chapter 4, we explained that there is no way to use entangled particles to send a message from one location to another faster than the speed of light. If a succession of entangled photons were each measured at widely separated locations, the data collected at either detector would be a random sequence of results (with the overall frequency of spinning one way or another being consistent with the particles' probability waves). The entanglement would become evident only on comparing the two lists of results, and seeing, remarkably, that they were identical. But that comparison requires some kind of ordinary, slower-than-light-speed communication. And since before the comparison no trace of the entanglement could be detected, no faster than light-speed signal could be sent.

Nevertheless, even though entanglement can't be used for superluminal communication, one can't help feeling that long-distance correlations between particles are so bizarre that they've got to be useful for something extraordinary. In 1993, Bennett and his collaborators discovered one such possibility. They showed that quantum entanglement could be used for quantum teleportation. You might not be able to send a message at a speed greater than that of light, but if you'll settle for slower-than-light teleportation of a particle from here to there, entanglement's the ticket.

The reasoning behind this conclusion, while mathematically straightforward, is cunning and ingenious. Here's the flavor of how it goes.

Imagine I want to teleport a particular photon, one I'll call Photon A, from my home in New York to my friend Nicholas in London. For simplicity, let's see how I'd teleport the exact quantum state of the photon's spin—that is, how I'd ensure that Nicholas would acquire a photon whose probabilities of spinning one way or another were identical to Photon A's.

I can't just measure the spin of Photon A, call Nicholas, and have him manipulate a photon on his end so its spin matches my observation; the result I find would be affected by the observation I make, and so would not reflect the true state of Photon A before I looked. So what can I do? Well, according to Bennett and colleagues, the first step is to ensure that Nicholas and I each have one of two additional photons, let's call them Photons B and C, which are entangled. How we get these photons is not particularly important. Let's just assume that Nicholas and I are certain that even though we are on opposite sides of the Atlantic, if I were to measure Photon B's spin about any given axis, and he were to do the same for Photon C, we would find exactly the same result.

The next step, according to Bennett and coworkers, is
not
to directly measure Photon A—the photon I hope to teleport—since that turns out to be too drastic an intervention. Instead, I should measure a
joint
feature of Photon A and the entangled Photon B. For instance, quantum theory allows me to measure whether Photons A and B have the same spin about a vertical axis, without measuring their spins individually. Similarly, quantum theory allows me to measure whether Photons A and B have the same spin about a horizontal axis, without measuring their spins individually. With such a joint measurement, I do not learn Photon A's spin, but I do learn how Photon A's spin is related to Photon B's. And that's important information.

The distant Photon C is entangled with Photon B, so if I know how Photon A is related to Photon B, I can deduce how Photon A is related to Photon C. If I now phone this information to Nicholas, communicating how Photon A is spinning relative to his Photon C, he can determine how Photon C must be manipulated so that its quantum state will match Photon A's. Once he carries out the necessary manipulation, the quantum state of the photon in his possession will be identical to that of Photon A, and that's all we need to declare that Photon A has been successfully teleported. In the simplest case, for example, should my measurement reveal that Photon B's spin is identical to Photon A's, we would conclude that Photon C's spin is also identical to Photon A's, and without further ado, the teleportation would be complete. Photon C would be in the same quantum state as Photon A, as desired.

Well, almost. That's the rough idea, but to explain quantum teleportation in manageable steps, I've so far left out an absolutely crucial element of the story, one I'll now fill in. When I carry out the joint measurement on Photons A and B, I do indeed learn how the spin of Photon A is related to that of Photon B. But, as with all observations, the measurement itself affects the photons. Therefore, I do
not
learn how Photon A's spin was related to Photon B's before the measurement. Instead, I learn how they are related after they've both been disrupted by the act of measurement. So, at first sight, we seem to face the same quantum obstacle to replicating Photon A that I described at the outset: the unavoidable disruption caused by the measurement process. That's where Photon C comes to the rescue. Because Photons B and C are entangled, the disruption I cause to Photon B in New York will also be reflected in the state of Photon C in London. That is the wondrous nature of quantum entanglement, as elaborated in Chapter 4. In fact, Bennett and his collaborators showed mathematically that through its entanglement with Photon B, the disruption caused by my measurement is imprinted on the distant Photon C.

And that's fantastically interesting. Through my measurement, we are able to learn how Photon A's spin is related to Photon B's, but with the prickly problem that both photons were disrupted by my meddling. Through entanglement, however, Photon C is tied in to my measurement—even though it's thousands of miles away—and this allows us to isolate the effect of the disruption and thereby have access to information ordinarily lost in the measurement process. If I now call Nicholas with the result of my measurement, he will learn how the spins of Photons A and B are related after the disruption, and, via Photon C, he will have access to the impact of the disruption itself. This allows Nicholas to use Photon C to, roughly speaking, subtract out the disruption caused by my measurement and thus skirt the obstacle to duplicating Photon A. In fact, as Bennett and collaborators show in detail, by at most a simple manipulation of Photon C's spin (based on my phone call informing him how Photon A is spinning relative to Photon B) Nicholas will ensure that Photon C, as far as its spin goes, exactly replicates the quantum state of Photon A
prior to
my measurement.
Moreover, although spin is only one characteristic of a photon, other features of Photon A's quantum state (such as the probability that it has one energy or another) can be replicated similarly. Thus, by using this procedure, we could teleport Photon A from New York to London.
³

As you can see, quantum teleportation involves two stages, each of which conveys critical and complementary information. First, we undertake a joint measurement on the photon we want to teleport with one member of an entangled pair of photons. The disruption associated with the measurement is imprinted on the distant partner of the entangled pair through the weirdness of quantum nonlocality. That's Stage 1, the distinctly quantum part of the teleportation process. In Stage 2, the result of the measurement itself is communicated to the distant reception location by more standard means (telephone, fax, e-mail . . .) in what might be called the classical part of the teleportation process. In combination, Stage 1 and Stage 2 allow the exact quantum state of the photon we want to teleport to be reproduced by a straightforward operation (such as a rotation by a certain amount about particular axes) on the distant member of the entangled pair.

Notice, as well, a couple of key features of quantum teleportation. Since Photon A's original quantum state was disrupted by my measurement, Photon C in London is now the only one in that original state. There aren't two copies of the original Photon A and so, rather than calling this quantum faxing, it is indeed more accurate to call this quantum teleportation.
⁴
Furthermore, even though we teleported Photon A from New York to London—even though the photon in London becomes indistinguishable from the original photon we had in New York—we do not learn Photon A's quantum state. The photon in London has exactly the same probability of spinning in one direction or another as Photon A did before my meddling, but we do not know what that probability is. In fact, that's the trick underlying quantum teleportation. The disruption caused by measurement prevents us from determining Photon A's quantum state, but in the approach described,
we don't need to know the photon's quantumstate in order to teleport it.
We need to know only an aspect of its quantum state—what we learn from the joint measurement with Photon B. Quantum entanglement with distant Photon C fills in the rest.

Implementing this strategy for quantum teleportation was no small feat. By the early 1990s, creating an entangled pair of photons was a standard procedure, but carrying out a joint measurement of two photons (the joint measurement on Photons A and B described above, technically called a
Bell-state measurement
) had never been attained. The achievement of both Zeilinger's and De Martini's groups was to invent ingenious experimental techniques for the joint measurement and to realize them in the laboratory.
⁵
By 1997 they had achieved this goal, becoming the first groups to achieve the teleportation of a single particle.

Since you and I and a DeLorean and everything else are composed of many particles, the natural next step is to imagine applying quantum teleportation to such large collections of particles, allowing us to "beam" macroscopic objects from one place to another. But the leap from teleporting a single particle to teleporting a macroscopic collection of particles is staggering, and enormously far beyond what researchers can now accomplish and what many leaders in the field imagine achieving even in the distant future. But for kicks, here's how Zeilinger fancifully dreams we might one day go about it.

Imagine I want to teleport my DeLorean from New York to London. Instead of providing Nicholas and me with one member each of an entangled pair of photons (what we needed to teleport a single photon), we must each have a chamber of particles containing enough protons, neutrons, electrons, and so on to build a DeLorean, with all the particles in my chamber being quantum entangled with all those in Nicholas's chamber (see Figure 15.1). I also need a device that measures joint properties of all the particles making up my DeLorean with those particles flitting to and fro within my chamber (the analog of measuring joint features of Photons A and B). Through the entanglement of the particles in the two chambers, the impact of the joint measurements I carry out in New York will be imprinted on Nicholas's chamber of particles in London (the analog of Photon C's state reflecting the joint measurement of A and B). If I call Nicholas and communicate the results of my measurements (it'll be an expensive call, as I'll be giving Nicholas some 10
³⁰
results), the data will instruct him on how to manipulate the particles in his chamber (much as my earlier phone call instructed him on how to manipulate Photon C). When he finishes, each particle in his chamber will be in precisely the same quantum state as each particle in the DeLorean (before it was subjected to any measurements) and so, as in our earlier discussion, Nicholas will now have the DeLorean.
⁴¹
Its teleportation from New York to London will be complete.

Figure 15.1 A fanciful approach to teleportation envisions having two chambers of quantum entangled particles at distant locations, and a means of carrying out appropriate joint measurements of the particles making up the object to be teleported with the particles in one of the chambers. The results of these measurements would then provide the necessary information to manipulate the particles in the second chamber to replicate the object, and complete the teleportation.

Note, though, that as of today, every step in this macroscopic version of quantum teleportation is fantasy. An object like a DeLorean has in excess of a billion billion billion particles. While experimenters are gaining facility with entangling more than a single pair of particles, they are extremely far from reaching numbers relevant for macroscopic entities.
⁶
Setting up the two chambers of entangled particles is thus absurdly beyond current reach. Moreover, the joint measurement of
two
photons was, in itself, a difficult and impressive feat. Extending this to a joint measurement of billions and billions of particles is, as of today, unimaginable. From our current vantage point, a dispassionate assessment would conclude that teleporting a macroscopic object, at least in the manner so far employed for a single particle, is eons—if not an eternity—away.

But, as the one constant in science and technology is the transcendence of naysaying prophesies, I'll simply note the obvious: teleportation of macroscopic bodies looks unlikely. Yet, who knows? Forty years ago, the Enterprise's computer looked pretty unlikely too.
⁷