粒子的自旋似乎脱离了物体而移动——这一奇怪的实验观察引发了争论
量子物理学经常与猫有关——这里也是如此。
Seregraff/Getty图片社
物理学家似乎对猫很着迷。电动力学之父詹姆斯·克拉克·麦克斯韦研究了猫科动物的下落，以研究它们在下落时如何转身。许多物理老师用猫的皮毛和硬橡胶棒来解释摩擦电现象。埃尔温·薛定谔用一个涉及一只既不死也不活的猫的思想实验来说明量子物理学的奇异性。
因此，物理学家在 2013 年发表于《新物理学杂志》的一篇论文中再次将猫科动物作为新发现的量子现象的名称，这并不令人惊讶。他们的研究摘要只有三句话：“在这篇论文中，我们展示了一只量子柴郡猫。在预先选定和后选定的实验中，我们发现这只猫在一个地方，而它的笑容在另一个地方。猫是一个光子，而笑容是它的圆偏振。”
这一新发现的现象是，某些粒子的特征会选择与它们本身不同的路径，很像《爱丽丝梦游仙境》中柴郡猫的微笑。这部作品的作者是刘易斯·卡罗尔（数学家查尔斯·路特维奇·道奇森的笔名），出版于 1865 年。迄今为止，已有多个实验证明了这种奇怪的量子效应。但这一想法也引发了强烈的质疑。批评者更关心对证据的解读，而非理论计算或实验的严谨性。日本广岛大学物理学家霍尔格·霍夫曼说：“谈论无形传输在我看来有点大胆。我们应该重新审视粒子的概念。”
最近，查普曼大学的雅基尔·阿哈罗诺夫 (Yakir Aharonov) 领导的研究人员将这场争论推向了新的高度。阿哈罗诺夫是第一篇提出量子柴郡效应的论文的合著者。现在，在预印本服务器 arXiv.org 上，他和他的同事发布了一篇理论工作的描述，他们认为这表明量子特性可以在没有任何粒子的情况下移动——就像一个无形的笑容在世界中飘荡并影响周围环境——以一种绕过过去提出的关键问题的方式。
无猫一笑
几年前，阿哈罗诺夫和他的同事第一次遇到他们的量子柴郡猫，当时他们正在思考量子力学的一个最基本原理：没有什么是可以明确预测的。与经典物理学不同，同一个量子力学实验在完全相同的条件下可能会有不同的结果。因此，不可能预测单个实验的确切结果——只能预测其具有一定概率的结果。“没有人理解量子力学。它太违反直觉了。我们知道它的定律，但我们总是感到惊讶，”英国布里斯托尔大学的物理学家桑杜·波佩斯库 (Sandu Popescu) 说，他与阿哈罗诺夫合作发表了 2013 年的论文和新的预印本。
但阿哈罗诺夫并不满足于这种不确定性。因此，自 20 世纪 80 年代以来，尽管量子力学是基于概率的，但他一直在探索研究基本过程的方法。现年 92 岁的阿哈罗诺夫采用了一种方法，即密集重复实验、对结果进行分组，然后检查实验前后的结果并将这些事件相互关联。“要做到这一点，你必须了解量子力学中的时间流动，”波佩斯库解释说。“我们开发了一种全新的方法来结合实验前后测量的信息。”
研究人员通过这种方法偶然发现了几个惊喜——包括他们理论上的柴郡猫。他们的想法乍一看很简单：将粒子送入一种称为干涉仪的光学工具，这会导致每个粒子沿着两条路径中的一条移动，最后两条路径再次合并。阿哈罗诺夫和他的同事推测，如果设置和测量操作得当，就可以证明粒子在干涉仪中行进的路径与其偏振路径不同。换句话说，他们声称粒子的属性可以在一条路径上测量，即使粒子本身走的是另一条路径——就好像咧嘴笑和猫分开了一样。
受这一理论的启发，当时就职于维也纳技术大学的 Tobias Denkmayr 领导的团队在 2014 年发表的一项研究中进行了中子实验。该团队表明，干涉仪内的中性粒子遵循的路径与其自旋不同，自旋是粒子的量子力学特性，类似于角动量：Denkmayr 和他的同事确实找到了柴郡猫理论的证据。两年后，波特兰大学的 Maximilian Schlosshauer 领导的研究人员成功地用光子进行了同样的实验。科学家们看到了光粒子在干涉仪中走的路径与其偏振不同的证据。
弱测量和错觉
但并非所有人都相信这一点。“这种分离毫无意义。粒子的位置本身就是粒子的属性，”霍夫曼说。“更准确的说法是位置和极化之间存在着不寻常的相关性。”去年 11 月，霍夫曼和他的同事根据广为人知的量子力学效应提出了另一种解释。
巴西米纳斯吉拉斯联邦大学的 Pablo Saldanha 和他的同事对柴郡猫实验结果进行了另一种解释，他们认为这些发现可以用波粒二象性来解释。“如果你换个角度看，你会发现并不存在悖论，”Saldanha 说，“但所有结果都可以用传统量子力学的简单干涉效应来解释。”
大部分争议都围绕着这些实验中检测粒子特性和位置的方式。扰动粒子可能会改变其量子力学性质。因此，无法使用普通探测器在干涉仪内记录光子或中子。相反，科学家必须诉诸阿哈罗诺夫于 1988 年开发的弱测量原理。弱测量可以非常轻微地扫描粒子而不会破坏其量子态。然而，这是有代价的：弱测量结果极其不准确。（因此，这些实验必须重复多次，以弥补每次测量都具有高度不确定性的事实。）
在量子柴郡猫实验中，沿着干涉仪的一条路径进行弱测量，然后路径合并，用普通探测器测量出现的粒子。沿着干涉仪的一条路径，可以对粒子的位置进行弱测量，沿着另一条路径，可以对其自旋进行弱测量。利用探测器，物理学家可以更明确地表征穿过干涉仪的粒子，并可能重建粒子旅程中发生的情况。例如，只有某些粒子会出现在某些探测器中，这有助于物理学家拼凑出中子或光子之前走的路径。根据 Aharonov、Popescu 及其同事的说法，柴郡猫实验最终表明，即使在另一条路径上测量了粒子的极化或自旋，也可以在一条路径上确认粒子的位置。
Melissa Thomas Baum/Buckyball Design；来源：Tobias Denkmayr 等人撰写的《物质波干涉仪实验中对量子柴郡猫的观察》，载于《自然通讯》第 5 卷，文章编号 4492；2014 年 7 月 29 日
萨尔达尼亚和他的合著者断言，鉴于目前的测量结果，不可能对过去的量子系统做出断言。换句话说，在最终探测器中测量的光子和中子无法告诉我们它们之前的轨迹。相反，穿过干涉仪路径的粒子的波函数可能会重叠，这将使我们无法追踪粒子所走的路径。“最终，这些矛盾的行为与波粒二象性有关，”萨尔达尼亚说。但他断言，在报告量子柴郡猫证据的论文中，这些发现“经过了复杂的处理，掩盖了这种更简单的解释。”
与此同时，霍夫曼强调，如果你用不同的方式测量系统，结果会有所不同。这种现象在量子物理学中是众所周知的：例如，如果你先测量一个粒子的速度，然后测量它的位置，结果可能会与先测量同一粒子的位置，然后测量它的速度不同。因此，他和他的同事认为，阿哈罗诺夫和他的团队的结论本身是正确的——粒子沿着一条路径移动，极化则沿着另一条路径移动——但这种不同的路径并不同时适用。
正如霍夫曼的合著者、同样来自广岛大学的琼特·汉斯 (Jonte Hance)告诉《新科学家》杂志的那样， “它只是看上去像（粒子和极化）分离了，因为你在一个地方测量其中一种属性，在另一个地方测量另一种属性，但这并不意味着这些属性在一个地方，在另一个地方，这意味着实际测量本身会以某种方式影响它，使它看起来像在一个地方，又在另一个地方。”
捕捉柴郡猫的新方法？
但这些批评“没有抓住要点”，波佩斯库说。他同意萨尔达尼亚和霍夫曼各自团队的工作和推理是正确的——但他补充说，检验任何解释的最佳方法是从每个解释中生成可检验的预测。“据我所知，没有直接的方法可以基于它们做出预测，”波佩斯库在谈到这些替代解释时说。“他们看待事物的方式有点非常老式：存在矛盾，所以你不要再做数学运算了。”
在最近的预印本论文中，阿哈罗诺夫和波佩斯库与布里斯托尔大学的物理学家丹尼尔·柯林斯一起描述了粒子的自旋如何完全独立于粒子本身移动——而无需使用弱测量。在他们新的实验装置中，一个粒子位于一个细长的两部分圆柱体的左半部分，圆柱体的外边缘被密封。由于中间有一堵高反射壁，粒子隧穿到圆柱体右侧的概率微乎其微。在论文中，研究人员提供了一个证明，即使粒子在几乎所有情况下都停留在左侧区域，仍然应该有可能在右侧外壁测量粒子自旋的转移。“这很神奇，不是吗？”柯林斯说。“你认为粒子有自旋，自旋应该和粒子在一起。但自旋在没有粒子的情况下穿过了盒子。”
在一个旨在观察量子柴郡猫的新型思想实验中，物理学家将能够测量细长圆柱体的两个腔之一中的粒子的属性，尽管该粒子本身被包含在另一个腔中。
阿曼达·蒙塔内斯
这种方法将解决迄今为止提出的几个关键问题。物理学家不需要弱测量。他们也不需要将实验结果分组来得出时间结论。（话虽如此，分组结果仍会改善测量结果，因为由于海森堡不确定性原理，墙本身的角动量无法明确确定。）但在这种情况下，唯一涉及的物理原理是守恒定律，例如能量守恒定律或动量和角动量守恒定律。Popescu 和 Collins 解释说，他们希望其他团队能够实施实验以在实验室中观察效果。
这项新研究激起了霍夫曼的兴趣。“这个场景令人兴奋，因为极化和粒子运动之间的相互作用产生了一种特别强烈的量子效应，这显然与粒子图像相矛盾，”他说。
但他仍然不认为这是无形（无粒子）自旋转移的证据。“对我来说，这首先意味着假设独立于测量的现实是错误的，”霍夫曼说。相反，量子力学允许粒子的驻留延伸到圆柱体的右侧区域，即使左侧区域的驻留似乎在逻辑上令人信服。“我认为 Aharonov、Collins 和 Popescu 很清楚，墙壁前的空间并不是真正空的，”他补充道。
与此同时，萨尔达尼亚仍然认为研究人员过于复杂化了可以解释为传统量子干涉效应的现象。在讨论粒子进入实验装置右侧的概率非常低时，他解释说，“当我们提到波时，我们必须小心‘微不足道的概率’。”粒子的波函数也可以扩展到装置的右侧，从而影响壁面的角动量。“即使没有如此戏剧性的结论，也可以做出同样的预测，”他说。
针对这些批评，波佩斯库表示：“这当然是另一种思考方式。问题在于这种解释是否有用。”无论对事件的哪种解释是正确的，量子柴郡猫都可以实现新的技术应用。例如，它可以用来传输信息或能量，而无需移动物理粒子——无论是物质还是光。
然而，对于波佩斯库来说，物理学的基本问题发挥着更重要的作用。“这一切都始于我们思考时间在量子力学中如何传播，”他说。“突然间，我们能够发现守恒定律的一些基本原理。”
本文最初发表于《科学光谱》，经许可转载。
A New Quantum Cheshire Cat Thought Experiment Is Out of the Box
The spin of a particle seems to detach and move without a body—a strange experimental observation that’s stirring up debate
Quantum physics is often about cats—and this is also the case here.
Seregraff/Getty Images
Physicists seem to be obsessed with cats. James Clerk Maxwell, the father of electrodynamics, studied falling felines to investigate how they turned as they fell. Many physics teachers have used a cat’s fur and a hard rubber rod to explain the phenomenon of frictional electricity. And Erwin Schrödinger famously illustrated the strangeness of quantum physics with a thought experiment involving a cat that is neither dead nor alive.
So it hardly seems surprising that physicists turned to felines once again to name a newly discovered quantum phenomenon in a paper published in the New Journal of Physics in 2013. Their three-sentence study abstract reads, “In this paper we present a quantum Cheshire Cat. In a pre- and post-selected experiment we find the Cat in one place, and its grin in another. The Cat is a photon, while the grin is its circular polarization.”
The newfound phenomenon was one in which certain particle features take a different path from their particle—much like the smile of the Cheshire Cat in Alice’s Adventures in Wonderland, written by Lewis Carroll—a pen name of mathematician Charles Lutwidge Dodgson—and published in 1865. To date, several experiments have demonstrated this curious quantum effect. But the idea has also drawn significant skepticism. Critics are less concerned about the theoretical calculations or experimental rigor than they are about the interpretation of the evidence. “It seems a bit bold to me to talk about disembodied transmission,” says physicist Holger Hofmann of Hiroshima University in Japan. “Instead we should revise our idea of particles.”
Recently researchers led by Yakir Aharonov of Chapman University took the debate to the next level. Aharonov was a co-author of the first paper to propose the quantum Cheshire effect. Now, on the preprint server arXiv.org, he and his colleagues have posted a description of theoretical work that they believe demonstrates that quantum properties can move without any particles at all—like a disembodied grin flitting through the world and influencing its surroundings—in ways that bypass the critical concerns raised in the past.
A GRIN WITHOUT A CAT
Aharonov and his colleagues first encountered their quantum Cheshire cat several years ago as they were pondering one of the most fundamental principles of quantum mechanics: nothing can be predicted unambiguously. Unlike classical physics, the same quantum mechanical experiment can have different outcomes under exactly the same conditions. It is therefore impossible to predict the exact outcome of a single experiment—only its outcome with a certain probability. “Nobody understands quantum mechanics. It’s so counterintuitive. We know its laws, but we are always surprised,” says Sandu Popescu, a physicist at the University of Bristol in England, who collaborated with Aharonov on the 2013 paper and the new preprint.
But Aharonov was not satisfied with this uncertainty. So, since the 1980s, he has been exploring ways to investigate fundamental processes despite the probability-based nature of quantum mechanics. Aharonov—now age 92—employs an approach that involves intensively repeating an experiment, grouping results and then examining what came out before and after the experiment and relating these events to each other. “To do this, you have to understand the flow of time in quantum mechanics,” Popescu explains. “We developed a completely new method to combine information from measurements before and after the experiment.”
The researchers have stumbled across several surprises with this method—including their theoretical Cheshire cat. Their idea sounds simple at first: send particles through an optical tool called an interferometer, which causes each particle to move through one of two paths that ultimately merge again at the end. If the setup and measurements were carried out skillfully, Aharonov and his colleagues theorized, it could be shown that the particle traveled a path in the interferometer that differed from the path of its polarization. In other words, they claimed the property of the particle could be measured on one path even though the particle itself took the other—as if the grin and the cat had come apart.
Inspired by this theory, a team led by Tobias Denkmayr, then at the Vienna University of Technology, implemented the experiment with neutrons in a study published in 2014. The team showed that the neutral particles inside an interferometer followed a different path from that of their spin, a quantum mechanical property of particles similar to angular momentum: Denkmayr and his colleagues had indeed found evidence of the Cheshire cat theory. Two years later researchers led by Maximilian Schlosshauer of the University of Portland successfully implemented the same experiment with photons. The scientists saw evidence that the light particles took a different path in the interferometer than their polarization did.
WEAK MEASUREMENTS AND ILLUSIONS
But not everyone is convinced. “Such a separation makes no sense at all. The location of a particle is itself a property of the particle,” Hofmann says. “It would be more accurate to talk about an unusual correlation between location and polarization.” Last November Hofmann and his colleagues provided an alternative explanation based on widely known quantum mechanical effects.
And in another interpretation of the Cheshire cat results, Pablo Saldanha of the Federal University of Minas Gerais in Brazil and his colleagues argue that the findings can be explained with wave-particle duality. “If you take a different view, there are no paradoxes,” Saldanha says, “but all results can be explained with traditional quantum mechanics as simple interference effects.”
Much of the controversy surrounds the way in which particles’ properties and positions are detected in these experiments. Disturbing a particle could alter its quantum mechanical properties. For that reason, the photons or neutrons cannot be recorded inside the interferometer using an ordinary detector. Instead scientists must resort to a principle of weak measurementdeveloped by Aharonov in 1988. A weak measurement makes it possible to scan a particle very lightly without destroying its quantum state. This comes at a price, however: the weak measurement result is extremely inaccurate. (Thus, these experiments must be repeated many times over, to compensate for the fact that each individual measurement is highly uncertain.)
In the quantum Cheshire cat experiments, a weak measurement is made along a path in the interferometer, the paths then merge, and the emerging particles are measured with an ordinary detector. Along one path of the interferometer, a weak measurement of the particle’s position can be taken and, along the other, its spin. Using detectors, physicists can more definitively characterize the particles that traveled through the interferometer and potentially reconstruct what occurred during the particle’s journey. For example, only certain particles will appear in certain detectors, helping the physicists piece together which path their neutron or photon previously took. According to Aharonov, Popescu and their colleagues, the Cheshire cat experiments ultimately reveal that the particle’s position can be confirmed on one path even as its polarization or spin was measured on the other.
Melissa Thomas Baum/Buckyball Design; Source: “Observation of a Quantum Cheshire Cat in a Matter-Wave Interferometer Experiment,” by Tobias Denkmayr et al., in Nature Communications, Vol. 5, Article No. 4492; July 29, 2014
Saldanha and his co-authors assert that it is impossible to make claims about quantum systems in the past given their measurements in the present. In other words, the photons and neutrons measured in the final detectors cannot tell us much about their previous trajectory. Instead the wave functions of particles passing through the paths of the interferometer could overlap, which would make it impossible to trace which path a particle had taken. “Ultimately, the paradoxical behaviors are related to the wave-particle duality,” Saldanha says. But in the papers that report evidence of the quantum Cheshire cat, he asserts, the findings “are processed in a sophisticated way that obscures this simpler interpretation.”
Hofmann, meanwhile, has stressed that the results will differ if you measure the system in a different way. This phenomenon is well-known in quantum physics: if, for example, you first measure the speed of a particle and then its position, the result can be different than it would be if you first measured the position of the same particle and then its speed. He and his colleagues therefore contend that Aharonov and his team’s conclusions were correct in themselves—that the particle moved along one path and the polarization followed the other—but that such differing paths do not apply simultaneously.
As Hofmann’s co-author Jonte Hance, also at Hiroshima University, told New Scientist,“It only looks like [the particle and polarization are] separated because you’re measuring one of the properties in one place and the other property in the other place, but that doesn’t mean that the properties are in one place and the other place, that means that the actual measuring itself is affecting it in such a way that it looks like it’s in one place and the other place.”
A NEW WAY TO CATCH A CHESHIRE CAT?
But these critiques are “missing the point,” Popescu says. He agrees that the work and reasoning put forward by Saldanha and Hofmann’s respective groups are correct—but adds that the best way to test any interpretation is to generate testable predictions from each. “As I understand it, there is no direct way to make predictions based on them,” Popescu says in reference to these alternative explanations. “They kind of have a very old-fashioned way of looking at things: there are contradictions, so you stop doing the math.”
With their recent preprint paper, Aharonov and Popescu, together with physicist Daniel Collins of the University of Bristol, have now described how a particle’s spin can move completely independently of the particle itself—without employing a weak measurement. In their new experimental setup, a particle is located in the left half of an elongated two-part cylinder that is sealed at the outer edges. Because of a highly reflective wall in the middle, the particle has a vanishingly small probability of tunneling through to the right-hand side of the cylinder. In their paper, the researchers provide a proof that even if the particle remains in the left-hand area in almost all cases, it should still be possible to measure a transfer of the particle’s spin at the right-hand outer wall. “It’s amazing, isn’t it?” Collins says. “You think the particle has a spin and the spin should stay with the particle. But the spin crosses the box without the particle.”
In a new thought experiment designed to observe the quantum Cheshire cat, physicists would be able to measure the property of a particle in one of two chambers of an elongated cylinder despite the fact that the particle itself would be contained in the other chamber.
Amanda Montañez
This approach would address several of the critical concerns raised thus far. The physicists don't need weak measurements. Nor do they need to group their experimental results to draw temporal conclusions. (That being said, grouping results would still improve the measurements, given that the angular momentum of the wall itself cannot be determined unambiguously because of the Heisenberg uncertainty principle.) But in this scenario, the only physical principles involved are conservation laws, such as the conservation of energy or the conservation of momentum and angular momentum. Popescu and Collins explain that they hope other groups will implement the experiment to observe the effects in the laboratory.
The new work has piqued Hofmann’s interest. “The scenario is exciting because the interaction between polarization and particle motion produces a particularly strong quantum effect that clearly contradicts the particle picture,” he says.
But he still does not see this as proof of disembodied (particle-free) spin transfer. “For me, this means, above all, that it is wrong to assume a measurement-independent reality,” Hofmann says. Instead quantum mechanics allows a particle’s residence to extend to the right-hand region of the cylinder, even if a residence in the left-hand region seems logically compelling. “I think it is quite clear to Aharonov, Collins and Popescu that the space in front of the wall is not really empty,” he adds.
Saldanha, meanwhile, still sees the researchers as overcomplicating what could be explained as traditional quantum interference effects. When discussing the particle’s very low probability of entering the right-hand side of the experimental setup, he explains, “we have to be careful about a ‘vanishingly small probability’ when we refer to waves.” The wave function of the particle could also expand into the right-hand side of the setup and thus influence the angular momentum of the wall. “The same predictions can be made without such dramatic conclusions,” he says.
In response to these critiques, Popescu says, “This is of course another way of thinking about it. The question is whether this interpretation is useful.” Regardless of which interpretation of the events is correct, the quantum Cheshire cat could enable new technological applications. For example, it could be used to transfer information or energy without moving a physical particle—whether made of matter or light.
For Popescu, however, the fundamental questions of physics play a more important role. “It all started when we thought about how time propagates in quantum mechanics,” he says. “And suddenly we were able to discover something fundamental about the laws of conservation.”
This article originally appeared in Spektrum der Wissenschaft and was reproduced with permission.
