Fundamentally contradictory but inescapably joined dual attributes, wave and particle, remain a conceptually unsettling element at the heart of quantum mechanics. It was a career-long unanswered challenge for Bohr to rationalize quantum duality. The conceptual dilemma it presents has remained an open issue, a topic of continued discussion, ever since. Here we report the discovery of an experimentally manageable route to control the weirdness of duality. Ironically, entanglement, the other conceptually challenging weirdness of quantum theory, will be shown to be in control of duality. We establish a simple identity through which entanglement prescribes quantitatively the degree of duality, of combined waveness and particleness, that can be recorded in any one-quantum two-path coherence experiment.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. MOST WEIRD
In the 1920s, the joining of two incompatible words as a label, wave mechanics, was an accurate indication that the newly emerging physics would be very unusual. The identification of particles as waves was a contradiction of all previous experience and was the first counterintuitive weirdness of quantum theory to appear. One can argue that it is still the central weirdness. In the years after his invention of complementarity in 1928 , as an answer to this strange duality of de Broglie , Bohr was never satisfied that he had found the right resolution of its paradoxical character . Quantum duality remains as weird today as in the 1920s. Academic, philosophical, and historical discussion of the conceptual dilemma of duality has never stopped [4,5]. Attention to it by the physics community has moved through several stages of increasing sophistication, but after 90 years of reflection, there has been no resolution. Here we describe a resolution. It is noteworthy that our resolution also applies to quantum duality’s classical analog, which is the ray-wave dichotomy of optics. This is appropriate, since optical physics and single-particle quantum mechanics are both theories founded on coherence and interference in linear vector spaces, and they share the issues inherent in duality.
2. INTERFERENCE COHERENCE
The two-path experimental scenario leading to the first quantification of the coherence of light is well known. It was invented by Thomas Young for his famous demonstration of optical interference . It is still well suited for discussions of coherence because the specific context for a Young-type observation can be modified to expose different coherences. The role of context is illustrated in Fig. 1, but its importance is often not recognized . This is especially true in the discussion of so-called “hidden” coherences, a number of which have been reported and analyzed recently [8–14].
The basis for our analysis will be the classical optical field. The results are new, and we also report a (classical) tomographic examination that provides experimental confirmation. An optical field has more than 3 degrees of freedom, but 3 degrees of freedom allow us to utilize a sufficiently general theory of vector-mode coherence . Classical optics is a suitable ground for discussing these issues, but all results to be developed below have one-quantum counterparts, as should become clear.
3. DUALITY BALANCE
In 1979, after 50 years of discussion and debate about wave and particle interpretations of duality [3,4,9,16–31], Wootters and Zurek  drew attention to the fact that duality need not represent an absolute incompatibility. As they indicated, Young-type scenarios can allow an amount of particleness to be given up for an increase in waveness, and vice versa, implying a kind of “duality balance” between wave and particle (or wave and ray) interpretations.
Many theoretical examinations since 1979 confirm that both particleness and waveness are open to analysis and fully support the balance implied by Wootters and Zurek. There is agreement on a central result, an inequality that quantifies the “amount” of duality present, the sum of two commonly accepted markers, visibility for waveness and which-path distinguishability , for particleness. The multiply rederived and accepted quantification of their combination is this:
One naturally asks, if previous treatments have arrived at a quantification formula, what can remain to be found? Here we are pointing out that two oversights are built into the interpretation of inequalities such as Eq. (1). They are valid as they stand, but are still short of the complete story. To examine the oversights, and also prepare the way for our experimental setup, we first rederive Eq. (1) as a standard but incomplete quantification.
4. VECTOR MODE COHERENCE
Our field’s complex amplitude depends on 3 degrees of freedom: (i) its two-dimensional transverse coordinate , hereafter denoted simply as ; (ii) time ; and (iii) intrinsic polarization , which we will call by the shorthand term spin. The required field coherence function in the Young source plane (cf. Fig. 1) is made from the fields at the two slit locations, and . Those two fields have arbitrary amplitudes and unit spins: and . Thus, the two fields received at point on the detection screen are33], Section 3.1).
There are three detectable signals (intensities) at the screen, one from each slit separately and one from the combination of the two fields at . The total intensity is the sum of the individual intensities and the cross term33]
The visibility of the measured fringes has its usual expression, based on the extreme values of the phase cosine function, which leads to
The which-path distinguishability , i.e., the degree to which the light field at screen is coming from only one of the two slits, or , is taken to be given by the standard expression18]. By combining Eqs. (7) and (8), we recover Eq. (1), the agreed duality inequality:
5. CALLING OUT OVERSIGHTS
In a summary  of his attempts to get control of the duality concept, Bohr focused on two important criteria in any analysis (see ). These were exclusivity and completeness, and awareness of them guides attention to two oversights not previously called out, to the best of our knowledge. These oversights cause one to overlook weaknesses in the usual interpretations of Eq. (1).
First, one should note lack of completeness, because the Young scenario allows for the simultaneous vanishing of and , while a finite signal is nevertheless present. For example, if beams and are equally strong, the standard measure of distinguishability vanishes: . And if the equally strong beams are also oppositely polarized, there can be no interference and no fringes (cf. Fig. 1). Thus the absence of fringes can accompany the absence of distinguishability. A nonzero signal must be characterized by something more, in addition to and , a clear indication of incompleteness. Something is missing.
Another oversight is to fail to note that a recently identified coherence theorem connects and in a strict equality with degree of polarization (see ):
What is missing is attention to entanglement (quantum or classical ), confirming a suggestion made 20 years ago about quantum weirdness by Knight . Quantification of coherence in our own recent optical research [7,13,37–40] has exposed connections to entanglement. Other recent work has subjected coherence to quantification and examination of relations to entanglement [41–43]. The unifying quantum-classical aspect is the vector space foundation common to quantum states and classical wave field theories. We have already demonstrated the intimate connection of entanglement and polarization . They are constrained in Young-type two-slit configurations by an additional identity:44]. That is, and are Young-type perfect opposites. Gaining some of comes at the expense of losing an equal amount of , which is an unexpected possibility of rearrangement, since and have such different traditional origins and interpretations. As De Zela has correctly emphasized , this unusual kind of pairing is significant, not coincidental.
Now one immediately sees that by controlling via Eq. (11), entanglement automatically controls duality. That is, when we eliminate between Eqs. (10) and (11), the duality inequality becomes an equality:1)] into a tight equality, the identity, 13), nothing is extra, and nothing is missing. The addition of supplies the quantity that has been missing, both from the result [Eq. (1)] and from the earlier discussion, namely, entanglement, and in the amount needed via concurrence. Still, as a relation among three observables in a Young-type experiment, Eq. (13) is open to test. In the following section we provide the results of that test.
6. DESCRIPTION OF EXPERIMENT
Here we report tomographic observation of all three elements of Eq. (13): visibility, distinguishability, and entanglement. Our setup and laboratory results are sketched below. We engage a specific Young-analog context. By use of a monochromatic laser, the temporal dependence can be factored out completely, leaving2)–(6), and have shifted the meaning of the coefficients and in order to take the space-mode dependent functions and , which are the replacements here for and in Eqs. (2) and (3), to be unit-normalized. We will see that the important influence of the mutual coherence in the experiment will come from the two spin components, determined by their relative angle and phase , i.e., we will have 2, the entire scheme contains a preparation stage realized by the modified Mach–Zehnder interferometer (MZI) and a measurement stage with a tomography setup.
The preparation stage is designed to generate optical beams of the form given in Eq. (14). Specifically, a single-mode laser operating at 795 nm, with linear polarization, passes through a half-wave plate (HWP1) and changes into an arbitrary linearly polarized state . A polarizing beam splitter (PBS) splits the beam into and polarizations. The transmitted channel contains the signal in path . It then passes through a 50/50 beam splitter (BS1) that directs it to a mirror mounted on a translation stage (TS). The reflected channel (labeled ) of the PBS passes through a half-wave plate (HWP2) to become , where . The two spatial functions and obviously have no overlap. Then the light field is characterized exactly by Eq. (14).
Here the parameter ratio is controlled by HWP1 and is controlled by the combination of HWP2 in channel (in terms of ) and the TS in channel (in terms of phase ). The output 50/50 beam splitter (BS2) combines channels and to produce interference. The light beam then enters the measurement stage so that , , and can be registered.
Fringe visibility can be simply achieved by continuously registering the intensity at the output of BS2 while moving the TS. The systematic maximum visibility that is achievable for our MZI is 98.1%, obtained by producing equal intensities of the two channels and measuring interference intensities with a polarizer placed right after BS2. All measurements of visibilities in other arrangements are corrected by this systematic maximum. The which-way distinguishability can be obtained straightforwardly with intensity measurements by blocking one of the two channels, or . Measurement of entanglement, quantified by concurrence , between the spatial degree of freedom and polarization space , is realized by a tomography setup, as shown in Fig. 2. It is a joint measurement of the Stokes-like parameters in both degrees of freedom .
7. CONFIRMATION BY TOMOGRAPHY MEASUREMENT
Each degree of freedom has two effective “states,” i.e., , in the spatial domain and , in the spin polarization domain. In reality, the two paths may not be perfectly coherent, i.e., the contents of the modes are not perfectly pure. These cases result in a matrix description of the light beam in the basis , given as47].
Then, following the standard two-qubit tomography procedure , the output beams are measured jointly in the Stokes-like basis of both degrees of freedom to recover each matrix element in Eq. (16). In principle, one needs to project the light onto the following 36 joint bases, i.e., , six bases in each degree of freedom. The projections of light in the six spin polarization state bases are standard, by the use of quarter-wave and half-wave plates combined with PBSs, as shown in the tomography measurement section of the setup in Fig. 2.
The projection in the spatial basis is realized straightforwardly by blocking one of the two channels of the MZI. The projection onto the bases can be realized by leaving both channels open. BS2 is a unitary rotation operator of the two incoming paths. It operates as an effective projection with one output port in the basis and the other port in . Finally, the projection in the basis is achieved with the combination of BS2 and the TS that can introduce an additional phase in path .
With the above arrangements, one is able to realize measurements in all joint bases. Then the coherence matrices [Eq. (16)] for all 13 beams can be reconstructed in the standard basis , , , . The moduli of the measured matrix elements are shown, respectively, in Fig. 3.
The completed duality or complementarity identity in Eq. (13) calls to mind a unit sphere in a space with axes that are the independent degrees of freedom . The measured values, as well as the sum , are shown in Table 1, and they do fit on the surface of one octant of a unit sphere. This is shown in Fig. 4. One sees that the three-way duality identity [Eq. (13)] that connects visibility-distinguishability-entanglement is confirmed.
Our central result is, in the first case, the derivation of the equation , showing how entanglement controls the amount of duality available. Then the formulation of the identity in Eq. (13) is automatic. It was followed by its experimental validation. As we pointed out, previously proposed quantitative duality measures engaging single inequalities such as Eq. (1) cannot embody completeness or exclusivity. A fully developed examination [13,34,48,49] of polarization coherence (both ordinary and generalized) relative to entanglement (both quantum and classical), as in Eq. (11), has only more recently been attempted. Finally, although we suggest that represents the simplest possible closing of a 90-year discussion of duality, we do not conclude that physics is finished with complementarity. Recent work shows that it can also be formulated for more than one entity. The first steps, which have already been taken [12,18,20,26,27], open a window onto multi-entity coherences that are not yet well explored or even conventionally named.
We conclude with two remarks: (A) It is fascinating that the two elements of quantum theory most often said to be conceptually mysterious or weird, duality and entanglement, are intimately related, and in control of each other, as our classical derivation of Eq. (13) implies. (B) The fully quantum single-photon version of our derivation is easy to imagine and will be presented elsewhere . Its resolution of the duality paradox also shows that duality is controlled by entanglement (and vice versa), just as is described here.
National Science Foundation (NSF) (PHY-1203931, PHY-1505189, PHY-1539859); Army Research Office (ARO) (W911NF-16-1-0162); Office of Naval Research (ONR) (N00014-14-1-0260).
We are pleased to acknowledge communication with colleagues Iwo Bialynicki-Birula, Paul Brumer, Steven Cundiff, Justin Dressel, B.-G. Englert, Andrew Jordan, Peter Knight, Peter Milonni, Michael Raymer, Lukasz Rudnicki, and Juan P. Torres.
REFERENCES AND NOTES
1. Bohr’s multiply revised and delayed report of his address to an international Physics Congress at Como in 1927 appeared as “The Quantum Postulate and the Recent Development of Atomic Theory,” Nature121, 580–590 (1928). See also Bohr’s book Atomic Theory and the Description of Nature (Cambridge University, 1934), p. 10.
2. L. de Broglie, “Recherches sur la théorie des quanta,” Ann. Phys. 10, 22–128 (1925). English translation by A. F. Kracklauer, “On the Theory of Quanta,” available from aflb.ensmp.fr/LDB-oeuvres/De_Broglie_Kracklauer.pdf. [CrossRef]
3. Bohr was not the only one confused or frustrated. This is made clear in Jammer’s account of discussions following Bohr’s address at the Como conference, where he introduced complementarity to attendees including Born, Fermi, Heisenberg, Kramers, Pauli, Rosenfeld, von Neumann, and Wigner. See an overview by M. Jammer in The Conceptual Development of Quantum Mechanics (McGraw-Hill, 1966), p. 354. See also M. Jammer, The Philosophy of Quantum Mechanics (Wiley, 1974).
4. A. Whitaker, Einstein, Bohr and the Quantum Dilemma, 2nd ed. (Cambridge University, 2006).
5. D. Murdoch, Niels Bohr’s Philosophy of Physics (Cambridge University, 1987).
6. Th. Young, “An account of some cases of the production of colours, not hitherto described,” Philos. Trans. R. Soc. London 92, 387–397 (1802). [CrossRef]
7. J. H. Eberly, “Correlation, coherence and context,” Laser Phys. 26, 084004 (2016). [CrossRef]
8. K. H. Kagalwala, G. DiGiuseppe, A. F. Abouraddy, and B. E. A. Saleh, “Bell’s measure in classical optical coherence,” Nat. Photonics 7, 72–78 (2013). [CrossRef]
9. F. De Zela, “Relationship between the degree of polarization, indistinguishability, and entanglement,” Phys. Rev. A 89, 013845 (2014). [CrossRef]
10. J. H. Eberly, “Shimony-Wolf states and hidden coherences in classical light,” Contem. Phys. 56, 407–416 (2015). [CrossRef]
11. J. Svozilík, A. Vallés, J. Peřina Jr., and J. P. Torres, “Revealing hidden coherence in partially coherent light,” Phys. Rev. Lett. 115, 220501 (2015). [CrossRef]
12. W. F. Balthazar, C. E. R. Souza, D. P. Caetano, E. F. Galvão, J. A. O. Huguenin, and A. Z. Khoury, “Tripartite nonseparability in classical optics,” Opt. Lett. 41, 5797–5800 (2016). [CrossRef]
13. X.-F. Qian, T. Malhotra, A. N. Vamivakas, and J. H. Eberly, “Coherence constraints and the last hidden optical coherence,” Phys. Rev. Lett. 117, 153901 (2016). [CrossRef]
14. F. De Zela, “Hidden coherences and two-state systems,” Optica 5, 243–250 (2018). [CrossRef]
15. The term “vector mode” appears to have originated in the title of a study by Gori, et al.: F. Gori, M. Santarsiero, and R. Borghi, “Vector mode analysis of a Young interferometer,” Opt. Lett. 31, 858–860 (2006).
16. P. A. Schilpp, ed., Albert Einstein: Philosopher-Scientist (Library of the Living Philosophers, 1949).
17. L. Mandel, “Coherence and indistinguishability,” Opt. Lett. 16, 1882–1883 (1991). [CrossRef]
18. G. Jaeger, M. A. Horne, and A. Shimony, “Complementarity of one-particle and two-particle interference,” Phys. Rev. A 48, 1023–1027 (1993). [CrossRef]
19. B.-G. Englert, M. O. Scully, and H. Walther, “The duality in matter and light,” Sci. Am. 271, 86–92 (1994). [CrossRef]
20. G. Jaeger, A. Shimony, and L. Vaidman, “Two interferometric complementarities,” Phys. Rev. A 51, 54–67 (1995). [CrossRef]
21. B.-G. Englert, “Fringe visibility and which-way information: an inequality,” Phys. Rev. Lett. 77, 2154–2157 (1996). [CrossRef]
22. B.-G. Englert, M. O. Scully, and H. Walther, “Complementarity and uncertainty,” Nature 375, 367–368 (1995). [CrossRef]
23. S. Dürr, T. Nonn, and G. Rempe, “Origin of quantum-mechanical complementarity probed by a ‘which-way’ experiment in an atom interferometer,” Nature 395, 33–37 (1998). [CrossRef]
24. S. Dürr, T. Nonn, and G. Rempe, “Fringe visibility and which-way information in an atom interferometer,” Phys. Rev. Lett. 81, 5705–5709 (1998). [CrossRef]
25. B.-G. Englert, M. O. Scully, and H. Walther, “On mechanisms that enforce complementarity,” J. Mod. Opt. 47, 2213–2220 (2000). [CrossRef]
26. F. de Melo, S. P. Walborn, J. A. Bergou, and L. Davidovich, “Quantum nondemolition circuit for testing bipartite complementarity,” Phys. Rev. Lett. 98, 250501 (2007). [CrossRef]
27. M. Jakob and J. A. Bergou, “Quantitative complementarity relations in bipartite systems: entanglement as a physical reality,” Opt. Commun. 283, 827–830 (2010). [CrossRef]
28. M. Lahiri, “Wave-particle duality and polarization properties of light in single-photon interference experiments,” Phys. Rev. A 83, 045803 (2011). [CrossRef]
29. R. Menzel, D. Puhlmann, A. Heuer, and W. P. Schleich, “Wave-particle dualism and complementarity unraveled by a different mode,” Proc. Natl. Acad. Sci. USA 109, 9314–9319 (2012). [CrossRef]
30. H.-Y. Liu, J.-H. Huang, J.-R. Gao, M. S. Zubairy, and S.-Y. Zhu, “Relation between wave-particle duality and quantum uncertainty,” Phys. Rev. A 85, 022106 (2012). [CrossRef]
31. A. Norrman, K. Blomstedt, T. Setälä, and A. Friberg, “Complementarity and polarization modulation in photon interference,” Phys. Rev. Lett. 119, 040401 (2017). [CrossRef]
32. W. K. Wootters and W. H. Zurek, “Complementarity in the double-slit experiment: quantum nonseparability and a quantitative statement of Bohr’s principle,” Phys. Rev. D 19, 473–484 (1979). [CrossRef]
33. The formulation we have followed is closely modeled on Chap. 3 of Theory of Coherence and Polarization of Light, by E. Wolf (Cambridge University, 2007).
34. J. H. Eberly, X.-F. Qian, and A. N. Vamivakas, “Polarization coherence theorem,” Optica 4, 1113–1114 (2017). [CrossRef]
35. R. J. C. Spreeuw, “A classical analogy of entanglement,” Found. Phys. 28, 361–374 (1998), for a very early recognition that entanglement (nonseparability of superpositions of tensor products) also belongs to classical wave theory as well as to quantum physics. [CrossRef]
36. P. L. Knight, “Quantum mechanics: where the weirdness comes from,” Nature 395, 12–13 (1998). [CrossRef]
37. X.-F. Qian, B. Little, J. C. Howell, and J. H. Eberly, “Shifting the quantum-classical boundary: theory and experiment for statistically classical optical fields,” Optica 2, 611–615 (2015). [CrossRef]
38. X.-F. Qian, A. N. Vamivakas, and J. H. Eberly, “Emerging connections, classical and quantum optics,” Opt. Photon. News 28(10), 34–41 (2017). [CrossRef]
39. X.-F. Qian and J. H. Eberly, “Entanglement and classical polarization states,” Opt. Lett. 36, 4110–4112 (2011). [CrossRef]
40. J. H. Eberly, X.-F. Qian, A. Al Qasimi, H. Ali, M. A. Alonso, R. Gutiérrez-Cuevas, B. J. Little, J. C. Howell, T. Malhotra, and A. N. Vamivakas, “Quantum and classical optics—emerging links,” Phys. Scripta 91, 063003 (2016). [CrossRef]
41. T. Baumgratz, M. Cramer, and M. B. Plenio, “Quantifying coherence,” Phys. Rev. Lett. 113, 140401 (2014). [CrossRef]
42. A. Streltsov, U. Singh, H. S. Dhar, M. N. Bera, and G. Adesso, “Measuring quantum coherence with entanglement,” Phys. Rev. Lett. 115, 020403 (2015). [CrossRef]
43. A. Streltsov, G. Adesso, and M. B. Plenio, “Colloquium: quantum coherence as a resource,” Rev. Mod. Phys. 89, 041003 (2017). [CrossRef]
44. W. K. Wootters, “Entanglement of formation of an arbitrary state of two qubits,” Phys. Rev. Lett. 80, 2245–2248 (1998). Concurrence is designed for two-qubit entanglement, and so is particularly appropriate in qubit-analog contexts such as two-slit interference configurations. It quantifies entanglement (inseparability) and is strictly bounded, 0 ≤ C ≤ 1. [CrossRef]
45. F. De Zela, “Optical approach to concurrence and polarization,” Opt. Lett. 43, 2603–2606 (2018). [CrossRef]
46. D. F. V. James, P. G. Kwiat, W. J. Munro, and A. G. White, “On the measurement of qubits,” Phys. Rev. A 64, 052312 (2001). [CrossRef]
47. A. F. Abouraddy, K. H. Kagalwala, and B. E. A. Saleh, “Two-point optical coherency matrix tomography,” Opt. Lett. 39, 2411–2414 (2014). [CrossRef]
48. A. Luis, “Coherence, polarization and entanglement for classical light fields,” Opt. Commun. 282, 3665–3670 (2009). [CrossRef]
49. M. A. Alonso, X.-F. Qian, and J. H. Eberly, “Center-of-mass interpretation for bipartite purity analysis of N-party entanglement,” Phys. Rev. A 94, 030303 (2016). [CrossRef]
50. J. P. Torres, X.-F. Qian, S. K. Manikandan, A. N. Vamivakas, and J. H. Eberly, “Distinguishability, visibility and quantum entanglement,” (in preparation).