The laws of quantum physics govern the behavior of identical particles via the symmetry of the wave function, as dictated by the Pauli principle (). In particular, it has been known since Bose and Einstein () that the symmetry of the bosonic wave function favors the so-called bunching of identical bosons. A striking demonstration of bosonic statistics for a pair of identical bosons was achieved in 1987 in a seminal experiment by Hong, Ou, and Mandel (HOM) (), who observed the cancellation of coincident detections behind a 50:50 beam splitter (BS) when two indistinguishable photons impinge on its two input ports (Fig. 1A). This HOM effect follows from the destructive two-photon interference between the probability amplitudes for double transmission and double reflection at the BS (Fig. 1B). Together with the Hanbury Brown and Twiss effect (, ) and the violation of Bell inequalities (, ), it is often viewed as the most prominent genuinely quantum feature: it highlights the singularity of two-particle quantum interference as it cannot be understood in terms of classical wave interference (, ). It has been verified in numerous experiments over the last 30 y (see, e.g., refs. 1112–), even in case the single photons are simultaneously emitted by two independent sources (15–) or within a silicon photonic chip (, ). Remarkably, it has even been experimentally observed with He metastable atoms, demonstrating that this two-boson mechanism encompasses both light and matter ().
Here, we explore how two-boson quantum interference transforms under reversal of the arrow of time in one of the two bosonic modes (Fig. 2A). This operation, which we dub partial time reversal (PTR), is unphysical but nevertheless central as it allows us to exhibit a duality between the linear optical coupling effected by a BS and the nonlinear optical (Bogoliubov) transformation effected by a parametric amplifier. As a striking implication of these considerations, we predict a two-photon interferometric effect in a parametric amplifier of gain 2 (which is dual to a BS of transmittance 1/2). We argue that this unsuspected effect originates from the indistinguishability between photons from the past and future, which we coin “time-like” indistinguishability as it is the partial time-reversed version of the usual “space-like” indistinguishability that is at work in the HOM effect.
Since Bogoliubov transformations are ubiquitous in quantum physics, it is expected that this two-boson interference effect in time could serve as a test bed for a wide range of bosonic transformations. Furthermore, from a deeper viewpoint, it would be fascinating to demonstrate the consequence of time-like indistinguishability in a photonic or atomic platform as it would help in elucidating some heretofore overlooked fundamental property of identical quantum particles.
Hong–Ou–Mandel Effect
The HOM effect is a landmark in quantum optics as it is the most spectacular manifestation of boson bunching. It is a two-photon intrinsically quantum interference effect where the probability amplitude of both photons being transmitted cancels out the probability amplitude of both photons being reflected. A 50:50 BS effects the single-photon transformations (for details, see Materials and Methods, Gaussian Unitaries for a BS and PDC)
where
a and
b label the bosonic modes and
1a/b stands for a single-photon state in mode
a/b (here, “trans” stands for transmitted and “ref” for reflected). Hence, the state of two indistinguishable photons (one in each mode) transforms as
since the double-transmission term
1a1b cancels out the double-reflection term
1b1a. More generally, the probability for coincident detections is given by (for details, see
Materials and Methods, Two-Photon Interference in a BS and PDC)
where
UηBS denotes the unitary corresponding to a BS of transmittance
η. In a nutshell, the HOM effect boils down to
1,1U1/2BS1,1=0. Its experimental manifestation is the presence of a dip in the correlation function, witnessing that two photons cannot be coincidently detected at the two output ports when
η=1/2 (
Fig. 1C).
Partial Time Reversal
Bogoliubov transformations on two bosonic modes comprise passive and active transformations. The BS is the fundamental passive transformation, while parametric down conversion (PDC) gives rise to the class of active transformations (also called nondegenerate parametric amplification). Although the involved physics is quite different (a simple piece of glass makes a BS, while an optically pumped nonlinear crystal is needed to effect PDC), the Hamiltonians generating these two unitaries are amazingly close, namely
where
â and
b^ are mode operators. It is striking that a simple swap
b^↔b^† transforms
HBS into
HPDC, suggesting a deep duality between a BS and a PDC by reversing the arrow of time in mode
b^ (keeping mode
â untouched).
The underlying concept of PTR will be formalized in , but we first illustrate this duality between a BS and PDC with the simple example of Fig. 2A, where n photons impinge on port b^ of a BS (with vacuum on port â), resulting in the binomial output state
with
η=cos2θ (see
Gaussian Unitaries for a BS and PDC). The path where all photons are reflected (
k=n) is associated with the transition probability amplitude
n,0UηBS0,n=sinnθ. Reversing the arrow of time on mode
b^ (
Fig. 2A) leads us to consider the transition probability amplitude for a PDC of gain
g=cosh2r with vacuum state on its two inputs and
n photon pairs on its outputs, that is,
n,nUgPDC0,0=tanhnr/coshr (see
Gaussian Unitaries for a BS and PDC). Strikingly, the above two amplitudes can be made equal (up to a constant
coshr) if we set
sinθ=tanhr or equivalently,
η=1/g. Thus, we have
The case of
m photons (instead of vacuum) impinging on port
â and
n photons impinging on port
b^ is illustrated in
Materials and Methods, Example of PTR. Conditioning again the output port
b^′ on vacuum and reversing the arrow of time, we obtain the same output as for a PDC of gain
g=1/η with input state
m,0.
These examples reflect the existence of a general duality between a BS and PDC. Indeed, as demonstrated in Materials and Methods, Proof of PTR Duality, partial transposition in Fock basis gives rise to PTR duality
where indices
j and
m have been swapped (this can alternatively be expressed as or ). The PTR duality is nicely evidenced by the conservation rules exhibited by the BS and PDC transformations: the former conserves the total photon number, while the latter conserves the difference between the photon numbers. In , the only nonzero matrix elements for a BS are those satisfying
i+j=n+m. This directly implies that the only nonzero matrix elements for a PDC satisfy
i−m=n−j, as expected.
The notion of time reversal can be conveniently interpreted using the so-called “retrodictive” picture of quantum mechanics (). Along this line, PTR must be understood here as the fact that the “retrodicted” state of mode b^ propagates backward in time, while the state of mode â normally propagates forward in time (this is made precise in Materials and Methods, Retrodictive Picture of Quantum Mechanics). As shown in Fig. 2B, the PTR duality can be made operational by sending half of a so-called Einstein–Podoslky–Rosen (EPR) entangled state on mode b^, so that we access the output retrodicted state on the second entangled mode b^′′.
Two-Boson Interference in an Amplifier
Due to this duality, the HOM effect for a BS of transmittance 1/2 immediately suggests the possible existence of a related interferometric suppression effect in a PDC of gain 2, namely 1,1U2PDC1,1=0. This striking prediction can indeed be verified by examining the state at the output of a PDC of gain g=2 (see Two-Photon Interference in a BS and PDC):
As it appears, the component with a single photon on each output mode (
n=1) is indeed suppressed, which is reminiscent of the HOM effect. However, the structure of
Ψ is more complicated here as the PDC (unlike the BS) does not conserve the total photon number; hence, we observe terms with
n≥2 photons on each mode. The distribution of the photon pair number is illustrated in
Fig. 3A.
Fig. 3.
(A) Probability Pn of observing n photon pairs at the output of a PDC of gain g=2 when a single photon impinges on both the signal and idler input ports. The output state U2PDC1,1=12−0,0+14 2,2+12 3,3
+916 4,4+12 5,5+⋯ has a vanishing 1,1 component, owing from two-photon interferometric suppression. The significant components are the vacuum as well as the terms with 2 to ∼10 pairs (the next terms quickly decay to zero). (B) Corresponding distributions of the photon pair number for a gain g=3 (showing a dip at n=2) and g=4 (showing a dip at n=3). The distributions are shown as continuous curves in order to guide the eye, but only integer values of n are relevant. The curve for g=2 is also plotted for comparison.
The dependence of the probability of detecting a single pair (n=1) on the gain g of PDC is given by (see Two-Photon Interference in a BS and PDC)
confirming that the probability for coincident detections vanishes if the gain
g=2. Note that if we substitute
η by
1/g in and divide by
g, we get exactly , as implied by PTR duality. More generally, we show in
Materials and Methods, Extension to a PDC with Integer Gain that this interferometric suppression effect actually extends to any larger integer value of the gain (e.g.,
g=3, 4,⋯ ). As illustrated in
Fig. 3B, the probability of detecting
n photons simultaneously on each output port vanishes when the gain
g=n+1. Again, this is the consequence of PTR duality applied to an extended HOM effect.
Discussion and Conclusion
The role of time reversal in quantum physics has long been a fascinating subject of questioning (see, e.g., ref. and references therein), but the key idea of the present work is to consider a bipartite quantum system (two bosonic modes) with counterpropagating times. Incidentally, we note that the notion of time reversal has been exploited in the context of defining separability criteria (, ), but this seems to be unrelated to PTR duality. Further, the link between time reversal and optical-phase conjugation has been mentioned in the quantum optics literature (see, e.g., ref. ), but it exploits the fact that the complex conjugate of an electromagnetic wave is the time-reversed solution of the wave equation (the phase conjugation time-reversal mirror concerns one mode only). The PTR duality introduced here bears some resemblance with an early model of lasers () based on the coupling of an “inverted” harmonic oscillator (having a negative frequency ω) with a heat bath. The inverted harmonic oscillator (e−iωt→eiωt) can indeed be viewed as a time-reversed harmonic oscillator. Quantum amplification in this model occurs from a PDC-like coupling of this inverted harmonic oscillator, whereas quantum damping follows from the BS-like coupling of a usual harmonic oscillator with the bath. The PTR duality is also reminiscent of Klyshko’s so-called “advanced-wave picture” in PDC (), which provides an interpretation of coincidence-based two-photon experiments: the wave that is detected by one of the detectors behind PDC can be viewed as resulting from an “advanced wave” emitted by the second detector, so that PDC acts as a mirror (). This picture may indeed be interpreted as a special case of , namely 0,0UgPDCψ,ϕ=g−1/20,ϕ*U1/gBSψ,0. In the limit where the gain g→∞, the two outputs of PDC can be viewed as the input ψ and output ϕ* of a fully reflecting (phase-conjugating) mirror with η→0.
In this work, we have promoted PTR as the proper way to approach the duality between passive and active bosonic transformations. As a compelling application of PTR duality, we have unveiled a hitherto unknown quantum interference effect, which is a manifestation of quantum indistinguishability for identical bosons in active transformations (space-like indistinguishability, which is at the root of the HOM effect, transforms under PTR into time-like indistinguishability). The interferometric suppression of the coincident 1,1 term is induced by the indistinguishability between a photon pair originating from the past and a photon pair going to the future. Stated more dramatically, while the two photons may cross the amplification medium and be detected, the sole fact that they could instead be annihilated and replaced by two other photons makes the detection probability drop to zero when g=2.
The experimental verification of this effect can be envisioned with present technologies (see Experimental Scheme). A coincidence probability lower than 25% would be sufficient to rule out a classical interpretation, which could in principle be reached with a moderate gain of 1.28 (see Classical Baseline). Observing time-like two-photon interference in experiments involving active optical components would then be a highly valuable metrology tool given that the HOM dip is commonly used today as a method to benchmark the reliability of single-particle sources and mode matching. More generally, the interference of many photons scattered over many modes in a linear optical network has generated a tremendous interest in the recent years, given the connection with the “boson sampling” problem [i.e., the hardness of computing the permanent of a random matrix ()], and technological progress in integrated optics now makes it possible to access large optical circuits (see, e.g., ref. ). In this context, it would be exciting to uncover new consequences of PTR duality and time-like interference.
Finally, we emphasize that our analysis encompasses all bosonic Bogoliubov transformations, which are widespread in physics, appearing in quantum optics, quantum field theory, or solid-state physics, but also in black hole physics or even in the Unruh effect (describing an accelerating reference frame). This suggests that time-like quantum interference may occur in various physical situations where identical bosons participate in such a transformation. Beyond bosons, let us point out an intriguing connection with the notion of “crossing” in quantum electrodynamics (, ). Crossing symmetry refers to a substitution rule connecting two scattering matrix elements that are related by a Wick rotation (antiparticles being turned into particles going backward in time). For example, the scattering of a photon by an electron (Compton scattering) and the creation of an electron–positron pair by two photons are processes that are related to each other by such a substitution rule (see, e.g., ref. ). This is in many senses analogous to the PTR duality described here: since a photon (or truly neutral boson) is its own antiparticle, we may view PTR duality as a substitution rule connecting the BS diagram to the PDC diagram. We hope that this connection with quantum electrodynamics may open up even broader perspectives.
Materials and Methods
Gaussian Unitaries for a BS and PDC.
Passive and active Gaussian unitaries are effected by linear optical interferometry or parametric amplification, respectively (). The fundamental passive two-mode Gaussian unitary, namely the BS unitary
effects an energy-conserving linear coupling between modes
â and
b^ and acts in the Heisenberg picture as
where
â and
b^ are the mode operators and
η is the transmittance (
0≤η≤1). Similarly, the unitary
models the generation of pairs of entangled photons by PDC due to the optical pumping of a nonlinear crystal. It transforms the mode operators according to the Bogoliubov transformation
where
g is the parametric gain (
g≥1) and
r is the squeezing parameter. The photon number difference is conserved by the PDC transformation, namely
â′†â′−b^′†b^′=â†â−b^†b^.
The action of UηBS on Fock states can be expressed by using the decomposition of the exponential
Alternatively, it can easily be computed by using the canonical transformation, . For example, when
n photons impinge on one of the input ports, each photon may be transmitted or reflected, so we get the binomial state
or
The action of
UgPDC on Fock states can be conveniently expressed by use of the disentanglement formula,
For example, the two-mode squeezed vacuum state is obtained by applying
UgPDC onto the vacuum state, namely
where the
nth term in the right-hand side corresponds to the stimulated emission of
n photon pairs. In the more general case where
m photons are impinging into mode
â (with vacuum in the other input mode), we have
where each term corresponds again to the stimulated emission of
n photon pairs, with
m extra photons traveling in mode
â.
Example of PTR.
We illustrate the PTR duality between a BS and PDC by considering the additional example of a BS with m photons impinging on input port â and n photons impinging on input port b^ (Fig. 5A). If we condition on the vacuum on mode b^′ at the output of the BS, the (unnormalized) conditional output state of mode â′ is
Hence, we have the probability amplitude
The special case
m=0 is considered in the text (
Fig. 2) and corresponds to
ϕn=sinn θ n. By reversing the arrow of time on mode
b^, we compare the probability amplitude of with the probability amplitude
for a PDC of gain
g (). Now, if we make the substitution
sinθ=tanhr and
cosθ=(coshr)−1, which is equivalent to
g=1/η, we confirm that and are dual under PTR, namely
As shown in
Fig. 5B, if the input mode
b^ is entangled (that is, if we use the entangled state
Ψb,b′′∝∑n=0∞n,n), we get the output state
Ψa′,b′′∝∑n=0∞ϕn,n, which is precisely proportional to the output of a PDC when the signal and idler modes are initially in states
m and
0, respectively ().
Fig. 5.
(A) PTR duality in a more general case with m and n photons impinging on modes â and b^ of a BS. By conditioning the output mode b^′ on vacuum and reversing the arrow of time, we get the same transition probability amplitude (up to a constant) as for a PDC of gain g=1/η with input state m,0 and output state n+m,n. (B) Corresponding operational scheme using an entangled (EPR) state at the input of mode b^. This makes it possible to access the output retrodicted state on mode b^′′.
Proof of PTR Duality.
The PTR duality is illustrated in Table 1 for few photons. As expressed by , it can be viewed as a consequence of partial transposition of the state of mode b^ (leaving mode â unchanged), namely the fundamental relation
where
Tb stands for transposition in the Fock basis of the Hilbert space associated with mode
b^. This is sketched in
Fig. 6A and can also be interpreted by comparing the unitaries
UηBS and
UgPDC in and or their corresponding decompositions, and . In general terms, we may say that the (passive) linear coupling of two bosonic modes is dual, under PTR, to an (active) Bogoliubov transformation, which is expressed as
for any states
ψa,
ψb,
ϕa, and
ϕb, where
* denotes the complex conjugation in the Fock basis. In
Fig. 6B, the corresponding operational scheme is depicted, relying on the preparation of an entangled (EPR) state at the input of mode
b^ and the projection onto an entangled (EPR) state at the output of mode
b^′.
Illustration of PTR duality with few photons
| BS | PDC |
| 0,0UηBS0,0=1 | 0,0UgPDC0,0=1/g |
| 1,0UηBS1,0=η | 1,0UgPDC1,0=1/g |
| 0,1UηBS0,1=η | 0,1UgPDC0,1=1/g |
| 0,1UηBS1,0=−1−η | 0,0UgPDC1,1=−g−1/g |
| 1,0UηBS0,1=1−η | 1,1UgPDC0,0=g−1/g |
Fig. 6.
(A) General statement of the PTR duality as expressed in , when i and j photons impinge on input modes â and b^ of a BS, while n and m photons are detected in output modes â′ and b^′. We recover a PDC with input i,m and output n,j. (B) Corresponding operational scheme, where a PDC is emulated with a BS. The input state ψa of mode â simply propagates through the BS and is projected onto ϕa. The input state ψb of mode b^‴ is converted, by projection on an EPR state, into the retrodicted state ψb of mode b^′, which back propagates through the BS and is projected onto ϕb in mode b^. We obtain the corresponding output state ϕb on mode b^′′ by using an EPR pair. We thus recover a PDC with input ψa,ψb and output ϕa,ϕb, as encapsulated by .
We now prove PTR duality by reexpressing in the Heisenberg picture, namely
where
V≡UgPDC. Note that
â and
b^ are real operators in Fock basis; hence, the same is true for
V. Therefore, we have
â†=aT,
b^†=bT, and
V†=VT, where
T stands for transposition in Fock basis. Obviously, we have
(VTb)T=VTa. By using the identity
((A⊗B)M)Tb=(A⊗1)MTb(1⊗BT), where
1 represents the identity operator, we express
âPTR≡VTa â VTb as
Equivalently, using
((A⊗B)M)Ta=(1⊗B)MTa(AT⊗1), we may also reexpress
âPTR as
Defining the operators
b^PTR≡VTa b^ VTb,
W≡VTaVTb, and identifying with , we see that
We can perform a similar development starting from
b^PTR, resulting in
Now, solving and for
âPTR and
b^PTR, we get
where we have made the substitutions
coshr=(cosθ)−1 and
sinhr=tanθ.
Similar equations can be derived starting from VTa ↠VTb and VTa b^† VTb, which imply that the operator W also commutes with ↠and b^†. Hence, W is a scalar (proportional to 1), and it is sufficient to compute its diagonal matrix element in an arbitrary state (e.g., 0,0). Recalling that V†=VT, we have W=(V†)TbVTb, so that
Substituting
W with
1/g in , we get , which concludes the proof of .
Note that PTR duality can be reexpressed by using the identity
which is valid for any joint unitary
Uab and for any operators
X^a(b) and
Ŷa(b) acting on mode
a (
b). Plugging
Uab=UgPDC into and using implies the general relation
This equation is needed to interpret PTR in the context of the retrodictive picture of quantum mechanics.
Retrodictive Picture of Quantum Mechanics.
In the usual, predictive approach of quantum mechanics, one deals with the preparation of a quantum system followed by its time evolution and ultimately, its measurement. Specifically, one uses the prior knowledge on the state ρ^j (prepared with probability pj) in order to make predictions about the outcomes of a measurement {Π^m}. If the state evolves according to unitary U before being measured, Born’s rule provides the conditional probabilities P(m|j)=Tr[Uρ^jU†Π^m]. In contrast, in the retrodictive approach of quantum mechanics (), one postselects the instances where a particular measurement outcome m was observed, and one focuses on the probability of the preparation variable j conditionally on this measurement outcome. This can be interpreted as if the actually measured state had propagated backward in time to the preparer (Fig. 7A). Specifically, one associates a retrodicted state σ^m with the observed outcome m and makes retrodictions about the preparation by evolving σ^m according to U† and applying a measurement {Θ^j}, whose outcome j discriminates the prepared state ρ^j. In the simplest case (to which we restrict here) where there is no a priori information about the source, one sets ∑jpj ρ^j∝1. Then, by defining
one recovers the expected properties of a state (
σ^m≥0,
Tr σ^m=1) and of a measurement (
Θ^j≥0,
∑jΘ^j=1). Now, applying Born’s rule to the retrodicted state
σ^m having evolved according to
U† followed by a measurement
Θ^j, we get
which is consistent with Born’s rule in the forward direction combined with Bayes’ rule.
Fig. 7.
(A) Predictive (Left) and retrodictive (Right) pictures, describing the same experiment where state ρ^j is prepared, evolves according to U, and leads to measurement outcome associated with Π^m. The probability P(j|m) can be written as resulting from the retrodictive state σ^m back propagating according to U†, followed by measurement Θ^j. (B) Predictive (Left) and intermediate (Right) pictures for a bipartite system. In the latter picture, associated with PTR, the retrodictive state of subsystem b propagates backward in time, while the predictive state of subsystem a propagates forward in time. If Uab is a unitary, Vab=UabTb is not necessarily proportional to a unitary; however, it is the case when considering the BS vs. PDC duality.
The retrodictive picture can be successfully exploited in different situations [for example, to characterize the quantum properties of an optical measurement device ()], but it is always used in lieu of the predictive picture. Here, we instead combine it with the predictive picture in order to properly define PTR duality and describe a composite system that is propagated partly forward and partly backward in time, as represented in Fig. 7 B, Right. Specifically, we consider a composite system prepared in a product state ρ^ia⊗ρ^jb, then undergoing a unitary evolution Uab followed by a product measurement {Π^na⊗Π^mb}. In the fully predictive picture (Fig. 7 B, Left), the conditional probabilities are given by
In our intermediate picture, we postselect the instances where a particular measurement outcome
m was observed in subsystem
b when subsystem
a was prepared in state
ρ^ia and consider the joint probability of the preparation variable
j of subsystem
b together with the measurement outcome
n of subsystem
a. Bayes’ rule yields
where
pjb is the probability that subsystem
b is prepared in state
ρ^jb. Similarly as before, without information about the source, we set
∑jpjb ρ^jb∝1. We associate a retrodicted state
σ^mb with the observed outcome
m on subsystem
b, while subsystem
a is still considered in the initial state
ρ^ia. Similarly, we make retrodictions about the preparation of subsystem
b by applying measurement
{Θ^jb}, whose outcome
j discriminates the prepared state
ρ^jb, while still measuring subsystem
a according to
{Π^na}. Using identity () and defining
which are easily seen to behave as a proper state and measurement, we may reexpress as
where
Vab≡UabTb. This can be viewed as the evolution of state
ρ^ia⊗σ^mb according to
Vab, followed by the measurement of
Π^na⊗Θ^jb. In other words, in , the predictive picture is used for subsystem
a, while the retrodictive picture is used for subsystem
b.
In our analysis of a BS under PTR, we have Uab=UgPDC and Vab=U1/gBS/g, so that reduces to . Hence, Vab is unitary (up to a constant) and can be interpreted as the propagation of the retrodicted state of mode b^ backward in time through the BS, while the predictive state of mode â normally propagates forward in time through the BS. According to , the joint state is then shown to evolve according to a PDC. Note that it is not always possible to construct an operator Vab that is proportional to a unitary operator, as it is the case here.
Two-Photon Interference in a BS and PDC.
The HOM effect can be simply understood by calculating the probability amplitude for coincident detection
at the output of a BS. By using , it is simple to rewrite it as
c=cos2θ, which yields the coincidence probability
If the transmittance
η=1/2, no coincident detections can be observed as a result of destructive interference.
Now, we examine the corresponding quantum interferometric suppression in a PDC and its dependence in the parametric gain g. Let us calculate the probability amplitude for coincident detection
where
V is a short-hand notation for
UgPDC. Thus, we get the scalar product between
V†0,0, which is the state of up to changing the sign of
r, and the ket
V†âVV†b^Vâ†b^†0,0, which can be reexpressed as
by use of . This gives
c′=(1−sinh2r)/cosh3r, so that the probability for coincidence is written as
If the gain
g=2, the probability for coincident detections fully vanishes. More generally, the joint state at the output of a PDC of arbitrary gain
g when the input state is
1,1 reads
A parametric gain
g=2 corresponds to
coshr=2 and
sinhr=1, so we recover .
Extension to a PDC with Integer Gain.
We may also consider the case where the gain takes a larger integer value (e.g., g=3, 4,⋯ ). A closer look at reveals that the output term with g−1 photon pairs fully vanishes when g is an integer. The corresponding distribution of the output photon pair number
is displayed in
Fig. 3B for
g=3 and 4. This interferometric suppression
n,nUn+1PDC1,1=0,
∀n, can be interpreted as dual, under PTR, to the extended HOM effect
n,1U1/(n+1)BS1,n=0 for a BS of transmittance
1/(n+1) (
Fig. 8,
Left). The latter effect is easy to understand as the interference between the amplitude with all
n+1 photons being reflected and the amplitude with one photon of each mode being transmitted (the other
n−1 being reflected). Indeed, the operator
â†(b^†)n transforms into
The term proportional to
(a′†)n b′† vanishes as a result of the cancellation of the term where the photon on mode
â is reflected and the
n photons on mode
b^ are reflected, together with the term where the photon on mode
â is transmitted and one of the photons on mode
b^ is transmitted (the other
n−1 photons being reflected). This is sketched in
Fig. 8,
Right. More generally, the transition probability
1,n→n,1 for a BS of transmittance
η is given by
which is consistent with under PTR, namely
Fig. 8.
Extended quantum interferometric suppression in an amplifier where the detection of n photon pairs at the output is suppressed if the gain g=n+1. This is the PTR dual of the extended HOM effect when a single photon and n photons impinge on the two input ports of an unbalanced BS of transmittance η=1/(n+1). The interference at play in the extended HOM is again the cancellation between a double reflection and double transmission (with n−1 other photons being always reflected from mode b^ to mode â). Applying PTR, this translates into the cancellation between the amplitude for the two photons crossing the crystal and the amplitude for the stimulated annihilation combined with stimulated emission of the two photons (accompanied, in both cases, with the stimulated emission of n−1 pairs). In other words, the input pair may be up converted into the pump beam while n pump photons are down converted, or the input pair may simply be transmitted while n−1 pump photons are down converted. Since the two scenarios are time-like indistinguishable, they interfere destructively.
Experimental Scheme.
The HOM effect is considered a most spectacular evidence of genuinely quantum two-boson interference, and we expect the same for its PTR counterpart as it admits no classical interpretation. The experimental verification of our effect can be envisioned with present technologies, as sketched in Fig. 9. We would need two single-photon sources, which could be heralded by the detection of a trigger photon at the output of a PDC with low gain (the single photon being prepared conditionally on the detection of the trigger photon in the twin beam). The two single photons would impinge on a PDC of gain 2, whose output modes should be monitored: the probability of detecting exactly one photon on each mode should be suppressed as a consequence of time-like indistinguishability. In principle, photon number resolution would be needed in order to discriminate the output term with one photon pair (n = 1) from the terms with more pairs (n≥ 2). The ability of counting photons has become increasingly available over the last years (e.g., exploiting superconducting detectors), but this could also be achieved by splitting each of the two output modes into several modes followed by an array of on/off photodetectors. Experiments involving PDC in three coherently pumped crystals have already been achieved recently (, ), aiming at observing induced decoherence, so the proposed setup here should be implementable along the same lines. The squeezing needed to reach a gain 2 amounts to 7.66 dB, which is high but lies in the range of experimentally accessible values (in the continuous-wave regime). The experiment could alternatively be carried out with a lower gain (especially in the pulsed regime) provided the observed dip is sufficient to rule out a classical interpretation. As a matter of fact, a coincidence probability lower than 1/4 would be needed (see Classical Baseline), which could in principle be reached with a gain of 1.28 (i.e., a squeezing of 4.39 dB). Of course, the effect of losses should also be carefully analyzed in order to assess the feasibility of the scheme depicted in Fig. 9.
Fig. 9.
Schematic of a potential demonstration of two-photon quantum interference in the amplification of light with gain 2. Two heralded single-photon sources (exploiting an avalanche photodetector [APD]) are used to feed the signal and idler modes of a PDC, and two photon number-resolving (PNR) detectors are used to monitor the presence of a single photon in each output port. The two-photon interference effect would be demonstrated by measuring a depletion of fourfold coincidences between the two trigger photons (heralding the preparation of two single photons) and the two output photons (here, the detectors should filter out the output states with ≥2 photons on each mode). The dominant terms will then consist of the stimulated annihilation of the two input photons (witnessed by two trigger photons but no output photons) as well as the stimulated emission of a photon pair (witnessed by two output photons but no trigger photons). When the time lapse between the detection of the trigger and output photons is close to zero (which means a perfect match of the timing of the output photons originating from the input photons associated with the trigger photons) the two terms should interfere destructively.
Demonstrating this effect would be invaluable in view of the fact that the HOM dip is widely used to test the indistinguishability of single photons and to benchmark mode matching: it witnesses the fact that the photons are truly indistinguishable (they admit the same polarization and couple to the same spatiotemporal mode). For example, HOM experiments have been used to test the indistinguishability of single photons emitted by a semiconductor quantum dot in a microcavity (), while the interference of two single photons emitted by two independently trapped rubidium-87 atoms has been used as an evidence of their indistinguishability (). The HOM effect has also been generalized to three-photon interference in a three-mode optical mixer (), while the case of many photons in two modes has been analyzed in ref. , implying a possible application of the quantum Kravchuk–Fourier transform (). We anticipate that most of these ideas could extend to interferences in an active optical medium.
Classical Baseline.
The two-photon quantum interference effect in amplification cannot be interpreted within a classical model of PDC, where a pair can be annihilated or created with some probability. We have two possible indistinguishable paths (the photon pair either going through the crystal or being replaced by another one) with equal individual probabilities but opposite probability amplitudes; hence, the resulting probability vanishes (whereas the two probabilities would add for classical particles). In order to assess an experimental verification of this effect, it is necessary to establish a classical baseline, namely to determine the depletion of the probability of coincident detections that can be interpreted classically. As a guide, consider first a classical model of the HOM effect where the two input photons are distinguishable. We have to add the double-transmission probability |Adt|2 with the double-reflection probability |Adr|2 since these two paths can be distinguished. Then, the classical probability for coincident detections is
to be compared with
Pcoinc(η) of . For a 50:50 BS,
Pcl(1/2)=1/2; hence, a depletion below 50% ensures that that the dip is quantum. Similarly, in a classical model of PDC, we can distinguish the path where the two input photons are transmitted (probability
|Adt′|2) from the path where they are replaced by another pair (probability
|Ast′|2). Thus, the classical probability for coincident detections is
to be compared with
Pcoinc′(g) of . For a gain 2 PDC,
Pcl′(2)=1/4; hence, we need to have a coincidence probability less than 25% in order to exclude a classical interpretation of the dip. Interestingly, this suggests that a gain lower than two may be utilized for demonstrating this quantum effect, thus lowering the experimental requirements. Solving
Pcoinc′(g)=1/4, we obtain
g=1.28, implying that the partial indistinguishability between the two paths achieved with this lower gain would be sufficient for observing a quantum effect. With this gain, the probability amplitude corresponding to the transmission through the crystal is larger than (minus) the one corresponding to double-stimulated events, but the partial destructive interference between the two paths is sufficient to reduce the coincidence probability to 1/4.