Measurement of analogue Hawking radiation stimulated by a single photon Authors: Rodrigo Felipe-Elizarraras, Hector Cruz-Ramirez, Karina Garay-Palmett, Alfred U’Ren, David Bermudez Published: 2026 ABSTRACT AbstractHawking radiation is a fundamental spontaneous phenomenon in which the event horizon of a black hole transforms the vacuum state into a pair of correlated particles: the Hawking radiation and its partner. This emission can also be stimulated by a state other than the vacuum. A similar phenomenon, known as analogue Hawking radiation, has been observed across various physical systems, but experimental realizations have largely remained in the classical regime. This work presents an experimental verification of analogue Hawking radiation stimulated by a single-particle state, producing a verifiably single-photon Hawking output. Using heralded single photons generated via spontaneous four-wave mixing, we stimulate the analogue Hawking effect in an optical fiber. We characterize the resulting Hawking signal by measuring its spectral dependence, intensity scaling, and heralded second-order correlation. Our results establish a platform for exploring quantum correlations and entanglement in analogue gravity.Subject terms: Single photons and quantum effects, Nonlinear optics, Quantum simulation Introduction In 1974, Stephen Hawking proposed that black holes are not entirely black, but emit radiation through a process known as the Hawking effect1. This effect is spontaneous, i.e. it creates particles from the quantum vacuum. In this case, the event horizon of a black hole transforms the vacuum state into a pair of particles2: one that reaches an observer at infinity—the Hawking radiation (HR)—and another one that remains trapped by the event horizon—the Hawking partner (HP)—see Fig. 1a. This fascinating phenomenon has captivated both popular science and physicists, albeit for different reasons. For popular science, black holes are intriguing because they are the ultimate devourers: nothing can escape from them, not even light. For physicists, this phenomenon may provide clues on how to combine gravity with quantum mechanics in a system that can be solved analytically and suggest connections between different areas of physics3. However, Hawking emission is extremely weak in known astrophysical black holes (stellar and supermassive), making its measurement challenging. Still, its observational confirmation remains an active area of research4,5. Astrophysical Hawking radiation vs Single-photon stimulated analogue Hawking radiation in optics. a Spontaneous astrophysical Hawking radiation. The vacuum state 0 scatters into two particles with defined frequency ω: the Hawking radiation (HR) and the Hawking partner (HP); the black region represents the black hole interior. b Single-photon stimulated analogue of Hawking radiation in optics. The seed is a single-photon state 1 instead of the vacuum, and it scatters into two particles with defined comoving frequency ω′. The vertical dashed line marks the analogue of the event horizon of a black hole. In 1981, while teaching a course on fluid dynamics, William Unruh6 realized that a similar phenomenon could occur in hydrodynamics. Sound waves traveling in a transonic flow—which exceeds the speed of sound—are blocked, creating an analogue black hole from which no sound can escape. Quantizing the sound field leads to a similar Hawking effect in which phonons—quasiparticles of sound—are emitted. This analogy gave rise to a new field of study known as analogue gravity, where phenomena usually associated with gravity can be studied in other fields. Examples of alternative systems that exhibit Hawking radiation include water tanks7,8, Bose-Einstein condensates (BECs)9,10, fluids of light11,12, superconductive inductive elements13, and optical fibers14–16. Besides Hawking radiation, other effects studied under analogue gravity include the dynamical Casimir effect17, superradiance18,19, the cosmological creation of particles in theory20,21 and in experiments22–24, the Hubble tension25, false vacuum decay26, and curved spacetime signatures27,28. The spontaneous Hawking radiation seeded by the quantum vacuum is extremely weak and difficult to detect even in the controlled environments of laboratory analogues29. For this reason, most experiments to date have focused on measuring the stimulated case, in which the same Hawking effect is seeded not by a vacuum state, but by a classical one30. In this case, the resulting effect is classical rather than quantum, rendering it more intense and easier to detect. Experimentally, the most successful systems for observing analogue Hawking radiation are: (1) water tanks, the first system in which stimulated Hawking radiation was measured7,8,31; (2) BECs, for which the spontaneous version of the effect has been measured in phonons using correlation measurements10,32; and (3) optical fibers, for which only the stimulated version has been measured, including both the positive-frequency Hawking radiation14 and the negative-frequency Hawking partner15,33, as well as the interference between two Hawking signals16 and, recently, the backreaction of the Hawking emission34. A claimed 2010 spontaneous demonstration in optics35 was later identified as horizonless radiation emitted by a superluminal refractive-index perturbation36–38. This work explores a previously unobserved regime in analogue systems: the stimulation of the Hawking effect by a single-particle state, producing a verifiably single-photon Hawking signal. In quantum optics experiments, heralded single photons can be obtained from photon pairs generated through spontaneous parametric down-conversion (SPDC)39 or spontaneous four-wave mixing (SFWM)40; however, each process entails certain advantages and disadvantages41. In this work, we rely on an optical fiber to provide a single unified platform to: (i) generate the required single photons (from the SFWM process), and (ii) to produce the Hawking process. We present a measurement of the Hawking radiation effect stimulated by a single photon; see Fig. 1b. We send a pump pulse (red) with frequency ω0 through a photonic-crystal fiber (PCF) to generate signal ωs and idler ωi photon pairs via SFWM, see Fig. 2a. Energy and momentum conservation—known as the phase-matching conditions—determine the emission frequencies for the signal and idler photons, depending on the dispersion relation of the fiber. The resulting conditions are where βk(ω) = ωnk(ω)/c, nk(ω) is the refractive index, and {j, k} represent one of the two orthogonal polarization states of the fiber. The resulting wavelengths are shown in Fig. 3a for four polarization combinations of the pump and the signal/idler. Sequence of the experiment. a The pump pulse emits the initial signal (se) and idler (ie) photons via spontaneous four-wave mixing (SFWM). We use the emitted signal photon as a probe; its quantum nature leads to two possibilities after interacting with the pump pulse through the white-hole (WH) or black-hole (BH) horizon: (b1) it stimulates the analogue Hawking effect (HR), or (b2) it escapes as a signal (s) photon. c Comoving frequency ω′ conservation predicts the laboratory frequencies ω of all signals. The Kerr effect goes from zero to a maximum of δnmax. Matching of SFWM and analogue Hawking emissions. a The phase-matching conditions of the spontaneous four-wave mixing (SFWM) determine the signal and idler photon frequencies. b The group-velocity matching determines the horizon frequency. c The yx (fast-slow) configuration allows us to satisfy both conditions at a wavelength accessible with our laser (shaded region), and the resulting photon wavelengths are within the range of our detectors. We focus on the circled matching point. We have carefully designed our experiment, selecting the pump wavelength and polarization to simultaneously produce a second phenomenon: the analogue Hawking effect stimulated by a single photon. The Hawking effect requires bringing the probe signal to a standstill, i.e., making its group velocity zero in the comoving frame of the pump. This condition is fulfilled around the horizon frequency ωh that has a group-velocity matching to the pump pulse uj = vg,j(ω0), resulting in the condition vg,k(ωh) = uj, with vg,k(ω)=(∂βk/∂ω)−1. The resulting horizon wavelengths are shown in Fig. 3b. After interacting with the pump pulse, the signal photon has two possibilities: (1) it stimulates the Hawking process and produces a pair of particles, the Hawking radiation (HR) and the Hawking partner (not shown), see Fig. 2b1; or (2) it tunnels and continues as SFWM, see Fig. 2b2. We determine their optical frequencies using the conservation of the comoving frequency15,42: which again depends on the dispersion relation βj(ω). The measurable effect is a frequency shift of the signal photon ωs into the Hawking photon ωHR, as shown in Fig. 2c. We explain this figure in more detail in the Methods Section. To achieve both effects under the same experimental conditions, we match the frequency of the SFWM signal photon with the horizon frequency of the analogue Hawking effect: ωs = ωh. We can fulfill this condition for the yx polarization, i.e. with the pump polarized along the fast axis and the SFWM and Hawking photons polarized along the slow axis, see Fig. 3c. This matching determines the central conditions under which we perform our analogue Hawking radiation measurements. We demonstrate that this signal corresponds to the analogue HR by analyzing its dependence on the pump frequency and intensity, and comparing its behavior with our theoretical model (see the Methods section for more details). Additionally, we verify its quantum nature by measuring the three-photon second-order correlation function between the idler, signal, and Hawking photons. Stimulation by a single photon preserves the quantum nature of the measured signal, even in the absence of the quantum vacuum43. Ideally, we expect no coincidence detection events between the signal and Hawking modes, as a single signal photon interacting with the pump pulse either transforms into a Hawking photon or tunnels through the pulse and is transmitted while undergoing a slight frequency shift44. Our experiment constitutes a demonstration of an unobserved analogue Hawking radiation regime that combines the best of two worlds: (1) a quantum origin and (2) a signal sufficiently intense to be readily detected with single-photon detectors, with a verifiable single-photon nature. This platform opens up a paradigm for experiments with the optical analogue of Hawking radiation, enabling tests of several quantum properties—including entanglement, backreaction, and energy conservation—all of which are immediately accessible in tabletop optical experiments. We present our experimental setup in Fig. 4, which uses a 110 fs TiSa laser and a 90 cm length NKT Photonics NL-PM-750 photonic crystal fiber (PCF). We have previously used this setup to measure the classically stimulated analogue of HR16. We provide further details of our setup in the Methods section. Experimental setup. The experimental setup is divided into three parts: (1) the input stage for pump control, (2) the interaction stage for the photonic-crystal fiber (PCF), and (3) the detection stage for photon separation and measurement. The abbreviations of optical elements are defined in the text. We acknowledge the use of CAD models numbers: GTH10M, AHWP10M-980, M43L01, P3-1550A-FC-1, FEL0700, PS856, 354240-B provided by Thorlabs, Inc. Results Analogue Hawking effect We vary the pump wavelength between 802 and 838 nm, maintaining 1.5 mW of average power coupled to the PCF, and record 36 resulting single-photon-level spectra. Each spectrum is captured as a band on the 512 × 512 pixel array of the CCD at the monochromator output. We integrate the 120 central rows of pixels to obtain the signal. Separately, we integrate another band of 120 rows where no signal is present to obtain a noise measurement, which is then subtracted from the signal data. We also apply a three-point median filter to eliminate a few individual white and dark single pixels from the CCD array. We show six representative spectra of the raw data in Fig. 5. Spectral data for SFWM and analogue Hawking. Spectra of photons generated around the horizon λh at 644.5 nm for six different pump wavelengths: a–f 805, 811, 817, 823, 829, and 835 nm. The top three plots correspond to the generation of Hawking radiation at a black-hole (BH) horizon (redshift) and the bottom three at a white-hole (WH) horizon (blueshift). In each spectrum we mark the signal λs and the analogue Hawking λHR wavelengths. A pump at 820 nm creates a horizon at 644.5 nm. A key advantage of our experiment is the ability to observe both the black-hole horizon (BH), where the HR is redshifted relative to the signal (first row of Fig. 5), and the white-hole horizon (WH), where the HR is blueshifted (second row of Fig. 5). By fitting Gaussians to the raw data shown in Fig. 5, we obtain the central wavelengths and widths of the Hawking (HR) and SFWM (s) signals, and plot them in Fig. 6a. The expected wavelengths for the SFWM signal (yellow) and Hawking (red) photons are determined by the phase-matching conditions and the conservation of comoving frequency. The orange dashed line corresponds to the unmodified SFWM photon in Fig. 2c, i.e., the case without the nonlinear contribution of the Kerr effect. Considering that the photon is emitted and interacts with the pump, the Kerr contribution modifies the expected spectra, curving the orange line into the two yellow segments for the SFWM signal photon and the red lines for the Hawking photon. The resulting data for SFWM and HR exhibit a forbidden region produced by the Kerr effect with peak intensity δnmax, corresponding to the region between the two purple lines in Fig. 6a originating from the two purple dots in Fig. 2c. As expected, the SFWM and HR emission are concentrated at the peak intensity defined by δnmax and the purple lines. These results agree with our theory presented in the Methods section. Comparison between experimental results and theoretical model. a Experimental data for the signal photon from spontaneous four-wave mixing (SFWM) (λs in yellow) and the corresponding analogue Hawking emission (λHR in red). b Increasing pump power increases the spectral separation between the SFWM and Hawking photons at a fixed pump wavelength of 817 nm. The data points represent our experimental measurements, while the lines denote our model. Error bars represent the 1σ width obtained from Gaussian fits, as shown in Fig. 5. Increasing the pump power enhances the Kerr effect; accordingly, our theory predicts that the Hawking and SFWM signal photons should exhibit a greater spectral separation. We tested this prediction by fixing the pump wavelength at 817 nm and increasing the pump power from 1 to 4 mW. We show our experimental measurements in Fig. 6b. We also performed a second run of experiments in which we varied the pump wavelength with a higher pump power of 3 mW (data not shown). In all cases, the data agree with our theory: increasing the pump intensity increases the spectral separation of the SFWM and Hawking photons by the predicted amount. This observation indicates that our measurements correspond to SFWM and Hawking photons rather than some other nonlinear effect. Quantum nature We can model the analogue Hawking emission as a two-mode squeezer followed by a beam splitter43,45. This model implies that when a signal photon interacts with the pump pulse, it either transforms into a Hawking pair (comprising the Hawking radiation and its partner) or tunnels and continues as a transmitted signal photon. In our experiments, there is an additional two-mode squeezer from the SFWM process. Therefore, an anticorrelation measurement between the Hawking and signal photons can verify the quantum nature of the process. To improve the signal-to-noise ratio of this measurement, we use the idler photon to herald the signal and Hawking photons. We characterize the photon statistics with the second-order cross-correlation function46: where i, s, H denote the three channels idler, signal, and Hawking; the various N values represent the respective detection rates; single-index quantities represent simple counts, while two-index (three-index) quantities represent two-channel (three-channel) coincidence counts. Note that a 50:50 beam splitter is commonly used at the exit of a fiber to separate the photons for a g(2)(0) measurement44. In our experimental design we instead exploit the analogue Hawking process to separate the two photons spectrally, using a dichroic mirror to direct them to different detectors. We thus demonstrate the single-photon nature of the Hawking radiation mode and to ensure that only one photon exits the interaction with the pump—either a Hawking or an SFWM signal photon47. The values resulting from averaging 10 measurements are shown in Fig. 7 for both the SFWM and Hawking processes, with the pump power ranging from 0.16 to 1.8 mW. Note that both curves show a clearly quadratic behavior as expected from the theory of single-photon events41,48,49. Second-order correlation function for SFWM and analogue Hawking. Measurements of the second-order correlation function g(2)(0) for the spontaneous four-wave mixing (SFWM) and Hawking radiation (HR) processes showing values smaller than 1/2 and quadratic behavior. Error bars represent the 1σ width obtained from 10 different measurements for each point. The mechanism behind the classically stimulated analogue Hawking radiation effect in optics was described in the seminal paper by Philbin et al.14 and it is related to cross-phase modulation (XPM). In this work, we have demonstrated an experimental scheme to produce analogue Hawking radiation stimulated instead by a single photon. This radiation exhibits an unambiguously quantum nature in the sense that: (i) it is seeded by a verifiably single-photon state produced by SFWM, and (ii) it produces a Hawking signal in an equally verifiably single-photon state. These results demonstrate that frequency shifting converts a signal photon into a Hawking photon. We confirm the quantum nature of the Hawking radiation mode by the fact that all of our measurements satisfy g(2)(0) < 1. In fact, we also fulfill the single-photon criteria48 of g(2)(0) < 1/2. We reach values as low as g(2)(0) = 0.009 ± 0.028 (@ 0.16 mW) for the SFWM and g(2)(0) = 0.106 ± 0.018 (@ 0.72 mW) for the analogue Hawking effect. It is useful to compare the values of g(2)(0) between the SFWM and Hawking photons at the same pump intensity. The comparable g(2)(0) value for SFWM is g(2)(0) = 0.065 ± 0.014 (@ 0.72 mW). Remember the SFWM measurement uses a beam splitter on a single photon and the analogue Hawking measurement replaces the beam splitter by the Hawking frequency shifting on a single photon plus a dichroic mirror. Still, we can see that the Hawking effect does not significantly increase the value of g(2)(0). The low values of g(2)(0) prove that the detected signal is of quantum nature and dominated by single photons. Also, the mainly quadratic behavior of both g(2)(0) in terms of pump power shows that the signal is dominated by the single photon emission, instead of Raman noise or multiphoton events, see Fig. 7. This is further confirmed using our photon-number resolving detector, where the ratio of single-photon detection with respect to multi-photon detection was 100. Additionally, the coincidence-to-accidental ratio (CAR) for the SFWM process is 75 at the intensity where the Hawking effect is detectable (@ 0.72 mW). All these measurements confirm the single-photon nature of our Hawking radiation emission. We can conclude that our small non-zero value of g(2)(0) is due to contamination mainly by random Raman noise and a very small amount of double SFWM events. Furthermore, we achieve results comparable to our measurement of the SFWM effect and to other experiments where SFWM is measured under similar conditions50. The g(2)(0) measurement of the SFWM can reach values as low as 0.009 at lower pump power (@ 0.16 mW) with a CAR as high as 190; however, as predicted by our model, the efficiency of the Hawking effect in that regime is too low to produce a significant number of events. In that case, we do not observe the double peak of SFWM and analogue Hawking, but rather a single SFWM peak following the orange dashed line in Fig. 6a. A key advantage of the current configuration is that a single pump in a single fiber produces both the SFWM and the analogue Hawking effects, leading to a significant simplification of the experimental setup and the elimination of the need for temporal synchronization. However, a disadvantage is the reduced control over the experimental parameters; as mentioned above, the pump frequency and polarization constitute our only adjustable degrees of freedom. Discussion In this work we introduced an approach to measuring a quantum version of analogue Hawking radiation, stimulated by heralded single photons from a fiber-based spontaneous four-wave mixing (SFWM) process. A unique aspect of our work is the use of a single fiber for both processes, the same ultrashort pulse serves as the pump for the SFWM and provides the Kerr effect necessary to create the analogue black hole and white hole horizons. We employed cross-polarized SFWM processes to increase experimental flexibility and to ensure the simultaneous unfolding of both the SFWM and Hawking effects within the same optical system, further emphasizing the novelty of our approach. We have thus demonstrated for the first time an intermediate regime involving quantum stimulation by a single photon, sitting between the classically stimulated and the spontaneous cases. This regime effectively combines two key advantages: (1) heralded high photon counts that allow for reliable detection of the Hawking emission, and (2) the retention of the quantum nature of our Hawking radiation signal. This platform enables the experimental verification of quantum correlations and entanglement between the idler photon and the Hawking photon, the latter of which corresponds to a frequency-shifted signal photon. In this paper, we demonstrated that the heralded second-order cross-correlation value g(2)(0) of the analogue Hawking radiation mode is significantly less than 1, verifying the quantum nature of the effect. In fact, we attain a value as low as 0.106 ± 0.018. Furthermore, this value is relatively close to that of the SFWM signal mode, showing that the analogue Hawking effect preserves its quantum characteristics. The g(2)(0) for both SFWM and analogue Hawking radiation have a mainly quadratic behavior as expected for the single-photon case. These findings establish a regime of experiments in analogue Hawking radiation using heralded single photons rather than classical pulses. These experiments can be used to investigate properties such as correlation and entanglement, deepening our understanding of these processes. This work paves the way for further studies and applications of analogue Hawking radiation in optical fibers within the quantum regime, including the simultaneous measurement of the photon pair produced by the analogue Hawking process (the Hawking radiation and its partner), as well as further tests of entanglement, Bell inequalities, or quantum state tomography. Methods Theory Our experiment consists of two parts: (a) a pump pulse propagates through a fiber and generates photon pairs through SFWM, where the detection of an idler photon heralds a signal-mode single photon. (b) The pump pulse blocks the signal-mode single photons (reducing their group velocity in the comoving frame to zero) and converts some of these signal photons to Hawking photons via the frequency shift resulting from the analogue Hawking effect. The quantum nature of the signal photons leads to two possibilities after interacting with the classical pump pulse (see Fig. 2): (b1) they are frequency-shifted by the pump and reflected in the comoving frame, creating the analogue Hawking effect15,51, or (b2) they tunnel through the pump pulse and continue as signal photons. It is challenging to ensure that these two effects occur simultaneously and are successfully detected because we only have the pump frequency and polarization as degrees of freedom at our disposal. Our available lasers limit the pump frequency, and the output signals must lie within the spectral window of our detectors. In our approach, we take advantage of a polarization-maintaining fiber (exhibiting high birefringence) to utilize the two linear polarizations of our fiber and obtain slightly different dispersion relations. This method results in a distinct behavior for each combination of polarizations among the four different waves participating in the SFWM process52. This setup provides greater experimental control compared to the copolarized case. In the SFWM process, a pump pulse at frequency ω0 produces two photons: a signal ωs and an idler ωi. We describe this effect using the usual frequency and momentum conservation laws, i.e., the phase-matching conditions41: where βj(ω) = ωnj(ω)/c is the dispersion relation of the fiber with polarization j, refractive index nj(ω), and {j, k} = {x, y}, where x, y are the slow and fast axes, respectively52. The resulting signal and idler frequencies are shown in Fig. 3a as a function of the pump frequency for the various combinations of pump and signal/idler polarizations. The generation of the analogue Hawking effect in optics is localized around the kinematic blocking of modes14. This blocking is characterized by the frequency of the horizon ωh, which is group-velocity matched to the pump pulse speed uj = vg,j(ω0), where vg,j(ω)=(∂βj/∂ω)−1. This condition is expressed as considering the two polarizations j, k. The resulting horizon frequencies ωh as a function of the pump frequency for the four polarization combinations are shown in Fig. 3b. Finally, to observe both effects in the same fiber, the signal photons from the SFWM must be generated near the horizon frequency. We can set them equal to find the central conditions for our experiment, then This condition corresponds to the SFWM signal photon being produced exactly at the horizon frequency; we are interested in working near this frequency, probing both higher and lower frequencies (see Fig. 2). Figure 3c shows that the two effects, the generation of photons by SFWM and the analogue Hawking effect, can occur simultaneously with a pump at 820 nm and a horizon at 644.5 nm for the yx polarization combination (fast-slow), where the pump pulse is on the fast axis, and the SFWM and Hawking photons are on the slow axis. The pump wavelength is limited by the tuning range of our Ti:Sapphire laser, between 780 and 840 nm, indicated by the shaded region in Fig. 3c. The range of frequencies that satisfy the blockage condition depends on the peak intensity of the pump pulse, i.e. on the magnitude of the optical Kerr effect15,53. Note that we have also successfully measured the analogue Hawking effect at a different horizon, at 596 nm, shown in Fig. 3c when pumping with a different laser at 908 nm. This measurement further validates our theory. To predict the optical frequency at which the Hawking photons appear, we must consider the conservation of the comoving frequency ω′, including the Kerr effect15,16, which leads to where δn = n2I is the refractive index change induced by the Kerr effect, n2 is the nonlinear index of the fiber, and I is the intensity of the pump pulse. Considering that the SFWM signal photon is emitted within the pump where δne ≠ 0 (curved spacetime in the optical metric14) and subsequently measured where δn = 0 (flat spacetime), the predicted frequency of the HR is given by The resulting condition is illustrated in Fig. 2c, where the emitted signal photon ωse (orange) is created and moves towards the horizon. After its creation, it will propagate to a region where δn = 0 (flat spacetime). It then follows one of two paths: it can either frequency shift to ωHR (red)—the predicted frequency of the analogue Hawking radiation14,15—or tunnel and continue as a signal photon with a slight frequency shift to ωs (yellow), defined by ωx′(ωs,0)=ωx′(ωse,δne), see Fig. 6a. The dimensionless comoving group velocity vg′=−∂ω′/∂ω determines the direction of movement in the frame comoving with the pump pulse. From Fig. 2c, we can determine that the signal photons travel in the same direction as the emitted signal photon, whereas the HR photons travel in the opposite direction. Experiment We show our experimental setup in Fig. 4. Our Ti:Sapphire pulsed laser, with a 110 fs pulse duration and a repetition rate of 72 MHz, is controlled in power and polarization using the polarizer POL1 and the half-wave plates HWP1 and HWP2. The laser is then coupled with an aspherical lens into a 90-cm-long NKT Photonics NL-PM-750 photonic-crystal fiber (PCF), where the SFWM and analogue Hawking effects occur. Once the light exits the fiber, it is collimated by another aspherical lens and separated by three Semrock dichroic mirrors: (1) DM1, which reflects (transmits) wavelengths above (below) 925 nm to isolate the idler photon around 1140 nm, (2) DM2, which reflects (transmits) wavelengths above (below) 795 nm to filter out the pump pulses, and (3) DM3, which reflects (transmits) wavelengths above (below) 640 nm to split the signal and Hawking photons. Each outgoing signal from the dichroic mirrors passes through a set of spectral filters (SF), a half-wave plate, and a polarizer before being collected by a multimode fiber (MMF). To detect the visible photon spectra, we remove DM3 and couple the output to an Andor SR-500i monochromator fitted with an iStar intensified CCD camera. The infrared idler photons pass through a half-wave plate, a set of spectral filters, a polarizer, and a tuneable spectral filter made by a dispersive prism with a vertical movable slit and two lenses. This configuration allows us to select the precise photon wavelength and to suppress Raman noise. Afterwards, the photons are collected in a single-mode fiber and detected using an ID281 IDQ superconducting nanowire single-photon detector with photon-number resolution capability. To measure signal-Hawking coincidence events, we replace the monochromator and CCD camera with two Excelitas SPCM APDs connected to the two output modes of DM3. An electronic delay generator ensures that the signals from both the visible and infrared photons arrive simultaneously, which is monitored using an IDQ ID800 time tagger. Supplementary information Transparent Peer Review file Acknowledgements D. B. thanks Lorenzo Procopio and Scott Robertson for valuable discussions. Author contributions R. F.-E. and H. C.-R. designed and conducted the experiment. K. G.-P., A.U., and D.B. conceived the project. R. F.-E. and D. B. analyzed the experimental results, developed the theoretical model, and wrote the manuscript. A.U. and D.B. supervised the project. All authors reviewed and approved the final version of the manuscript. Peer review Peer review information Nature Communications thanks Maxime Jacquet, Carlos Sabin and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available. Funding D.B. acknowledges the support of the Secretariat of Science, Humanities, Technology and Innovation (Mexico) (Project 152574 and SNII 62547) and the Marcos Moshinsky Chair (2023). A.U. acknowledges support from PAPIIT-UNAM (IN105024). K.G.-P. and A.U. acknowledge support from Laboratorio Nacional de Materia Cuántica: Materia Ultrafría e Información Cuántica. Data availability The datasets generated in this study have been deposited in the Figshare repository in ref. 54. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary information The online version contains supplementary material available at 10.1038/s41467-026-73812-8. References Associated Data Supplementary Materials Transparent Peer Review file Data Availability Statement The datasets generated in this study have been deposited in the Figshare repository in ref. 54.