---
title: Entangled-Two-Photon Photoemission Overview
url: https://www.emergentmind.com/topics/entangled-two-photon-photoemission-etpp
type: topic
---

# Entangled-Two-Photon Photoemission Overview

Searching arXiv for the specified papers and closely related ETPP material.
arxiv_search.query({"search_query":"id:2604.13375 OR id:2012.06736 OR id:0712.2713","start":0,"max_results":10})
Search results retrieved.
arxiv_search.query({"search_query":"all:\"Photoemission and absorption under coherent and entangled-photon-pair illumination\"","start":0,"max_results":5})
Entangled-two-photon photoemission (ETPP) is a subthreshold photoemission process in which the absorption of an entangled-photon pair is followed by the emission of an electron into vacuum. In the contemporary review literature, ETPP is treated as the photoemissive analog of entangled-two-photon absorption (ETPA), and as one of three subthreshold photoemission mechanisms together with one-photon Fermi-tail photoemission (FTP) and ordinary two-photon photoemission (TPP) [2604.13375]. Its defining feature is not merely that two photons participate, but that the two photons belong to the same entangled pair and act as a correlated excitation event. In the ideal low-flux isolated-pair regime, this produces a linear dependence on photon flux or optical power, in contrast to the quadratic scaling characteristic of TPP from independent photons. The topic sits at the intersection of nonlinear photoemission, SPDC-based quantum optics, and surface-sensitive electron detection, and it is closely connected—conceptually but not identically—to the broader literatures on ordinary TPP in solids and entangled-photon nonlinear spectroscopy.

## 1. Placement within subthreshold photoemission

The review framework that organizes ETPP places it beside FTP and TPP as the three prominent forms of subthreshold photoemission [2604.13375]. In the band-diagram language used there, ETPP occurs when
$$
E_g < h\nu < (E_g+\chi) < 2h\nu,
$$
so that a single photon lies below the photoemission threshold, whereas two photons together can eject an electron. For the CsK\(_2\)Sb photocathode used in the key experiment, the quoted parameters are \(E_g \approx 1.0\) eV, \(\chi \approx 1.1\) eV, and therefore
$$
W = E_g+\chi \approx 2.1\ \text{eV},
$$
with the corresponding wavelength window
$$
\lambda/2 < 590\ \text{nm} < \lambda < 1240\ \text{nm}.
$$

The distinction among the three mechanisms is operational as well as physical. FTP is a one-photon process in which a subthreshold photon ejects an electron from a thermally populated tail state; it is linear in optical power and strongly wavelength- and temperature-dependent. TPP involves two independently absorbed photons and yields a quadratic rate. ETPP instead attributes the emission event to two photons from the same entangled pair; in the ideal low-flux regime, its rate is linear in photon flux because detection of one twin implies the other is present.

| Mechanism | Physical origin | Scaling signature |
|---|---|---|
| FTP | One subthreshold photon ejects an electron from a thermally populated tail state | Linear in optical power |
| TPP | Two photons are absorbed independently | Quadratic in optical power/intensity |
| ETPP | Two photons from the same entangled pair jointly cause emission | Linear in photon flux/power in the isolated-pair regime |

A persistent experimental difficulty follows immediately from this taxonomy. FTP and ETPP are both linear in power, so FTP can mask ETPP unless it is suppressed. TPP is easier to separate because its quadratic scaling differs from the low-flux ETPP signature.

## 2. Correlated-pair mechanism and scaling laws

The physical mechanism emphasized for ETPP is a pair-correlation effect rather than any increase of photon energy. Because the photons originate in the same SPDC event, the presence of one twin signals the presence of the other. If both arrive at the material within the relevant spatial and temporal overlap window, they can drive a two-photon transition that lifts an electron above the vacuum level [2604.13375]. The review identifies two source parameters as central:

- **Entanglement time** \(T_E\): the temporal width of the fourth-order temporal correlation.
- **Entanglement area** \(A_E\): the transverse spatial width over which one photon predicts the position of its twin.

Within the heuristic rate-equation picture, the total subthreshold photocurrent under entangled illumination is decomposed as
$$
\overline{\imath}_{TE} = i_D + \mathcal{F}\left[ \mathcal{R}_F(\mathcal{T}\overline{P_E}) + \mathcal{R}_E(\mathcal{T}^2\overline{P_E}) + \mathcal{L}_C(\mathcal{T}\overline{P_E})^2\Gamma/A \right].
$$
The corresponding count-rate form is
$$
\overline{\mu}_{TE} = \mu_D + \mu_F + \mu_E + \mu_C.
$$

The central scaling laws are stated explicitly as
$$
i_E \propto \phi A \propto \Phi \propto P,
$$
for ETPP, and
$$
i_C \propto \phi^2 A \propto \Phi^2/A \propto P^2/A \propto IP,
$$
for classical or independent-pair TPP. The review repeatedly stresses that the linearity of ETPP is not a sign of a one-photon process; it arises because the pair behaves as one correlated excitation event.

The same framework also describes the loss of the entangled signature at higher brightness. When twins “overcrowd” the interaction volume, some photons become effectively separated from their partners, producing “cousins” or singleton pairs. Then uncorrelated two-photon photoemission appears and the scaling becomes quadratic, as in ordinary TPP. This makes the isolated-pair regime a defining condition for observing ETPP as distinct from classical nonlinear background.

## 3. Quantum modeling and identification criteria

The principal quantum-mechanical description highlighted for ETPP is the model of Lissandrin et al., formulated for volume-initiated, surface-limited photoemission with direct interband transitions, Bloch-like initial and final states, spherical Fermi surfaces, type-I collinear SPDC photon pairs, entanglement time \(T_E\), entanglement area \(A_E\), and a monochromatic pump [2604.13375]. In that treatment, the detailed current expression contains a nontrivial overlap function \(F(k,T_E)\), but the main scaling extracted in the review is
$$
i_E \propto \frac{1}{A_E T_E}.
$$

Several consequences are made explicit. First, \(i_E\) is proportional to the photon flux \(\phi A\). Second, it is independent of illumination area if the beam fits within the entanglement area. Third, optical loss enters as \(\beta^2=(1-R)^2\), and more generally as \(\mathcal{T}^2\), because loss of either twin destroys the pair. The responsivity and quantum efficiency are defined by
$$
i_E = \mathcal{R}_E P, \qquad \eta_E = \frac{i_E/e}{\phi A},
$$
and
$$
\eta_E = \frac{hc}{e\lambda}\,\mathcal{R}_E,
$$
with the review stating directly that
$$
\mathcal{R}_E \propto \eta_E \propto \frac{1}{A_E T_E}.
$$

For experimental identification, the review introduces a linear-to-quadratic crossover intensity
$$
I_{EC} = \frac{\mathcal{R}_F + \mathcal{T}\mathcal{R}_E}{\mathcal{L}_C\Gamma},
$$
to be compared with the coherent-light FTP/TPP crossover
$$
I_{FC} = \frac{\mathcal{R}_F}{\mathcal{L}_C\Gamma}.
$$
This yields
$$
\frac{I_{EC}}{I_{FC}} = 1+\frac{\mathcal{T}\mathcal{R}_E}{\mathcal{R}_F}.
$$
The review interprets this as an enlargement of the linear regime by ETPP. It also gives the empirical relation
$$
\phi_{EC} = \frac{\mathcal{T}_0/\Gamma}{A_E T_E},
$$
which was used to infer an entanglement time from the observed crossover in the Kobayashi experiment.

The review further notes that a real intermediate state can participate. For CsK\(_2\)Sb, the transition can proceed through the conduction band, and this may help ETPP relative to a purely virtual transition. This suggests that the electronic structure of the photocathode can materially affect observability even when the central ETPP signature is defined by source correlations.

## 4. Experimental observation in CsK\(_2\)Sb photocathodes

The review treats the experiment of Kobayashi et al. (2007) as the first convincing observation of ETPP [2604.13375]. Its decisive methodological choice was the use of a **channel photomultiplier (CPM) module** rather than a conventional PMT. The quoted reason is that the CPM has much lower Fermi-tail noise, uses a small capillary-like electron multiplier, has a smaller photocathode and lower dark count, and “reduced Fermi-tail photoemission below detectability.” This directly addressed the main confounder for linear-in-power measurements.

The photocathode was **CsK\(_2\)Sb**, operated at \(27^\circ\)C in the CPM module. The optical system employed a **532-nm CW laser diode** pumping a **PPLN waveguide** that produced **1064-nm degenerate SPDC pairs** by type-I quasi-phase-matched downconversion. The setup further included a half-wave plate / Glan prism for pump control, filters to remove residual pump light, a focusing lens, and the CPM in photon-counting mode. The review emphasizes that the illumination was CW, so \(\Gamma=1\), which is advantageous because pulsed excitation enhances ordinary TPP and therefore the unwanted quadratic background.

ETPP was separated from competing signals by a combination of detector choice and operating regime: CPM detection to suppress FTP, low intensity to minimize TPP, CW illumination to avoid pulse-enhanced quadratic backgrounds, longer wavelength at 1064 nm to keep the Fermi-tail contribution low, and intensity-scaling analysis to distinguish linear ETPP from quadratic TPP. In the observed data, the high-intensity regime was quadratic, attributed to cousin/singleton-pair TPP, while the low-intensity regime was linear and not attributable to FTP. The reported crossover occurred around
$$
I_{EC}\approx 240\ \text{W/m}^2.
$$

The quantitative figures reported for the observed ETPP signal are
$$
\eta_E \approx 2.3\times 10^{-13}\ \text{electrons/photon}, \qquad \mathcal{R}_E \approx 2.0\times 10^{-13}\ \text{A/W},
$$
together with the normalized cross section
$$
\delta_E = \sigma_E A_E T_E \approx 2.1\times 10^{-54}\ \text{m}^4\text{s}.
$$
The Lissandrin model, renormalized to the experimental parameters, predicted
$$
\eta_E \approx 1.5\times 10^{-12},\qquad \mathcal{R}_E \approx 1.3\times 10^{-12}\ \text{A/W},
$$
approximately a factor of seven above the measured values. The review characterizes this as reasonably good agreement given uncertainties in \(T_E\), \(A_E\), \(\beta\), \(\mathcal{T}\), and the underlying model assumptions. Using the observed crossover, it infers
$$
T_E \approx 340\ \text{fs}.
$$

The same review contrasts these results with earlier PMT-based work by Booth et al. In that case, a linear photocurrent consistent with FTP was observed under both coherent and entangled-singleton illumination, with no ETPP detection because the Fermi-tail floor remained too high:
$$
\mathcal{T}_0 \eta_E < \eta_F.
$$
The quoted FTP efficiencies,
$$
\eta_F \approx 3.7\times 10^{-10}\ \text{electrons/photon}
$$
for entangled singletons at \(-20^\circ\)C and
$$
\eta_F \approx 4.8\times 10^{-10}
$$
for coherent light in the analog case, illustrate the scale of that background problem.

## 5. Relation to ordinary two-photon photoemission in solids

A distinct but closely related line of work studies two-photon photoemission in the ordinary nonlinear-optical sense rather than under entangled-photon excitation. The paper “Direct resolution of unoccupied states in solids via two photon photoemission” [0712.2713] is central in that literature. Its core claim is that a monochromatic second-order photoemission process can provide direct access to unoccupied electronic structure in solids because the detected photocurrent contains, besides the standard one-photon term, a resonant second-order term in which an electron is first promoted from an occupied initial state into an unoccupied intermediate state and then emitted into the final vacuum state.

The photocurrent is written as
$$
J \propto \sum_{k_a\le k_F} \left|h_{fa}\right|^2\delta(E_k^f-E_{k_a}^a-\hbar\omega) +\frac{1}{4}\sum_{k_a\le k_F}\delta(E_k^f-E_{k_a}^a-2\hbar\omega)\left|\sum_{k_z} \frac{h_{fz}h_{za}}{E_{k_z}^z-E^a_{k_a}-\hbar\omega-i\eta}\right|^2,
$$
with resonance when
$$
E_{k_z}^z - E_{k_a}^a \approx \hbar\omega.
$$
The review-level significance of this classical TPP result for ETPP is primarily conceptual: it establishes how an intermediate-state picture can endow two-photon photoemission with sharp resonant structure, strict selection rules, and sensitivity to unoccupied states.

The 2007 paper emphasizes exact selection rules of energy, crystal symmetry, and momentum. Energy conservation appears through the one- and two-photon delta functions, while the intermediate-state denominator yields Lorentzian-like peaks with finite width \(\eta=\eta_0+\eta_1\), where \(\eta_0\) collects intrinsic elastic and inelastic processes and \(\eta_1\) is additional strong-field broadening. Especially notable is its assertion that, for the bulk-like initial-to-intermediate transition, “full three-dimensional momentum is conserved,” in contrast to the usual one-photon uncertainty
$$
\Delta k_{\perp}=(2V_i)^{1/2}.
$$

The same paper demonstrates the framework with ab initio calculations for Si(001), obtaining very good agreement with previously published low-energy laser spectroscopy experiments after an upward shift of the unoccupied states by about \(0.3\) eV, interpreted as a scissor-operator correction. With that shift, the peak maximum moves by approximately \(0.15\) eV to a photon energy of \(3.95\) eV, matching experiment. A model calculation for Al(111) further shows very narrow resonances in UV angle-resolved photoemission geometries and particularly strong features at \(2\hbar\omega = 25.5\) eV and \(2\hbar\omega = 40\) eV when both transition steps satisfy momentum conditions.

ETPP is not identical to this ordinary TPP band-mapping framework. The former is defined by entangled-pair absorption and low-flux correlated scaling, whereas the latter concerns monochromatic second-order photoemission under classical illumination. A plausible implication is that if solid-state ETPP implementations are developed beyond photocathode detection, intermediate-state resonances and band-structure selectivity of the kind identified in ordinary TPP may become important design parameters.

## 6. Relation to entangled-two-photon absorption and observability limits

ETPP is commonly discussed in parallel with ETPA because both are entangled-photon-driven nonlinear processes, and the review literature explicitly treats ETPP as the photoemissive analog of ETPA [2604.13375]. An important external constraint on overly strong claims of entanglement-induced enhancement comes from the experimental study of isolated time-frequency-entangled pairs in molecular two-photon absorption [2012.06736].

That study used a type-0 SPDC source pumped by a 532 nm continuous-wave laser to produce time-frequency-entangled photon pairs centered around 1064 nm, and it deliberately operated in the regime of isolated, non-overlapping photon pairs. The paper stresses that this regime avoids ambiguity from accidental coincidences, prevents pair overlap from complicating scaling interpretation, and makes flux and enhancement easier to bound. In that context, the measured usable pair rate in the interaction region was inferred to be
$$
2.0(\pm0.2)\times10^9\,\mathrm{s}^{-1}
$$
at a pump power of 1 W.

Its most general lesson for ETPP is that entanglement does not automatically imply a large practical enhancement. The molecular work identifies two distinct ingredients—photon-number correlations and time-frequency entanglement / spectral correlations—and gives, for broadband molecular transitions, the approximate enhancement
$$
QEF \approx \frac{B}{F},
$$
where \(B\) is the entangled-photon-pair bandwidth and \(F\) is the photon-pair flux. In that experiment, the derived value
$$
QEF = 16{,}000 \pm 1{,}000
$$
was still insufficient to make the fluorescence detectable, and the resulting upper bound on any additional enhancement was
$$
E < 3.2(\pm0.7)\times10^{5}.
$$

The transferable methodological lessons were stated explicitly: the need for isolated photon pairs, the need to separate genuine pair-driven effects from single-photon backgrounds, the importance of scaling tests, the need to know the actual pair flux in the interaction region, the need to account for dispersion, spatial overlap, and collection efficiency, the likely dependence of enhancement on bandwidth-to-flux ratio, and the usefulness of log-log scaling analysis. The same paper also specifies what is not directly transferable: the Rhodamine-6G fluorescence model, the exact molecular cross section, the specific upper bound \(3.2\times10^{5}\), and the conclusion that broadband molecular TPA is too weak for detection at the tested flux.

For ETPP, the implication is narrower and more disciplined than claims of generic “quantum enhancement.” Entanglement can alter which terms dominate and under what flux conditions, but practical observability remains constrained by pair isolation, loss of either twin, backgrounds such as FTP, and the electronic structure available to mediate or accept the transition. The review literature therefore presents ETPP less as an unqualified enhancement effect than as a distinct low-flux correlated-emission regime whose signature is linearity, pair-conditioned loss behavior, and sensitivity to \(A_E\) and \(T_E\).

Source: https://www.emergentmind.com/topics/entangled-two-photon-photoemission-etpp