betaFeedback

WorksPhysics and Astronomy

Quantum free-electron laser oscillator

Abstract

If the quantum mechanical recoil of the electron due to its scattering from the undulator and laser fields dominates the dynamics, a regime of the free-electron laser emerges where quantum effects lead to a drastic change in the radiation properties. However, the large interaction length required for a single-pass quantum free-electron laser impedes the experimental realization. The quantum free-electron laser oscillator, proposed in the present article, is a possible scheme to resolve this issue. Here we show that this device features a photon statistics that is closer to a coherent state in comparison to existing classical free-electron lasers. The device can be even operated in such a way that a sub-Poissonian statistics is obtained. Beside the benefit of demonstrating this pure quantum effect, the narrowing of the photon distribution implies reduced intensity fluctuations of the emitted radiation, which in turn lead to decreased noise in imaging experiments or to an enhanced sensitivity in interferometric applications.

Introduction

The quantum free-electron laser[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15] (Quantum FEL) is a proposed regime of FEL operation where improved features of the emitted radiation are expected. So far, the research focused on single-pass Quantum FELs with the drawback that the required undulator length is rather large and ultimately reaches a fundamental limit that is determined by the counteracting effects of multiphoton transitions, spontaneous decay, and space charge[12].

We therefore propose a Quantum FEL oscillator, where the emitted radiation of many consecutive electron bunches is stored in a resonator and that can be operated in the low-gain regime with a drastically reduced undulator length compared to high-gain, single-pass FELs. Indeed, the quantum regime requires a high quantum mechanical recoil that translates to an FEL wavelength in the X-ray regime, where up-to-date no suitable high-quality resonators exist. However, in recent years a lot of research was devoted to (classical) X-ray FEL oscillators[16] (XFELO) based on Bragg diffraction from crystals[17] and a large progress was made in this and closely related fields[18, 19, 20, 21, 22, 23, 24].

In the classical limit[25, 26, 27, 28, 29, 30] many photons are emitted into or absorbed from the laser field[31] and the discrete nature[32, 33, 34, 35] of the electron dynamics is washed out[36]. However, for a large quantum mechanical recoil q=2ℏkq = 2\hbar k single-photon processes dominate[11] and the electron occupies only two resonant momentum levels, namely p≃q/2p \simeq q/2 and p−q≃−q/2p-q \simeq-q/2. (Note that we consider the Bambíni–Renìeri frame[37, 38], where the wavenumbers of the wiggler and laser mode are equal, that is kW=kL=kk_{\mathrm{W}} = k_{\mathrm{L}} = k, and the motion of the electrons is non-relativistic.) This definition of the Quantum FEL[11] can be quantified by the condition αn=gn/ωr≪1\alpha_n = g\sqrt{n}/\omega_{\mathrm{r}} \ll1 for the quantum parameter αn\alpha_n, where gg denotes the coupling strength of an electron of mass mm to the fields, nn is the number of laser photons, and ωr=q2/(2mℏ)\omega_{\mathrm{r}} = q^2/(2m\hbar) defines the recoil frequency. Moreover, we require that the initial momentum distribution of the electrons, ρ=ρ(p)\rho= \rho(p), has to be centered around p=q/2p = q/2 and that its width Δp\Delta p has to be small[3, 11], that is Δp<q\Delta p < q.

In an oscillator configuration many electron bunches, each consisting of NN electrons, are injected with a rate 1/τin1/\tau_{\mathrm{in}} and interact subsequently with the fields inside a cavity which simultaneously stores and damps the generated light field[39]. Thus, we identify two contributions of the laser field dynamics, that is (i) Rabi oscillations during the flight time TT of electrons through the undulator with the momentum-dependent Rabi frequency[11, 40]

Ωn(p)=g2(n+1)+ωr2(pq−12)2(1)\Omega_n(p) = \sqrt{g^2(n+1) + \omega_{\mathrm{r}}^2\left(\frac{p}{q} - \frac{1}{2}\right)^2} \tag*{(1)}

and (ii) cavity damping which is characterized by the quality QQ of the cavity. With the help of standard methods from laser and micromaser theory[39, 41, 42, 43, 44, 45] we derive the formal expression

Pn=∏n′=1n[θ2∫dp ρ(p) sinc⁡2(Ωn′−1(p)T)](2)P_n = \prod_{n'=1}^{n}\left[\theta^2 \int\mathrm{d}p\,\rho(p)\,\operatorname{sinc}^2\left(\Omega_{n'-1}(p)T\right)\right] \tag*{(2)}

for the photon statistics PnP_n at steady state, up to a normalization constant. Here we have introduced the pump parameter θ≡gTNa\theta\equiv gT\sqrt{N_a} as well as the inverse loss parameter Na≡NQ/(ωLτinj)N_a \equiv N_Q/(\omega_L\tau_{\mathrm{inj}}) with the laser frequency ωL=ckL\omega_L = ck_L in the laboratory frame. Compared to its classical counterpart we obtain a narrowed photon distribution in the quantum regime ranging from super- to sub-Poissonian behavior. This narrowing implies reduced intensity fluctuations of the emitted radiation and by that an increase of the signal-to-noise ratio in imaging schemes or to an enhancement in sensitivity in interferometric applications.

A shorter undulator length for an oscillator facilitates possible Quantum FEL experiments, for example with respect to the implementation of the necessary optical undulator[4, 6, 12], since the focal area and pulse duration of the required high-power laser are limited. However, similar to a single-pass Quantum FEL there are tight bounds on the electron beam quality, especially for the energy spread and the beam emittance. Indeed, the emittance of electrons from (photo-)cathodes cannot fall below an intrinsic thermal limit[46]. However, advanced schemes like for example ultracold electron sources[47, 48], that use ionization from ultracold atoms, or plasma wakefield acceleration[49] have the potential to provide the required beam quality in the future.

Results

In the following we briefly compare the properties of the Quantum and the classical FEL[50, 51, 52, 53, 54]. For the time being, we restrict ourselves to the small-signal limit, that is gTn≪1gT\sqrt{n} \ll1 in the quantum regime, leading to a change of the photon number which scales linearly with the initial photon number nn. A Gaussian approximation[42] of the photon statistics yields the expression[55]

σ2≅1δ(3)\sigma^2 \cong\frac{1}{\delta} \tag*{(3)}

for the Fano factor[45] σ2≡Var⁡(n^)/⟨n^⟩\sigma^2 \equiv\operatorname{Var}(\hat{n})/\langle\hat{n}\rangle, where δ\delta denotes the relative deviation δ≡(G1−ωLτinj/Q)/G1\delta\equiv(\mathcal{G}_1-\omega_L\tau_{\mathrm{inj}}/Q)/\mathcal{G}_1 of the losses ωLτinj/Q\omega_L\tau_{\mathrm{inj}}/Q from linear gain G1≡(gT)2N\mathcal{G}_1 \equiv(gT)^2N at p=q/2p=q/2. The FEL is operated above threshold and in the small-signal regime if 0<δ≪10 < \delta\ll1, resulting in a value for σ2\sigma^2 that is larger than unity. Hence, the Quantum FEL shows a super-Poissonian behavior in the small-signal limit.

In the classical regime the small-signal limit is characterized by the Madey gain[25, 56]

G1(cl)≅16π3ωrT(gT)2N=16π3ωrTG1(4)\mathcal{G}_1^{(\mathrm{cl})} \cong\frac{16}{\pi^3}\omega_{\mathrm{r}}T(gT)^2N = \frac{16}{\pi^3}\omega_{\mathrm{r}}T\mathcal{G}_1 \tag*{(4)}

for the ‘classical resonance’ 2kpT/m≃π2kpT/m \simeq\pi which gives the maximum of a smooth gain curve and thus differs from the sharp peak at p=q/2p = q/2 for a Quantum FEL. As a consequence, the Fano factor in the classical regime is given by[53, 55, 57]

σcl2≃π/4ωrT1δ(cl)(5)\sigma_{\mathrm{cl}}^2 \simeq\frac{\pi/4}{\omega_r T}\frac{1}{\delta^{(\mathrm{cl})}} \tag*{(5)}

which can be for example derived[55] with an ansatz relying on a Fokker-Planck equation[54, 58, 59]. Since Eq. (5) implies the classical limit of a small recoil, that is ωrT≪1\omega_r T \ll1, the Fano factor for the classical FEL is much larger than the corresponding value for the Quantum FEL given by Eq. (3). In other words, the photon statistics of the Quantum FEL in the small-signal regime is in principle closer to a Poissonian than the statistics of the classical FEL, as illustrated in Figure 1.

Image not extracted

Figure 1. Steady-state photon statistics PnP_n of an FEL in the quantum regime (left) with ωrT=20\omega_{\mathrm{r}}T = 20 and the corresponding distribution function fclf_{\mathrm{cl}} in the classical regime (middle) with ωrT=0.2\omega_{\mathrm{r}}T = 0.2, both depending on the photon number nn and the momentum pp of a peaked initial momentum distribution ρ(p′)=δ(p′−p)\rho(p') = \delta(p' - p). In order to compare quantum with classical regime we have chosen in both cases the same values for the relative deviation δ=δ(cl)=0.05\delta= \delta^{(\mathrm{cl})} = 0.05 of losses from gain and the same mean photon number ⟨n^⟩=3⋅103\langle\hat{n} \rangle= 3 \cdot10^3 (for resonance). This choice of parameters brings us to Na=2⋅104N_a = 2 \cdot10^4 in the quantum and Na=1.5⋅103N_a = 1.5 \cdot10^3 in the classical case, respectively, for the inverse loss parameter Na≡NQ/(ωLτinj)N_a \equiv N_Q/(\omega_L\tau_{\mathrm{inj}}). We observe that the curve in momentum space of a classical FEL covers multiples of the recoil qq and thus is broadened in comparison to its quantum counterpart which is sharply peaked around p=q/2p = q/2. Moreover, the statistics, PnP_n against nn, for a Quantum FEL is narrower as in the classical case which is even more pronounced in the picture on the very right. Here PnP_n and fclf_{\mathrm{cl}} for their respective maximum, at p=q/2p = q/2 and 2kpT/m=π2k p_T/m = \pi, are compared to a Poisson statistics. Both curves for a small-signal FEL show a super-Poissonian behavior, but due to the additional broadening, Eq. (5), of fclf_{\mathrm{cl}} with 1/(ωrT)≫11/(\omega_{\mathrm{r}}T) \gg1 the photon statistics of a Quantum FEL is closer to the Poisson distribution.

In Figure 2 we present the behavior of the mean value ⟨n^⟩\langle\hat{n}\rangle and the Fano factor σ2\sigma^2 from the photon statistics PnP_n of a Quantum FEL oscillator given by Eq. (2) also beyond the small-signal limit. Only momenta close to resonance p=q/2p = q/2 or small momentum widths Δp\Delta p lead to a nonzero mean photon number. In atomic scattering this resonance width is known as velocity-selectivity[60, 61, 62] and solely caused by a Doppler detuning of the momentum distribution. For an efficient operation we even have to go beyond the condition Δp<q\Delta p < q for the momentum spread Δp\Delta p and demand for

Image not extracted

Figure 2. Mean photon number ⟨n^⟩\langle\hat{n}\rangle divided by the inverse loss parameter NaN_a (left) and Fano factor σ2≡Var⁡(n^)/⟨n^⟩\sigma^2 \equiv\operatorname{Var}(\hat{n})/\langle\hat{n}\rangle (right) of a Quantum FEL oscillator for Na=150N_a = 150, and αNa=0.1\alpha N_a = 0.1. Top: as functions of the pump parameter θ\theta and the momentum pp for an initially peaked momentum distribution ρ(p′)=δ(p′−p)\rho(p') = \delta(p' - p) of the electrons. Bottom: as functions of θ\theta and the standard deviation Δp\Delta p of a Gaussian momentum distribution ρ(p′)=(2πΔp)−1exp⁡[−(p′−q/2)2/(2Δp2)]\rho(p') = (\sqrt{2\pi}\Delta p)^{-1}\exp[-(p' - q/2)^2/(2\Delta p^2)] centered at p=q/2p = q/2. For momenta pp deviating from resonance and increasing values of Δp\Delta p, respectively, the mean photon number ⟨n^⟩\langle\hat{n}\rangle decreases. However, this decrease is more slowly for higher values of the pump parameter θ\theta since there the interaction between electron and field is stronger. On the right-hand side, we observe situations with a sub-Poissonian statistics, that is σ2<1\sigma^2 < 1, which is illustrated by the white contour line at σ2=1\sigma^2 = 1.

Δp<αnq≪q,(6)\Delta p < \alpha_n q \ll q, \tag*{(6)}

where we have estimated the width of the resonance in momentum space by means of Eq. (1). However, increasing the pump parameter θ\theta leads to a stronger interaction between electrons and fields and at some point we also observe amplification for off-resonant momenta.

Another feature analogous to the micromaser[45] is the possibility of a sub-Poissonian photon statistics. Indeed, the Fano factor takes on values which are smaller than unity, that is σ2<1\sigma^{2}<1, which is highlighted in Figure 2 by the white contour line for σ2=1\sigma^{2}=1. We observe that this sub-Poisson behavior at resonance p=q/2p=q/2 is not fully destroyed by an increasing momentum spread Δp\Delta p. Moreover, we expect that also beyond the small-signal limit the photon distribution in the classical regime is very broad[53]. Hence, we identify the sub-Poissonian photon statistics in the quantum regime as a unique feature of a Quantum FEL.

Experimental challenges

To this end we discuss the challenges for a possible experimental realization of a Quantum FEL oscillator. For this purpose, we restrict ourselves again to the small-signal limit, that is gTn≪1gT\sqrt{n}\ll1. A large recoil can be achieved by a small wiggler wavelength λW\lambda_{\mathrm{W}} or a high electron energy γ0\gamma_{0} which becomes more apparent when we write the recoil parameter ωrT\omega_{\mathrm{r}}T in terms of the laboratory frame yielding[55]

1≪ωrT∝γ0LλW,(7)1 \ll\omega_{\mathrm{r}}T \propto\gamma_{0}\frac{L}{\lambda_{\mathrm{W}}}, \tag*{(7)}

where we have introduced the interaction length LL. On the other hand, we demand for a moderately high gain to observe amplification of the laser field. For a given value of the linear gain G1\mathcal{G}_{1} we thus obtain for resonance, p=q/2p=q/2, the relation

L=G1cgN∝γ02λWne(8)L=\sqrt{\mathcal{G}_{1}}\frac{c}{g\sqrt{N}}\propto\frac{\gamma_{0}^{2}}{\sqrt{\lambda_{\mathrm{W}}n_{\mathrm{e}}}} \tag*{(8)}

for the length of the wiggler with nen_{\mathrm{e}} denoting the electron density. For increasing values of γ0\gamma_{0} the wiggler length LL grows quadratically while it only scales with λW−1/2\lambda_{\mathrm{W}}^{-1/2} with the wiggler wavelength. Hence, for simultaneously being in the quantum regime and reducing the length of the wiggler we propose to operate a Quantum FEL oscillator with a moderately high electron energy while decreasing the wiggler wavelength. This need for a small wavelength quite naturally forces us to employ an optical undulator[63, 64, 65, 66] which has also been proposed previously[6].

We have to ensure[12] that apart from multiphoton effects the impact of (i) space charge and of (ii) spontaneous emission can be neglected. Hence, we aditionally require that kpL≪1k_{\mathrm{p}}L\ll1 as well as RspL≪1R_{\mathrm{sp}}L\ll1 where we have introduced the plasma wavenumber kp∝ne/γ03k_{\mathrm{p}}\propto\sqrt{n_{\mathrm{e}}/\gamma_{0}^{3}} and the spontaneous decay rate Rsp∝a02/λWR_{\mathrm{sp}}\propto a_{0}^{2}/\lambda_{\mathrm{W}} with a0a_{0} denoting the undulator parameter. However, these quantities are not independent from each other, but are connected

(kpL)2⋅(RspL)=2αf3G1⋅(ωrT)(9)(k_{\mathrm{p}}L)^{2}\cdot(R_{\mathrm{sp}}L)=\frac{2\alpha_{\mathrm{f}}}{3}\mathcal{G}_{1}\cdot(\omega_{\mathrm{r}}T) \tag*{(9)}

via the gain G1\mathcal{G}_{1} and the fine structure constant αf≈1/137\alpha_{\mathrm{f}}\approx1/137. In contrast to a single-pass FEL[12], we straightforwardly identify a regime, where all conditions can be simultaneously satisfied. If we for example demand for a gain G1\mathcal{G}_{1} of 10% and set the recoil parameter to ωrT=2π\omega_{\mathrm{r}}T=2\pi, Eq. (9) allows for kpL=RspL≈0.145k_{\mathrm{p}}L=R_{\mathrm{sp}}L\approx0.145. This more advantageous behavior is due to the low gain which comes in hand with a decreased gain. Furthermore, the low gain will have the freedom to adjust two parameters to fully determine the set of fundamental parameters γ0\gamma_{0}, nen_{\mathrm{e}}, λW\lambda_{\mathrm{W}}, a0a_{0}, and λL\lambda_{\mathrm{L}}. From the choice λL=1\lambda_{\mathrm{L}}=1 Å and λW=1.064 μm\lambda_{\mathrm{W}}=1.064\,\mu\mathrm{m} we derive the values that are listed on the right-hand side of Table 1.

laser wavelengthλL\lambda_{\mathrm{L}} (Å)1.0gain bandwidthΓ\Gamma7.45⋅10−57.45 \cdot10^{-5}
undulator wavelengthλW\lambda_{\mathrm{W}} (µm)1.064electron energyγ0\gamma_{0}51.8
recoil parameterωrT\omega_{\mathrm{r}}T2π2\piundulator lengthLL (mm)1.14
spontaneous emissionRspLR_{\mathrm{sp}}L0.145undulator parametera0a_{0}0.0944
space chargekpLk_{\mathrm{p}}L0.145electron densitynen_{\mathrm{e}} (µm−3^{-3})6.39⋅1046.39 \cdot10^{4}
linear gainG1\mathcal{G}_{1}0.1

Table 1. Proposed fundamental parameters for a Quantum FEL oscillator experiment in the small-signal regime.

Besides such fundamental constraints we have to take geometrical considerations into account. We assume that the optical undulator can be described as a Gaussian laser beam with waist w0w_{0} and Rayleigh length zR≡πw02/λWz_{\mathrm{R}}\equiv\pi w_{0}^{2}/\lambda_{\mathrm{W}}. Following other studies[12] we require w0≥2πσew_{0}\geq\sqrt{2\pi}\sigma_{\mathrm{e}}, where σe\sigma_{\mathrm{e}} denotes the radius of the electron beam. Moreover, the Rayleigh length zRz_{\mathrm{R}} has to be at least as long as half of the interaction length, that is 2zR≥L2z_{\mathrm{R}}\geq L. Similar to the Rayleigh length of a light beam, the quantity

β∗=σe2γ0/εn\beta^* = \sigma_e^2\gamma_0/\varepsilon_n describes the typical length for the divergence of an electron beam. The normalized transverse emittance εn\varepsilon_n is connected to the area of the beam in phase space[46]. From 2β∗≥L2\beta^* \ge L we derive[12] the condition εn≤σe2γ0/L\varepsilon_n \le\sigma_e^2\gamma_0/L.

We also have to relate the dimensions of the electron beam to the emitted radiation[12] characterized by the beam emittance λL/(4π)\lambda_L/(4\pi). A compromise between a strongly diverging laser beam (λL/(4π)>εn/γ0\lambda_L/(4\pi) > \varepsilon_n/\gamma_0) and the loss of transverse coherence (λL/(4π)≪εn/γ0\lambda_L/(4\pi) \ll\varepsilon_n/\gamma_0) is typically found in the regime[12]

0.5γ0λL2π<εn≤10γ0λL2π.(10)0.5\frac{\gamma_0\lambda_L}{2\pi} < \varepsilon_n \le10\frac{\gamma_0\lambda_L}{2\pi}. \tag*{(10)}

Indeed, this relation that ensures transverse coherence was originally derived[67, 68] for self-amplified spontaneous emission (SASE). However, the radiation in an oscillator also starts up from vacuum and reaches steady state after passages of many electron bunches. There is no mechanism evident inducing transverse coherence and we consequently demand that Eq. (10) is also satisfied for an oscillator, assuming transverse coherence for the start-up from vacuum.

In Table 1 we have also included the value for the gain bandwidth Γ\Gamma. In the small-signal limit the condition from momentum selectivity, Eq. (6), reads Δp≤q/(ωrT)\Delta p \le q/(\omega_r T) and we obtain

Δγ0γ0≤Γ≡λW4πL.(11)\frac{\Delta\gamma_0}{\gamma_0} \le\Gamma\equiv\frac{\lambda_W}{4\pi L}. \tag*{(11)}

which results in a challenging value for the electron energy spread at the order of ∼10−4\sim10^{-4}. We can relate fluctuations and deviations of several parameters to the gain bandwidth Γ\Gamma. For this purpose, we note the fundamental relation[12, 69]

λL=λW2(1−cos⁡ϕ)γ02(1+α02+γ02ϑ2),(12)\lambda_L = \frac{\lambda_W}{2(1-\cos\phi)\gamma_0^2}\left(1+\alpha_0^2+\gamma_0^2\vartheta^2\right), \tag*{(12)}

where ϑ\vartheta is the emission angle and ϕ\phi denotes the angle between the propagation directions of electrons and optical undulator. For example, we find the conditions ΔλW/λW≤2Γ\Delta\lambda_{\mathrm{W}}/\lambda_{\mathrm{W}} \leq2\Gamma and ΔI0/I0≤2Γ/a02\Delta I_0/I_0 \leq2\Gamma/a_0^2 for the bandwidth and the intensity fluctuations of the undulator laser with intensity I0∝a02I_0 \propto a_0^2. Divergence of the electron beam leads to deviations from the head-on geometry (ϕ=π\phi= \pi and ϑ=0\vartheta= 0) resulting in the condition$[12] εn≤σe2Γ\varepsilon_{\mathrm{n}} \leq\sigma_{\mathrm{e}}\sqrt{2\Gamma}, where we have set γ0εn=δϕσe\gamma_0\varepsilon_{\mathrm{n}} = \delta\phi\sigma_{\mathrm{e}}.

Moreover, one can derive requirements for the available interaction area due to longitudinal and transversal intensity variations of the undulator laser close to the focus, resulting in Δz/zR≤2Γ/a02\Delta z/z_{\mathrm{R}} \leq\sqrt{2\Gamma/a_0^2}, and, Δx/w0≤Γ/a02\Delta x/w_0 \leq\sqrt{\Gamma/a_0^2}, respectively$[12]. For a minimized beam waist we thus derive a maximum radius of σmax⁡=[Γ/(2a02)]1/4λWL/(2π)\sigma_{\max} = [\Gamma/(2a_0^2)]^{1/4}\sqrt{\lambda_{\mathrm{W}}L/(2\pi)} for the electron beam.

For the sake of completeness, we have to exclude Fermionic effects by demanding$[5] that the number NN of electrons is smaller than the phase space volume of the electron bunch divided by ℏ3\hbar^3 leading to the condition

εn>σe(λC2π)3πγ0neΓ,(13)\varepsilon_{\mathrm{n}} > \sigma_{\mathrm{e}}\sqrt{\left(\frac{\lambda_{\mathrm{C}}}{2\pi}\right)^3\frac{\pi\gamma_0 n_{\mathrm{e}}}{\Gamma}}, \tag*{(13)}

where λC\lambda_{\mathrm{C}} denotes the Compton wavelength of an electron. We also have to ensure that our one-dimensional approach is correct and require$[50] σe>σ1D≡LλL\sigma_{\mathrm{e}} > \sigma_{\mathrm{1D}} \equiv\sqrt{L\lambda_{\mathrm{L}}}.

We visualize the conditions for the radius σe\sigma_{\mathrm{e}} and the normalized emittance εn\varepsilon_{\mathrm{n}} of the electron beam on the left-hand side of Figure 3 at hand of the parameters in Table 1. The regime which allows for a realization of a Quantum FEL oscillator is indicated by a green shading. We realize that the required beam radius lies in the µm range with a normalized emittance at the order of $10^{-2}$ mm mrad, which are extremely challenging requirements for an experiment. On the right-hand side of Figure 3 we sketch the longitudinal and transverse dimensions of electron beam and undulator laser, when (i) only the geometrical requirements are met, that is zR=β∗=L/2z_{\mathrm{R}} = \beta^* = L/2 (top), and (ii) all conditions are satisfied leading to zR>β∗>L/2z_{\mathrm{R}} > \beta^* > L/2 (bottom). We observe that zRz_{\mathrm{R}} and w0w_0 increase in the latter case, but at the same time the electron beam occupies only a small area of the undulator beam. To improve the situation, a traveling-wave Thomson-scattering scheme (TWTS) has been proposed$[12, 66], where the undulator laser has a tilted front and copropagates with the electron beam under an certain angle. While this method relaxes the conditions for the optical undulator, it has no influence on the strict requirements for the electron-beam quality.

Image not extracted

Figure 3. Experimental challenges for a Quantum FEL oscillator: On the left-hand side present different constraints on the transverse normalized emittance εn\varepsilon_n of the electron beam electron beam as functions of the beam radius σe\sigma_e. Only parameter pairs within the green shaded area fulfill these conditions leading to a situation with εn∼10−2\varepsilon_n \sim10^{-2} mm mrad and σ∼μ\sigma\sim\mum, which is extremely challenging for an experimental realization. On the right hand side we have shown the longitudinal, zz, and transversal, xx, dimensions of the undulator and of the electron beam, when (i) only geometrical considerations are taken into account and when (ii) all conditions are satisfied. While in the former case, both, the Rayleigh length zRz_R of the optical undulator and its counterpart β∗\beta^* for the electron beam coincide wit half of the interaction length LL, that is zR=β∗=L/2z_R = \beta^* = L/2, we observe in the latter case an increased value of zRz_R and the relation L/2<β∗<zRL/2 < \beta^* < z_R. Hence, the electron beam occupies only a small area of the undulator beam.

Similar to the Rayleigh length of the undulator, its pulse duration τ0\tau_0 has to be larger than twice the interaction length divided by cc, that is τ0>2L/c∼10\tau_0 > 2L/c \sim10 ps. Further, neglecting slippage implies$[50] that the bunch length cτec\tau_{\mathrm{e}} is larger than the number L/λWL/\lambda_{\mathrm{W}} of undulator periods times the emitted wavelength λL\lambda_{\mathrm{L}} which means for our case that the beam duration τe\tau_{\mathrm{e}} has to be larger than approximately 0.3 fs. The beam duration is the missing parameter to determine the bunch charge and with that the number of emitted photons. For XFELOs a gain reduction is expected when the beam duration is at the order of the inverse bandwidth of the cavity crystals that is typically in the sub-picosecond regime$[20]. We therefore require a beam duration larger than 1 ps. A set of possible parameters for electron beam and optical undulator of a Quantum FEL oscillator is listed in Table 2.

Electron beamOptical undulator
electron energyγ0\gamma_052undulator wavelengthλW\lambda_{\mathrm{W}} (µm)1.064
electron densitynen_{\mathrm{e}} (µm−3^{-3})6.4⋅1046.4 \cdot10^{4}undulator parametera0a_00.094
bunch radiusσe\sigma_{\mathrm{e}} (µm)0.68beam waistw0w_0 (µm)38.6
trans. norm. emittanceεn\varepsilon_{\mathrm{n}} (mm mrad)0.0082Rayleigh lengthzRz_{\mathrm{R}} (mm)4.4
beam durationτe\tau_{\mathrm{e}} (ps)1.2pulse durationτ0\tau_0 (ps)7.6
peak currentIpI_{\mathrm{p}} (A)8.9intensityI0I_0 (PW cm−2^{-2})21.7
bunch chargeQQ (pC)11laser powerP0P_0 (TW)0.5
energy spreadΔγ0γ0\frac{\Delta\gamma_0}{\gamma_0} (‰)0.075undulator linewidthΔλWλW\frac{\Delta\lambda_{\mathrm{W}}}{\lambda_{\mathrm{W}}} (‰)0.15
intensity fluctuationsΔI0I0\frac{\Delta I_0}{I_0} (%)1.7

Table 2. Proposed parameters for the electron beam and the optical undulator of a low gain Quantum FEL oscillator based on the values in Table 1. The values are chosen such that 10610^{6} photons are outcoupled from a cavity with reflectivity R=95.05R = 95.05 % at the wavelength λL=1\lambda_{\mathrm{L}} = 1 Å.

A detailed study of the x-ray cavity lies outside the scope of the present article. Instead, we briefly summarize in the following the most important relations and conditions: The back and forward reflected FEL pulses in the resonator have to arrive at the same time as the electron bunches at the interaction area. Hence, we require Lcav=c/(2frep)L_{\mathrm{cav}} = c/(2f_{\mathrm{rep}}) for the cavity length, where we have introduced the repetition rate frep=1/τinjf_{\mathrm{rep}} = 1/\tau_{\mathrm{inj}} of the accelerator. In experimental implementations this rate is limited and consequently the cavity becomes large. For example, a very optimistic repetition rate of 10 MHz implies a cavity length of 15 m. Despite the small dimensions of the optical undulator, a Quantum FEL oscillator would thus be far away from being a compact device.

The steady-photon number nst≡⟨n^⟩n_{\mathrm{st}} \equiv\langle\hat{n} \rangle can be determined via the relation nst=3δN/Gn_{\mathrm{st}} = 3\delta N/\mathcal{G}. Further, the quality of the resonator is fundamentally tied to its reflectivity RR and its length LcavL_{\mathrm{cav}}[59] and with Lcav=c/(2frep)L_{\mathrm{cav}} = c/(2f_{\mathrm{rep}}) we find the relation

R=1−ωLτinj2Q.(14)R = 1 - \frac{\omega_L \tau_{\mathrm{inj}}}{2Q}. \tag*{(14)}

A relative deviation δ\delta of gain from losses of 1% (corresponding to gTnst≈0.17gT\sqrt{n_{\mathrm{st}}} \approx0.17) thus translates to a reflectivity of 95.05%. We assume that there are no losses due to absorption so that 4.95 % of the photons are transmitted through a thin mirror. For our set of parameters this ratio translates to 10610^6 outcoupled photons. We note that focusing the cavity mode to the size of the electron beam at its waist ∼ ⁣μm\sim\!\mu\mathrm{m} would lead to a Rayleigh length in the cm-regime which is well above the interaction length.

To ensure that the cavity is in resonance with the FEL mode we further require that fluctuations in the cavity length LcavL_{\mathrm{cav}} due to vibrations or analogously effective length changes from timing errors due to fluctuations in the accelerator[19] are small. By means of Eq. (12) we derive ΔLcav/Lcav≤2Γ\Delta L_{\mathrm{cav}}/L_{\mathrm{cav}} \leq2\Gamma which translates to a length change in the mm to cm regime depending on the cavity length. However, the timing error has to be also much smaller than the inverse bandwidth of the cavity which typically leads to a maximum length change in the μm\mu\mathrm{m} regime[19]. Moreover, the x rays produce a heat load on the crystals affecting their reflectivity so that a cooling of the mirrors to a temperature below 100 K becomes necessary[70].

Discussion

The steady-state photon statistics of a Quantum FEL oscillator is closer to the Poisson distribution of a coherent state and eventually shows a sub-Poissonian behavior. Beside the benefit of demonstrating this pure quantum effect, a narrowed photon statistics would lead to smaller intensity fluctuations. In imaging application, this reduction implies for a fixed value of ⟨n^⟩\langle\hat{n} \rangle a smaller signal-to-noise ratio, which is connected to the inverse Fano factor. Besides intensity fluctuations limiting imaging, the sensitivity of interferometric applications depends on the photon statistics of the interference pattern and therefore benefits from a narrower photon distribution.

The interaction length of a low-gain FEL oscillator is smaller compared to a single-pass device. As a result, the conditions on multiphoton effects, spontaneous emission, and space charge can be simultaneously fulfilled. Although the shorter interaction length also improves the challenges for an optical undulator, the limits due to focusing are still prominent, at least for head-on geometries. However, TWTS schemes[66] could help to overcome this issue. Unfortunately, a Quantum FEL oscillator can be hardly reduced in size to be a compact device, since the cavity is much longer than the interaction length due to a limited electron bunch repetition rate. For the quality of the electron beam there are very strict criteria (also because the gain bandwidth for a Quantum FEL oscillator is very small). At least, the shorter interaction length might reduce the effect that the macroscopic beam dynamics[12] impairs the beam quality along the undulator.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Acknowledgements

We are grateful to W. P. Schleich for his stimulating input and continuing support. We also thank C. M. Carmesin, A. Debus, M. A. Efremov, M. Gühr, M. Oppold, R. Sauerbrey, K. Steiniger and A. Wolf for many fruitful discussions. P.K. acknowledges funding by the German Aerospace Center.

Author contributions statement

P.K.: Conceptualization (equal), Data curation (lead), Formal Analysis (equal), Methodology (equal), Validation (supporting), Visualization (lead), Writing – original draft (lead), Writing – review & editing (supporting); E.G.: Conceptualization (equal), Data curation (supporting), Formal Analysis (equal), Methodology (equal), Validation (lead), Visualization (supporting), Writing – original draft (supporting), Writing – review & editing (lead); all authors reviewed the manuscript.

Additional information

The authors declare no competing interests.

References

  1. [1]Schroeder, C. B., Pellegrini, C. & Chen, P. Quantum effects in high-gain free-electron lasers. Phys. Rev. E 64, 056502, DOI: 10.1103/PhysRevE.64.056502 (2001).DOI
  2. [2]Bonifacio, R., Piovella, N., Robb, G. R. M. & Schiavi, A. Quantum regime of free electron lasers starting from noise. Phys. Rev. ST Accel. Beams 9, 090701, DOI: 10.1103/PhysRevSTAB.9.090701 (2006).DOI
  3. [3]Piovella, N. & Bonifacio, R. Inhomogeneous effects in the quantum free electron laser. Nucl. Instrum. & Methods A 560, 240–244, DOI: 10.1016/j.nima.2006.01.128 (2006).DOI
  4. [4]Bonifacio, R., Piovella, N., Cola, M. M. & Volpe, L. Experimental requirements for X-ray compact free electron lasers with a laser wiggler. Nucl. Instrum. & Methods A 577, 745–750, DOI: https://doi.org/10.1016/j.nima.2007.03.024 (2007).
  5. [5]Bonifacio, R., Piovella, N. & Robb, G. R. M. The quantum free electron laser: A new source of coherent, short-wavelength radiation. Fortschr. Phys. 57, 1041–1051, DOI: 10.1002/prop.200900097 (2009).DOI
  6. [6]Bonifacio, R., Fares, H., Ferrario, M., McNeil, B. W. J. & Robb, G. R. M. Design of sub-angström compact free-electron laser source. Opt. Commun. 382, 58–63, DOI: 10.1016/j.optcom.2016.07.007 (2017).DOI
  7. [7]Serbeto, A., Monteiro, L. F., Tsui, K. H. & Mendonça, J. T. Quantum plasma fluid model for high-gain free-electron lasers. Plasma Phys. Control. Fus. 51, 124024, DOI: 10.1088/0741-3335/51/12/124024 (2009).DOI
  8. [8]Brown, M. S., Henderson, J. R., Campbell, L. T. & McNeil, B. W. J. An extended model of the quantum free-electron laser. Opt. Express 25, 33429–33438, DOI: 10.1364/OE.25.033429 (2017).DOI
  9. [9]Anisimov, P. M. Quantum theory for 1D X-ray free electron laser. J. Mod. Opt. 65, 1370–1377, DOI: 10.1080/09500340.2017.1375567 (2018).DOI
  10. [10]Schaap, B. H., Schouwen­aars, S. & Luiten, O. J. A Raman quantum free-electron laser model. Phys. Plasmas 29, 113302, DOI: 10.1063/5.0106439 (2022).DOI
  11. [11]Kling, P. et al. What defines the quantum regime of the free-electron laser? New J. Phys. 17, 123019, DOI: 10.1088/1367-2630/17/12/123019 (2015).DOI
  12. [12]Debus, A., Steiniger, K., Kling, P., Carmesin, C. M. & Sauerbrey, R. Realizing quantum free-electron lasers: a critical analysis of experimental challenges and theoretical limits. Phys. Scr. 94, 074001, DOI: 10.1088/1402-4896/aaf951 (2019).
  13. [13]Kling, P., Giese, E., Carmesin, C. M., Sauerbrey, R. & Schleich, W. P. High-gain quantum free-electron laser: Emergence and exponential gain. Phys. Rev. A 99, 053823, DOI: 10.1103/PhysRevA.99.053823 (2019).DOI
  14. [14]Kling, P., Giese, E., Carmesin, C. M., Sauerbrey, R. & Schleich, W. P. High-gain quantum free-electron laser: Long-time dynamics and requirements. Phys. Rev. Res. 3, 033232, DOI: 10.1103/PhysRevResearch.3.033232 (2021).DOI
  15. [15]Kling, P. & Giese, E. Multiphoton processes and higher resonances in the quantum regime of the free-electron laser. Phys. Rev. Res. 5, 033057, DOI: 10.1103/PhysRevResearch.5.033057 (2023).
  16. [16]Kim, K.-J., Shvyd’ko, Y. & Reiche, S. A proposal for an X-ray free-electron laser oscillator with an energy-recovery linac. Phys. Rev. Lett. 100, 244802, DOI: 10.1103/PhysRevLett.100.244802 (2008).
  17. [17]Colella, R. & Luccio, A. Proposal for a free electron laser in the X-ray region. Opt. Commun. 50, 41–44, DOI: https://doi.org/10.1016/0030-4018(84)90009-9 (1984).
  18. [18]Huang, Z. & Ruth, R. D. Fully coherent X-ray pulses from a regenerative-amplifier free-electron laser. Phys. Rev. Lett. 96, 144801, DOI: 10.1103/PhysRevLett.96.144801 (2006).
  19. [19]Lindberg, R. R. & Kim, K.-J. Mode growth and competition in the x-ray free-electron laser oscillator start-up from noise. Phys. Rev. ST Accel. Beams 12, 070702, DOI: 10.1103/PhysRevSTAB.12.070702 (2009).
  20. [20]Lindberg, R. R., Kim, K.-J., Shvyd’ko, Y. & Fawley, W. M. Performance of the x-ray free-electron laser oscillator with crystal cavity. Phys. Rev. ST Accel. Beams 14, 010701, DOI: 10.1103/PhysRevSTAB.14.010701 (2011).
  21. [21]Shvyd’ko, Y., Stoupin, S. & Terentyev, S. Near-100% Bragg reflectivity of X-rays. Nat. Photonics 5, 539–542, DOI: https://doi.org/10.1038/nphoton.2011.197 (2011).
  22. [22]Marcus, G. & Decker, F.-J. Cavity-based free-electron laser research and development: A joint Argonne National Laboratory and SLAC National Laboratory collaboration. In Proceedings of the 39th International Free Electron Laser Conference, Hamburg, Germany, DOI: 10.18429/JACoW-FEL2019-TUD04 (2019).
  23. [23]Rauer, P. et al. Cavity based x-ray free electron laser demonstrator at the European X-ray Free Electron Laser facility. Phys. Rev. Accel. Beams 26, 020701, DOI: 10.1103/PhysRevAccelBeams.26.020701 (2023).
  24. [24]Margraf, R. et al. Low-loss stable storage of 1.2 Å x-ray pulses in a 14 m Bragg cavity. Nat. Photonics 17, 878–882, DOI: 10.1038/s41566-023-01267-0 (2023).
  25. [25]Madey, J. M. J. Simulated emission of bremsstrahlung in a periodic magnetic field. J. Appl. Phys. 42, 1906–1913, DOI: 10.1063/1.1660466 (1971).
  26. [26]Hopf, F. A., Meystre, P., Scully, M. O. & Louisell, W. H. Classical theory of a free-electron laser. Opt. Commun. 18, 413 – 416, DOI: 10.1016/0030-4018(76)90283-2 (1976).
  27. [27]Colson, W. B. One-body electron dynamics in a free electron laser. Phys. Lett. A 64, 190 – 192, DOI: 10.1016/0375-9601(77)90712-5 (1977).
  28. [28]Sprangle, P. & Smith, R. A. Theory of free-electron lasers. Phys. Rev. A 21, 293, DOI: 10.1103/PhysRevA.21.293 (1980).DOI
  29. [29]Bonifacio, R., Pellegrini, C. & Narducci, L. M. Collective instabilities and high-gain regime in a free electron laser. Opt. Commun. 50, 373–378, DOI: 10.1016/0030-4018(84)90105-6 (1984).
  30. [30]Huang, Z. & Kim, K.-J. Review of x-ray free-electron laser theory. Phys. Rev. ST Accel. Beams 10, 034801, DOI: 10.1103/PhysRevSTAB.10.034801 (2007).DOI
  31. [31]McIver, J. K. & Fedorov, M. V. Quantum theory of stimulated processes in a free-electron laser in a strong field. J. Exp. Theor. Phys 49, 1012 – 1019 (1979).
  32. [32]Friedman, A., Gover, A., Kurizki, G., Ruschin, S. & Yariv, A. Spontaneous and stimulated emission from quasifree electrons. Rev. Mod. Phys. 60, 471, DOI: 10.1103/RevModPhys.60.471 (1988).DOI
  33. [33]Becker, W. & McIver, J. K. Fully quantized many-particle theory of a free-electron laser. Phys. Rev. A 27, 1030, DOI: 10.1103/PhysRevA.27.1030 (1983).DOI
  34. [34]Gea-Banacloche, J. Quantum theory of the free-electron laser: Large gain, saturation, and photon statistics. Phys. Rev. A 31, 1607, DOI: 10.1103/PhysRevA.31.1607 (1985).
  35. [35]Ciocci, F., Dattoli, G., Renieri, A. & Torre, A. About the mathematical aspects of the FEL and their relationship to quantum optics. Phys. Rep. 141, 1 – 50, DOI: 10.1016/0370-1573(86)90106-7 (1986).DOI
  36. [36]Carmesin, C. M., Kling, P., Giese, E., Sauerbrey, R. & Schleich, W. P. Quantum and classical phase-space dynamics of a free-electron laser. Phys. Rev. Res. 2, 023027, DOI: 10.1103/PhysRevResearch.2.023027 (2020).DOI
  37. [37]Bambini, A. & Renieri, A. The free electron laser: A single-particle classical model. Lett. Nuovo Cimento 21, 399–404, DOI: 10.1007/BF02762613 (1978).DOI
  38. [38]Bambini, A., Renieri, A. & Stenholm, S. Classical theory of the free-electron laser in a moving frame. Phys. Rev. A 19, 2013 – 2025, DOI: 10.1103/PhysRevA.19.2013 (1979).DOI
  39. [39]Sargent III, M., Scully, M. O. & Lamb Jr., W. E. Laser Physics (Addison-Wesley, Reading, 1974).
  40. [40]Jaynes, E. T. & Cummings, F. W. Comparison of quantum and semiclassical radiation theories with application to the beam maser. Proc. IEEE 51, 89 – 109, DOI: 10.1109/PROC.1963.1664 (1963).DOI
  41. [41]Meschede, D., Walther, H. & Müller, G. One-atom maser. Phys. Rev. Lett. 54, 551, DOI: 10.1103/PhysRevLett.54.551 (1985).DOI
  42. [42]Lugiato, L. A., Scully, M. O. & Walther, H. Connection between microscopic and macroscopic maser theory. Phys. Rev. A 36, 740, DOI: 10.1103/PhysRevA.36.740 (1987).DOI
  43. [43]Filipowicz, P., Javanainen, J. & Meystre, P. Theory of a microscopic maser. Phys. Rev. A 34, 3077, DOI: 10.1103/PhysRevA.34.3077 (1986).DOI
  44. [44]Guzman, A., Meystre, P. & Wright, E. Semiclassical theory of a micromaser. Phys. Rev. A 40, 2471, DOI: 10.1103/PhysRevA.40.2471 (1989).DOI
  45. [45]Schleich, W. P. Quantum Optics in Phase Space (Wiley-VCH, Weinheim, 2001).
  46. [46]Reiser, M. Theory and Design of Charged Particle Beams (Wiley-VCH, Weinheim, 2008).
  47. [47]Claessens, B. J., van der Geer, S. B., Taban, G., Vredenbregt, E. J. D. & Luiten, O. J. Ultracold electron source. Phys. Rev. Lett. 95, DOI: 10.1103/PhysRevLett.95.164801 (2005).DOI
  48. [48]Franssen, J. G. H., de Raadt, T. C. H., van Nijnhuis, M. A. W. & Luiten, O. J. Compact ultracold electron source based on a grating magneto-optical trap. Phys. Rev. Accel. Beams 22, 023401, DOI: 10.1103/PhysRevAccelBeams.22.023401 (2019).DOI
  49. [49]Habib, A. F. et al. Attosecond-angström free-electron-laser towards the cold beam limit. Nat. communications 14, 1054, DOI: 10.1038/s41467-023-36592-z (2023).DOI
  50. [50]Schmüser, P., Dohlus, M. & Rossbach, J. Ultraviolet and Soft X-Ray Free-Electron Lasers (Springer, Heidelberg, 2008).
  51. [51]Becker, W. & Zubairy, M. S. Photon statistics of a free-electron laser. Phys. Rev. A 25, 2200, DOI: 10.1103/PhysRevA.25.2200 (1982).DOI
  52. [52]Becker, W., Scully, M. O. & Zubairy, M. S. Quantum statistical properties of radiation in a free-electron laser. Coherence Quantum Opt. V, L. Mandel E. Wolf, eds 811–818 (1984).DOI
  53. [53]Gea-Banacloche, J. Steady-state photon statistics of a free-electron laser. Phys. Rev. A 33, 1448, DOI: 10.1103/PhysRevA.33.1448 (1986).
  54. [54]Orszag, M. Free-electron-laser linewidth obtained from a master Fokker-Planck equation. Phys. Rev. A 36, 189, DOI: 10.1103/PhysRevA.36.189 (1987).DOI
  55. [55]Kling, P. Theory of the free-electron laser: From classical to quantum, DOI: 10.18725/OPARU-5238 (2017).DOI
  56. [56]Madey, J. M. J., Schwettman, H. A. & Fairbank, W. M. A free electron laser. IEEE Transactions on Nucl. Sci. 20, 980–983, DOI: 10.1109/TNS.1973.4327304 (1973).DOI
  57. [57]Becker, W. & McIver, J. K. Many-particle quantum theory for a class of free-electron devices. Phys. Reports 154, 205–245, DOI: 10.1016/0370-1573(87)90068-8 (1987).DOI
  58. [58]Risken, H. Distribution-and correlation-functions for a laser amplitude. Z. Phys. 186, 85–98, DOI: 10.1007/BF01383512 (1965).DOI
  59. [59]Haken, H. Light, Volume 2 – Laser Light Dynamics (North-Holland Physics Publishing, Amsterdam, 1985).
  60. [60]Giese, E., Roura, A., Tackmann, G., Rasel, E. M. & Schleich, W. P. Double Bragg diffraction: A tool for atom optics. Phys. Rev. A 88, 053608, DOI: 10.1103/PhysRevA.88.053608 (2013).DOI
  61. [61]Giltner, D. M., McGowan, R. W. & Lee, S. A. Theoretical and experimental study of the Bragg scattering of atoms from a standing light wave. Phys. Rev. A 52, 3966 – 3972, DOI: 10.1103/PhysRevA.52.3966 (1995).DOI
  62. [62]Szigeti, S. S., Debs, J. E., Hope, J. J., Robins, N. P. & Close, J. D. Why momentum width matters for atom interferometry with Bragg pulses. New J. Phys. 14, 023009, DOI: 10.1088/1367-2630/14/2/023009 (2012).DOI
  63. [63]Schlicher, R. R., Scully, M. O. & Walther, H. Vorschlag für einen kompakten Freie-Elektronen Laser mit einem elektromagnetischen Undulator für den Infrarot- und weichen Röntgenbereich. Tech. Rep., Max-Planck Institut für Quantenoptik (1987).
  64. [64]Gea-Banacloche, J., Moore, G. T., Schlicher, R. R., Scully, M. O. & Walther, H. Soft X-ray free-electron laser with a laser undulator. IEEE J. Quantum Electron. 23, 1558 – 1570, DOI: 10.1109/JQE.1987.1073559 (1987).
  65. [65]Sprangle, P., Hafizi, B. & Peñano, J. R. Laser-pumped coherent x-ray free-electron laser. Phys. Rev. ST Accel. Beams 12, 050702, DOI: 10.1103/PhysRevSTAB.12.050702 (2009).DOI
  66. [66]Steiniger, K. et al. Optical free-electron lasers with traveling-wave Thomson-scattering. J. Phys. 47, 234011, DOI: 10.1088/0953-4075/47/23/234011 (2014).DOI
  67. [67]Saldin, E. L., Schneidmiller, E. A. & Yurkov, M. V. Coherence properties of the radiation from X-ray free electron laser. Opt. Commun. 281, 1179–1188, DOI: https://doi.org/10.1016/j.optcom.2007.10.044 (2008).
  68. [68]Saldin, E. L., Schneidmiller, E. A. & Yurkov, M. V. Output power and degree of transverse coherence of X-ray free electron lasers. Opt. Commun. 281, 4727–4734, DOI: https://doi.org/10.1016/j.optcom.2008.05.033 (2008).
  69. [69]Ride, S. K., Esarey, E. & Baine, M. Thomson scattering of intense lasers from electron beams at arbitrary interaction angles. Phys. Rev. E 52, 5425–5442, DOI: 10.1103/PhysRevE.52.5425 (1995).DOI
  70. [70]Kim, K.-J., Shvyd’ko, Y. & Lindberg, R. R. An X-ray free-electron laser oscillator for record high spectral purity, brightness, and stability. Synchrotron Radiat. News 25, 25–31, DOI: 10.1080/08940886.2012.645421 (2012).DOI

Paper details

vlnc:W_bwgaxce2
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 4

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents