WorksPhysics and Astronomy
Generation of Radio Frequency Radiation by Femtosecond Filaments
Abstract
Recent experiments have shown that femtosecond filamentation plasmas generate ultra-broadband radio frequency radiation (RF). We show that a combination of plasma dynamics is responsible for the RF: a plasma wake field develops behind the laser pulse, and this wake excites (and copropagates with) a surface wave on the plasma column. The surface wave proceeds to detach from the end of the plasma and propagates forward as the RF pulse. We have developed a four stage model of these plasma wake surface waves and find that it accurately predicts the RF from a wide range of experiments, including both 800 nm and 3.9 m laser systems.
The development of chirped pulse amplification [1] has enabled an array of exotic physics [2, 3, 4], including atmospheric filamentation [5]. Filamentation in turn produces interesting and useful effects, from supercontinuum generation [6, 7], broadband THz radiation [8, 9], and RF pulses, which have recently been explored in great detail [10, 11, 12, 13, 14, 15]. We have discovered that a novel combination of a plasma wake field and a surface wave is responsible for this filament RF. Femtosecond lasers are becoming important spectroscopic tools (both in the lab [16] and for remote sensing [17]), and the added production of pulsed RF promises to expand the utility of these systems [18, 19, 20], and to provide insight into the physics of filamentation.
During typical nm filamentation Kerr self-focusing and plasma defocusing are in rough balance, resulting a plasma column with a density of [21, 22]. This electron density corresponds to a plasma frequency of rad/s, which is closely linked to the generation of THz radiation. In the single color THz theory of [8] the freshly ionized electrons receive a ponderomotive push from the laser pulse, thus exciting a coherent longitudinal current in the plasma (which is quickly damped by the high collision frequency). This short current pulse translates at behind the laser pulse, thereby producing a conical shell of radially polarized THz radiation.
The RF resembles the THz radiation in some ways, and differs critically in others. It has a similar conical shell spatial profile and radial polarization, which indicates that a GHz scale longitudinal current pulse is also generated within the plasma. However it also has a well defined broadband peak in the 5-15 GHz range (and is not the tail of the THz radiation [11, 13, 14]) and the RF amplitude grows strongly with decreasing pressure [11, 13], in contrast to the THz which falls off sharply [23]. A distinct physical mechanism thus must be responsible for the RF.
Exploratory Particle In Cell (PIC) simulations [24, 25, 26, 27, 28, 29, 30] (see section 1 in [31] for more details) provided the key insights that explain the RF generation. They revealed that a hot shell of electrons expands off the plasma surface into the surrounding atmosphere over roughly 50 ps (as anticipated by [32]), and that the corresponding current density depends strongly on the electron-neutral collision frequency. Subsequent simulations demonstrated that this Plasma Wake field (PW) excites a Surface Wave (SW) on the outer boundary of the plasma (namely, a long wavelength Surface Plasmon Polariton (SPP) [33, 34], with a speed approaching ). Finally the SW was found to effectively detach from the end of the plasma, with most of the energy being then converted into a broadband RF pulse.
The PWSW numerical model is split into 4 stages, as length scales spanning 7 orders of magnitude need to be resolved. First the initial electron distribution function is determined. In a typical nm experiment a linearly polarized, single color, 40 m pulse is focused into a quartz cell with a mirror, which sets the beam waist to 30 m (10 mJ and focusing are common for m). The resulting Keldysh parameter is close to 1, so we use the general ionization rate from [35], which is supported by experimental data [36] near atmospheric pressure (other ionization mechanisms will likely be needed for 10 m lasers [37]). A laser intensity of generates enough plasma at 1 atm for intensity clamping [21] to occur (with ), which leads to an estimated plasma length of 25 cm, and a radius of mm (see section 2 of [31] for more details). At 1 Torr increases to , which ionizes much of the , and the plasma is reduced to cm and mm. In turn, the lower energy 3.9 m laser plasmas are shorter, with ranging from about 1 to 2 cm.
These estimates match well (within 25%) with photographic measurements of the plasma taken at different pressures (available in section 2 of [31]). Simulations of filamentation with these parameters have also been performed using a Unidirectional Pulse Propagation Equations (UPPE) solver [38]. These agree with the overall dimensions of the plasma column, but also indicate that there is rich fine scale structure throughout the plasma (as expected as we are in the multifilamentary regime [39, 40, 41]). For this letter we simplify and do not include fine scale plasma perturbations. The plasma column is approximated as having a constant density central core due to intensity clamping, and then a 100 radial ramp down to zero density at the outer radius .
The velocity distribution is determined by our PIC code: given the air is ionized at the rate [35] and the new electrons are accelerated through the remainder of the pulse [28] (code to perform these calculations is included in [31]). In general the velocity distribution is influenced by both the details of strong field ionization near (e.g. [42]), and by the deformation of the laser pulse during filamentation. For this work we further simplify and assume that the electrons are ionized with zero initial velocity and are then accelerated by the remainder of a Gaussian pulse (with polarization and propagating in ). As a whole the initial is highly non-Maxwellian, and at 100 Torr has a peak kinetic energy eV, and an average kinetic energy eV, while these increase at 1 Torr to eV and eV. For 3.9 lasers the kinetic energy is times larger, as the laser intensities are comparable and the energy scales as .
We next consider the evolution of the plasma column. We construct thin transverse slices of plasma given , with periodic boundary conditions used in the longitudinal direction (which is valid to leading order as the electron velocities are a small fraction of ) and simulate the radial evolution with our PIC code. The Debye length is quite small: nm, so we use an energy conserving method [43] to calculate the Lorentz force. The electron-neutral elastic collision frequency depends on the cross sections for and , which for our energies is roughly 10 [44]. In turn the electron-ion momentum-transfer collision frequency is given by , where [45]. The resulting radial current density and electron density are then recorded as functions of radius and time (more detail can be found in section 3 of [31]). These are well resolved with a mesh resolution of , and a macro-particle weight of on the outer edge of the plasma.
The electron number density for nm simulations of the PW at 100, 10, and 1 Torr can be seen in Figure 1. The outer edge of the plasma at has a simplified step function profile, with at radius mm. Thus, in addition to the PW being launched off of the edge of this plasma, a coherent radial plasma frequency oscillation of GHz is stimulated in the interior [46] and a SPP at GHz is excited on the surface [33, 34, 47]. The PW that expands out into the neutral atmosphere () is insensitive to the density of the outer edge of the plasma (see Figure 2), as opposed to the excited surface and volume plasmons.
Figure 1. Log plots of the electron density showing the plasma wake variation at pressures of 100 (a), 10 (b), and 1 Torr (c) for a nm laser pulse. The plasma outer radius has been set to mm for all 3 pressures for comparison purposes.
Figure 2. Log plots of the radial current density at 1 Torr (800 nm source), as it evolves in a transverse slice of plasma. The initial density at the edge of the plasma has been given a step function profile, with a value of in the top plot and on the bottom. The profiles of the plasma wakes outside the column ( mm) are quite similar, while at the surface SPPs are excited at roughly and GHz respectively, along with higher frequency volume plasmons in the interior.
Analytic approximations provide a useful complement to the simulations. Consider the late time electron density at a small distance off the surface of the plasma at radius . Using Gauss’s law and approximating as for we find the that the number of electrons per unit length that escape to scales as:
where is the kinetic energy of the hot electrons that lead the escape. The radial electrostatic field that corresponds to has a simple form: . At nm and 1 Torr eV and m, giving V/m. This allows for an estimate of the PW evolution timescale: an electron with energy eV in an electric field of this magnitude follows a parabolic trajectory over 100 ps. This compares well with the spread of electrons seen in Figure 1. The magnitude of the radial current density can likewise be approximated as:
where the effective velocity is the minimum of the tail velocity and the Fickian diffusion speed . At 1 Torr this gives A/m. Finally we note that caution is needed for small values: Eqs. (1) and (2) only hold up to densities that are comparable to the original plasma edge density.
We next switch from these transverse PIC simulations to a continuum FDTD-Drude model in a 2D axisymmetric coordinate system [26] that spans the length of the plasma. The Drude model: is integrated into the code via an auxiliary differential equation [48]. A mixed resolution of m and m suffices to resolve the SWs. The PW current profile is driven across the surface of the plasma column in the longitudinal direction at the speed of light. In this work only the external radial currents are used to drive the SW and subsequent RF (see section 4 of [31] for more detail). The higher frequency volume plasmons can also excite waves in neighboring lower density plasma, but we suspect that including fine scale plasma structure will act as surface roughness [49] and cause these to continuously detach.
As expected the PW current excites a broadband surface wave with similar frequency content. The frequency of the wave is considerably lower than the limit SPP frequency and it thus travels at velocity which approaches the speed of light. The resulting phase-matched copropagation leads to a steady growth in the surface wave intensity (see also [50, 51]).
The SW is well approximated by the Sommerfeld-Goubau [52, 53] solution for SPPs on a cylinder of finite conductivity. The external radial component of the electric field has the form:
for waves of frequency propagating in the direction, with an amplitude of at the surface. The complex wavenumber describes the SW wavelength and attenuation length scale, is the free space wavenumber , is a Hankel function of the first kind with order 1, and . For radii the Hankel function with complex argument is approximately , and for larger radii it falls off exponentially (as is typical for SPPs).
Recent work [54, 55] has simplified the calculation of through the use of the Lambert function. Given a typical plasma conductivity a complex can be defined:
(where is the Euler-Mascheroni constant), and is then given by:
where the first branch of is used. At 1 Torr the plasma has a radius mm and conductivity S/m, and thus for a 10 GHz SW we find . The dispersive SW thus travels at about 0.988, attenuates by a factor of every 30 cm (if not being actively excited), and has a transverse envelope out to cm, all of which compares well with simulations.
We next consider the excitation of the surface wave by the plasma wake currents. An isolated Hertzian dipole with uniform current along an antenna of length will radiate electromagnetic waves at a wavelength with power: . In our case we have a radial current distribution of approximate length that is normal to a conducting cylinder of radius . When there is a separation of length scales and an overall radial current of then the power that is primarily radiated into SWs is:
(where we have simplified with ). We note that the dimensionless term (compare to for the Hertzian dipole), has different asymptotic scalings in other length scale regimes. Equation (6) applies both to a stationary radial antenna and the wavetrains it emits, and to our case, where the antenna translates at with its comoving SW pulse.
The total effective current is found by integrating the current density from 0 to , and absent dissipation the amplitude of the SW in (3) would grow linearly (for constant ). However there is dissipation due to the finite conductivity of the plasma, and the attenuation coefficient is similar in magnitude to . The SW thus grows linearly initially and then saturates after several , and the effective total radial current after propagating a distance is:
We express the amplitude of the final SW in terms of its longitudinal current, as . Integrating (3) to find the SW peak power and setting this equal to (6) we find:
For the nm laser this gives A at the end of the plasma at 1 Torr, a current of A at 10 Torr, and A at 100 Torr, which are close to the measured currents in the corresponding simulations.
When the surface wave reaches the end of plasma the majority of the energy detaches and propagates forward as the RF pulse, with the reflected wave decaying as it travels back down the plasma. Figure 3 shows an example of this, where the SW has just detached from a 2 cm plasma column produced by a m laser source at 10 Torr. In this way the system resembles a surface wave end-fire antenna [56, 57]. The second half of the primarily one cycle RF pulse forms as the SW separates from the plasma, with some variation in the profile dependent on the abruptness of the detachment. The recorded currents from the SW simulation are used in a m laboratory scale axisymmetric FDTD model, with resolution mm. Plots of the resulting Fourier transformed RF field strengths at 10 Torr for both nm and m simulations can be seen in Figure 4 (see section 5 of [31] for more detail).
Figure 3. Log plot of electric field strength shortly after the detachment of a m, 10 Torr surface wave from a 2 cm plasma column, as simulated by the axisymmetric FDTD-Drude model.
Figure 4. Plots of simulated electric field strength as a function of frequency and angle for RF pulses from a 40 mJ, , 800 nm laser source (left, [11]) and a 12 mJ, , 3.9 m source (right, [13]), with both cases run at 10 Torr. The numerical profiles and amplitudes closely match the experimental results.
To directly compare with lab results the simulated electric fields are converted to peak to peak voltages via a frequency dependent calibration factor (which captures the horn antenna response [11]). We adapt an equation for energy spectral density radiated per unit solid angle derived for THz radiation [8]:
to construct voltage profiles for our analytic approximations. The average amplitude of (8) over the length
of the plasma is applied to a 10 GHz pulse, which is Fourier transformed for use with (9) (along with the calibration factor), and then converted back to an effective (code that implements all of these analytic approximations is available in [31]).
In general simulations with varying pressures, focal geometries, and laser pulse energies and wavelengths all reproduce RF that nicely matches recent experimental results [11, 12, 13, 14]. The magnitude of the simulated RF is a bit lower than the lab data for the runs shown in Figure 5, but is within the errors for both the simulations and in the experimental data. Subtle changes in Ti:Sapphire laser alignment generates amplitude variations at a given pressure (with the spatial and frequency content being much more stable). It is suspected that the slight laser pulse variations are amplified during filamentation, leading to fine scale changes in plasma radius, density, and electron velocities. Plausible variations in the same parameters lead to comparable changes in the simulated RF amplitude as well.
Figure 5. Polar plots of the detector peak to peak Voltage for the nm laser source as a function of angle and pressure as measured in the lab (left), and as calculated by both the simulation framework (right, triangles), and by the associated analytic approximations (right, ticks) that culminate in (9).
We have demonstrated that a plasma wake field that excites a surface wave is responsible for the generation of filament RF. The rich internal dynamics of the plasma merits further research: we expect that stochastic higher frequency RF may inform on the fine scale structure of the plasma and the nonlinear optics that generate it. The ionization model needs to be expanded to include more physical effects (e.g. [37, 42, 58]), and strong field ionization can be further explored with noble gas filament RF. The plasma can also be driven into the strongly coupled regime with harder focusing, and we intend to model resulting RF with the use of [59, 60]. Finally we note that it may be possible to recreate the core mechanism – an antenna that copropagates with the wave that it excites – in other settings and thereby drive strong coherent radiation.
Acknowledgments
The authors thank Edward Ruden, Serge Kalmykov and Charles Armstrong for useful discussions, and the Air Force Office of Scientific Research (AFOSR) for support via Laboratory Tasks No. FA9550-18RDCOR0009 and No. FA9550-19RDCOR027. This work was supported in part by high-performance computer time and resources from the DoD High Performance Computing Modernization Program. Approved for public release; distribution is unlimited. Public Affairs release approval No. AFRL-2020-0489.
References
References
- [1]Donna Strickland and Gerard Mourou, “Compression of amplified chirped optical pulses,” Optics communications 56, 219–221 (1985).
- [2]Jérôme Faure, Yannick Glinec, A Pukhov, S Kiselev, S Gordienko, E Lefebvre, J-P Rousseau, F Burgy, and Victor Malka, “A laser–plasma accelerator producing monoenergetic electron beams,” Nature 431, 541–544 (2004).pubmed.ncbi.nlm.nih.gov/15457253
- [3]X Liu, D Du, and G Mourou, “Laser ablation and micromachining with ultrashort laser pulses,” IEEE journal of quantum electronics 33, 1706–1716 (1997).DOI
- [4]A Di Piazza, C Müller, KZ Hatsagortsyan, and Ch H Keitel, “Extremely high-intensity laser interactions with fundamental quantum systems,” Reviews of Modern Physics 84, 1177 (2012).DOI
- [5]A Braun, G Korn, X Liu, D Du, J Squier, and G Mourou, “Self-channeling of high-peak-power femtosecond laser pulses in air,” Optics letters 20, 73–75 (1995).
- [6]A Brodeur and SL Chin, “Ultrafast white-light continuum generation and self-focusing in transparent condensed media,” JOSA B 16, 637–650 (1999).DOI
- [7]VP Kandidov, OG Kosareva, IS Golubtsov, W Liu, A Becker, N Aközbek, CM Bowden, and SL Chin, “Self-transformation of a powerful femtosecond laser pulse into a white-light laser pulse in bulk optical media (or supercontinuum generation),” Applied Physics B 77, 149–165 (2003).DOI
- [8]CD Amico, Aurélien Houard, Selcuk Akturk, Yi Liu, J Le Bloas, Michel Franco, Bernard Prade, Arnaud Couairon, VT Tikhonchuk, and André Mysyrowicz, “Forward thz radiation emission by femtosecond filamentation in gases: theory and experiment,” New Journal of Physics 10, 013015 (2008).DOI
- [9]VA Andreeva, OG Kosareva, NA Panov, DE Shipilo, PM Solyankin, MN Esaulkov, P González de Alaiza Martínez, AP Shkurinov, VA Makarov, L Bergé, et al., “Ultrabroad terahertz spectrum generation from an air-based filament plasma,” Physical review letters 116, 063902 (2016).DOI
- [10]Benjamin Forestier, Aurélien Houard, Magali Durand, Yves-Bernard André, Bernard Prade, J-Y Dauvignac, Franck Perret, Ch Pichot, Michel Pellet, and André Mysyrowicz, “Radiofrequency conical emission from femtosecond filaments in air,” Applied Physics Letters 96, 141111 (2010).DOI
- [11]Alexander Englesbe, Jennifer Elle, Remington Reid, Adrian Lucero, Hugh Pohle, Matthew Domonkos, Serge Kalmykov, Karl Krushelnick, and Andreas Schmitt-Sody, “Gas pressure dependence of microwave pulses generated by laser-produced filament plasmas,” Optics Letters 43, 4953–4956 (2018).DOI
- [12]A Janicek, E Thornton, T Garrett, A Englesbe, J Elle, and A Schmitt-Sody, “Length dependence on broadband microwave emission from laser-generated plasmas,” IEEE Transactions on Plasma Science 48, 1979–1983 (2020).DOI
- [13]AV Mitrofanov, DA Sidorov-Biryukov, MM Nazarov, AA Voronin, MV Rozhko, AB Fedotov, and AM Zheltikov, “Coherently enhanced microwave pulses from midinfrared-driven laser plasmas,” Optics Letters 46, 1081–1084 (2021).DOI
- [14]Alexander Englesbe, Jennifer Elle, Robert Schwartz, Travis Garrett, Daniel Woodbury, Dogeun Jang, Ki-Yong Kim, Howard Milchberg, Remington Reid, Adrian Lucero, et al., “Ultrabroadband microwave radiation from near-and mid-infrared laser-produced plasmas in air,” Physical Review A 104, 013107 (2021).DOI
- [15]AV Mitrofanov, AA Voronin, MV Rozhko, DA Sidorov-Biryukov, MM Nazarov, AB Fedotov, and AM Zheltikov, “Polarization and spatial mode structure of mid-infrared-driven terahertz-to-microwave radiation,” ACS Photonics 8, 1988–1996 (2021).DOI
- [16]Tobias Kampfrath, Koichiro Tanaka, and Keith A Nelson, “Resonant and nonresonant control over matter and light by intense terahertz transients,” Nature Photonics 7, 680–690 (2013).
- [17]Timur A Labutin, Vasily N Lednev, Alexey A Ilyin, and Andrey M Popov, “Femtosecond laser-induced breakdown spectroscopy,” Journal of Analytical Atomic Spectrometry 31, 90–118 (2016).DOI
- [18]G Barratt Park and Robert W Field, “Perspective: The first ten years of broadband chirped pulse fourier transform microwave spectroscopy,” The Journal of chemical physics 144, 200901 (2016).DOI
- [19]Andrin Doll, Stephan Pribitzer, René Tschäggelar, and Gunnar Jeschke, “Adiabatic and fast passage ultrawideband inversion in pulsed epr,” Journal of magnetic resonance 230, 27–39 (2013).DOI
- [20]José Campany and Dalma Novak, “Microwave photonics combines two worlds,” Nature photonics 1, 319–330 (2007).
- [21]Arnaud Couairon and André Mysyrowicz, “Femtosecond filamentation in transparent media,” Physics reports 441, 47–189 (2007).DOI
- [22]See Leang Chin, Femtosecond laser filamentation, Vol. 55 (Springer, 2010).DOI
- [23]George Rodriguez and Georgi L Dakovski, “Scaling behavior of ultrafast two-color terahertz generation in plasma gas targets: energy and pressure dependence,” Optics express 18, 15130–15143 (2010).DOI
- [24]Charles K Birdsall and Bruce Langdon, Plasma physics via computer simulation (CRC press, 2004).
- [25]Robert E Peterkin and John W Luginsland, “A virtual prototyping environment for directed-energy concepts,” Computing in Science & Engineering 4, 42–49 (2002).DOI
- [26]Allen Taflove and Susan C Hagness, Computational electrodynamics: the finite-difference time-domain method (Artech house, 2005).
- [27]Kane Yee, “Numerical solution of initial boundary value problems involving maxwell’s equations in isotropic media,” IEEE Transactions on antennas and propagation 14, 302–307 (1966).DOI
- [28]Jay P Boris, “Relativistic plasma simulation-optimization of a hybrid code,” in Proc. Fourth Conf. Num. Sim. Plasmas (1970) pp. 3–67.
- [29]John Villasenor and Oscar Buneman, “Rigorous charge conservation for local electromagnetic field solvers,” Computer Physics Communications 69, 306–316 (1992).DOI
- [30]Jean-Pierre Berenger, “A perfectly matched layer for the absorption of electromagnetic waves,” Journal of computational physics 114, 185–200 (1994).DOI
- [31]See the Supplemental Material at http://link.aps.org/supplemental/10.1103/PhysRevE.104.L063201 for details on the numerical methods used in this Letter, along with code that implements the analytic approximations, and simulation movies.
- [32]Bing Zhou, Aurélien Houard, Yi Liu, Bernard Prade, André Mysyrowicz, Arnaud Couairon, Patrick Mora, Christopher Smeenk, Ladan Arissian, and Paul Corkum, “Measurement and control of plasma oscillations in femtosecond filaments,” Physical review letters 106, 255002 (2011).pubmed.ncbi.nlm.nih.gov/21770648
- [33]JM Pitarke, VM Silkin, EV Chulkov, and PM Echenique, “Theory of surface plasmons and surface-plasmon polaritons,” Reports on progress in physics 70, 1 (2006).DOI
- [34]Stefan Fasold, Amrit Jha, Steve R Andrews, L Martin-Moreno, and FJ Garcia-Vidal, “Terahertz surface plasmon-polariton propagation and focusing on periodically corrugated metal wires,” Physical Review Letters 97, 176805 (2006).
- [35]SV Popruzhenko, VD Mur, VS Popov, and D Bauer, “Strong field ionization rate for arbitrary laser frequencies,” Physical review letters 101, 193003 (2008).DOI
- [36]Animesh Sharma, Mikhail N Slipchenko, Mikhail N Shneider, Xingxing Wang, Kazi Arafat Rahman, and Alexey Shashurin, “Measurements of electron numbers in femtosecond laser induced plasmas using rayleigh microwave scattering,” in 2018 AIAA Aerospace Sciences Meeting (2018) p. 0177.DOI
- [37]Sergei Tochitsky, Eric Welch, Mikhail Polyanskiy, Igor Pogorelsky, Paris Panagiotopoulos, Miroslav Kolesik, Ewan M Wright, Stephan W Koch, Jerome V Moloney, Jeremy Pigeon, et al., “Megafilament in air formed by self-guided terawatt long-wavelength infrared laser,” Nature Photonics 13.
- [38]Miroslav Kolesik and Jerome V Moloney, “Nonlinear optical pulse propagation simulation: From maxwell’s to unidirectional equations,” Physical Review E 70, 036604 (2004).pubmed.ncbi.nlm.nih.gov/15524653
- [39]L Bergé, S Skupin, F Lederer, G Méjean, J Yu, J Kasparian, E Salmon, JP Wolf, M Rodriguez, L Wöste, et al., “Multiple filamentation of terawatt laser pulses in air,” Physical Review Letters 92, 225002 (2004).pubmed.ncbi.nlm.nih.gov/15245231
- [40]Wahb Ettoumi, Jérôme Kasparian, and J-P Wolf, “Laser filamentation as a new phase transition universality class,” Physical Review Letters 114, 063903 (2015).DOI
- [41]Wahb Ettoumi, Jérôme Kasparian, and J-P Wolf, “Spin-glass model governs laser multiple filamentation,” Physical review letters 115, 033902 (2015).arxiv.org/abs/1505.07041
- [42]Paul B Corkum, “Plasma perspective on strong field multiphoton ionization,” Physical review letters 71, 1994 (1993).pubmed.ncbi.nlm.nih.gov/10054556
- [43]Timothy D Pointon, “Second-order, exact charge conservation for electromagnetic particle-in-cell simulation in complex geometry,” Computer Physics Communications 179, 535–544 (2008).
- [44]Yukikazu Itikawa, “Cross sections for electron collisions with nitrogen molecules,” Journal of physical and chemical reference data 35, 31–53 (2006).DOI
- [45]Kip S Thorne and Roger D Blandford, Modern Classical Physics: Optics, Fluids, Plasmas, Elasticity, Relativity, and Statistical Physics (Princeton University Press, 2017).
- [46]John M Dawson, “Nonlinear electron oscillations in a cold plasma,” Physical review 113, 383 (1959).DOI
- [47]Osamu Sakai, Akinori Iwai, Yoshiharu Omura, Satoshi Ito, and Teruki Naito, “Wave propagation in and around negative-dielectric-constant discharge plasma,” Physics of Plasmas 25, 031901 (2018).DOI
- [48]M Okoniewski, M Mrozowski, and MA Stuchly, “Simple treatment of multi-term dispersion in fdtd,” IEEE Microwave and Guided Wave Letters 7, 121–123 (1997).
- [49]E Kretschmann, “The angular dependence and the polarisation of light emitted by surface plasmons on metals due to roughness,” Optics Communications 5, 331–336 (1972).DOI
- [50]Stephen Judson Smith and EM Purcell, “Visible light from localized surface charges moving across a grating,” Physical Review 92, 1069 (1953).DOI
- [51]Raphael Mahan, Saar Nemet, Michael Shentcis, Ori Reinhardt, Yuval Adiv, Xihang Shi, Orr Be’er, Morgan H Lynch, Yaniv Kurman, Kangpeng Wang, et al., “Resonant phase-matching between a light wave and a free-electron wavefunction,” Nature Physics 16, 1123–1131 (2020).
- [52]Arnold Sommerfeld, “Ueber die fortpflanzung elektrodynamischer wellen längs eines drahtes,” Annalen der Physik 303, 233–290 (1899).DOI
- [53]Georg Goubau, “Surface waves and their application to transmission lines,” Journal of Applied Physics 21, 1119–1128 (1950).DOI
- [54]Denis Jaisson, “Simple formula for the wave number of the goubau line,” Electromagnetics 34, 85–91 (2014).DOI
- [55]J Ricardo G Mendonça, “Electromagnetic surface wave propagation in a metallic wire and the lambert w function,” American Journal of Physics 87, 476–484 (2019).DOI
- [56]JB Andersen, “Radiation from surface-wave antennas,” Electronics Letters 3, 251–252 (1967).DOI
- [57]Michael Sporer, Robert Weigel, and Alexander Koelpin, “A 24 ghz dual-polarized and robust dielectric rod antenna,” IEEE Transactions on Antennas and Propaga- tion 65, 6952–6959 (2017).
- [58]Kasra Amini, Jens Biegert, Francesca Calegari, Alexis Chacón, Marcelo F Ciappina, Alexandre Dauphin, Dmitry K Efimov, Carla Figueira de Morisson Faria, Krzysztof Giergiel, Piotr Gniewek, et al., “Symphony on strong field approximation,” Reports on Progress in Physics 82, 116001 (2019).
- [59]Scott D Baalrud and Jérôme Daligault, “Effective potential theory for transport coefficients across coupling regimes,” Physical review letters 110, 235001 (2013).pubmed.ncbi.nlm.nih.gov/25167502
- [60]Zoltán Donkó, “Molecular dynamics simulations of strongly coupled plasmas,” Journal of Physics A: Mathematical and Theoretical 42, 214029 (2009).




