Part IV: High-Energy Astrophysics

High-energy astrophysics explores the most violent and energetic phenomena in the universe: relativistic jets, gamma-ray bursts, cosmic ray acceleration, and the radiation processes that power them. This field bridges particle physics and astrophysics, probing conditions where matter and radiation interact at extreme energies, from keV X-rays to ZeV (10²¹ eV) ultra-high-energy cosmic rays. Understanding these phenomena requires mastery of relativistic radiation theory, particle acceleration mechanisms, and the physics of magnetized plasmas under extreme conditions.

1. Radiation Processes

The three fundamental radiation mechanisms in high-energy astrophysics are bremsstrahlung (free-free emission), synchrotron radiation, and inverse Compton scattering. Each produces a characteristic spectrum and dominates under different physical conditions. Together, they explain the broadband emission from virtually every high-energy astrophysical source.

1.1 Bremsstrahlung (Free-Free Emission)

Bremsstrahlung ("braking radiation") is produced when a charged particle, typically an electron, is accelerated in the Coulomb field of another charged particle, usually an ion. This is the dominant emission mechanism in hot, ionized plasmas such as galaxy clusters, the intracluster medium (ICM), and the coronae of accretion disks. Because no bound states are involved, it is also called free-free emission.

For a thermal plasma at temperature $T$, the volume emissivity (energy radiated per unit volume per unit time per unit frequency) is:

$$\epsilon_{\nu}^{ff} = \frac{2^5 \pi e^6}{3 m_e c^3} \left(\frac{2\pi}{3 k_B m_e}\right)^{1/2} T^{-1/2} n_e n_i Z^2 \bar{g}_{ff} \, e^{-h\nu / k_B T}$$

where $\bar{g}_{ff}$ is the velocity-averaged Gaunt factor (a quantum-mechanical correction of order unity), $n_e$ and $n_i$ are the electron and ion number densities, and $Z$ is the ionic charge. The key features are:

The spectrum is flat for $h\nu \ll k_BT$ and cuts off exponentially at $h\nu \gtrsim k_BT$
The total (frequency-integrated) emissivity scales as $\epsilon_{ff} \propto n_e n_i Z^2 T^{1/2}$
The bremsstrahlung cooling time is $t_{ff} \sim 2.5 \times 10^{11} \left(\frac{T}{10^8\,\text{K}}\right)^{1/2} \left(\frac{n_e}{10^{-3}\,\text{cm}^{-3}}\right)^{-1}$ yr
For galaxy clusters with ICM temperatures of 2–10 keV, bremsstrahlung dominates the X-ray emission

Relativistic Bremsstrahlung

When the electron kinetic energy exceeds $m_ec^2 \approx 511$ keV, the non-relativistic Larmor formula must be replaced by the relativistic generalization. The cross section scales as $\sigma \propto Z^2 \alpha r_e^2 \ln(2\gamma)$ where $\alpha$ is the fine-structure constant and $r_e$ is the classical electron radius. Relativistic bremsstrahlung is important in AGN coronae and gamma-ray burst fireballs.

1.2 Synchrotron Radiation

Synchrotron radiation is emitted by relativistic charged particles spiraling in a magnetic field. It is the dominant emission mechanism in radio galaxies, pulsar wind nebulae, supernova remnants, and relativistic jets. Named after the General Electric synchrotron accelerator where it was first observed (1947), it is arguably the most important radiation process in high-energy astrophysics.

A relativistic electron with Lorentz factor $\gamma$ moving in a uniform magnetic field $B$ radiates with total power (averaged over pitch angle):

$$P_{\text{synch}} = \frac{4}{3} \sigma_T c \frac{B^2}{8\pi} \gamma^2 = \frac{2 e^4 B^2 \gamma^2}{3 m_e^2 c^3}$$

The radiation is beamed into a narrow cone of half-angle $\sim 1/\gamma$ in the direction of the electron's instantaneous velocity. An observer sees a series of short pulses, each of duration $\Delta t \sim 1/\gamma^3 \omega_B$ where $\omega_B = eB/(m_ec\gamma)$ is the relativistic gyrofrequency. The Fourier transform of these pulses gives the emission spectrum, which peaks near the critical frequency:

$$\nu_c = \frac{3 e B \gamma^2}{4\pi m_e c} \approx 4.2 \times 10^6 \left(\frac{B}{1\,\text{G}}\right) \gamma^2 \;\text{Hz}$$

The single-electron spectrum is described by the function $F(x) = x \int_x^\infty K_{5/3}(\xi)\,d\xi$ where $x = \nu/\nu_c$ and $K_{5/3}$ is a modified Bessel function. The asymptotic behaviors are $F(x) \propto x^{1/3}$ for $x \ll 1$ and $F(x) \propto x^{1/2} e^{-x}$ for $x \gg 1$.

Spectrum from a Power-Law Electron Distribution

In most astrophysical sources, electrons follow a power-law energy distribution: $\frac{dN}{d\gamma} = K \gamma^{-p}$ for $\gamma_{\min} \leq \gamma \leq \gamma_{\max}$. Integrating the single-electron emissivity over this distribution yields an optically thin synchrotron spectrum:

$$F_\nu \propto \nu^{-(p-1)/2} \equiv \nu^{-\alpha}$$

where $\alpha = (p-1)/2$ is the spectral index. For the canonical $p = 2.5$ from diffusive shock acceleration, $\alpha = 0.75$. This power-law spectrum is one of the most robust predictions in astrophysics and is observed across many decades of frequency in radio galaxies and SNR.

Synchrotron Cooling Time

An electron with Lorentz factor $\gamma$ cools on a timescale $t_{\text{synch}} = \frac{6\pi m_e c}{\sigma_T B^2 \gamma}$. Numerically:

$$t_{\text{synch}} \approx 1.3 \times 10^{10} \left(\frac{B}{10^{-5}\,\text{G}}\right)^{-2} \gamma^{-1}\;\text{s}$$

Cooling Break

Electrons above the cooling Lorentz factor $\gamma_c$ (where $t_{\text{synch}} = t_{\text{age}}$) lose their energy efficiently. The spectrum steepens by $\Delta \alpha = 0.5$ above the cooling frequency $\nu_c = \nu_c(\gamma_c)$, a feature observed in many GRB afterglows and pulsar wind nebulae.

1.3 Synchrotron Self-Absorption

At low frequencies, the synchrotron-emitting plasma becomes optically thick to its own radiation. This occurs because the absorption coefficient for synchrotron radiation is related to the emissivity by Kirchhoff's law. For a power-law electron distribution, the self-absorption coefficient scales as $\alpha_\nu \propto \nu^{-(p+4)/2}$, which increases steeply toward low frequencies.

Below the self-absorption frequency $\nu_a$, the spectrum becomes:

$$F_\nu \propto \nu^{5/2} \quad (\nu < \nu_a)$$

This $\nu^{5/2}$ spectrum is characteristic of synchrotron self-absorbed emission and is observed in compact radio sources, the cores of AGN jets, and the early phases of GRB afterglows. The self-absorption frequency depends on the magnetic field strength, electron density, and source size: $\nu_a \propto (K B)^{2/(p+4)} R^{2/(p+4)}$, where $R$ is the source radius. This relationship can be inverted to estimate $B$ from the observed turnover frequency and flux, a technique called equipartition analysis.

1.4 Inverse Compton Scattering

In inverse Compton (IC) scattering, a relativistic electron transfers energy to a low-energy photon, boosting it to high energy. This is the inverse of ordinary Compton scattering. For an electron with Lorentz factor $\gamma$ scattering a photon of frequency $\nu_0$, the scattered photon has characteristic frequency:

$$\nu_{\text{IC}} \approx \frac{4}{3} \gamma^2 \nu_0$$

This result holds in the Thomson regime, where $\gamma h\nu_0 \ll m_e c^2$. In this regime, the total IC power for a single electron in an isotropic radiation field with energy density $U_{\text{rad}}$ is:

$$P_{\text{IC}} = \frac{4}{3} \sigma_T c U_{\text{rad}} \gamma^2$$

Compton Y-Parameter

The Compton y-parameter quantifies the fractional energy change of photons passing through a hot electron gas: $y = \frac{4 k_B T_e}{m_e c^2} \max(\tau, \tau^2)$ where $\tau$ is the optical depth to Thomson scattering. When $y \gg 1$, repeated scatterings produce a Wien spectrum; for $y \ll 1$, the spectrum is barely modified. This is crucial for understanding the Sunyaev-Zel'dovich effect in galaxy clusters and the spectra of X-ray binaries.

Klein-Nishina Regime

When $\gamma h\nu_0 \gtrsim m_e c^2$, quantum effects become important and the cross section decreases from the Thomson value. The Klein-Nishina cross section is $\sigma_{\text{KN}} \approx \frac{3}{8} \sigma_T \frac{m_ec^2}{\gamma h\nu_0} \left[\ln\left(\frac{2\gamma h\nu_0}{m_ec^2}\right) + \frac{1}{2}\right]$. This suppression is critical for modeling the spectra of blazars, where the highest-energy electrons interact with UV/X-ray photons, and for understanding the maximum energy of IC-scattered photons from pulsar wind nebulae.

1.5 Synchrotron Self-Compton (SSC) and the Compton Dominance

In synchrotron self-Compton (SSC), the synchrotron photons produced by relativistic electrons serve as the seed photon field for inverse Compton scattering by the same electron population. This produces a second spectral component at higher energies. The ratio of IC to synchrotron luminosities is given by a beautifully simple result:

$$\frac{P_{\text{IC}}}{P_{\text{synch}}} = \frac{U_{\text{rad}}}{U_B}$$

where $U_B = B^2/(8\pi)$ is the magnetic energy density. When $U_{\text{rad}} > U_B$, the Compton component dominates and the electrons cool predominantly via IC scattering. This leads to the Compton catastrophe: if the synchrotron radiation field is too intense, each generation of IC scattering produces more photons, creating a runaway cooling process. This sets a maximum brightness temperature $T_b \lesssim 10^{12}$ K for incoherent synchrotron sources (the Kellermann-Pauliny-Toth limit).

The SSC model is the standard explanation for the broadband spectral energy distribution (SED) of BL Lac objects, which show two broad peaks: one in the radio-to-X-ray range (synchrotron) and one in the X-ray-to-gamma-ray range (SSC). External Compton (EC) models, where seed photons come from the accretion disk, broad-line region, or dusty torus, are needed for flat-spectrum radio quasars (FSRQs) that show higher Compton dominance.

2. Particle Acceleration Mechanisms

How are particles accelerated to the extreme energies observed in cosmic rays and astrophysical jets? This question, first posed by Fermi in 1949, has led to some of the most elegant results in theoretical astrophysics. The primary mechanisms are diffusive shock acceleration (Fermi I), stochastic acceleration (Fermi II), and magnetic reconnection.

2.1 Diffusive Shock Acceleration (Fermi I)

Diffusive shock acceleration (DSA), also known as first-order Fermi acceleration, is the leading mechanism for accelerating cosmic rays at astrophysical shocks such as supernova remnant blast waves. A charged particle scatters off magnetic irregularities (Alfvén waves) on both sides of a shock front, gaining energy at each crossing. The average fractional energy gain per shock crossing is $\langle \Delta E / E \rangle \propto v_{\text{shock}}/c$, which is first-order in the velocity — hence the name.

The remarkable result of DSA theory is that, independent of the details of the scattering, the accelerated particle spectrum is a universal power law:

$$\frac{dN}{dE} \propto E^{-p}, \quad p = \frac{r+2}{r-1}$$

where $r = \rho_2/\rho_1 = v_1/v_2$ is the shock compression ratio. For a strong, non-relativistic shock in a monatomic ideal gas ($\gamma_{\text{ad}} = 5/3$), the Rankine-Hugoniot conditions give $r = 4$, yielding $p = 2$. This is remarkably close to the observed cosmic ray spectral index of $p \approx 2.7$ (the difference is attributed to energy-dependent escape from the Galaxy). For the synchrotron spectral index, this gives $\alpha = (p-1)/2 = 0.5$, consistent with observations of young SNR.

Derivation Sketch

Let $P_{\text{esc}}$ be the probability of escape per cycle and $\langle \Delta E/E \rangle = \beta$ the fractional energy gain. After $k$ cycles, $E = E_0(1+\beta)^k$ and the fraction of particles remaining is $(1-P_{\text{esc}})^k$. Eliminating $k$:

$$N(>E) \propto \left(\frac{E}{E_0}\right)^{-\ln(1-P_{\text{esc}})/\ln(1+\beta)}$$

For a non-relativistic shock, $\beta = (4/3)(v_1-v_2)/c$ and $P_{\text{esc}} = 4v_2/(c)$, giving $p = 1 + \ln(1-P_{\text{esc}})/\ln(1+\beta) \approx 1 + P_{\text{esc}}/\beta = (r+2)/(r-1)$.

2.2 Stochastic Acceleration (Fermi II)

In Fermi's original (1949) mechanism, particles scatter off randomly moving magnetic clouds in the interstellar medium. Head-on collisions (energy gain) are slightly more probable than overtaking collisions (energy loss), giving a net energy gain that is second-order in the cloud velocity: $\langle \Delta E / E \rangle \propto (v_{\text{cloud}}/c)^2$.

This mechanism is slow compared to DSA but may operate in turbulent environments such as the lobes of radio galaxies, galaxy cluster turbulence, and solar flares. The acceleration rate can be described as a diffusion in momentum space with coefficient $D_{pp} \propto p^2 v_A^2 / \kappa$, where $v_A$ is the Alfvén speed and $\kappa$ is the spatial diffusion coefficient. The resulting spectrum depends on the balance between acceleration and escape, and is generally steeper than the DSA spectrum.

2.3 Magnetic Reconnection

Magnetic reconnection converts magnetic energy into particle kinetic energy by changing the topology of magnetic field lines. When oppositely directed field lines are brought together (e.g., in a current sheet), they can "break" and reconnect, releasing the stored magnetic energy. The reconnection rate in the Sweet-Parker model is slow ($v_{\text{rec}} \sim v_A / \sqrt{S}$ where $S$ is the Lundquist number), but Petschek reconnection and plasmoid-mediated reconnection can achieve rates $v_{\text{rec}} \sim 0.01\text{--}0.1\,v_A$.

Reconnection is thought to power solar flares, magnetar bursts, rapid variability in blazar jets, and the striped wind of pulsars. Recent particle-in-cell (PIC) simulations have shown that relativistic reconnection can produce hard power-law particle spectra with indices as flat as p ~ 1.5, harder than the DSA prediction.

2.4 Maximum Energy: The Hillas Criterion

The maximum energy a particle can achieve in an accelerator of size $R$ and magnetic field $B$ is set by the requirement that the particle's Larmor radius must fit inside the acceleration region. This gives the Hillas criterion:

$$E_{\max} = Z e B R \frac{v}{c} \approx 10^{18} Z \left(\frac{B}{1\,\mu\text{G}}\right) \left(\frac{R}{1\,\text{kpc}}\right) \left(\frac{v}{c}\right) \;\text{eV}$$

where $Z$ is the particle charge number and $v$ is the velocity of the scattering centers (or the shock velocity). The Hillas diagram plots astrophysical sources on the $B$–$R$ plane, showing which sources can accelerate particles to a given energy. Only a few source classes — AGN jets, galaxy cluster shocks, GRBs, and magnetars — can potentially reach the highest observed cosmic ray energies ($\sim 10^{20}$ eV).

In practice, the maximum energy is often set by energy losses (synchrotron, IC, or adiabatic) rather than by the Hillas limit. The acceleration timescale must be shorter than both the cooling time and the escape time: $t_{\text{acc}} < \min(t_{\text{cool}}, t_{\text{esc}})$. For DSA at a parallel shock, $t_{\text{acc}} \approx 20 \kappa/(v_s^2)$ where $\kappa$ is the diffusion coefficient.

3. Cosmic Rays

Cosmic rays are charged particles — predominantly protons (about 89%), helium nuclei (about 10%), heavier nuclei (about 1%), and a small fraction of electrons and positrons — that arrive at Earth from space with energies spanning over 12 orders of magnitude, from $\sim 10^9$ eV to beyond $10^{20}$ eV. Discovered by Victor Hess in 1912 via balloon flights, cosmic rays remain one of the most active research areas in astroparticle physics.

3.1 The Cosmic Ray Energy Spectrum

The all-particle cosmic ray spectrum follows a remarkable power law spanning many decades in energy, with several notable features:

The Knee

At $E \sim 3 \times 10^{15}$ eV (3 PeV), the spectrum steepens from $E^{-2.7}$to $E^{-3.1}$. This is likely the maximum energy for proton acceleration at galactic SNR shocks. The knee energy scales with charge $Z$, so heavier nuclei contribute increasingly above the proton knee.

The Ankle

At $E \sim 3 \times 10^{18}$ eV (3 EeV), the spectrum flattens back to $\sim E^{-2.7}$. This marks the transition from galactic to extragalactic cosmic rays. An alternative interpretation is a "dip" caused by electron-positron pair production on the CMB.

GZK Cutoff

Above $E \sim 5 \times 10^{19}$ eV (50 EeV), protons interact with CMB photons via photopion production: $p + \gamma_{\text{CMB}} \to \Delta^+ \to p + \pi^0$(or $n + \pi^+$). This limits the distance from which UHECRs can reach us to ~100 Mpc, the Greisen-Zatsepin-Kuzmin (GZK) horizon. The suppression has been confirmed by the Pierre Auger Observatory and Telescope Array.

3.2 Cosmic Ray Propagation

Galactic cosmic rays propagate diffusively through the interstellar medium, scattered by magnetic field irregularities. The transport is described by the diffusion-loss equation:

$$\frac{\partial N}{\partial t} = \nabla \cdot (D \nabla N) + \frac{\partial}{\partial E}\left[b(E) N\right] + Q(E, \mathbf{r}, t) - \frac{N}{\tau_{\text{esc}}}$$

where $D(E) \sim D_0 (E/E_0)^\delta$ is the energy-dependent diffusion coefficient (with $\delta \approx 0.3\text{--}0.6$ from observations of secondary-to-primary ratios like B/C), $b(E) = -dE/dt$ accounts for energy losses, $Q$ is the source term, and $\tau_{\text{esc}}$ is the escape timescale. The observed spectral index at Earth ($\sim 2.7$) is steeper than the source spectrum ($\sim 2.0\text{--}2.3$) because higher-energy particles diffuse out of the Galaxy faster.

3.3 Cosmic Ray Sources and Galactic Winds

The primary sources of galactic cosmic rays are supernova remnants (SNR). The energetics argument is compelling: the cosmic ray energy density in the Galaxy is $U_{\text{CR}} \sim 1$ eV/cm$^3$, the CR escape time is $\tau_{\text{esc}} \sim 10^7$ years, and the Galactic disk volume is $V \sim 10^{66}$ cm$^3$. The required power is $L_{\text{CR}} = U_{\text{CR}} V / \tau_{\text{esc}} \sim 10^{41}$ erg/s. With a supernova rate of ~3 per century and $E_{\text{SN}} \sim 10^{51}$ erg, the required efficiency is only ~10%, consistent with DSA theory. Direct evidence comes from gamma-ray observations of SNR (e.g., by Fermi-LAT) showing the characteristic pion-decay spectral signature.

For ultra-high-energy cosmic rays (UHECRs) above the ankle, the leading candidate sources are AGN jets and gamma-ray bursts. The Pierre Auger Observatory has reported a correlation between UHECR arrival directions and nearby starburst galaxies and AGN, though the angular deflections by extragalactic and galactic magnetic fields complicate the identification of individual sources.

Cosmic ray pressure is dynamically important in galaxies. The CR pressure $P_{\text{CR}} \approx U_{\text{CR}}/3 \sim 0.3$ eV/cm$^3$ is comparable to the thermal gas pressure, magnetic pressure, and turbulent pressure in the ISM. CRs can drive galactic winds, particularly in starburst galaxies, because their long mean free path allows them to exert a force over large scales. Simulations show that CR-driven winds can regulate star formation and enrich the circumgalactic medium with metals.

4. Jets and Relativistic Outflows

Collimated, bipolar outflows — jets — are among the most spectacular phenomena in astrophysics. They are observed from protostars (non-relativistic), X-ray binaries (mildly relativistic), and AGN (highly relativistic, with bulk Lorentz factors up to ~50). AGN jets can extend over megaparsec scales, far exceeding the size of their host galaxies, and carry enormous kinetic power that shapes the evolution of galaxy clusters.

4.1 Relativistic Beaming

Relativistic motion profoundly affects the observed properties of jets through Doppler boosting. For a source moving at velocity $\beta c$ at angle $\theta$ to the line of sight, the Doppler factor is:

$$\delta = \frac{1}{\Gamma(1 - \beta \cos\theta)}$$

where $\Gamma = (1-\beta^2)^{-1/2}$ is the bulk Lorentz factor. The observed flux density is boosted by a factor $\delta^{n+\alpha}$ where $n = 2$ for a continuous jet and $n = 3$ for a discrete blob, and $\alpha$ is the spectral index. For a jet with $\Gamma = 10$ viewed at the optimal angle $\theta = 1/\Gamma$, the Doppler factor is $\delta \approx 2\Gamma = 20$, and the flux can be boosted by factors of $\sim 10^4\text{--}10^6$ relative to the intrinsic emission. The counter-jet, moving away from the observer, is correspondingly de-boosted, explaining the jet/counter-jet asymmetry in radio galaxies.

4.2 Apparent Superluminal Motion

One of the most striking consequences of relativistic jet motion is apparent superluminal motion: components in VLBI maps of AGN jets appear to move across the sky faster than the speed of light. This is a geometric projection effect. The apparent transverse velocity is:

$$\beta_{\text{app}} = \frac{\beta \sin\theta}{1 - \beta \cos\theta}$$

This is maximized when $\cos\theta = \beta$ (i.e., $\theta = 1/\Gamma$), giving $\beta_{\text{app,max}} = \beta\Gamma \approx \Gamma$ for $\Gamma \gg 1$. Superluminal motions with $\beta_{\text{app}} \sim 5\text{--}50$ have been measured in many quasars and BL Lac objects using VLBI at radio wavelengths, providing direct evidence for bulk relativistic motion.

4.3 Jet Power and Composition

The total kinetic power of a relativistic jet is:

$$L_{\text{jet}} = \pi R_j^2 \Gamma^2 \beta c \left(\rho c^2 + \frac{\hat{\gamma}}{\hat{\gamma}-1} P + \frac{B^2}{4\pi}\right)$$

where $R_j$ is the jet radius, $\rho$ is the rest-mass density, $P$ is the pressure, and $\hat{\gamma}$ is the adiabatic index. The jet composition — whether dominated by electron-positron pairs ($e^\pm$) or electron-proton ($e^-p$) plasma — remains debated. Pair-dominated jets have lower inertia for a given power, enabling higher Lorentz factors. Observational constraints come from Faraday rotation measurements (sensitive to the presence of thermal protons) and circular polarization.

Jet powers estimated from X-ray cavities in galaxy clusters range from $\sim 10^{42}$ to $\sim 10^{46}$ erg/s, comparable to or exceeding the radiative luminosity of the AGN. The Blandford-Znajek mechanism extracts rotational energy from a spinning black hole threaded by magnetic field lines, providing jet power $L_{\text{BZ}} \sim \frac{\kappa}{4\pi c} \Phi_{\text{BH}}^2 \Omega_H^2$ where $\Phi_{\text{BH}}$ is the magnetic flux threading the horizon and $\Omega_H$ is the angular velocity of the horizon.

4.4 Blazars and Radio Galaxy Classification

Blazars are AGN with jets pointed nearly along our line of sight, producing extreme Doppler boosting. They are subdivided into:

BL Lac Objects

Weak or absent emission lines, SED well-described by SSC model. Further classified by the synchrotron peak frequency: Low-frequency peaked (LBL), Intermediate (IBL), and High-frequency peaked (HBL). HBLs are the primary extragalactic TeV gamma-ray sources detected by Cherenkov telescopes (H.E.S.S., MAGIC, VERITAS).

FSRQs (Flat-Spectrum Radio Quasars)

Strong broad emission lines, high Compton dominance (gamma-ray luminosity exceeds synchrotron luminosity). The gamma-ray emission is explained by external Compton scattering of photons from the broad-line region or dusty torus. FSRQs are generally more luminous but have lower synchrotron peak frequencies than BL Lacs.

Radio galaxies (jets viewed at larger angles) are classified by the Fanaroff-Riley scheme:

FR I: Edge-darkened, decelerating jets (e.g., M87). Lower radio luminosity ($L_{178\,\text{MHz}} \lesssim 2 \times 10^{25}$ W/Hz). Jets are disrupted by entrainment of ambient gas.
FR II: Edge-brightened, with powerful hotspots at the terminus of collimated jets (e.g., Cygnus A). Higher luminosity. Jets remain relativistic to large distances and terminate in strong shocks. FR IIs are the parent population of FSRQs.

5. Gamma-Ray Bursts

Gamma-ray bursts (GRBs) are the most luminous electromagnetic events in the universe, with isotropic equivalent luminosities reaching $L_{\text{iso}} \sim 10^{52}\text{--}10^{54}$ erg/s during the prompt phase. Discovered serendipitously by the Vela satellites in 1967 (published 1973), GRBs remained one of the deepest mysteries in astrophysics until the BeppoSAX satellite enabled the first X-ray and optical afterglow detections in 1997, placing them at cosmological distances.

5.1 Short vs. Long GRBs

GRBs divide into two classes based on their prompt emission duration (measured by $T_{90}$, the time during which 90% of the fluence is detected):

Long GRBs (T₉₀ > 2 s)

Produced by the core collapse of massive stars (collapsars). Associated with Type Ic-BL supernovae. Found in star-forming regions of galaxies. Typical redshift z ~ 1–3. The first confirmed GRB-SN connection was GRB 030329 / SN 2003dh.

Short GRBs (T₉₀ < 2 s)

Produced by neutron star mergers (NS-NS or NS-BH). Associated with kilonovae (r-process nucleosynthesis). Found in both young and old stellar populations. GW170817 / GRB 170817A was the landmark multi-messenger detection confirming the merger origin.

5.2 The Fireball Model

The standard model for GRBs is the fireball model, where an enormous amount of energy ($\sim 10^{51}\text{--}10^{52}$ erg) is released in a compact region ($\sim 10^7$ cm), producing an opaque fireball of photons, electrons, positrons, and a small baryon load. The fireball expands ultra-relativistically with Lorentz factors $\Gamma \sim 100\text{--}1000$.

The prompt gamma-ray emission is produced by internal shocks: the central engine ejects shells with variable Lorentz factors; faster shells catch slower ones, producing shocks where particles are accelerated and emit synchrotron radiation. The observed variability timescale (~ms) reflects the engine activity. An alternative model (photospheric emission) attributes the prompt emission to thermal radiation from the photosphere, modified by sub-photospheric dissipation.

5.3 Afterglow Theory

The afterglow is produced when the relativistic blast wave decelerates in the circumburst medium, driving an external shock. The shock-accelerated electrons radiate synchrotron emission across the electromagnetic spectrum (X-ray, optical, radio). The blast wave dynamics follow the Blandford-McKee self-similar solution, with Lorentz factor evolving as:

$$\Gamma(t) \propto \left(\frac{E_{\text{iso}}}{n_0 m_p c^5 t^3}\right)^{1/8} \propto t^{-3/8}$$

for a constant-density ISM environment. The afterglow light curves show characteristic power-law decays $F_\nu \propto t^{-\alpha} \nu^{-\beta}$ with closure relations linking the temporal and spectral indices that depend on whether the observation frequency is above or below the characteristic frequencies $\nu_m$ (peak of electron distribution) and $\nu_c$ (cooling frequency).

Jet Break

When the blast wave decelerates to $\Gamma \sim 1/\theta_j$ (where $\theta_j$ is the jet half-opening angle), the edge of the jet becomes visible and the jet begins to spread sideways. This produces an achromatic steepening of the light curve — the jet break — typically occurring at $t_{\text{jet}} \sim 1\text{--}10$ days. The jet break allows the true (beaming-corrected) energy to be estimated: $E_{\text{jet}} = E_{\text{iso}} (1 - \cos\theta_j) \approx E_{\text{iso}} \theta_j^2/2$. Remarkably, the beaming-corrected energies cluster around $\sim 10^{51}$ erg, suggesting a standard energy reservoir despite the wide range of isotropic equivalent energies.

5.4 Multi-Messenger Connections

The connection between GRBs and other transient phenomena has transformed our understanding of stellar death and nucleosynthesis:

Kilonova (Short GRBs)

Neutron star mergers eject $\sim 0.01\text{--}0.1\,M_\odot$ of neutron-rich material that undergoes rapid neutron capture (r-process) nucleosynthesis, producing heavy elements (Au, Pt, U, Th). The radioactive decay of these elements powers a thermal transient — the kilonova — peaking in the infrared at ~1 week. AT2017gfo, the kilonova associated with GW170817, confirmed that mergers are a major site of r-process nucleosynthesis.

GRB-Supernova Connection (Long GRBs)

Long GRBs are associated with broad-lined Type Ic supernovae (Ic-BL), confirming their origin in the core collapse of massive, stripped-envelope stars. The supernova component emerges in the optical light curve at ~2–3 weeks post-burst, as the afterglow fades. The SN ejecta velocities are ~30,000 km/s, indicating highly aspherical explosions driven by a central engine (black hole or magnetar).

6. Active Galactic Nuclei

Active galactic nuclei (AGN) are powered by accretion onto supermassive black holes ($M_{\text{BH}} \sim 10^6\text{--}10^{10}\,M_\odot$) at the centers of galaxies. They are the most luminous persistent sources in the universe, with bolometric luminosities up to $\sim 10^{48}$ erg/s. AGN are observed across the entire electromagnetic spectrum and exhibit a bewildering variety of phenomenology that is unified by a single physical model.

6.1 The AGN Unified Model

The unified model of AGN (Antonucci 1993; Urry & Padovani 1995) proposes that the diversity of AGN types arises primarily from orientation effects. The key components are:

Accretion disk: A geometrically thin, optically thick disk (Shakura-Sunyaev) emitting thermal UV/optical radiation (the "big blue bump"). The temperature profile is T(r) ~ r^-3/4, with peak temperatures ~10⁵ K for SMBH masses.
Hot corona: A cloud of hot (~10⁹ K) electrons above the disk that Comptonizes disk photons to produce the X-ray power-law continuum.
Broad-line region (BLR): Dense gas clouds (n ~ 10⁹–10¹¹ cm^-3) at ~0.01–0.1 pc, moving at ~3,000–10,000 km/s, producing broad emission lines (H-alpha, MgII, CIV). The BLR size scales with luminosity, enabling reverberation mapping to measure BH masses.
Dusty torus: A parsec-scale toroidal structure of molecular gas and dust that obscures the central engine and BLR when viewed edge-on. The inner radius is set by the dust sublimation temperature (~1500 K).
Narrow-line region (NLR): Lower-density gas (n ~ 10³–10⁵ cm^-3) at ~100 pc–1 kpc, producing narrow emission lines (~500 km/s FWHM). The NLR is visible from all viewing angles.
Relativistic jets: Present in ~10% of AGN (radio-loud). Launched by the Blandford-Znajek mechanism from the ergosphere of a spinning black hole.

6.2 AGN Classification

Under the unified model, the major AGN types correspond to different viewing angles and radio loudness:

Seyfert 1 / Type 1 QSO

Viewed face-on (direct view of BLR). Both broad and narrow emission lines visible. Strong UV/X-ray continuum. Radio-quiet. Seyferts are lower-luminosity AGN in spiral galaxies; QSOs (quasars) are the high-luminosity counterparts.

Seyfert 2 / Type 2 QSO

Viewed edge-on (BLR obscured by torus). Only narrow emission lines visible in direct light. Broad lines can be detected in polarized (scattered) light, as demonstrated by Antonucci & Miller (1985) for NGC 1068 — the key observation that established the unified model.

Radio Galaxies

Radio-loud AGN viewed at intermediate angles. Broad-line radio galaxies (BLRGs) are the radio-loud analogs of Seyfert 1; narrow-line radio galaxies (NLRGs) are analogs of Seyfert 2. FR I and FR II morphologies correspond to different jet powers and accretion rates.

Blazars

Radio-loud AGN with jets pointed at us. Dominated by non-thermal jet emission. BL Lacs (parent population: FR I) and FSRQs (parent population: FR II).

6.3 Accretion Regimes and Eddington Ratio

The Eddington luminosity sets the maximum luminosity for spherically symmetric accretion:

$$L_{\text{Edd}} = \frac{4\pi G M_{\text{BH}} m_p c}{\sigma_T} \approx 1.26 \times 10^{38} \left(\frac{M_{\text{BH}}}{M_\odot}\right) \;\text{erg/s}$$

The Eddington ratio $\lambda = L_{\text{bol}}/L_{\text{Edd}}$ determines the accretion regime:

Radiatively efficient ($\lambda \gtrsim 0.01$): Standard thin disk (Shakura-Sunyaev). Radiative efficiency $\eta \sim 0.06\text{--}0.42$ depending on BH spin. Quasars and luminous Seyferts operate in this regime.
Radiatively inefficient ($\lambda \lesssim 0.01$): Advection-dominated accretion flow (ADAF/RIAF). The accretion energy is advected into the black hole rather than radiated. Low-luminosity AGN (LLAGN), LINERs, and Sgr A* operate in this regime. The SED is dominated by synchrotron and IC emission rather than thermal disk emission.
Super-Eddington ($\lambda \gg 1$): The disk becomes geometrically thick ("slim disk"), with photon trapping and powerful radiation-driven outflows. Thought to occur in narrow-line Seyfert 1 galaxies, tidal disruption events, and possibly ultraluminous X-ray sources.

6.4 AGN Feedback

AGN feedback is one of the most important concepts in modern galaxy evolution theory. The energy released by the accreting SMBH can regulate or quench star formation in the host galaxy, explaining the observed correlations between BH mass and galaxy properties ($M_{\text{BH}}\text{--}\sigma$ relation) and the exponential cutoff at the bright end of the galaxy luminosity function.

Radiative (Quasar) Mode

At high accretion rates ($\lambda \gtrsim 0.01$), radiation pressure and/or AGN-driven winds (with velocities ~0.1c observed as ultrafast outflows in X-rays) can sweep gas out of the galaxy, suppressing star formation. This is "quasar mode" or "radiative mode" feedback, thought to be episodic and particularly important at high redshift during the peak of quasar activity.

Kinetic (Radio/Jet) Mode

At low accretion rates ($\lambda \lesssim 0.01$), jets inflate cavities (bubbles) in the hot gas halos of galaxies and clusters, providing a gentle, continuous heating that prevents the gas from cooling and forming stars. This "maintenance mode" feedback is directly observed as X-ray cavities in galaxy clusters (e.g., Perseus cluster) and solves the classic "cooling flow problem."

Simulation: Synchrotron Spectrum from Power-Law Electrons

Compute and plot the synchrotron emission spectrum from a power-law electron distribution in a magnetic field, including the synchrotron self-absorption turnover at low frequencies. The left panel shows the emitted spectrum (optically thin vs. self-absorbed), and the right panel shows the electron energy distribution with the critical frequency mapping. Adjust the magnetic field strength, electron spectral index, and Lorentz factor range to explore different astrophysical regimes.

Synchrotron Spectrum

Python

Parameters

B Field (G)

Electron Index p

Min Lorentz Factor

Max Lorentz Factor

script.py144 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
from scipy.special import gamma as gamma_func

# --- User Parameters ---
BFIELD = 1e-4
ELEC_INDEX = 2.5
GAMMA_MIN = 10.0
GAMMA_MAX = 1e6

# --- Physical Constants ---
e = 4.803e-10       # electron charge (esu)
m_e = 9.109e-28     # electron mass (g)
c = 2.998e10        # speed of light (cm/s)
sigma_T = 6.652e-25 # Thomson cross section (cm^2)

B = BFIELD  # Gauss
p = ELEC_INDEX
g_min = GAMMA_MIN
g_max = GAMMA_MAX

# Characteristic frequency for electron with Lorentz factor gamma
def nu_c(gamma):
    return 3.0 * e * B * gamma**2 / (4.0 * np.pi * m_e * c)

# Single-electron synchrotron power (total, averaged over pitch angle)
def P_synch(gamma):
    return (4.0/3.0) * sigma_T * c * (B**2 / (8*np.pi)) * gamma**2

# Approximate synchrotron function F(x) using Rybicki & Lightman fit
def F_sync(x):
    # Approximation: F(x) ~ 1.8 * x^(1/3) * exp(-x) for x > 0
    return 1.8 * x**(1.0/3.0) * np.exp(-x) * (1.0 + 0.884 * x**(1.0/3.0))

# Frequency grid (log-spaced)
nu_min_plot = 0.1 * nu_c(g_min)
nu_max_plot = 10.0 * nu_c(g_max)
nu = np.logspace(np.log10(nu_min_plot), np.log10(nu_max_plot), 500)

# Power-law electron distribution: dN/dgamma = K * gamma^(-p) for g_min < gamma < g_max
K = 1.0  # normalization (arbitrary units)

# Compute emissivity by integrating over electron distribution
# j_nu ~ integral from g_min to g_max of (dN/dgamma) * P_single(nu, gamma) dgamma
N_gamma_pts = 300
gamma_arr = np.logspace(np.log10(g_min), np.log10(g_max), N_gamma_pts)

j_nu = np.zeros_like(nu)
for i, freq in enumerate(nu):
    # For each frequency, integrate over electron Lorentz factors
    x_arr = freq / nu_c(gamma_arr)
    integrand = K * gamma_arr**(-p) * F_sync(x_arr) / gamma_arr**2
    # Use trapezoidal integration in log space
    j_nu[i] = np.trapezoid(integrand * gamma_arr, np.log(gamma_arr))

# Normalize
j_nu_max = np.max(j_nu[j_nu > 0])
j_nu_norm = j_nu / j_nu_max

# Synchrotron self-absorption: tau_ssa ~ nu^(-(p+4)/2) at low frequencies
# Below the self-absorption frequency, spectrum goes as nu^(5/2)
alpha_ssa = (p + 4.0) / 2.0
nu_ssa_est = nu_c(g_min) * 0.5  # rough SSA turnover

# Simple SSA model: I_nu ~ j_nu/alpha_nu * (1 - exp(-tau))
# tau ~ (nu/nu_ssa)^(-(p+4)/2)
tau_ssa = (nu / nu_ssa_est)**(-(p + 4.0) / 2.0)
tau_ssa = np.clip(tau_ssa, 0, 50)
I_nu = j_nu_norm * np.where(tau_ssa > 1e-3, (1.0 - np.exp(-tau_ssa)) / tau_ssa, 1.0)

# Optically thin power-law slope
alpha_thin = -(p - 1.0) / 2.0

# Plotting
fig, axes = plt.subplots(1, 2, figsize=(14, 6), facecolor='#0f172a')

# Left: Synchrotron spectrum with and without SSA
ax1 = axes[0]
ax1.set_facecolor('#1e293b')
ax1.loglog(nu, j_nu_norm, color='#60a5fa', linewidth=2, label='Optically thin', alpha=0.7)
ax1.loglog(nu, I_nu, color='#f87171', linewidth=2.5, label='With self-absorption')

# Mark characteristic frequencies
ax1.axvline(nu_c(g_min), color='#34d399', linestyle='--', alpha=0.6, label=f'nu_c(gamma_min={g_min:.0f})')
ax1.axvline(nu_c(g_max), color='#fbbf24', linestyle='--', alpha=0.6, label=f'nu_c(gamma_max={g_max:.0e})')
ax1.axvline(nu_ssa_est, color='#c084fc', linestyle=':', alpha=0.6, label=f'SSA turnover')

# Reference slopes
nu_ref = nu[(nu > 3*nu_ssa_est) & (nu < 0.1*nu_c(g_max))]
if len(nu_ref) > 10:
    slope_ref = (nu_ref / nu_ref[0])**alpha_thin * I_nu[nu > 3*nu_ssa_est][0] if np.any(nu > 3*nu_ssa_est) else None
    if slope_ref is not None and len(slope_ref) == len(nu_ref):
        ax1.loglog(nu_ref, slope_ref, 'w--', alpha=0.4, linewidth=1, label=f'slope = {alpha_thin:.2f}')

ax1.set_xlabel('Frequency (Hz)', color='#cbd5e1', fontsize=12)
ax1.set_ylabel('Intensity (normalized)', color='#cbd5e1', fontsize=12)
ax1.set_title(f'Synchrotron Spectrum (B={B:.1e} G, p={p:.1f})', color='#60a5fa', fontsize=13, fontweight='bold')
ax1.legend(fontsize=8, facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1')
ax1.tick_params(colors='#cbd5e1')
for spine in ax1.spines.values():
    spine.set_color('#334155')
ax1.set_ylim(1e-8, 10)

# Right: Electron distribution and characteristic frequencies
ax2 = axes[1]
ax2.set_facecolor('#1e293b')
dNdg = gamma_arr**(-p)
ax2.loglog(gamma_arr, dNdg / dNdg[0], color='#f87171', linewidth=2.5, label=f'dN/d\u03b3 ~ \u03b3^(-{p:.1f})')
ax2.set_xlabel('Lorentz factor \u03b3', color='#cbd5e1', fontsize=12)
ax2.set_ylabel('dN/d\u03b3 (normalized)', color='#cbd5e1', fontsize=12)
ax2.set_title('Electron Energy Distribution', color='#f87171', fontsize=13, fontweight='bold')

# Add second y-axis for critical frequency
ax2b = ax2.twinx()
ax2b.loglog(gamma_arr, nu_c(gamma_arr), color='#34d399', linewidth=1.5, linestyle='--', alpha=0.7)
ax2b.set_ylabel('Critical freq \u03bd_c (Hz)', color='#34d399', fontsize=11)
ax2b.tick_params(colors='#34d399')

ax2.legend(fontsize=10, facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1')
ax2.tick_params(colors='#cbd5e1')
for spine in ax2.spines.values():
    spine.set_color('#334155')
for spine in ax2b.spines.values():
    spine.set_color('#334155')

fig.suptitle('Synchrotron Radiation from Power-Law Electron Distribution', color='white', fontsize=14, fontweight='bold', y=1.02)
plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0f172a')
plt.close()

# Print key results
print(f"=== Synchrotron Spectrum Parameters ===")
print(f"Magnetic field B = {B:.2e} G")
print(f"Electron spectral index p = {p:.2f}")
print(f"Lorentz factor range: {g_min:.0f} - {g_max:.0e}")
print(f"Critical freq at gamma_min: {nu_c(g_min):.3e} Hz")
print(f"Critical freq at gamma_max: {nu_c(g_max):.3e} Hz")
print(f"Optically thin spectral index alpha = -(p-1)/2 = {alpha_thin:.3f}")
print(f"Self-absorption spectral index (low freq): +5/2 = 2.500")
print(f"Single electron power at gamma=1000: {P_synch(1000):.3e} erg/s")
print(f"Synchrotron cooling time at gamma=1000: {m_e*c*1000/P_synch(1000):.3e} s")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Simulation: GRB Afterglow Light Curves

Compute the broadband afterglow emission from a decelerating relativistic blast wave using the Blandford-McKee self-similar solution. The simulation tracks the blast wave dynamics, characteristic synchrotron frequencies, and multi-band (X-ray, optical, radio) light curves including the jet break. Adjust the isotropic equivalent energy, ambient density, and jet opening angle to explore different GRB environments.

GRB Afterglow

Python

Parameters

Isotropic Energy (erg)

Ambient Density (cm^-3)

Jet Opening Angle (rad)

script.py199 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# --- User Parameters ---
E_ISO = 1e52
N_AMB = 1.0
THETA_JET = 0.1

# --- Physical Constants ---
c = 3e10           # cm/s
m_p = 1.673e-24   # proton mass (g)
sigma_T = 6.652e-25  # Thomson cross-section (cm^2)
m_e = 9.109e-28   # electron mass (g)
e_charge = 4.803e-10  # electron charge (esu)

# Microphysical parameters (standard afterglow values)
epsilon_e = 0.1    # fraction of energy in electrons
epsilon_B = 0.01   # fraction of energy in magnetic field
p_elec = 2.3       # electron power-law index
z = 1.0            # redshift
d_L = 6.6e27 * (1+z)  # luminosity distance (cm), approximate

E_iso = E_ISO       # isotropic equivalent energy (erg)
n_0 = N_AMB         # ambient density (cm^-3)
theta_j = THETA_JET  # jet half-opening angle (rad)

# Deceleration radius and time
Gamma_0 = (3*E_iso / (4*np.pi*n_0*m_p*c**2))**(1.0/8.0) * 100  # rough initial Gamma
R_dec = (3*E_iso / (4*np.pi*n_0*m_p*c**2 * Gamma_0**2))**(1.0/3.0)
t_dec = R_dec / (2 * Gamma_0**2 * c)  # deceleration time in source frame

# Jet break time (when 1/Gamma ~ theta_j)
t_jet = 0.5 * (E_iso / (n_0 * m_p * c**5))**(1.0/3.0) * theta_j**(8.0/3.0) * (1+z)

# Time array (observer frame), from 10s to 1000 days
t_obs = np.logspace(1, np.log10(1000*86400), 500)  # seconds
t_days = t_obs / 86400.0

# Blandford-McKee self-similar solution: Gamma(t) ~ t^(-3/8) (ISM)
# R(t) ~ (17 E t / (4 pi n m_p c))^(1/4)
def blast_wave(t):
    R = (17.0 * E_iso * t / (4.0 * np.pi * n_0 * m_p * c))**0.25
    Gamma = np.sqrt(17.0 * E_iso / (1024.0 * np.pi * n_0 * m_p * c**5 * t**3))**0.5
    Gamma = np.clip(Gamma, 1.01, 1e4)
    return R, Gamma

R_arr, Gamma_arr = blast_wave(t_obs / (1+z))

# Downstream magnetic field
B_down = np.sqrt(32.0 * np.pi * epsilon_B * n_0 * m_p * c**2) * Gamma_arr

# Minimum Lorentz factor of electrons
gamma_m = epsilon_e * (p_elec - 2.0) / (p_elec - 1.0) * (m_p / m_e) * Gamma_arr

# Cooling Lorentz factor
t_source = t_obs / (1+z)
gamma_c = 6.0 * np.pi * m_e * c / (sigma_T * B_down**2 * Gamma_arr * t_source)

# Characteristic frequencies
nu_m = 0.5 * (e_charge * B_down) / (2*np.pi*m_e*c) * gamma_m**2 * Gamma_arr
nu_c = 0.5 * (e_charge * B_down) / (2*np.pi*m_e*c) * gamma_c**2 * Gamma_arr

# Peak flux density (mJy)
N_e = (4.0/3.0) * np.pi * R_arr**3 * n_0  # total swept-up electrons
P_nu_max = m_e * c**2 * sigma_T * B_down / (3.0 * e_charge)  # single electron peak power
F_nu_max = N_e * P_nu_max * (1+z) / (4.0 * np.pi * d_L**2)  # flux
F_nu_max_mJy = F_nu_max * 1e26  # convert to mJy

# Broadband spectrum: compute flux at given frequency
def flux_at_nu(nu_obs, nu_m, nu_c, F_max):
    F = np.zeros_like(nu_obs, dtype=float)
    # Slow cooling: nu_m < nu_c
    slow = nu_m < nu_c
    fast = ~slow

# For each time step
    result = np.zeros(len(nu_m))
    for i in range(len(nu_m)):
        nu = nu_obs
        if nu_m[i] < nu_c[i]:
            # Slow cooling
            if nu < nu_m[i]:
                result[i] = F_max[i] * (nu / nu_m[i])**(1.0/3.0)
            elif nu < nu_c[i]:
                result[i] = F_max[i] * (nu / nu_m[i])**(-(p_elec-1)/2.0)
            else:
                result[i] = F_max[i] * (nu_c[i]/nu_m[i])**(-(p_elec-1)/2.0) * (nu/nu_c[i])**(-p_elec/2.0)
        else:
            # Fast cooling
            if nu < nu_c[i]:
                result[i] = F_max[i] * (nu / nu_c[i])**(1.0/3.0)
            elif nu < nu_m[i]:
                result[i] = F_max[i] * (nu / nu_c[i])**(-0.5)
            else:
                result[i] = F_max[i] * (nu_m[i]/nu_c[i])**(-0.5) * (nu/nu_m[i])**(-p_elec/2.0)
    return result

# Observation bands
nu_xray = 2.4e17    # 1 keV X-ray
nu_optical = 4.3e14  # R-band optical
nu_radio = 8.5e9     # 8.5 GHz radio

F_xray = flux_at_nu(nu_xray, nu_m, nu_c, F_nu_max_mJy)
F_optical = flux_at_nu(nu_optical, nu_m, nu_c, F_nu_max_mJy)
F_radio = flux_at_nu(nu_radio, nu_m, nu_c, F_nu_max_mJy)

# Apply jet break: steepen light curves after t_jet
# Post-jet-break: extra t^(-3/4) steepening
for F_arr in [F_xray, F_optical, F_radio]:
    mask = t_obs > t_jet
    if np.any(mask):
        t_ratio = t_obs[mask] / t_jet
        F_arr[mask] *= t_ratio**(-0.75)  # jet break steepening

# Plotting
fig, axes = plt.subplots(2, 2, figsize=(14, 11), facecolor='#0f172a')

# Panel 1: Multi-band light curves
ax1 = axes[0, 0]
ax1.set_facecolor('#1e293b')
ax1.loglog(t_days, F_xray, color='#818cf8', linewidth=2.5, label='X-ray (1 keV)')
ax1.loglog(t_days, F_optical, color='#34d399', linewidth=2.5, label='Optical (R-band)')
ax1.loglog(t_days, F_radio, color='#f87171', linewidth=2.5, label='Radio (8.5 GHz)')
ax1.axvline(t_jet/86400, color='#fbbf24', linestyle='--', alpha=0.8, linewidth=1.5, label=f'Jet break ({t_jet/86400:.2f} d)')
ax1.set_xlabel('Time (days)', color='#cbd5e1', fontsize=11)
ax1.set_ylabel('Flux Density (mJy)', color='#cbd5e1', fontsize=11)
ax1.set_title('GRB Afterglow Light Curves', color='#818cf8', fontsize=13, fontweight='bold')
ax1.legend(fontsize=9, facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1')
ax1.tick_params(colors='#cbd5e1')
ax1.set_xlim(1e-2, 1000)
for spine in ax1.spines.values():
    spine.set_color('#334155')

# Panel 2: Blast wave dynamics
ax2 = axes[0, 1]
ax2.set_facecolor('#1e293b')
ax2.loglog(t_days, Gamma_arr, color='#60a5fa', linewidth=2.5)
ax2.set_xlabel('Time (days)', color='#cbd5e1', fontsize=11)
ax2.set_ylabel('Bulk Lorentz Factor \u0393', color='#cbd5e1', fontsize=11)
ax2.set_title('Blast Wave Deceleration', color='#60a5fa', fontsize=13, fontweight='bold')
ax2.tick_params(colors='#cbd5e1')
ax2.axvline(t_jet/86400, color='#fbbf24', linestyle='--', alpha=0.6)
# Reference slope t^{-3/8}
t_ref = t_days[(t_days > 0.01) & (t_days < t_jet/86400)]
if len(t_ref) > 2:
    G_ref = Gamma_arr[np.argmin(np.abs(t_days - t_ref[0]))] * (t_ref / t_ref[0])**(-3.0/8.0)
    ax2.loglog(t_ref, G_ref, 'w--', alpha=0.4, linewidth=1, label='t^(-3/8)')
    ax2.legend(fontsize=9, facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1')
for spine in ax2.spines.values():
    spine.set_color('#334155')

# Panel 3: Characteristic frequencies
ax3 = axes[1, 0]
ax3.set_facecolor('#1e293b')
ax3.loglog(t_days, nu_m, color='#f87171', linewidth=2, label='\u03bd_m (peak)')
ax3.loglog(t_days, nu_c, color='#60a5fa', linewidth=2, label='\u03bd_c (cooling)')
ax3.axhline(nu_xray, color='#818cf8', linestyle=':', alpha=0.5, label='X-ray band')
ax3.axhline(nu_optical, color='#34d399', linestyle=':', alpha=0.5, label='Optical band')
ax3.axhline(nu_radio, color='#fbbf24', linestyle=':', alpha=0.5, label='Radio band')
ax3.axvline(t_jet/86400, color='#fbbf24', linestyle='--', alpha=0.6)
ax3.set_xlabel('Time (days)', color='#cbd5e1', fontsize=11)
ax3.set_ylabel('Frequency (Hz)', color='#cbd5e1', fontsize=11)
ax3.set_title('Characteristic Frequencies', color='#f87171', fontsize=13, fontweight='bold')
ax3.legend(fontsize=8, facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1', loc='upper right')
ax3.tick_params(colors='#cbd5e1')
for spine in ax3.spines.values():
    spine.set_color('#334155')

# Panel 4: Downstream B-field and electron Lorentz factors
ax4 = axes[1, 1]
ax4.set_facecolor('#1e293b')
ax4.loglog(t_days, gamma_m, color='#f87171', linewidth=2, label='\u03b3_m (minimum)')
ax4.loglog(t_days, gamma_c, color='#60a5fa', linewidth=2, label='\u03b3_c (cooling)')
ax4.axvline(t_jet/86400, color='#fbbf24', linestyle='--', alpha=0.6)
ax4.set_xlabel('Time (days)', color='#cbd5e1', fontsize=11)
ax4.set_ylabel('Electron Lorentz Factor', color='#cbd5e1', fontsize=11)
ax4.set_title('Electron Energy Scales', color='#34d399', fontsize=13, fontweight='bold')
ax4.legend(fontsize=9, facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1')
ax4.tick_params(colors='#cbd5e1')
for spine in ax4.spines.values():
    spine.set_color('#334155')

fig.suptitle(f'GRB Afterglow: E_iso={E_iso:.1e} erg, n={n_0:.1f} cm\u207b\u00b3, \u03b8_j={np.degrees(theta_j):.1f}\u00b0',
             color='white', fontsize=14, fontweight='bold', y=1.02)
plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0f172a')
plt.close()

print(f"=== GRB Afterglow Parameters ===")
print(f"Isotropic energy: E_iso = {E_iso:.2e} erg")
print(f"Ambient density: n = {n_0:.2f} cm^-3")
print(f"Jet opening angle: theta_j = {np.degrees(theta_j):.2f} deg")
print(f"Jet break time: t_jet = {t_jet/86400:.3f} days")
print(f"Beaming-corrected energy: E_jet = {E_iso * (1 - np.cos(theta_j)):.2e} erg")
print(f"Initial Gamma estimate: {Gamma_0:.1f}")
print(f"Microphysics: eps_e={epsilon_e}, eps_B={epsilon_B}, p={p_elec}")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Key Equations Summary

Synchrotron Power

$$P_{\text{synch}} = \frac{2e^4 B^2 \gamma^2}{3m_e^2 c^3}$$

Critical Frequency

$$\nu_c = \frac{3eB\gamma^2}{4\pi m_e c}$$

IC/Synchrotron Ratio

$$\frac{P_{\text{IC}}}{P_{\text{synch}}} = \frac{U_{\text{rad}}}{U_B}$$

DSA Spectral Index

$$p = \frac{r+2}{r-1}$$

Hillas Criterion

$$E_{\max} = ZeBR\frac{v}{c}$$

Doppler Factor

$$\delta = [\Gamma(1-\beta\cos\theta)]^{-1}$$

Superluminal Motion

$$\beta_{\text{app}} = \frac{\beta\sin\theta}{1-\beta\cos\theta}$$

Eddington Luminosity

$$L_{\text{Edd}} = \frac{4\pi G M m_p c}{\sigma_T}$$

MIT: Cosmic Origin of the Chemical Elements

Selected lectures from MIT (RES.8-007) on heavy element formation through the r-process in neutron star mergers and other extreme astrophysical events.

Ep. 9: Formation of the Heaviest Elements

Ep. 8: Spectroscopy

Ep. 10: Telescopes and Observing

← Part III: Galactic Dynamics Part V: Gravitational Waves →

Part IV: High-Energy Astrophysics

1. Radiation Processes

1.1 Bremsstrahlung (Free-Free Emission)

Relativistic Bremsstrahlung

1.2 Synchrotron Radiation

Spectrum from a Power-Law Electron Distribution

Synchrotron Cooling Time

Cooling Break

1.3 Synchrotron Self-Absorption

1.4 Inverse Compton Scattering

Compton Y-Parameter

Klein-Nishina Regime

1.5 Synchrotron Self-Compton (SSC) and the Compton Dominance

2. Particle Acceleration Mechanisms

2.1 Diffusive Shock Acceleration (Fermi I)

Derivation Sketch

2.2 Stochastic Acceleration (Fermi II)

2.3 Magnetic Reconnection

2.4 Maximum Energy: The Hillas Criterion

3. Cosmic Rays

3.1 The Cosmic Ray Energy Spectrum

The Knee

The Ankle

GZK Cutoff

3.2 Cosmic Ray Propagation

3.3 Cosmic Ray Sources and Galactic Winds

4. Jets and Relativistic Outflows

4.1 Relativistic Beaming

4.2 Apparent Superluminal Motion

4.3 Jet Power and Composition

4.4 Blazars and Radio Galaxy Classification

BL Lac Objects

FSRQs (Flat-Spectrum Radio Quasars)

5. Gamma-Ray Bursts

5.1 Short vs. Long GRBs

Long GRBs (T90 > 2 s)

Short GRBs (T90 < 2 s)

5.2 The Fireball Model

5.3 Afterglow Theory

Jet Break

5.4 Multi-Messenger Connections

Kilonova (Short GRBs)

GRB-Supernova Connection (Long GRBs)

6. Active Galactic Nuclei

6.1 The AGN Unified Model

6.2 AGN Classification

Seyfert 1 / Type 1 QSO

Seyfert 2 / Type 2 QSO

Radio Galaxies

Blazars

6.3 Accretion Regimes and Eddington Ratio

6.4 AGN Feedback

Radiative (Quasar) Mode

Kinetic (Radio/Jet) Mode

Simulation: Synchrotron Spectrum from Power-Law Electrons

Synchrotron Spectrum

Simulation: GRB Afterglow Light Curves

GRB Afterglow

Key Equations Summary

Synchrotron Power

Critical Frequency

IC/Synchrotron Ratio

DSA Spectral Index

Hillas Criterion

Doppler Factor

Superluminal Motion

Eddington Luminosity

MIT: Cosmic Origin of the Chemical Elements

Long GRBs (T₉₀ > 2 s)

Short GRBs (T₉₀ < 2 s)