---
title: 'LPM 1.0: In-Medium QCD Splitting Rates'
url: https://www.emergentmind.com/topics/lpm-1-0
type: topic
---

# LPM 1.0: In-Medium QCD Splitting Rates

LPM 1.0 refers to the analytic framework for the computation of in-medium QCD splitting and joining rates—specifically in the deep Landau-Pomeranchuk-Migdal (LPM) regime—where the suppression of nearly collinear bremsstrahlung (and reverse processes) due to quantum interference between multiple scatterings is dominant. The formalism, as assembled by Arnold and Doğan, provides leading-logarithmic and next-to-leading-logarithmic (LL+NLL) closed-form expressions for parton branching rates in a thermal (or isotropic) plasma, fully incorporating all color, kinematic, and running-coupling dependencies, and systematically generalizes to arbitrary isotropic non-equilibrium backgrounds. The LPM 1.0 approach supplies all necessary parameter values and does not require additional numerical fitting, unifying both theoretical foundation and phenomenological recipes for QCD matter at high energies [0804.3359].

## 1. Deep-LPM Physical Regime

The LPM effect in QCD arises when a hard parton (energy $E$) traverses a dense medium (typically a quark-gluon plasma at temperature $T$), undergoing multiple soft scattering events as it attempts to radiate a daughter gluon or quark. In the *deep-LPM regime*, the following conditions are met:
- $E \gg T$
- $\ln(E/T) \gg 1$
- $\alpha_s\ln(E/T) \ll 1$ (so perturbation theory remains controlled)
- The formation time of the radiation $t_{\mathrm{form}} \sim E/Q_\perp^2$ exceeds the mean free time between scatterings.
Consequently, radiation is suppressed relative to independent Bethe-Heitler emissions; successive scatterings act coherently over the formation interval, leading to strong destructive interference [0804.3359].

The total transferred transverse momentum during formation is $Q_\perp^2 \sim (g^4 T^3 E\,\ln(E/T))^{1/2}$, highlighting the nontrivial scaling in $(E,T)$ of the suppression.

## 2. Leading-Log and Next-to-Leading-Log Splitting Rates

The central analytic results of the LPM 1.0 framework are the closed-form expressions for the rate at which a high-energy parton of species $a$ splits into daughters $b$ and $c$ with energy fractions $xE$ and $(1-x)E$:
- **Leading-Log Rate**
  $$
  \frac{d\Gamma^{\rm LL}_{a \to bc}}{dx} =
  \frac{(2\pi)^3}{E\,\nu_a}\;\gamma^{\rm LL}_{a \leftrightarrow bc}(E;\,xE,\,(1-x)E)
  $$
  The splitting kernel is
  $$
  \gamma^{\rm LL}_{a \leftrightarrow bc} =
  \frac{d_a\,C_a\,\alpha_s}{(2\pi)^4\sqrt{2}}\,\mu_\perp^2\,
  [\text{DGLAP kernel}]
  $$
  with, for $g \to gg$,
  $$
  \gamma^{\rm LL}_{g\to gg}(E; xE, (1-x)E) =
  \frac{16\,\alpha_s\,\mu_\perp^2}{(2\pi)^4\sqrt{2}}
  \frac{1 + x^4 + (1-x)^4}{x^2(1-x)^2}
  $$
  The transverse-momentum scale is
  $$
  \mu_\perp^2 = m_D^2\,\hat\mu^2,\quad
  m_D^2 = (C_A + N_f/2)\frac{g^2 T^2}{3}
  $$
  $$
  \hat\mu^2 \simeq \frac{gT}{m_D}
  \sqrt{\frac{2}{\pi}\,x(1-x)\frac{E}{T}
  \sqrt{[x^2 + (1-x)^2]\ln(E/T)}}
  $$

- **Next-to-Leading-Log Correction**
  Extending to NLL order introduces corrections modifying the argument of the logarithm:
  $$
  \Gamma_{\text{LL+NLL}} = \Gamma_{\text{LL}}\,[1 + \delta_\mathrm{NLL}(E/T)]
  $$
  $$
  \delta_{\rm NLL}(E/T) =
  \frac{1}{2}
  \frac{\sum_{i=1}^3 A_i\,\ln[2\xi H^2/(m_D^2 x_i^2)]}
      {\sum_{i=1}^3 A_i\,\ln(E/T)}
   - \frac{1}{2}
  $$
  with $A_i = C_{s_j} + C_{s_k} - C_{s_i}$, $\xi = e^{2-\gamma_E + \pi/4} \simeq 9.10$, and $H^2 \sim g^2 T E \sum_i A_i x_i^2 \ln(E/T)$. Alternatively, the NLL-corrected scale $\mu_\perp$ is set implicitly via:
  $$
  \mu_\perp^2 = m_D^2 \hat\mu^2 =
  \frac{gT}{m_D} \sqrt{\frac{2 x (1-x) E}{\pi T}
    \left[\sum_{i=1}^3 \frac{1}{2} A_i x_i^2
        \ln[\xi \mu_\perp^2 / (m_D^2 x_i^2)]\right]^{1/2}}
  $$
The scaling behavior is $\gamma^{\rm LL}\propto \alpha_s (ET)^{1/2} x^{-3/2}(1-x)^{-3/2} \sqrt{\ln(E/T)}$.

## 3. Running Coupling and Applicability

When $\ln(E/T)$ becomes so large that $\alpha_s\ln(E/T)\sim \mathcal{O}(1)$, 1-loop running of the coupling must be resummed. In the LPM integral equation, all occurrences of $g^2$ are replaced by $g^2(q_\perp)$, with the dominant $q_\perp$-scales running from $T$ up to $Q_\perp\sim(E T)^{1/4}$:
$$
g^2(\mu) = \frac{1}{-\bar\beta_0 \ln(\mu^2/\Lambda_{\mathrm{QCD}}^2)}
$$
The relevant Coulomb log is
$$
\int_{q_\perp \sim T}^{Q_\perp} dq_\perp\, g^2(q_\perp) \mathcal{N} q_\perp
\simeq \mathcal{N} \frac{g^2(T) - g^2(Q_\perp)}{-\bar\beta_0}
$$
with $\mathcal{N} \sim (\zeta(3)/\zeta(2))(1 + N_f/4) T^3$. The bremsstrahlung vertex is evaluated at $g^2(Q_\perp)$.

The formal regime of validity is
$$
T \ll E \ll \frac{T}{\alpha_s^2 \ln(1/\alpha_s)}
$$
Numerically, the NLL analytic formulas agree with the full numerical solution to within $\sim20\%$ for $E\gtrsim 10T$, improving to $\sim5\%$ at $E\sim 100T$ [0804.3359].

## 4. Generalization to Isotropic Non-Equilibrium Plasmas

For an arbitrary isotropic distribution $f(p)$ (not evolving appreciably over the formation time), the LPM 1.0 rates generalize by substituting the following:
$$
T^2\to 2g^2\sum_s \nu_s t_{R_s} \int \frac{d^3p}{(2\pi)^3}\frac{f_s(p)}{p}
$$
and
$$
T^* = \frac{\sum_s \nu_s t_{R_s} \int d^3p\, f_s(1\pm f_s)}
       {2 \sum_s \nu_s t_{R_s} \int d^3p\, f_s/p}
$$
in the screening correlator $\mathcal{A}(q_\perp)$, and using $T\to T^*$, $m_D^2 \to 2g^2\sum_s \nu_s t_{R_s} \int (d^3p/2\pi^3) f_s/p$ elsewhere. LL+NLL rate formulas and running-coupling prescriptions apply identically in this context [0804.3359].

## 5. Parameters, Implementation, and Library Design

All required parameters for LPM 1.0 are specified:
- $\xi,\;\bar\beta_0,\;m_D^2,\,\mathcal{N}$ (see above)
- Casimir factors ($C_A$, $C_F$)
- Color-spin degeneracies $\nu_a$

The sequence for a practical “LPM 1.0” code is:
- Implement LL formula for $\gamma_{a\to bc}$
- Include the NLL correction either additively ($\delta_{\text{NLL}}$) or via the implicit $\mu_\perp$ prescription
- Employ the running-$\alpha_s$ extension at very large $E/T$
- Generalize to any isotropic $f(p)$ via the substitutions above
No further numerical fits are needed—the formulas are closed, modular, and suitable for both theoretical studies and numerical simulation frameworks [0804.3359].

## 6. Impact and Legacy

The “LPM 1.0” formalism constitutes a systematic, unambiguous basis for the calculation of in-medium parton splitting rates dominated by the quantum-interference phenomenon first elucidated by Landau, Pomeranchuk, and Migdal, and ported to QCD by the BDMPS-Z group. The architecture enables rapid evaluation of rates crucial for heavy-ion phenomenology—particularly in the modeling of jet quenching and energy loss in the quark-gluon plasma—while providing an extensible platform for future higher-order corrections, inclusion of full running-coupling effects, and generalizations to non-equilibrium scenarios. Its compact, analytic formulas have been foundational in the implementation of splitting kernels in modern Monte Carlo and transport simulations of QCD matter at extreme conditions [0804.3359].

Source: https://www.emergentmind.com/topics/lpm-1-0