Spin-averaged $B_c$ Spectrum in a Cornell-type Potential Using VMC Baseline and GFMC Evolution

This paper computes the spin-averaged $B_c$ meson spectrum using a nonrelativistic Cornell potential calibrated to experimental data, employing a two-stage Variational and Fixed Node Green's function Monte Carlo approach to achieve mass predictions accurate to within tens of MeV.

The Gist

Imagine the universe is filled with tiny, invisible Lego bricks called quarks. Usually, these bricks come in pairs of the same color (like two red bricks or two blue bricks) to form particles called mesons. But sometimes, nature gets creative and snaps together two different colored bricks: a heavy "bottom" brick and a heavy "charm" brick. When they stick together, they form a special particle called the $B_c$ meson.

This paper is like a master builder trying to figure out exactly how strong the glue is between these two specific bricks, and how they vibrate and dance around each other.

Here is the story of the paper, broken down into simple concepts:

1. The Problem: The Invisible Spring

Physicists know that quarks are held together by a force that acts like a spring.

• The Short-Range Glue: When the bricks are close, they attract each other strongly (like a magnet).
• The Long-Range Rubber Band: If you try to pull them apart, the force gets stronger and stronger, like stretching a rubber band, until it snaps (or creates new bricks).

This "spring" is described by a famous mathematical recipe called the Cornell Potential. It has a few knobs you can turn to make the spring tighter or looser. The goal of this paper is to find the perfect setting for those knobs so that the math matches what we see in real experiments.

2. The Challenge: It's Too Hard to Solve with a Pen

You can't just write down an equation and solve it on a piece of paper to find the exact energy of these dancing bricks. The math is too messy. So, the authors used a computer simulation that acts like a virtual laboratory.

They used a two-step process, which they call VMC and GFMC. Let's use an analogy to understand them:

• Step 1: The Sketch (VMC - Variational Monte Carlo)
  Imagine you are an artist trying to draw a portrait of a moving dancer. You don't know the exact pose, so you make a quick, rough sketch based on your best guess. In the computer, this is the "trial state." It's a good guess, but it's not perfect. It's like a blurry photo.

• Step 2: The High-Definition Video (GFMC - Green's Function Monte Carlo)
  Now, imagine you take that rough sketch and run it through a super-smart video editor that simulates time passing. The editor slowly refines the image, removing the blur and the mistakes, until the dancer is perfectly clear. This step "projects" the rough guess into the true, exact state of the particle.

3. The Calibration: Tuning the Radio

The authors had to tune the "Cornell knobs" (the strength of the glue and the rubber band).

• They started with a specific, well-known note: the 1S state (the ground state, or the lowest energy level). This is like the "A" note on a piano.
• They adjusted the knobs until their computer simulation hit that exact "A" note perfectly.
• Once the "A" note was perfect, they didn't touch the knobs again. They just let the simulation run to see what notes the higher vibrations (excited states) would produce.

4. The Results: A Perfect Chord

When they looked at the results, they found something amazing:

• The computer predicted the "notes" (masses) of the higher vibrations (like the 2S, 3S, 1P, etc.) with incredible accuracy.
• The predictions were off by only a tiny amount (a few tens of MeV), which is like being off by a fraction of a millimeter on a football field.
• This proved that the simple "Cornell spring" model is actually a very good description of how these heavy quarks behave, even without needing complex, messy corrections.

5. Why This Matters

Think of this paper as building a calibrated ruler.
Before this, different scientists used different rulers to measure the $B_c$ meson, and some were slightly too long or too short. This paper says, "Here is a ruler that we built using a very precise, step-by-step computer method. It matches the ground truth perfectly."

Now, other scientists can use this "ruler" as a baseline. If they want to study more complex things later (like how the spin of the quarks affects the particle, or how relativity changes things), they can compare their new results against this solid foundation.

Summary

In short, the authors built a virtual microscope to look at a heavy particle made of two different quarks. They tuned their mathematical model until it perfectly matched the known weight of the particle. Then, they used that model to predict the weights of the particle's excited states, and the predictions turned out to be spot-on. It's a victory for using simple physics combined with powerful computer simulations to understand the building blocks of our universe.

Technical Summary

1. Problem Statement

The paper addresses the calculation of the spin-averaged mass spectrum of the $B_c$ meson (a bound state of a charm quark and a bottom antiquark) within a nonrelativistic, spin-independent framework.

• Context: While the $B_c$ meson is a unique system for testing nonrelativistic QCD (NRQCD) due to its mixed-flavor nature, previous studies have relied on various methods (relativistic quark models, lattice QCD, semi-analytic iteration methods, and neural networks) that often yield disparate results or lack rigorous control over numerical systematics.
• Goal: The authors aim to establish a computationally controlled baseline for the $B_c$ spectrum using a minimal Cornell potential. The objective is not to propose a new potential form but to demonstrate that a standard Cornell Hamiltonian, when solved with high-precision Monte Carlo methods, can reproduce experimental data within a few tens of MeV without systematic drifts across radial and orbital excitations.

2. Methodology

The study employs a two-stage Monte Carlo approach to solve the radial Schrödinger equation for the $B_c$ system.

A. Theoretical Framework

• Hamiltonian: The system is treated as a nonrelativistic two-body problem with reduced mass $\mu = m_b m_c / (m_b + m_c)$. The interaction is modeled by the Cornell potential:
  $$V(r) = \sigma r - \frac{\kappa}{r} + V_0$$
  where $\sigma$ is the string tension (confinement), $\kappa$ is the Coulomb strength (one-gluon exchange), and $V_0$ is an additive constant fixing the energy origin.
• Parameters: Fixed quark masses are used ($m_c = 1.2730$ GeV, $m_b = 4.183$ GeV). The parameters $(\sigma, \kappa, V_0)$ are the variables to be determined.

B. Numerical Algorithms

1. Variational Monte Carlo (VMC):

   • Role: Generates optimized radial trial wavefunctions ($\Psi_T$) with the correct nodal structure for each $(n, \ell)$ channel (1S, 1P, 1D, 2S, etc.).
   • Process: Minimizes the expectation value of the Hamiltonian. Orthogonality between excited states and lower levels is enforced via Gram-Schmidt projection.
   • Output: Provides "fixed-node" templates and reference energies free of time-step artifacts.

2. Green's Function Monte Carlo (GFMC):

   • Role: Projects the VMC trial states onto the true ground state of each nodal sector using imaginary time evolution ($\tau = it$).
   • Process: Uses importance sampling with the VMC trial function as a guide. Walkers undergo drift-diffusion moves and branching (weight updates) based on the local energy.
   • Systematic Control: The authors rigorously control numerical errors by scanning:
     • Time step ($\Delta\tau$): Extrapolating to the $\Delta\tau \to 0$ limit to remove discretization bias.
     • Projection time ($\tau_{tot}$): Ensuring sufficient steps ($N_{steps}$) to reach the asymptotic plateau where higher excited state contributions are exponentially damped.
     • Radial Grid: Verifying convergence with respect to grid size ($N_r$) and cutoff ($r_{max}$).

C. Calibration Strategy

• Anchoring: The additive constant $V_0$ is not a free fit parameter. For every pair $(\sigma, \kappa)$, $V_0$ is uniquely determined by anchoring the theoretical 1S mass to the experimental 1S centroid.
• Optimization: A grid scan over $(\sigma, \kappa)$ is performed. The quality of the fit is evaluated using a Root Mean Square Error (RMSE) calculated from the absolute masses of the entire tower (1S through 4S).
• Selection Criteria: The "best" parameters are selected not just by minimizing global RMSE, but by ensuring no systematic drift in residuals across the excitation ladder (i.e., the model shouldn't consistently overestimate higher states).

3. Key Contributions

1. Numerical Rigor: The paper establishes a benchmark for $B_c$ spectroscopy by explicitly quantifying and controlling numerical systematics (time-step, projection time, population size) in a VMC+GFMC framework, a level of detail often missing in potential model studies.
2. Validation of Minimal Cornell Model: It demonstrates that a simple, spin-independent Cornell potential is sufficient to describe the spin-averaged $B_c$ spectrum with high precision, provided the parameters are calibrated directly to the $B_c$ data rather than borrowed from charmonium or bottomonium.
3. Parameter Space Mapping: The authors map the "low-RMSE valley" in the $(\sigma, \kappa)$ parameter space, showing that the optimal $B_c$ parameters lie in a region consistent with, but distinct from, canonical charmonium and bottomonium fits.

4. Results

• Best-Fit Parameters: The optimal point in the low-RMSE valley is identified as:
  • $\sigma = 0.1625 \text{ GeV}^2$
  • $\kappa = 0.6125$
  • $V_0 = 0.901875 \text{ GeV}$
• Spectral Accuracy:
  • The predicted spin-averaged masses agree with experimental centroids to within ~22 MeV (average offset ~ -2 MeV).
  • The agreement is consistent across the entire tower (1S to 4S), with no systematic drift observed for higher excitations (3S, 3P, 4S), which were not used in the calibration.
• Comparison with Literature:
  • The results fall within the interquartile range of various other approaches (neural network solvers, asymptotic iteration methods, relativistic quark models).
  • The GFMC results are generally closer to the median of the literature values than many other specific models.
• Parameter Consistency: The derived parameters ($\sigma \approx 0.16$, $\kappa \approx 0.61$) are physically sensible: a slightly softer string tension and stronger Coulomb term compared to pure $c\bar{c}$ or $b\bar{b}$ systems, reflecting the mixed-flavor reduced mass of the $B_c$.

5. Significance

• Baseline for Future Work: This work provides a numerically controlled reference point for future studies. Because the spin-independent baseline is now established with high precision, subsequent work can reliably add spin-dependent corrections (hyperfine splitting, spin-orbit coupling) and relativistic corrections to isolate the effects of these terms without ambiguity from the static potential.
• Methodological Benchmark: It validates the VMC+GFMC approach as a robust tool for heavy quark spectroscopy, offering an alternative to semi-analytic methods and demonstrating its capability to handle complex nodal structures in excited states.
• QCD Insight: The results reinforce the view that the Cornell potential, despite its simplicity, captures the essential nonperturbative dynamics of heavy quark systems, provided the parameters are tuned to the specific flavor system.

In summary, the paper successfully demonstrates that a minimal Cornell Hamiltonian, solved with rigorous Monte Carlo techniques, can reproduce the $B_c$ spin-averaged spectrum with high fidelity, establishing a solid foundation for more complex NRQCD analyses.