Impact of dynamic electrostatic disorder on hole mobility in rubrene: a nonadiabatic molecular dynamics investigation

By incorporating dynamic electrostatic disorder into nonadiabatic molecular dynamics simulations using the damped shifted-force method, this study demonstrates that electrostatic interactions significantly increase reorganization energy and site energy disorder in rubrene, thereby reducing the predicted hole mobility from 35 to 21 cm² V⁻¹ s⁻¹ and achieving close agreement with experimental values.

The Gist

The Big Picture: Moving Through a Crowded Room

Imagine a high-tech organic crystal (like rubrene) as a massive, crowded dance floor made of thousands of identical molecules. In this dance floor, a "hole" (which acts like a positive electric charge) is trying to move from one side of the room to the other. This movement is what we call mobility, and it determines how fast electronic devices made from these materials can work.

For a long time, scientists had a hard time predicting exactly how fast this "hole" moves. They knew the dance floor wasn't perfectly still; the molecules were jiggling and vibrating due to heat. But there was a specific type of "noise" they were ignoring: electrostatic disorder.

Think of electrostatic disorder like the invisible static electricity or the subtle "personal space" feelings between people in a crowd. Even though the molecules in rubrene are mostly neutral (apolar), they still have tiny internal electrical charges that push and pull on each other.

The Problem: Ignoring the "Static"

In previous computer simulations, scientists often turned off these tiny electrical interactions to save computing power. It seemed like a safe shortcut because the molecules aren't strongly charged.

However, the authors of this paper asked: "What if we actually turn the static back on?"

They wanted to see if ignoring these tiny electrical pushes and pulls was making their predictions about how fast the charge moves inaccurate.

The Method: A New Way to Count

Simulating these electrical interactions is usually very slow and expensive, like trying to calculate the weight of every grain of sand on a beach individually.

The researchers used a clever new math trick called the DSF method (Damped Shifted-Force).

• The Analogy: Imagine you are trying to count how many people are in a stadium. The old way (Ewald summation) is like asking every single person to shout their name so you can count them, which takes forever. The new DSF method is like using a smart camera that estimates the crowd density quickly and accurately without needing everyone to shout. This allowed them to run a massive simulation including all those tiny electrical interactions for the first time.

What They Found

1. The "Reorganization Energy" (The Cost of Moving)
When a charge moves from one molecule to another, the molecules around it have to wiggle and adjust to the new presence. This adjustment costs energy, called reorganization energy.

• The Result: When they included the electrical interactions, the cost to move went up by about 20–25%.
• The Analogy: It's like walking through a room where the floor is slightly sticky. If you ignore the stickiness, you think you can run fast. But once you account for the sticky floor, you realize it takes more effort to take each step.

2. The "Site Energy Disorder" (The Uneven Floor)
The electrical interactions made the energy levels of the molecules fluctuate more wildly.

• The Result: The "floor" became bumpier. Some spots became harder to stand on, and others easier, purely because of the shifting electrical forces.
• The Analogy: Imagine walking on a trampoline. If the springs are perfectly even, you bounce smoothly. But if the springs are slightly uneven (due to the electrical noise), your path gets bumpy and unpredictable.

3. The "Hole" Gets Less Confident (Localization)
Because the floor became bumpier, the charge carrier (the hole) couldn't spread out as easily.

• The Result: In the old simulations (without electricity), the hole was spread out over about 13 molecules at once. In the new simulations (with electricity), it got "stuck" or localized over only about 9 molecules.
• The Analogy: Without the electrical noise, the hole felt like a confident surfer gliding over a smooth wave, covering a lot of ground. With the noise, the hole became more like a surfer trying to balance on choppy, unpredictable waves, staying in one spot longer.

4. The Final Speed (Mobility)
This is the most important part. Because the hole got "stuck" more often and the floor was bumpier, it moved slower.

• The Result:
  • Old Prediction (Ignoring electricity): The hole moved at 35 units of speed.
  • New Prediction (Including electricity): The hole moved at 21 units of speed.
  • Real World Experiment: The actual speed measured in labs is about 20 units.

The Conclusion

The paper concludes that you cannot accurately predict how fast electricity moves in these materials unless you include the tiny, dynamic electrical interactions.

By using their new, efficient math method to include these interactions, their computer simulation finally matched the real-world experiments almost perfectly. They proved that even in materials that seem "neutral," these invisible electrical tugs and pulls are essential for understanding how the material works.

In short: They fixed a broken simulation by turning the "static electricity" back on, and suddenly, their computer predictions matched reality.

Technical Summary

Technical Summary: Impact of Dynamic Electrostatic Disorder on Hole Mobility in Rubrene

Problem Statement
High-mobility organic molecular crystals, such as rubrene, are critical for organic electronics, yet achieving a quantitatively predictive description of their charge transport remains challenging. Charge transport in these materials occurs in an intermediate regime between small-polaron hopping and band transport, characterized by transient delocalization where the charge carrier wavefunction is modulated by thermal fluctuations. While mixed quantum-classical nonadiabatic molecular dynamics (specifically Fragment Orbital-based Surface Hopping, or FOB-SH) offers a promising route to simulate this regime without assuming a specific transport mechanism, previous large-scale simulations of apolar crystals like rubrene have typically neglected dynamic electrostatic disorder. This omission was driven by the computational cost of evaluating long-range Coulombic interactions (typically requiring Ewald summation) and the assumption that such interactions are negligible in apolar systems. However, neglecting these interactions removes electrostatic contributions to reorganization energy and suppresses site energy fluctuations, potentially leading to inaccurate predictions of carrier delocalization and mobility.

Methodology
This study integrates dynamic electrostatic disorder into FOB-SH simulations of room-temperature hole transport in rubrene. The authors employ the Damped Shifted-Force (DSF) real-space electrostatic summation method, combined with an efficient addition-subtraction scheme, to overcome the computational bottlenecks associated with evaluating electrostatic contributions for multiple charge-localized states.

Key methodological components include:

• Electrostatic Treatment: The DSF method approximates long-range Coulomb interactions using a damped real-space interaction shifted to zero at a finite cutoff ($R_{cut} = 16$ Å). An addition-subtraction scheme decomposes the electrostatic energy of a charge-localized state into the energy of the neutral system plus correction terms, reducing computational scaling from $O(N_{mol})$ to near $O(1)$ relative to a single calculation.
• Polarization Effects: To account for electronic polarization without the cost of explicit polarizable force fields, an implicit scheme based on charge scaling (factor $\gamma = 0.80$) was applied to atomic partial charges.
• Simulation Protocol: FOB-SH simulations were performed at 300 K using a supercell of $50 \times 13 \times 1$ unit cells (1300 molecules). The simulations propagated the hole wavefunction quantum mechanically while evolving nuclei classically. Results were compared against a baseline "zero-charge" model (where partial charges were set to zero) and validated against Smooth Particle-Mesh Ewald (SPME) calculations for reorganization energy.
• Analysis: The study quantified reorganization energy ($\lambda$), site energy disorder ($\sigma_E$), electronic coupling fluctuations, the Inverse Participation Ratio (IPR) to measure wavefunction delocalization, and charge carrier mobility ($\mu$) derived from the mean-square displacement (MSD).

Key Results

1. Reorganization Energy: Including electrostatic interactions (via DSF or SPME) increases the reorganization energy for hole transfer by approximately 20–40 meV (from a baseline of 152 meV to ~175–191 meV) compared to the zero-charge model. Spectral analysis reveals that this increase arises from enhanced coupling across both low-frequency lattice motions (<200 cm$^{-1}$) and higher-frequency intramolecular modes. The DSF method captures the majority of this electrostatic contribution, showing good agreement with SPME benchmarks.
2. Site Energy Disorder: The inclusion of electrostatics significantly increases the root-mean-square fluctuations of site energies ($\sigma_E$), rising from 63.5 meV (zero-charge) to 69.2 meV (DSF). This increase is quantitatively consistent with the increased reorganization energy, confirming that electrostatics primarily modify diagonal energetic disorder while leaving off-diagonal electronic couplings largely unchanged.
3. Wavefunction Delocalization: The enhanced site energy disorder reduces the spatial extent of the hole wavefunction. The average Inverse Participation Ratio (IPR) decreases from 13 (zero-charge) to 9 (DSF), indicating a more localized charge carrier.
4. Charge Mobility: The reduction in delocalization leads to a significant decrease in predicted hole mobility. Along the high-mobility $a$-direction, the mobility drops from 34.8 cm$^2$V$^{-1}$s$^{-1}$ (zero-charge) to 21.1 cm$^2$V$^{-1}$s$^{-1}$ (DSF). This value is in close agreement with the experimental reference of 20.2 cm$^2$V$^{-1}$s$^{-1}$. Furthermore, the inclusion of electrostatics improves the predicted anisotropy ratio ($\mu_a/\mu_b$), bringing it closer to experimental observations.

Significance and Claims
The paper demonstrates that dynamic electrostatic disorder is an essential component for quantitatively predictive simulations of charge transport in molecular organic semiconductors, even those composed of apolar molecules like rubrene. By explicitly including these interactions via the computationally efficient DSF scheme, the authors show that previous models neglecting electrostatics overestimated carrier delocalization and mobility. The study establishes that electrostatic site energy fluctuations are the primary mechanism responsible for the dynamic localization observed in high-mobility organic crystals, and their inclusion is necessary to bridge the gap between theoretical simulations and experimental reality. The work validates the DSF method as a viable, scalable alternative to Ewald summation for large-scale nonadiabatic dynamics in periodic systems.