🔬 mesoscale physics

Enhanced Graphene-Water Thermal Transport via Edge Functionalization without Compromising In-Plane Thermal Conductivity

This study demonstrates that selectively functionalizing the edges of graphene nanoribbons with hydroxyl groups significantly enhances graphene-water interfacial thermal conductance by over eightfold while largely preserving intrinsic in-plane thermal conductivity, offering a superior alternative to surface functionalization.

Original authors: John Crosby, Haoran Cui, Mehrab Lotfpour, Yan Wang, Lei Cao

Published 2026-02-19

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: John Crosby, Haoran Cui, Mehrab Lotfpour, Yan Wang, Lei Cao

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: The "Super-Runner" and the "Swimming Pool"

Imagine graphene as a super-fast runner. In its pure form, this runner is incredibly efficient at sprinting (conducting heat) across a track. However, this runner is also very shy and hydrophobic (water-repelling). If you put this runner in a swimming pool (water), they don't want to interact with the water. They stay dry, and because they aren't touching the water well, they can't dump their heat into the pool efficiently.

In many technologies (like solar steam generators or cooling computer chips), we need this runner to get hot and then instantly dump that heat into the water. But right now, the "handshake" between the runner and the water is weak.

The Problem: The "Bad Fix"

Scientists have tried to fix this by "gluing" special stickers (chemical groups like hydroxyls) all over the runner's body.

The Good: These stickers make the runner very friendly to water. The water grabs onto the stickers, and heat transfers super fast.
The Bad: These stickers are heavy and awkward. They trip the runner up. The runner can no longer sprint fast across the track. Their speed (thermal conductivity) drops by over 90%.

The Dilemma: How do we make the runner friendly to the water without tripping them up and slowing them down?

The Solution: "The Edge Strategy"

This paper proposes a clever new strategy: Don't put stickers on the runner's body; put them only on the runner's hands and feet (the edges).

The researchers used a super-smart computer simulation (powered by Artificial Intelligence) to test this. They compared two types of graphene nanoribbons (tiny strips of graphene):

Surface-Functionalized: Stickers all over the body (The "Bad Fix").
Edge-Functionalized: Stickers only on the very edges of the strip.

What They Found

1. The "Magic 10%" Boost

When they put stickers on just 10% of the edges, something amazing happened:

Heat Transfer to Water: The ability to dump heat into the water increased by 8 times (more than 800%!).
Running Speed: The runner's speed (in-plane thermal conductivity) stayed almost exactly the same.

The Analogy: Imagine a highway (the graphene). If you put construction cones all over the lanes (surface functionalization), traffic grinds to a halt. But if you just put a few friendly toll booths at the very exit ramps (edge functionalization), cars can still drive fast down the highway, but they can now easily exit and transfer their cargo to the waiting trucks (water) much faster.

2. The "Goldilocks" Effect

The researchers found a weird, non-linear relationship.

Too few stickers: The edges are still a bit rough and messy, scattering the heat.
Just the right amount (around 10-20%): The stickers "smooth out" the rough edges. They fill in the gaps where the carbon atoms are dangling. This makes the edge a perfect bridge for heat to jump into the water.
Too many stickers: Once you cover the edges completely, you don't get much more benefit because the water is already packed as tightly as it can be against the edge.

3. Why Edges Work Better Than Surfaces

Surface Functionalization: Breaking the perfect honeycomb pattern of the graphene surface is like tearing holes in a trampoline. It ruins the bounce (heat flow).
Edge Functionalization: The main body of the graphene stays perfect and smooth. The edges act like a "passive guard." They stop the heat from getting stuck at the edge (phonon localization) and instead guide it smoothly into the water.

The Real-World Impact

This discovery is a game-changer for engineering.

Before: You had to choose between a material that cooled well (but didn't conduct heat internally) or a material that conducted heat well (but didn't cool well).
Now: You can have both. By simply treating the edges of graphene strips, you create a material that is a super-conductor of heat and a super-dumper of heat into water.

Summary

Think of graphene as a high-speed train.

Old way: Glue sandpaper to the whole train to make it stick to the track. It stops the train.
New way: Just grease the wheels (the edges). The train goes just as fast, but now it can transfer its energy to the environment much more efficiently.

This paper proves that less is more: a tiny bit of chemical treatment on the very edges of graphene creates a massive improvement in cooling performance without sacrificing the material's superpowers.

1. Problem Statement

Graphene is a promising material for thermal management due to its exceptional intrinsic in-plane thermal conductivity ( $\kappa > 1000$ W/m·K). However, in aqueous applications (e.g., solar vapor generation, laser processing), the interfacial thermal conductance ( $G$ ) between graphene and water is a critical bottleneck.

The Conflict: While chemical functionalization (adding groups like hydroxyl or epoxide) can significantly enhance $G$ by improving wettability and phonon coupling, traditional surface functionalization (oxidizing the basal plane) disrupts the $sp^2$ carbon network. This introduces $sp^3$ defects that severely degrade intrinsic $\kappa$ (often by >90%), negating graphene's primary advantage.
The Gap: There was a lack of systematic understanding regarding edge functionalization (modifying only the ribbon edges) as an alternative strategy. It was unclear if edge functionalization could boost interfacial heat transfer without sacrificing in-plane thermal transport, nor was the underlying mechanism of phonon scattering at functionalized edges fully understood.

2. Methodology

The authors employed a multi-scale computational approach combining Deep Neural Network (DNN) Molecular Dynamics (MD) with equilibrium and non-equilibrium simulations.

DNN Interatomic Potential:
- Developed using the DeePMD framework, trained on a comprehensive dataset of ab initio molecular dynamics (AIMD) simulations generated via VASP.
- Dataset Expansion: Unlike previous models limited to surface-functionalized graphene, this dataset included Graphene Nanoribbons (GNRs) with various edge terminations (bare, H-passivated, hydroxyl, carboxyl) in both dry and water-immersed states to ensure transferability.
Equilibrium MD (EMD):
- Used to calculate in-plane thermal conductivity ( $\kappa$ ) of Graphene Nanoribbons Oxides (GNROs) using the Green-Kubo formalism.
- Investigated the effects of varying edge functionalization ratios (0–100%) and ribbon widths (2.32 nm to 36.6 nm).
Transient MD & Lumped Capacitance Model:
- Simulated rapid laser heating of GNROs immersed in water followed by cooling.
- Extracted Interfacial Thermal Conductance ( $G$ ) by fitting the temperature decay curves to a lumped capacitance analytical solution.
Cross-Sectional Heat Flux Analysis:
- Utilized non-equilibrium MD to map heat flux distribution across the ribbon width, quantifying the spatial impact of edges on phonon transport.

3. Key Contributions

Development of a Specialized DNN Potential: Created a high-fidelity interatomic potential capable of accurately modeling the complex interactions between oxidized graphene edges, water, and the basal plane, overcoming limitations of traditional force fields.
Discovery of Non-Monotonic $\kappa$ Behavior: Identified a competing mechanism where edge functionalization initially suppresses $\kappa$ due to disorder but subsequently recovers it by passivating dangling bonds, leading to a non-monotonic dependence on functionalization ratio.
Strategic Trade-off Analysis: Demonstrated that edge functionalization offers a unique "sweet spot": it significantly enhances interfacial thermal conductance ( $G$ ) while largely preserving the high in-plane thermal conductivity ( $\kappa$ ), unlike surface functionalization.

4. Key Results

A. In-Plane Thermal Conductivity ( $\kappa$ )

Surface vs. Edge: Surface functionalization causes a catastrophic drop in $\kappa$ (e.g., >90% reduction at 5% functionalization) due to $sp^2$ to $sp^3$ conversion and symmetry breaking. In contrast, edge functionalization at 5% reduces $\kappa$ by only ~4–34% depending on width.
Non-Monotonic Trend: For narrow ribbons, $\kappa$ $κ$ initially decreases as edge functionalization increases (due to phonon scattering and localization). However, beyond ~20–30% functionalization, $\kappa$ $κ$ begins to recover.
- Mechanism: Low functionalization introduces disorder. High functionalization saturates dangling bonds at the edges, reducing edge-induced phonon scattering and localization.
Width Dependence: $\kappa$ increases with ribbon width for both pristine and functionalized ribbons because the relative contribution of edge scattering diminishes as the width approaches the phonon mean free path.

B. Interfacial Thermal Conductance ( $G$ )

Enhancement via Edge Functionalization: Functionalizing just 10% of the ribbon edges with hydroxyl groups increases $G$ $G$ by more than 8-fold (from ~0.215 to ~2.05 $\times 10^8$ $\times 1 0^{8}$ W/m²·K) compared to pristine bulk graphene.
- Mechanism: Polar hydroxyl groups at the edges create strong hydrogen bonds with water, drastically improving wettability and interfacial coupling.
Saturation: $G$ plateaus at higher functionalization ratios (>10%) as the interfacial water packing density reaches a maximum and parallel heat transfer pathways compete.
Comparison: While surface functionalization can increase $G$ by nearly an order of magnitude, it destroys $\kappa$ . Edge functionalization provides a massive boost to $G$ (8x) with minimal penalty to $\kappa$ .

C. Structural Insights

Edge Passivation: Bare GNR edges are naturally hydrophilic due to dangling bonds, but functionalized edges (hydroxyl) create even stronger interactions.
Heat Flux Distribution: Analysis of heat flux ( $\alpha$ ) across the ribbon width shows that edges act as scattering centers. However, at optimal functionalization levels, the passivation of edges reduces this scattering, allowing for more uniform heat flow compared to partially functionalized or bare edges.

5. Significance and Implications

This work establishes edge functionalization as a superior design strategy for graphene-based thermal management systems in liquid environments.

Optimization: It resolves the trade-off between interfacial heat dissipation and intrinsic thermal transport. Engineers can now enhance heat transfer to water without sacrificing the lateral heat-spreading capability of graphene.
Applications: This is critical for applications such as:
- Solar steam generation.
- Laser processing of graphene in solution.
- Bio-thermal interfaces where efficient heat dissipation and structural integrity are both required.
Fundamental Physics: The study provides a mechanistic understanding of how edge chemistry modulates phonon transport, revealing that edge passivation can counteract the scattering effects of defects, a phenomenon previously overlooked in graphene thermal studies.

In conclusion, the paper demonstrates that selectively modifying the edges of graphene nanoribbons is a highly effective, low-cost strategy to maximize the performance of graphene in aqueous thermal applications.