Edge‐Grafted Molecular Junctions between Graphene Nanoplatelets: Applied Chemistry to Enhance Heat Transfer in Nanomaterials

Abstract

The edge-functionalization of graphene nanoplatelets (GnP) is carried out exploiting diazonium chemistry, aiming at the synthesis of edge decorated nanoparticles to be used as building blocks in the preparation of engineered nanostructured materials for enhanced heat transfer. Indeed, both phenol functionalized and dianiline-bridged GnP (GnP-OH and E-GnP, respectively) are assembled in nanopapers exploiting the formation of non-covalent and covalent molecular junctions, respectively. Molecular dynamics allow to estimate the thermal conductance for the two different types of molecular junctions, suggesting a factor 6 between conductance of covalent vs noncovalent junctions. Furthermore, the chemical functionalization is observed to drive the self-organization of the nanoflakes into the nanopapers, leading to a 20% enhancement of the thermal conductivity for GnP-OH and E-GnP while the cross-plane thermal conductivity is boosted by 190% in the case of E-GnP. The application of chemical functionalization to the engineering of contact resistance in nanoparticles networks is therefore validated as a fascinating route for the enhancement of heat exchange efficiency on nanoparticle networks, with great potential impact in low-temperature heat exchange and recovery applications.

Flexible and High Thermal Conductivity Composites Based on Graphite Nanoplates Paper Impregnated with Polydimethylsiloxane

“…Thanks to the densification of the sample, the cross-plane conductivity turned out to be practically equal between the pristine and pressed samples, while in in-plane direction, it more than quadrupled (Table 1) to approximately 40 W/m•K. It is worth mentioning that this value is still relatively low compared with previously reported GNP nanopapers [26,31,37] owing to the relatively low density. Besides the thermal conductivity performance, it has to be mentioned that pristine GNP nanopapers are intrinsically brittle, which limits their practical application as heat spreaders.…”

Dimension Effects of Graphene Sheets as Building Blocks: Implications for Thermal Conductivity Improvement of Graphene Films

“…Herein, we develop a multiscale modeling strategy combining MD and FE methods to predict the κ eff of thermally conductive films prepared by graphene sheets with specific dispersity in lateral size and thickness. In previous reports using the FE methods to evaluate heat dissipation, graphene films are generally considered bulk materials with a given thermal conductivity value. − Only a few works have studied the thermal conductivity of graphene laminates using a multiscale scheme. , However, in their models, all graphene flakes share the same thermal conductivity, neglecting the differences in intrinsic thermal properties due to sheet size and thickness. Therefore, we establish an RVE modeling that can take into account the dimension dispersity based on the atomistic level MD results of various graphene sheets.…”

“…43−46 Only a few works have studied the thermal conductivity of graphene laminates using a multiscale scheme. 24,39 However, in their models, all graphene flakes share the same thermal conductivity, neglecting the differences in intrinsic thermal properties due to sheet size and thickness. Therefore, we establish an RVE modeling that can take into account the dimension dispersity based on the atomistic level MD results of various graphene sheets.…”

Crystallization of Polycaprolactone within Nanopapers Based on Graphene-Related Materials

“…A Shirley background function was utilized to adjust the spectra’s background. The C 1s peak fitting used the software CasaXPS and accounted for the contribution of C–C bonds with sp 2 -like characteristics using an asymmetric peak (Doniach-Šunjić shape), , as previously calculated on freshly cleaved highly oriented pyrolytic graphite (HOPG) (ZYH grade, Mikromasch), with an obtained asymmetry index (α) of 0.115. Curve fitting employed a Gaussian (80%)-Lorentzian (20%) peak shape by minimizing the total square-error fit.…”