3D printing of hexagonal boron nitride nanosheets/polylactic acid nanocomposites for thermal management of electronic devices

Mustafa Caner Gorur, Doga Doganay, Mete Batuhan Durukan, Melih Ogeday Cicek, Yunus Eren Kalay, Cem Kincal, Nuri Solak, Husnu Emrah Unalan*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)


The integration of electronic devices into various fields, from daily life appliances to industrial applications, has increased significantly. Thermal management of these devices has become crucial to improve their performance, efficiency, and lifespan. In this context, exfoliated hexagonal boron nitride nanosheets (h-BNNS) stand out as promising candidates due to their superior thermal conductivity despite being electrically insulating. In this work, h-BNNS were used as fillers in polylactic acid (PLA) matrix nanocomposites for 3D-printing. First, a nanocomposite filament was prepared and then used for 3D-printing of the heat sink and LED bulb holder. The thermal conductivity of 3D-printed PLA was found to increase by 400% with the addition of 40 vol. % h-BNNS. Adding h-BNNS to PLA performed almost on par with the commercial aluminum (Al) heat sink, while improving the heat dissipation by 220% compared to bare PLA. In addition, 3D-printed h-BNNS/PLA nanocomposite LED bulb holders dissipated the excess heat from the LEDs much more efficiently than the commercial product. The results shown here have proven that h-BNNS/PLA nanocomposites have great potential for the thermal management of electronic devices.

Original languageEnglish
Article number110955
JournalComposites Part B: Engineering
Publication statusPublished - Oct 2023

Bibliographical note

Publisher Copyright:
© 2023 Elsevier Ltd


This work was supported by Middle East Technical University , Scientific Research Projects program [grant number: GAP-308-2021-10740 ]. Autodesk Inventor Professional 2018 software was used for the design of the samples and Prusa Slicer 2.3.0 was used to set the printing parameters of these designs. Prusa i3 MKS3 3D Printer was used for the 3D-printing of filaments. Nozzle temperatures for PLA and h-BNNS/PLA filament were 215 °C, and print bed temperatures were 70 °C for PLA and 60 °C for h-BNNS/PLA respectively. The layer height, infill rate, and printing orientation were 0.1 mm, 100% and ±45°, respectively. Moreover, the rectilinear pattern was chosen for filling, and a 1 mm brim was used. The cooling settings for bare PLA were 35% minimum and 100% maximum fan speed and the fan was always on. Meanwhile, the fan speed was configured for the heat sink and LED bulb holder printing (for h-BNNS/PLA filament) to be 20% for minimum and 50% for maximum. In addition, support layers were also used for LED bulb holder printing. A commercially available Ecolite LED bulb was disassembled without damaging the bulb, photos of which are provided in Fig. S2a. 3D-printed PLA, h-BNNS/PLA, and commercial bulb holders were assembled with LED chips and sockets. In the final step, a thermal conductive paste was applied to the junction points between the LED chips and bulb holders to facilitate effective heat transfer from the LED chip to the holder (Fig. S2b).Morphologies of the bare PLA, h-BN/PLA, and h-BNNS/PLA composites, h-BN dispersion uniformity were investigated via SEM (FEI Nova Nano SEM 430). SEM samples were gold coated beforehand. The conservation of the (002) plane in the h-BNNS structure was analyzed by TEM (JEOL 2100F 200 kV RTEM). Labelling of the planes was performed by calculating the corresponding interplanar spacing values using the GATAN (GMS 3) software. TEM sample was prepared by drop casting of aqueous diluted h-BNNS solution over a holey carbon supported copper TEM grid. Atomic Force Microscopy (AFM) characterization was conducted to calculate the number of layers of h-BNNS using non-tapping mode via Veeco Multimode V AS-12 device. The crystal structure was analyzed through X-ray diffraction (XRD) using RIGAKU D/MAX 2200 ULTIMA/PC. It was used to monitor the changes in thickness. Raman spectroscopy was performed using an EMCCD camera (Andor Newton). This method was used to observe the changes in the E2g phonon mode caused by the stretching mode between boron and nitrogen atoms. Raman signals were analyzed with an ANDOR SR750 device. TGA characterization was conducted via Exstar SII TG/DTA 7300 to calculate the weight percent of h-BN and h-BNNS following the filament production. It was also used to compare the amount of loss in h-BN/PLA and the h-BNNS/PLA following filament production. This analysis was performed between 30 °C and 550 °C with a heating rate of 10 °C.min−1 under nitrogen atmosphere. Laser Flash Analysis (LFA) method was deployed to measure thermal diffusivities (α) of 3D-printed bare PLA and h-BNNS/PLA samples. Measurements were performed under ambient conditions via Netzsch LFA 457 Microflash device. Two sides of samples were coated with a thin layer of graphite in order to absorb incident laser beam for bottom surface and keep emissivity on the top surface same for all samples. In addition, the specific heat capacity (Cp) values of these samples were also measured by means of Laser Flash method. POCO graphite with 2.483 mm thickness was used as a reference sample for Cp calculations. Finally, Archimedes method was used to measure the densities of the samples. The obtained values were used for calculating the thermal conductivity.Following exfoliation, the morphologies of the h-BN and h-BNNS were analyzed using SEM. SEM image of h-BN is provided in Fig. 2a. Agglomerated, thick, and non-transparent h-BN layers were observed. Folding of the layers can be observed for the h-BNNS, which proved the presence of few-layered and exfoliated h-BNNS (Fig. 2b). Furthermore, electron transparent layers and h-BNNS with clear hexagonal morphology can be seen in this image. The reduction in the number of layers as a result of the exfoliation of h-BN was also monitored through AFM analysis. The thickness of a monolayer h-BNNS is 0.33 nm [50]. The AFM image and corresponding layer numbers are provided in Fig. 2c. The measured thicknesses were 0.29 nm and 0.65 nm, corresponding to 1 and 2 layers, respectively. Wang et al. reported thickness values of 1.8 nm and 2.3 nm corresponding to 6–7 layers of h-BNNS using IPA and Li+ ions as a support for exfoliation [56]. In a different study, Liu et al. reduced the thickness of h-BN to 3.5 nm by ball-milling. It was claimed that this corresponds to 3–4 layers [51]. In the current study, the flake thickness values were smaller than most of the values provided in the literature [50–55]. However, it can be noticed that the lateral size of h-BNNS was smaller than that observed in SEM images. The reason behind this size difference can be explained by the preparation method of AFM samples. AFM samples should be homogeneous and prepared from a dilute solution. This is required for the probe tip to get close to nanosheets on the surface. It was possible that the small-sized h-BN sheets stuck to the silicon wafer surface during spin coating, while larger sheets spread out of the silicon wafer surface. This situation has also been encountered in the literature [55].This work was supported by Middle East Technical University, Scientific Research Projects program [grant number: GAP-308-2021-10740].

FundersFunder number
Association Française contre les Myopathies
The Ministry of Economic Affairs and EmploymentSR750
Orta Doğu Teknik ÜniversitesiGAP-308-2021-10740


    • 3D printing
    • Hexagonal boron nitride nanosheets
    • Nanocomposite filament
    • Polylactic acid
    • Thermal management


    Dive into the research topics of '3D printing of hexagonal boron nitride nanosheets/polylactic acid nanocomposites for thermal management of electronic devices'. Together they form a unique fingerprint.

    Cite this