Thermal conductivity reduction of multilayer graphene with fine grain sizes
- 460 Downloads
The thermal conductivities of monolayer graphene with 4.1 μm grain size and multilayer graphene with 0.3 μm grain sizes are measured by optothermal Raman technique. The number of layers and average grain sizes of graphene are controlled by copper thickness and synthesis conditions in chemical vapor deposition system. In addition, the graphene samples are suspended on a thorough 8-μm hole substrate via PMMA transfer method to avoid substrate effects. As a result, the thermal conductivity graphene is significantly reduced from 3000 to 660 W/m K at 320 K with increasing graphene layers and decreasing grain sizes due to the enhanced phonon scattering. The multilayer graphene with fine grain sizes that has the suppressed thermal conductivity can be useful for application in various field, such as thermoelectric material and thermal rectification.
KeywordsThermal conductivity reduction CVD graphene Multilayer Grain size
List of symbols
An integral function of r0 and R
Average grain size
The number of grains
Hole-suspended graphene of radius
Laser beam radius
Measured graphene temperature
Thickness of graphene
Graphene , that is a two-dimensional material of carbon atoms in the form of hexagonal lattice, has many novel properties, such as high electrical mobility , optical transparency , and superior mechanical strength . Especially, graphene has a extraordinarily high thermal conductivity  due to long phonon mean free path affiliated with the strong carbon bondings in two-dimensional lattice [6, 7]. The high thermal conductivity is significantly reduced with increasing the number of layers  and decreasing grain sizes , respectively. However, the reduced thermal conductivity of graphene can be useful for the application in various fields, such as thermoelectric material  and thermal rectification . Therefore, we tried to intentionally reduce the thermal conductivity of graphene by increasing the number of graphene layer and decreasing grain sizes for the application.
In this study, the thermal conductivity of multilayer graphene with sub-μm grain size is measured and compared to monolayer graphene. In CVD system, monolayer graphene with μm grain size and multilayer graphene with sub-μm grain size are synthesized by controlling synthesis conditions and the thickness of copper catalyst. Then, graphene samples are transferred on a through 8-μm hole via PMMA transfer method  to avoid substrate effects. Finally, the thermal conductivities of graphene samples are measured and compared by optothermal Raman technique .
2 Sample preparation and experiment
2.1 Sample preparation
The purity of copper and synthesis conditions to obtain graphene samples with different grain sizes and number of layers
Grain size (μm)
Number of layers
The purity of copper (%)
Synthesis temperature (°C)
CH4:H2 pressure ratio
The number of graphene layers was estimated by the measured absorbance because the optical absorbance linearly increases with increasing the number of graphene layers . The optical absorbance of graphene was determined from the transmission measurements with and without graphene laid on a thorough hole pattern. The measured absorbance of graphene samples were \(3.32 \pm 0.33\) and \(8.86 \pm 1.02\). The graphene with grain size of 4.1 μm seems to monolayer because the measured absorbance falls within the known absorbance range of monolayer CVD graphene [19, 20, 21, 22]. On the other hand, the measured absorbance of graphene with grain size of 0.3 μm is much higher than monolayer about 3 times, so the effective number of graphene layers is estimated to 3 layers.
As shown in Fig. 1b, all graphene samples were suspended on the 8-μm hole pattern substrate via poly(methyl methacrylate) transfer method  to avoid substrate effect on thermal transport in graphene sheets. And the hole diameter of 8 μm is sufficiently large so that the absorbed heat at the center region of the suspended graphene can be assumed to transport to the hole edge by diffusion .
Also, there is no D peak around 1350 cm−1, showing the sheet quality of the graphene is very excellent [24, 25]. However, D peak is more pronounced and G peak to 2D peak intensity ratio decreases due to multilayers  and small grain sizes  for the multilayer graphene with 0.3 μm grain size.
3 Results and discussions
The thermal conductivities of two graphene samples decrease with increasing temperature due to the enhanced Umklapp scattering at higher temperature [7, 29]. Grain boundary scattering largely depends on grain boundary density or grain sizes unlike Umklapp scattering that has strong temperature dependency . As a result, the negative temperature dependence of thermal conductivities is weakened with decreasing grain sizes from ~ T−1.91 to ~ T−0.86 in Fig. 3, because the heat is transported by largely grain boundary scattering rather than Umklapp scattering for graphene with smaller grain sizes.
In summary, the measured thermal conductivities of graphene with different layers and grain sizes were compared to effectively reduce thermal transport in graphene. By manipulation of synthesis conditions and copper thickness, monolayer graphene with 4.1 μm grain size and multilayer graphene with 0.3 μm grain size were synthesized in CVD system. The measured thermal conductivities of multilayer graphene with 0.3 μm grain size (660–330 W/m K for 320 K < T < 550 K) is significantly lower than those of monolayer graphene with 4.1 μm grain size (3000–1280 W/m K for 320 K < T < 550 K).
This research was primarily supported by the Nano-Material Technology Development Program (R2011-003-2009) and Basic Science Research Program (2017R1A2B3005706), and Global Frontier R&D Program on Center for Multiscale Energy System (Grant No. 2012-054172) through the National Research Foundation of Korea funded by the Ministry of Science, ICT and Future Planning, and was also partially supported by the Magnavox Professorship fund from the University of Tennessee (R0-1137-3164) and Institute of Engineering Research at Seoul National University.