Boosting Thermal Conductivity by Surface Plasmon Polaritons Propagating along a Thin Ti Film

We experimentally demonstrate boosted in-plane thermal conduction by surface plasmon polaritons (SPPs) propagating along a thin Ti film on a glass substrate. Due to the lossy nature of metal, SPPs can propagate over centimeter-scale distances even along a supported metal film, and the resulting ballistic heat conduction can be quantitatively validated. Further, for a 100-nm-thick Ti film on a glass substrate, a significant enhancement of in-plane thermal conductivity compared to bulk value ($\sim 25\%$) is experimentally shown. This Letter will provide a new avenue to employ SPPs for heat dissipation along a supported thin film, which can be readily applied to mitigate hot-spot issues in microelectronics.

Figure 1

Experimental setup for measuring the in-plane thermal conductivity of Ti films. (a) Schematic of a custom-built steady-state thermoreflectance setup. (b) Temperature rise at the sample surface corresponding to the applied laser irradiation with power $Q$. The rise of the temperature of the heated region drops if there exists a ballistic thermal transport via SPPs supported along a thin Ti film. (c) Schematic of a set of measurement samples. Ti films were patterned in circular shapes with different radius values $r$ on a single glass (i.e., amorphous ${\mathrm{SiO}}_2$) substrate. Inset shows the cross section of the deposited Ti film with a thickness of 100 nm.

Figure 2

SPP-enhanced thermal conductivity, $k_{\parallel ,\mathrm{spp}}$. (a) Measured probe reflectance response with respect to the lock-in magnitude of the pump laser photodetector response for samples with different thicknesses and radii. Every plotted point was obtained by averaging the data over 5 seconds with a time interval of 50 ms. The standard deviation of each point is within 1.5%. (b) $k_{\parallel ,\mathrm{spp}}$ of the Ti film with respect to the radius of the film. A sample with a radius of $200\mathrm{\mu }\mathrm{m}$ was used for calibration; that is, we set $k_{\parallel ,\mathrm{spp}}=0$ with $r=200\mathrm{\mu }\mathrm{m}$ for all samples. Every point is the averaged value of three repeated measurements, and dashed curves represent the theoretically predicted values of $k_{\parallel ,\mathrm{spp}}$.

Figure 3

SPP dispersion curves for Ti films on a glass substrate with film thicknesses $d=50\mathrm{nm}$, 100 nm, and 300 nm. (a) Real part of in-plane wave vector (${\beta }_R$). (b) Skin depth at $\mathrm{air}/\mathrm{Ti}$ interface penetrating into air. (c) Propagation length ($\mathrm{\Lambda }$) of SPP with respect to film thickness. The shaded region indicates the frequency range where the real part of the dielectric function of the glass substrate is negative. In other words, SPPs originating from a $\mathrm{glass}/\mathrm{Ti}$ interface cannot be excited in the shaded region. (d) Calculated $k_{\parallel ,\mathrm{spp}}$ with respect to the film thickness for different values of radius.