Effect of film properties for non-linear DPL model in a nanoscale MOSFET with high-k material: ZrO2/HfO2/La2O3

• Highly non-linear transient DPL model with Neumann and Robin BCs is developed.
• Four different cases of MOSFETs, including high k/metal gate MOSFETs are studied using our new engineering model.
• Temperature/size dependent thermal properties of Silicon are considered.
• This model helps us to predict what is happening in a real MOSFET.

Numerical simulation of non-linear non-Fourier heat conduction within a nano-scale metal–oxide–semiconductor field-effect transistor (MOSFET) is presented under the framework of Dual-Phase-Lag model including the boundary phonon scattering. The MOSFET is modeled in four cases of: (I) thin silicon slab, (II) including uniform heat generation, (III) double-layered buried oxide MOSFET with uniform heat generation in silicon-dioxide layer, and (IV) high-k/metal gate transistor. First, four cases are studied under four conditions of (a) constant bulk and (b) constant film thermal properties, (c) temperature-dependent properties of bulk silicon, and (d) temperature-dependent thermal properties of film silicon. The heat source and boundary conditions are similar to what existed in a real MOSFET. It is concluded that in all cases, considering the film properties lowers the temperature jump due to the reduction of the Knudsen number. Furthermore, the speed of heat flux penetration for film properties is less than that of the cases concerning bulk properties. Also, considering the temperature-dependent properties drastically changes the temperature and heat flux distributions within the transistor, which increases the diffusion speed and more, decreases the steady state time. Calculations for case (III) presents that all previous studies have underestimated the value of the peak temperature rise by considering the constant bulk properties of silicon. Also, it is found that among the high-k dielectrics investigated in case (IV), zirconium dioxide shows the least peak temperature rise. This presents that zirconium dioxide is a good candidate as far as the thermal issues are concerned.