Illustrated guide to high-k dielectrics and metal gate electrodes

The Penryn processor debuted Intel’s 45nm fabrication, and was the first to utilize high-k gate dielectrics and metal gate electrodes. This change in technology was significant for a number of reasons: the processes used were far in advance of Intel’s competition it provided a basis on which Intel could build and scale down still further high-k dielectrics and metal gates allow decreases in transistor size and power consumption, and increases in speed and efficiency As transistors get smaller the thickness of the silicon dioxide (Gate Oxide on the diagram below) needs to reduce in order to increase capacitance. Increased capacitance means increased current and improved performance, while reduced transistor size means more transistors on a die. More transistors on a die means more computing power. Processor technology means that getting smaller and working faster and more efficiently complement perfectly. The problem arises when the thickness of silicon decreases past a certain level and it gets so small that current begins to leak out. The electrons undergo a process known as tunnelling: they escape the transistor and dissipate which means the gate is less efficient, resulting in increased power consumption and reduced reliability. This loss in efficiency is one of the reasons chips (think later Pentiums) that had many transistors crammed onto them would produce such high TDP. After a certain point further reduction in size becomes impractical, the physical limit of Moore’s law is reached, and a new approach is needed. Enter high-k dielectrics. A high-k dielectric is a material with a high dielectric constant (k) which, in this case, replaces the silicon dioxide layer of the transistor. A...

Pin It on Pinterest