Substrate Thickness

It is set on the wafer and is the same for every site on the array; it is not the shank width, which is a design parameter of the layout.

Thinner displaces less. A 15 µm shank parts tissue with less injury and less of the glial response that raises impedance over the weeks of a chronic implant, which is why it is the default for cortex and for most rodent work. Its cost is stiffness: a thin shank bends more easily and can buckle against the dura or in a long trajectory. The 50 µm substrate is for that case: deeper targets, longer shanks, tougher tissue, or an insertion through intact dura — and beyond what 50 µm can reach, the Vector design moves the array onto a stainless-steel support body rather than thickening the silicon further. The choice is made at order: most catalog designs are offered on both. NeuroNexus also offers ultra-thin 5 µm designs, which reduce tissue displacement and the encapsulation response further (Seymour and Kipke (2007)); the evidence is in NeuroNexus Probe Technology: Questions and Answers.

Polyimide serves the opposite requirement. At 12 µm it is flexible enough to conform to the surface of the cortex, a nerve or the heart, which is why it is the substrate for thin-film grids and for the flexible cable of a Hybrid package; it does not penetrate. Both are laid down in the same thin-film process; what differs is the material and the use of the finished array.