# How thin a practical capacitor can be?

According to the formula of a $parallel$ $plate$ capacitor...

$C$ $=$ $\dfrac{\epsilon_0 A}{d}$

The thinner the capacitor, more the charge it will be able to store...

And hence the graph should somewhat look like (excluding the negative part):

So if we keep on decreasing the distance between the two plate the C should increase... But how small the d can practically be???

It cannot be in nanometres or something like that...

• Gate dielectrics for modern CMOS devices are in the few nanometer thicknesses. Since these are (functionally) capacitors, d can certainly be small. But, high-k dielectrics had to be introduced for further scaling to reduce tunneling and thus leakage currents. Oct 2, 2014 at 14:42
• @JonCuster So any limit must be there...? Oct 2, 2014 at 14:44
• 0 is a limit (!). However, real materials have a dielectric breakdown limit, i.e. a limit to the voltage gradient supported by the material. If placing one charge on the top plate results in a voltage gradient exceeding the dielectric breakdown than you no longer have a capacitor. Note that as CMOS has scaled down the gate thickness the supply voltage has had to come down as well. Oct 2, 2014 at 14:54