The Metal-Dielectric Interface in a Capacitor

[chuckschuldiner]

Hi everyone,

I have been reading a recent Nature paper (Stengel and Spaldin, 2006) where ab initio simulations on SrRuO_3/SrTiO_3/SrRu_O3 nanocapacitors have been performed to investigate the origin of the so-called dead layer effect in thin-film nanocapacitors. They arrive at several interesting results which i will try to summarise

1) SrRuO_3 electrodes were found to have poor electron screening which resulted in high charge penetration inside the electrodes. This also resulted in a large depolarization field inside the SrTiO_3 dielectric layer which resulted in the observed capacitance being around 6 times lesser than the expected capacitance

2) The results also show that at the interface of the metal and the dielectric, the dielectric constant is continuos. Specifically, for the metal the dielectric constant is the same as that for the dielectric at the interface and increases to infinity at a large enough distance from the dielectric.

My question is, is the continuity in dielectric constant anything special? Will all physical properties be continuos at the interface? Also, is there any way i can model the spatial behavior of the dielectric constant of the metal?

 

[petr1243]

This schematic is nothing but a Josephson junction, taking into assumption that SrRuO_3 is a superconductor. SrTiO_3(Strontium Titanium Oxide) will be the dielectric(insulating barrier), so this insulating barrier is treated as the junction capacitance. Properties of this insulating barrier:Now based on whether the electrostatic energy is greater/less than the energy due to josephson coupling, we can find out whether we have strong or weak tunneling of cooper pairs(in other words this will tell us whether our junction capacitance is small/big). This is all true if SrTiO_3 is a superconductor. I'm not too experienced in this subject YET, so I apologize if there are a couple of misnomers in my reply.

 

[f95toli]

SRO is a superconductor, but only at very low temperatures (Tc is pretty low)
.
But SRO is also an oxide which is used to make what is sometimes called "functional interfaces". As a matter a fact these interfaces ARE somewhat related to Josephson junctions in that they are made (grown) using methods originally developed to fabricate trilayer high-Tc Josephson junctions (which are also oxides); you will find that many groups working on SRO also work on high-Tc.
However, this is the only connection; this has nothing direcly to do with JJ.

Unfortunately, I can't answer any of the originaly questions. However, I think Jochen Mannhart from University of Augsburg has published (or will publish soon) a short review on the subject which might be helpfull.