Scanning Tunneling Microscope I-V curve

[Edge5]

I know that scanning tunneling microscopy (STM) I-V curve is exponential for a conducting sample.
How would it look like if I had a semiconductor and a non-conductor sample?

 

[Lord Jestocost]

Scanning tunneling microscopy is based on the flow of an electrical current. Thus, this method isn’t suited to directly image electrically insulating material. Regarding metals and semiconductors, the current-voltage characteristics depends on the density of electron states of the tip and sample material. Have a look at http://hoffman.physics.harvard.edu/research/STMtechnical.php