Formulation for Soret effect is missing 1 / dz in the gradient term

What steps will reproduce the problem?

As was pointed out previously by Anatoli Mokhov 
(http://tech.groups.yahoo.com/group/cantera/message/1671) and in the thesis of 
Laurent Benezech (2008), in the formulation for thermal diffusion, it seem 
seems that the formula for the gradient of the logarithm of temperature, 
gradlogT, is incorrect:

On line 979 of StFlow.cpp, instead of “gradlogT = 2.0 ∗ (T(x, m+1)− T(x, 
m))/(T(x, m+ 1)+ T(x, m))”, it should be “gradlogT = 2.0 ∗ (T(x, m+1)− 
T(x, m))/(T(x, m+ 1)+ T(x, m))/(z(m+ 1)− z(m))".

What is the expected output? What do you see instead?

I've found that the missing multiplication factor, 1/dz =1/(z(m+ 1)−z(m)), is 
important especially when a finer mesh is applied to the flame region.

What version of the product are you using? On what operating system?

All versions

Please provide any additional information below.

Original issue reported on code.google.com by watso...@gmail.com on 12 Dec 2012 at 10:45

dbirch997 / cantera

Formulation for Soret effect is missing 1 / dz in the gradient term #132