question about Li-H2-Li two-probe system

[qqquser]

hi, I set up a Li-H2-Li two-probe system, followingthe VNL version 200810 tutorial,and the results about tran-spectrum is very well,but when I change Li to AU ,tran-spectrum shows that tran-coefficient is 0, this is reason I do not kown ,why??? please help me,thanks ..

[Anders Blom]

Most likely the calculation has converged to a state with zero charge in the entire system. Au is trickier than Li and requires not only a bit higher temperature but most likely also a longer electrode.

[qqquser]

thank you , I will try .........

[qqquser]

hi, I calculate the tran-spectrum for two-probe system, but I often find some questions, such as:
1.# Warning : Simple partition used, performance issues may occur!
2. runtime_parameters = runtime_parameters
    NLValueError: Increase the basis set size. The number of basis functions is less or equal to the number of electrons.
3.ATKError: Exceeded maximum number of self-consistent iterations.
   

          I want to correct these mistakes,  how to do?       thank you.........

[Nordland]

1) The first error indicates that your indicates that the system you are calculate on having a very strange geometry or else something is incorrect in your setup. I would advise you to inspect the structure in VNL first.
2) You are calculating on a system with all closed shells and therefore you can not use the SZ basis - you have to increase your basis set by using at least DZ.
3) Look for tips on converging twoprobes on this forum and you will find a lot of help for avoid this error.