QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: wot19920302 on October 20, 2016, 11:53

Title: magnetism or nonmagnetism for carbon nanotube?
Post by: wot19920302 on October 20, 2016, 11:53

Dear Quantumstaff:
           I use GGA and SGGA to calculate  electron structure of [6,0]carbon nanotube , i.e, zigzag nanotube and inspect magnetism. when using SGGA, I set initial_spin as  ferromagnetic state(1,1,1....) and  antiferromagnetic state(1,-1,1,-1.....),respectivly. The final results as following:
         Totoal energy:      NM   -3776.24077 eV
                                            FM    -3776.24582 eV
                                            AFM   -3776.24582 eV     
         it seems that FM and AFM state have the same  energy and are lower than NM. 
         However, Mulliken population shows that electron in both spin channel  of FM state(so is AFM state) are the same. (attachment one and attachment two are FM Mulliken population  and AFM Mulliken population )
        what's more, the energy band and transmission spectrums of NM, FM and AFM states are  same (fig1 )
        For my part, I think carbon nanotube shows nonferromagnetism because there is no net spin in atoms, but according to the total energy, it seems that NM state is unstable compared to FM or AFM. Is [6,0]carbon nanotube NM,FM or AFM? It confuses me :-[
      Look forward to your reply.
     Best       
      p.s.   Fig2 is calculated carbon nanotube
         

Title: Re: magnetism or nonmagnetism for carbon nanotube?
Post by: Daniele Stradi on October 20, 2016, 12:10

I would say that that simply means that the CNT is non magnetic, as it should  :)

The difference you see is probably due to numerical accuracy, which has to be increased if you want to compare spin-polarized and non-spin polarized calculations.

Title: Re: magnetism or nonmagnetism for carbon nanotube?
Post by: wot19920302 on October 20, 2016, 15:10

thanks for your reply！ ;D