MRCI vs FCI

[Gileadc]

Dear all

following an example from the user manual on MRCI I have noticed something that boggles me, when comparing MRCI and FCI. Here are the inputs I used:

--->>>mrci calculation

***,BH singlet Sigma and Delta states

r=2.1

geometry={b;h,b,r}

{hf;occ,3;wf,6,1;}

{multi;

occ,3,1,1;frozen,1;wf,6,1;state,3;lquant,0,2,0;wf,6,4;lquant,2;

tran,lz;

expec2,lzlz;}

! Sigma states:- energies -25.20509620 -24.94085861

{ci;occ,3,1,1;core,1;wf,6,1;state,2,1,3;}

e_sig_1 = energy(1)

e_sig_2 = energy(2)

! Delta states:- energies -24.98625171

{ci;occ,3,1,1;core,1;wf,6,1;state,1,2;}

e_delta_1 = energy(1)

! Delta state:- xy component

{ci;occ,3,1,1;core,1;wf,6,4;}

e_delta_2 = energy(1)

{table,e_sig_1,e_sig_2,e_delta_1,e_delta_2

head,e_sig_1,e_sig_2,e_delta_1,e_delta_2

save,energy_mrci.txt}

--- 

--->>>FCI calculation

***,BH singlet Sigma and Delta states

MEMORY,1,G    

r=2.1

geometry={b;h,b,r}

{hf;occ,3;wf,6,1;}

{multi;

occ,3,1,1;frozen,1;wf,6,1;state,3;lquant,0,2,0;wf,6,4;lquant,2;

tran,lz;

expec2,lzlz;}

! Sigma states:- energies -25.20509620 -24.94085861 Delta states:- energies -24.98625171

{fci;wf,6,1,0;state,3;}

e_sig_1 = energy(1)

e_delta_1 = energy(2)

e_sig_2 = energy(3)

! Delta state:- xy component

{fci;wf,6,4,0;}

e_delta_2 = energy(1)

{table,e_sig_1,e_sig_2,e_delta_1,e_delta_2

head,e_sig_1,e_sig_2,e_delta_1,e_delta_2

save,energy_fci.txt}

---

In the MRCI calculation, both delta states have the same exact energy, while in the FCI calculation, they do not. why is that, and which calculation is correct? If both states are degenerate shouldn't the FCI calculation converge to the same value? And if they are not degenerate, why the MRCI calculation is wrong? Is something wrong with my input?

Best regards and thanks for the help
Gilead