From f6b18f3c2a40ead3ac1d78805605eb43c5f5a57e Mon Sep 17 00:00:00 2001 From: pfloos Date: Wed, 18 Sep 2024 18:08:24 +0200 Subject: [PATCH] eomGW --- src/GW/RGW.f90 | 3 +- src/GW/ccRG0W0.f90 | 6 +- src/GW/eomRG0W0.f90 | 315 ++++++++++++++++++++++++++++++++++++++++++++ src/GW/ufRG0W0.f90 | 8 +- 4 files changed, 325 insertions(+), 7 deletions(-) create mode 100644 src/GW/eomRG0W0.f90 diff --git a/src/GW/RGW.f90 b/src/GW/RGW.f90 index a48c645..768dca8 100644 --- a/src/GW/RGW.f90 +++ b/src/GW/RGW.f90 @@ -132,6 +132,7 @@ subroutine RGW(dotest,doG0W0,doevGW,doqsGW,doufG0W0,doufGW,maxSCF,thresh,max_dii call wall_time(start_GW) ! TODO call ufRG0W0(dotest,TDA_W,nBas,nOrb,nC,nO,nV,nR,nS,ENuc,ERHF,ERI_MO,eHF) + call eomRG0W0(dotest,nBas,nOrb,nC,nO,nV,nR,nS,ENuc,ERHF,ERI_MO,eHF) call wall_time(end_GW) t_GW = end_GW - start_GW @@ -161,7 +162,7 @@ subroutine RGW(dotest,doG0W0,doevGW,doqsGW,doufG0W0,doufGW,maxSCF,thresh,max_dii ! Perform CC-based G0W0 calculation !------------------------------------------------------------------------ - doccG0W0 = .true. + doccG0W0 = .false. if(doccG0W0) then diff --git a/src/GW/ccRG0W0.f90 b/src/GW/ccRG0W0.f90 index 2c67fe3..a8b8d54 100644 --- a/src/GW/ccRG0W0.f90 +++ b/src/GW/ccRG0W0.f90 @@ -161,7 +161,8 @@ subroutine ccRG0W0(maxSCF,thresh,max_diis,nBas,nOrb,nC,nO,nV,nR,ERI,ENuc,ERHF,eH do k=nC+1,nO do c=1,nV-nR - res(i,a,j) = res(i,a,j) - 2d0*ERI(j,nO+c,nO+a,k)*amp(i,c,k) + res(i,a,j) = res(i,a,j) - 2d0*ERI(j,nO+c,nO+a,k)*amp(i,c,k) & + - 2d0*ERI(i,nO+c,nO+a,k)*amp(k,c,j) end do end do @@ -179,7 +180,8 @@ subroutine ccRG0W0(maxSCF,thresh,max_diis,nBas,nOrb,nC,nO,nV,nR,ERI,ENuc,ERHF,eH do k=nC+1,nO do c=1,nV-nR - res(i,a,nO+b) = res(i,a,nO+b) + 2d0*ERI(nO+a,k,i,nO+c)*amp(k,c,nO+b) + res(i,a,nO+b) = res(i,a,nO+b) + 2d0*ERI(nO+a,k,i,nO+c)*amp(k,c,nO+b) & + + 2d0*ERI(nO+b,k,i,nO+c)*amp(k,a,nO+c) end do end do diff --git a/src/GW/eomRG0W0.f90 b/src/GW/eomRG0W0.f90 new file mode 100644 index 0000000..c85ee9a --- /dev/null +++ b/src/GW/eomRG0W0.f90 @@ -0,0 +1,315 @@ +subroutine eomRG0W0(dotest,nBas,nOrb,nC,nO,nV,nR,nS,ENuc,ERHF,ERI,eHF) + +! EOM version of G0W0 + + implicit none + include 'parameters.h' + +! Input variables + + logical,intent(in) :: dotest + + integer,intent(in) :: nBas + integer,intent(in) :: nOrb + integer,intent(in) :: nC + integer,intent(in) :: nO + integer,intent(in) :: nV + integer,intent(in) :: nR + integer,intent(in) :: nS + double precision,intent(in) :: ENuc + double precision,intent(in) :: ERHF + double precision,intent(in) :: ERI(nOrb,nOrb,nOrb,nOrb) + double precision,intent(in) :: eHF(nOrb) + +! Local variables + + integer :: p + integer :: s + integer :: i,j,k,l + integer :: a,b,c,d + integer :: jb,kc,ia,ja + integer :: klc,kcd,ija,ijb,iab,jab + + logical :: print_W = .false. + logical :: dRPA + integer :: isp_W + double precision :: EcRPA + integer :: n2h1p,n2p1h,nH + double precision,external :: Kronecker_delta + double precision,allocatable :: H(:,:) + double precision,allocatable :: cGW(:,:) + double precision,allocatable :: eGW(:) + double precision,allocatable :: Z(:) + integer,allocatable :: order(:) + + logical :: verbose = .false. + double precision,parameter :: cutoff1 = 0.01d0 + double precision,parameter :: cutoff2 = 0.01d0 + double precision :: eF + double precision,parameter :: window = 2.5d0 + + double precision :: start_timing,end_timing,timing + +! Output variables + +! Hello world + + write(*,*) + write(*,*)'***********************************' + write(*,*)'* Restricted EOM-G0W0 Calculation *' + write(*,*)'***********************************' + write(*,*) + +! Dimension of the supermatrix + + n2h1p = nO*nO*nV + n2p1h = nV*nV*nO + nH = 1 + n2h1p + n2p1h + +! Memory allocation + + allocate(H(nH,nH),eGW(nH),cGW(nH,nH),Z(nH),order(nH)) + +! Initialization + + dRPA = .true. + EcRPA = 0d0 + + eF = 0.5d0*(eHF(nO+1) + eHF(nO)) + +!-------------------------! +! Main loop over orbitals ! +!-------------------------! + + do p=nO,nO + + H(:,:) = 0d0 + +!-----------------------------------------! +! Compute BSE supermatrix ! +!-----------------------------------------! +! ! +! | A V2h1p V2p1h 0 0 | ! +! | | ! +! | V2h1p A2h2p 0 B2h1p 0 | ! +! | | ! +! H = | V2p1h 0 A2p2h 0 B2p1h | ! +! | | ! +! | 0 0 0 0 0 | ! +! | | ! +! | 0 0 0 0 0 | ! +! ! +!-----------------------------------------! + + call wall_time(start_timing) + + !---------! + ! Block F ! + !---------! + + H(1,1) = eHF(p) + + !-------------! + ! Block V2h1p ! + !-------------! + + ija = 0 + do i=nC+1,nO + do j=nC+1,nO + do a=nO+1,nOrb-nR + ija = ija + 1 + + H(1 ,1+ija) = sqrt(2d0)*ERI(p,a,i,j) + H(1+ija,1 ) = sqrt(2d0)*ERI(p,a,i,j) +! H(1+n2h1p+n2p1h+ija,1 ) = sqrt(2d0)*ERI(p,a,i,j) +! H(1+ija,1+n2h1p+n2p1h ) = sqrt(2d0)*ERI(p,a,i,j) + + end do + end do + end do + + !-------------! + ! Block V2p1h ! + !-------------! + + iab = 0 + do i=nC+1,nO + do a=nO+1,nOrb-nR + do b=nO+1,nOrb-nR + iab = iab + 1 + + H(1 ,1+n2h1p+iab) = sqrt(2d0)*ERI(p,i,b,a) + H(1+n2h1p+iab,1 ) = sqrt(2d0)*ERI(p,i,b,a) +! H(1 ,1+2*n2h1p+n2p1h+iab) = sqrt(2d0)*ERI(p,i,b,a) +! H(1+2*n2h1p+n2p1h+iab,1 ) = sqrt(2d0)*ERI(p,i,b,a) + + end do + end do + end do + + !-------------! + ! Block A2h1p ! + !-------------! + + ija = 0 + do i=nC+1,nO + do j=nC+1,nO + do a=nO+1,nOrb-nR + ija = ija + 1 + + klc = 0 + do k=nC+1,nO + do l=nC+1,nO + do c=nO+1,nOrb-nR + klc = klc + 1 + + H(1+ija,1+klc) & + = ((eHF(i) + eHF(j) - eHF(a))*Kronecker_delta(j,l)*Kronecker_delta(a,c) & + - 2d0*ERI(j,c,a,l))*Kronecker_delta(i,k) + +! H(1+n2h1p+n2p1h+ija,1+n2h1p+n2p1h+klc) & +! = ((eHF(i) + eHF(j) - eHF(a))*Kronecker_delta(j,l)*Kronecker_delta(a,c) & +! - 2d0*ERI(j,c,a,l))*Kronecker_delta(i,k) + + end do + end do + end do + + end do + end do + end do + + !-------------! + ! Block A2p1h ! + !-------------! + + iab = 0 + do i=nC+1,nO + do a=nO+1,nOrb-nR + do b=nO+1,nOrb-nR + iab = iab + 1 + + kcd = 0 + do k=nC+1,nO + do c=nO+1,nOrb-nR + do d=nO+1,nOrb-nR + kcd = kcd + 1 + + H(1+n2h1p+iab,1+n2h1p+kcd) & + = ((eHF(a) + eHF(b) - eHF(i))*Kronecker_delta(i,k)*Kronecker_delta(a,c) & + + 2d0*ERI(a,k,i,c))*Kronecker_delta(b,d) +! H(1+2*n2h1p+n2p1h+iab,1+2*n2h1p+n2p1h+kcd) & +! = ((eHF(a) + eHF(b) - eHF(i))*Kronecker_delta(i,k)*Kronecker_delta(a,c) & +! + 2d0*ERI(a,k,i,c))*Kronecker_delta(b,d) + + end do + end do + end do + + end do + end do + end do + + !-------------! + ! Block B2h1p ! + !-------------! + + ija = 0 + do i=nC+1,nO + do j=nC+1,nO + do a=nO+1,nOrb-nR + ija = ija + 1 + + kcd = 0 + do k=nC+1,nO + do c=nO+1,nOrb-nR + do d=nO+1,nOrb-nR + kcd = kcd + 1 + + H(1+ija,1+n2h1p+kcd) = - 2d0*ERI(j,k,a,c) + + end do + end do + end do + + end do + end do + end do + + !-------------! + ! Block B2p1h ! + !-------------! + + iab = 0 + do i=nC+1,nO + do a=nO+1,nOrb-nR + do b=nO+1,nOrb-nR + iab = iab + 1 + + klc = 0 + do k=nC+1,nO + do l=nC+1,nO + do c=nO+1,nOrb-nR + klc = klc + 1 + + H(1+n2h1p+iab,1+klc) = - 2d0*ERI(a,c,i,l) + + end do + end do + end do + + end do + end do + end do + + !-------------------------! + ! Diagonalize supermatrix ! + !-------------------------! + + call wall_time(start_timing) + + call diagonalize_general_matrix(nH,H,eGW,cGW) + + do s=1,nH + order(s) = s + end do + + call quick_sort(eGW,order,nH) + call set_order(cGW,order,nH,nH) + + call wall_time(end_timing) + timing = end_timing - start_timing + + write(*,*) + write(*,'(A65,1X,F9.3,A8)') 'Total CPU time for construction of supermatrix = ',timing,' seconds' + write(*,*) + + !-----------------! + ! Compute weights ! + !-----------------! + + do s=1,nH + Z(s) = cGW(1,s)**2 + end do + + write(*,*)'-------------------------------------------' + write(*,'(1X,A32,I3,A8)')'| G0W0 energies (eV) for orbital',p,' |' + write(*,*)'-------------------------------------------' + write(*,'(1X,A1,1X,A3,1X,A1,1X,A15,1X,A1,1X,A15,1X,A1,1X,A15,1X)') & + '|','#','|','e_QP','|','Z','|' + write(*,*)'-------------------------------------------' + + do s=1,nH +! if(eGW(s) < eF .and. eGW(s) > eF - window) then + if(Z(s) > cutoff1) then + write(*,'(1X,A1,1X,I3,1X,A1,1X,F15.6,1X,A1,1X,F15.6,1X,A1,1X)') & + '|',s,'|',eGW(s)*HaToeV,'|',Z(s),'|' + end if + end do + + write(*,*)'-------------------------------------------' + write(*,*) + + end do ! Loop on the orbital in the e block + +end subroutine diff --git a/src/GW/ufRG0W0.f90 b/src/GW/ufRG0W0.f90 index 6548151..bf89929 100644 --- a/src/GW/ufRG0W0.f90 +++ b/src/GW/ufRG0W0.f90 @@ -33,7 +33,7 @@ subroutine ufRG0W0(dotest,TDA_W,nBas,nOrb,nC,nO,nV,nR,nS,ENuc,ERHF,ERI,eHF) logical :: print_W = .false. logical :: dRPA - integer :: ispin + integer :: isp_W double precision :: EcRPA integer :: n2h1p,n2p1h,nH double precision,external :: Kronecker_delta @@ -90,14 +90,14 @@ subroutine ufRG0W0(dotest,TDA_W,nBas,nOrb,nC,nO,nV,nR,nS,ENuc,ERHF,ERI,eHF) ! Spin manifold - ispin = 1 + isp_W = 1 ! Memory allocation allocate(Om(nS),Aph(nS,nS),Bph(nS,nS),XpY(nS,nS),XmY(nS,nS),rho(nOrb,nOrb,nS)) - call phLR_A(ispin,dRPA,nOrb,nC,nO,nV,nR,nS,1d0,eHF,ERI,Aph) - call phLR_B(ispin,dRPA,nOrb,nC,nO,nV,nR,nS,1d0,ERI,Bph) + call phLR_A(isp_W,dRPA,nOrb,nC,nO,nV,nR,nS,1d0,eHF,ERI,Aph) + call phLR_B(isp_W,dRPA,nOrb,nC,nO,nV,nR,nS,1d0,ERI,Bph) call phLR(TDA_W,nS,Aph,Bph,EcRPA,Om,XpY,XmY)