ViewVC Help

View File | Revision Log | Show Annotations | View Changeset | Root Listing

root/group/trunk/OOPSE-4/src/UseTheForce/DarkSide/electrostatic.F90

Comparing trunk/OOPSE-4/src/UseTheForce/DarkSide/electrostatic.F90 (file contents):
Revision 2309 by chrisfen, Sun Sep 18 20:45:38 2005 UTC vs.
Revision 3013 by chrisfen, Thu Sep 21 18:25:17 2006 UTC

#	Line 47 | Line 47 | module electrostatic_module
47		use vector_class
48		use simulation
49		use status
50	+	use interpolation
51		#ifdef IS_MPI
52		use mpiSimulation
53		#endif
#	Line 54 | Line 55 | module electrostatic_module
55
56		PRIVATE
57
58	+
59		#define __FORTRAN90
60	+	#include "UseTheForce/DarkSide/fInteractionMap.h"
61		#include "UseTheForce/DarkSide/fElectrostaticSummationMethod.h"
62	+	#include "UseTheForce/DarkSide/fElectrostaticScreeningMethod.h"
63
64	+
65		!! these prefactors convert the multipole interactions into kcal / mol
66		!! all were computed assuming distances are measured in angstroms
67		!! Charge-Charge, assuming charges are measured in electrons
#	Line 71 | Line 76 | module electrostatic_module
76		!! This unit is also known affectionately as an esu centi-barn.
77		real(kind=dp), parameter :: pre14 = 69.13373_dp
78
79	<	!! variables to handle different summation methods for long-range electrostatics:
79	>	real(kind=dp), parameter :: zero = 0.0_dp
80	>
81	>	!! conversions for the simulation box dipole moment
82	>	real(kind=dp), parameter :: chargeToC = 1.60217733e-19_dp
83	>	real(kind=dp), parameter :: angstromToM = 1.0e-10_dp
84	>	real(kind=dp), parameter :: debyeToCm = 3.33564095198e-30_dp
85	>
86	>	!! number of points for electrostatic splines
87	>	integer, parameter :: np = 100
88	>
89	>	!! variables to handle different summation methods for long-range
90	>	!! electrostatics:
91		integer, save :: summationMethod = NONE
92	+	integer, save :: screeningMethod = UNDAMPED
93		logical, save :: summationMethodChecked = .false.
94		real(kind=DP), save :: defaultCutoff = 0.0_DP
95	+	real(kind=DP), save :: defaultCutoff2 = 0.0_DP
96		logical, save :: haveDefaultCutoff = .false.
97		real(kind=DP), save :: dampingAlpha = 0.0_DP
98	+	real(kind=DP), save :: alpha2 = 0.0_DP
99	+	real(kind=DP), save :: alpha4 = 0.0_DP
100	+	real(kind=DP), save :: alpha6 = 0.0_DP
101	+	real(kind=DP), save :: alpha8 = 0.0_DP
102		logical, save :: haveDampingAlpha = .false.
103	<	real(kind=DP), save :: dielectric = 0.0_DP
103	>	real(kind=DP), save :: dielectric = 1.0_DP
104		logical, save :: haveDielectric = .false.
83	–	real(kind=DP), save :: constERFC = 0.0_DP
105		real(kind=DP), save :: constEXP = 0.0_DP
106	<	logical, save :: haveDWAconstants = .false.
107	<	real(kind=dp), save :: rcuti = 0.0_dp
108	<	real(kind=dp), save :: rcuti2 = 0.0_dp
109	<	real(kind=dp), save :: rcuti3 = 0.0_dp
110	<	real(kind=dp), save :: rcuti4 = 0.0_dp
106	>	real(kind=dp), save :: rcuti = 0.0_DP
107	>	real(kind=dp), save :: rcuti2 = 0.0_DP
108	>	real(kind=dp), save :: rcuti3 = 0.0_DP
109	>	real(kind=dp), save :: rcuti4 = 0.0_DP
110	>	real(kind=dp), save :: alphaPi = 0.0_DP
111	>	real(kind=dp), save :: invRootPi = 0.0_DP
112	>	real(kind=dp), save :: rrf = 1.0_DP
113	>	real(kind=dp), save :: rt = 1.0_DP
114	>	real(kind=dp), save :: rrfsq = 1.0_DP
115	>	real(kind=dp), save :: preRF = 0.0_DP
116	>	real(kind=dp), save :: preRF2 = 0.0_DP
117	>	real(kind=dp), save :: erfcVal = 1.0_DP
118	>	real(kind=dp), save :: derfcVal = 0.0_DP
119	>	type(cubicSpline), save :: erfcSpline
120	>	logical, save :: haveElectroSpline = .false.
121	>	real(kind=dp), save :: c1 = 1.0_DP
122	>	real(kind=dp), save :: c2 = 1.0_DP
123	>	real(kind=dp), save :: c3 = 0.0_DP
124	>	real(kind=dp), save :: c4 = 0.0_DP
125	>	real(kind=dp), save :: c5 = 0.0_DP
126	>	real(kind=dp), save :: c6 = 0.0_DP
127	>	real(kind=dp), save :: c1c = 1.0_DP
128	>	real(kind=dp), save :: c2c = 1.0_DP
129	>	real(kind=dp), save :: c3c = 0.0_DP
130	>	real(kind=dp), save :: c4c = 0.0_DP
131	>	real(kind=dp), save :: c5c = 0.0_DP
132	>	real(kind=dp), save :: c6c = 0.0_DP
133	>	real(kind=dp), save :: one_third = 1.0_DP / 3.0_DP
134
135	<
135	>	#if defined(__IFC) || defined(__PGI)
136	>	! error function for ifc version > 7.
137	>	real(kind=dp), external :: erfc
138	>	#endif
139	>
140		public :: setElectrostaticSummationMethod
141	+	public :: setScreeningMethod
142		public :: setElectrostaticCutoffRadius
143	<	public :: setDampedWolfAlpha
143	>	public :: setDampingAlpha
144		public :: setReactionFieldDielectric
145	+	public :: buildElectroSpline
146		public :: newElectrostaticType
147		public :: setCharge
148		public :: setDipoleMoment
#	Line 101 | Line 151 | module electrostatic_module
151		public :: doElectrostaticPair
152		public :: getCharge
153		public :: getDipoleMoment
104	–	public :: pre22
154		public :: destroyElectrostaticTypes
155	+	public :: self_self
156	+	public :: rf_self_excludes
157	+	public :: accumulate_box_dipole
158
159		type :: Electrostatic
160		integer :: c_ident
#	Line 119 | Line 171 | contains
171
172		type(Electrostatic), dimension(:), allocatable :: ElectrostaticMap
173
174	+	logical, save :: hasElectrostaticMap
175	+
176		contains
177
178		subroutine setElectrostaticSummationMethod(the_ESM)
#	Line 129 | Line 183 | contains
183		endif
184
185		summationMethod = the_ESM
186	+
187		end subroutine setElectrostaticSummationMethod
188
189	<	subroutine setElectrostaticCutoffRadius(thisRcut)
189	>	subroutine setScreeningMethod(the_SM)
190	>	integer, intent(in) :: the_SM
191	>	screeningMethod = the_SM
192	>	end subroutine setScreeningMethod
193	>
194	>	subroutine setElectrostaticCutoffRadius(thisRcut, thisRsw)
195		real(kind=dp), intent(in) :: thisRcut
196	+	real(kind=dp), intent(in) :: thisRsw
197		defaultCutoff = thisRcut
198	+	defaultCutoff2 = defaultCutoff*defaultCutoff
199	+	rrf = defaultCutoff
200	+	rt = thisRsw
201		haveDefaultCutoff = .true.
202		end subroutine setElectrostaticCutoffRadius
203
204	<	subroutine setDampedWolfAlpha(thisAlpha)
204	>	subroutine setDampingAlpha(thisAlpha)
205		real(kind=dp), intent(in) :: thisAlpha
206		dampingAlpha = thisAlpha
207	+	alpha2 = dampingAlpha*dampingAlpha
208	+	alpha4 = alpha2*alpha2
209	+	alpha6 = alpha4*alpha2
210	+	alpha8 = alpha4*alpha4
211		haveDampingAlpha = .true.
212	<	end subroutine setDampedWolfAlpha
212	>	end subroutine setDampingAlpha
213
214		subroutine setReactionFieldDielectric(thisDielectric)
215		real(kind=dp), intent(in) :: thisDielectric
#	Line 149 | Line 217 | contains
217		haveDielectric = .true.
218		end subroutine setReactionFieldDielectric
219
220	+	subroutine buildElectroSpline()
221	+	real( kind = dp ), dimension(np) :: xvals, yvals
222	+	real( kind = dp ) :: dx, rmin, rval
223	+	integer :: i
224	+
225	+	rmin = 0.0_dp
226	+
227	+	dx = (defaultCutoff-rmin) / dble(np-1)
228	+
229	+	do i = 1, np
230	+	rval = rmin + dble(i-1)*dx
231	+	xvals(i) = rval
232	+	yvals(i) = erfc(dampingAlpha*rval)
233	+	enddo
234	+
235	+	call newSpline(erfcSpline, xvals, yvals, .true.)
236	+
237	+	haveElectroSpline = .true.
238	+	end subroutine buildElectroSpline
239	+
240		subroutine newElectrostaticType(c_ident, is_Charge, is_Dipole, &
241		is_SplitDipole, is_Quadrupole, is_Tap, status)
242
#	Line 176 | Line 264 | contains
264		return
265		end if
266
267	<	if (.not. allocated(ElectrostaticMap)) then
180	<	allocate(ElectrostaticMap(nAtypes))
181	<	endif
267	>	allocate(ElectrostaticMap(nAtypes))
268
269		end if
270
#	Line 196 | Line 282 | contains
282		ElectrostaticMap(myATID)%is_Quadrupole = is_Quadrupole
283		ElectrostaticMap(myATID)%is_Tap = is_Tap
284
285	+	hasElectrostaticMap = .true.
286	+
287		end subroutine newElectrostaticType
288
289		subroutine setCharge(c_ident, charge, status)
#	Line 207 | Line 295 | contains
295		status = 0
296		myATID = getFirstMatchingElement(atypes, "c_ident", c_ident)
297
298	<	if (.not.allocated(ElectrostaticMap)) then
298	>	if (.not.hasElectrostaticMap) then
299		call handleError("electrostatic", "no ElectrostaticMap was present before first call of setCharge!")
300		status = -1
301		return
#	Line 237 | Line 325 | contains
325		status = 0
326		myATID = getFirstMatchingElement(atypes, "c_ident", c_ident)
327
328	<	if (.not.allocated(ElectrostaticMap)) then
328	>	if (.not.hasElectrostaticMap) then
329		call handleError("electrostatic", "no ElectrostaticMap was present before first call of setDipoleMoment!")
330		status = -1
331		return
#	Line 267 | Line 355 | contains
355		status = 0
356		myATID = getFirstMatchingElement(atypes, "c_ident", c_ident)
357
358	<	if (.not.allocated(ElectrostaticMap)) then
358	>	if (.not.hasElectrostaticMap) then
359		call handleError("electrostatic", "no ElectrostaticMap was present before first call of setSplitDipoleDistance!")
360		status = -1
361		return
#	Line 297 | Line 385 | contains
385		status = 0
386		myATID = getFirstMatchingElement(atypes, "c_ident", c_ident)
387
388	<	if (.not.allocated(ElectrostaticMap)) then
388	>	if (.not.hasElectrostaticMap) then
389		call handleError("electrostatic", "no ElectrostaticMap was present before first call of setQuadrupoleMoments!")
390		status = -1
391		return
#	Line 328 | Line 416 | contains
416		integer :: localError
417		real(kind=dp) :: c
418
419	<	if (.not.allocated(ElectrostaticMap)) then
419	>	if (.not.hasElectrostaticMap) then
420		call handleError("electrostatic", "no ElectrostaticMap was present before first call of getCharge!")
421		return
422		end if
#	Line 346 | Line 434 | contains
434		integer :: localError
435		real(kind=dp) :: dm
436
437	<	if (.not.allocated(ElectrostaticMap)) then
437	>	if (.not.hasElectrostaticMap) then
438		call handleError("electrostatic", "no ElectrostaticMap was present before first call of getDipoleMoment!")
439		return
440		end if
#	Line 365 | Line 453 | contains
453		call handleError("checkSummationMethod", "no Default Cutoff set!")
454		endif
455
456	<	rcuti = 1.0d0 / defaultCutoff
456	>	rcuti = 1.0_dp / defaultCutoff
457		rcuti2 = rcuti*rcuti
458		rcuti3 = rcuti2*rcuti
459		rcuti4 = rcuti2*rcuti2
460
461	<	if (summationMethod .eq. DAMPED_WOLF) then
462	<	if (.not.haveDWAconstants) then
463	<
376	<	if (.not.haveDampingAlpha) then
377	<	call handleError("checkSummationMethod", "no Damping Alpha set!")
378	<	endif
379	<
380	<	if (.not.haveDefaultCutoff) then
381	<	call handleError("checkSummationMethod", "no Default Cutoff set!")
382	<	endif
383	<
384	<	constEXP = exp(-dampingAlpha*dampingAlpha*defaultCutoff*defaultCutoff)
385	<	constERFC = erfc(dampingAlpha*defaultCutoff)
386	<
387	<	haveDWAconstants = .true.
461	>	if (screeningMethod .eq. DAMPED) then
462	>	if (.not.haveDampingAlpha) then
463	>	call handleError("checkSummationMethod", "no Damping Alpha set!")
464		endif
465	+
466	+	if (.not.haveDefaultCutoff) then
467	+	call handleError("checkSummationMethod", "no Default Cutoff set!")
468	+	endif
469	+
470	+	constEXP = exp(-alpha2*defaultCutoff2)
471	+	invRootPi = 0.56418958354775628695_dp
472	+	alphaPi = 2.0_dp*dampingAlpha*invRootPi
473	+
474	+	c1c = erfc(dampingAlpha*defaultCutoff) * rcuti
475	+	c2c = alphaPi*constEXP*rcuti + c1c*rcuti
476	+	c3c = 2.0_dp*alphaPi*alpha2 + 3.0_dp*c2c*rcuti
477	+	c4c = 4.0_dp*alphaPi*alpha4 + 5.0_dp*c3c*rcuti2
478	+	c5c = 8.0_dp*alphaPi*alpha6 + 7.0_dp*c4c*rcuti2
479	+	c6c = 16.0_dp*alphaPi*alpha8 + 9.0_dp*c5c*rcuti2
480	+	else
481	+	c1c = rcuti
482	+	c2c = c1c*rcuti
483	+	c3c = 3.0_dp*c2c*rcuti
484	+	c4c = 5.0_dp*c3c*rcuti2
485	+	c5c = 7.0_dp*c4c*rcuti2
486	+	c6c = 9.0_dp*c5c*rcuti2
487		endif
488
489		if (summationMethod .eq. REACTION_FIELD) then
490	<	if (.not.haveDielectric) then
491	<	call handleError("checkSummationMethod", "no reaction field Dielectric set!")
490	>	if (haveDielectric) then
491	>	defaultCutoff2 = defaultCutoff*defaultCutoff
492	>	preRF = (dielectric-1.0_dp) / &
493	>	((2.0_dp*dielectric+1.0_dp)*defaultCutoff2*defaultCutoff)
494	>	preRF2 = 2.0_dp*preRF
495	>	else
496	>	call handleError("checkSummationMethod", "Dielectric not set")
497		endif
498	+
499		endif
500
501	+	if (.not.haveElectroSpline) then
502	+	call buildElectroSpline()
503	+	end if
504	+
505		summationMethodChecked = .true.
506		end subroutine checkSummationMethod
507
508
509	<
402	<	subroutine doElectrostaticPair(atom1, atom2, d, rij, r2, sw, &
509	>	subroutine doElectrostaticPair(atom1, atom2, d, rij, r2, rcut, sw, &
510		vpair, fpair, pot, eFrame, f, t, do_pot)
511
512		logical, intent(in) :: do_pot
#	Line 407 | Line 514 | contains
514		integer, intent(in) :: atom1, atom2
515		integer :: localError
516
517	<	real(kind=dp), intent(in) :: rij, r2, sw
517	>	real(kind=dp), intent(in) :: rij, r2, sw, rcut
518		real(kind=dp), intent(in), dimension(3) :: d
519		real(kind=dp), intent(inout) :: vpair
520	<	real(kind=dp), intent(inout), dimension(3) :: fpair
520	>	real(kind=dp), intent(inout), dimension(3) :: fpair
521
522	<	real( kind = dp ) :: pot, swi
522	>	real( kind = dp ) :: pot
523		real( kind = dp ), dimension(9,nLocal) :: eFrame
524		real( kind = dp ), dimension(3,nLocal) :: f
525	+	real( kind = dp ), dimension(3,nLocal) :: felec
526		real( kind = dp ), dimension(3,nLocal) :: t
527
528		real (kind = dp), dimension(3) :: ux_i, uy_i, uz_i
#	Line 432 | Line 540 | contains
540		real (kind=dp) :: cx_i, cy_i, cz_i
541		real (kind=dp) :: cx_j, cy_j, cz_j
542		real (kind=dp) :: cx2, cy2, cz2
543	<	real (kind=dp) :: ct_i, ct_j, ct_ij, a1
543	>	real (kind=dp) :: ct_i, ct_j, ct_ij, a0, a1
544		real (kind=dp) :: riji, ri, ri2, ri3, ri4
545		real (kind=dp) :: pref, vterm, epot, dudr, vterm1, vterm2
546		real (kind=dp) :: xhat, yhat, zhat
547		real (kind=dp) :: dudx, dudy, dudz
548	<	real (kind=dp) :: scale, sc2, bigR, switcher, dswitcher
548	>	real (kind=dp) :: scale, sc2, bigR
549	>	real (kind=dp) :: varEXP
550	>	real (kind=dp) :: pot_term
551	>	real (kind=dp) :: preVal, rfVal
552	>	real (kind=dp) :: c2ri, c3ri, c4rij
553	>	real (kind=dp) :: cti3, ctj3, ctidotj
554	>	real (kind=dp) :: preSw, preSwSc
555	>	real (kind=dp) :: xhatdot2, yhatdot2, zhatdot2
556	>	real (kind=dp) :: xhatc4, yhatc4, zhatc4
557
442	–	if (.not.allocated(ElectrostaticMap)) then
443	–	call handleError("electrostatic", "no ElectrostaticMap was present before first call of do_electrostatic_pair!")
444	–	return
445	–	end if
446	–
558		if (.not.summationMethodChecked) then
559		call checkSummationMethod()
560		endif
561
451	–
562		#ifdef IS_MPI
563		me1 = atid_Row(atom1)
564		me2 = atid_Col(atom2)
#	Line 459 | Line 569 | contains
569
570		!! some variables we'll need independent of electrostatic type:
571
572	<	riji = 1.0d0 / rij
573	<
572	>	riji = 1.0_dp / rij
573	>
574		xhat = d(1) * riji
575		yhat = d(2) * riji
576		zhat = d(3) * riji
577
468	–	swi = 1.0d0 / sw
469	–
578		!! logicals
579		i_is_Charge = ElectrostaticMap(me1)%is_Charge
580		i_is_Dipole = ElectrostaticMap(me1)%is_Dipole
#	Line 500 | Line 608 | contains
608		if (i_is_SplitDipole) then
609		d_i = ElectrostaticMap(me1)%split_dipole_distance
610		endif
611	<
611	>	duduz_i = zero
612		endif
613
614		if (i_is_Quadrupole) then
#	Line 531 | Line 639 | contains
639		cx_i = ux_i(1)*xhat + ux_i(2)*yhat + ux_i(3)*zhat
640		cy_i = uy_i(1)*xhat + uy_i(2)*yhat + uy_i(3)*zhat
641		cz_i = uz_i(1)*xhat + uz_i(2)*yhat + uz_i(3)*zhat
642	+	dudux_i = zero
643	+	duduy_i = zero
644	+	duduz_i = zero
645		endif
646
647		if (j_is_Charge) then
#	Line 553 | Line 664 | contains
664		if (j_is_SplitDipole) then
665		d_j = ElectrostaticMap(me2)%split_dipole_distance
666		endif
667	+	duduz_j = zero
668		endif
669
670		if (j_is_Quadrupole) then
#	Line 583 | Line 695 | contains
695		cx_j = ux_j(1)*xhat + ux_j(2)*yhat + ux_j(3)*zhat
696		cy_j = uy_j(1)*xhat + uy_j(2)*yhat + uy_j(3)*zhat
697		cz_j = uz_j(1)*xhat + uz_j(2)*yhat + uz_j(3)*zhat
698	+	dudux_j = zero
699	+	duduy_j = zero
700	+	duduz_j = zero
701		endif
702
703	<	!!$    switcher = 1.0d0
704	<	!!$    dswitcher = 0.0d0
705	<	!!$    ebalance = 0.0d0
706	<	!!$    ! weaken the dipole interaction at close range for TAP water
592	<	!!$    if (j_is_Tap .and. i_is_Tap) then
593	<	!!$      call calc_switch(rij, mu_i, switcher, dswitcher)
594	<	!!$    endif
595	<
596	<	epot = 0.0_dp
597	<	dudx = 0.0_dp
598	<	dudy = 0.0_dp
599	<	dudz = 0.0_dp
703	>	epot = zero
704	>	dudx = zero
705	>	dudy = zero
706	>	dudz = zero
707
601	–	dudux_i = 0.0_dp
602	–	duduy_i = 0.0_dp
603	–	duduz_i = 0.0_dp
604	–
605	–	dudux_j = 0.0_dp
606	–	duduy_j = 0.0_dp
607	–	duduz_j = 0.0_dp
608	–
708		if (i_is_Charge) then
709
710		if (j_is_Charge) then
711	+	if (screeningMethod .eq. DAMPED) then
712	+	! assemble the damping variables
713	+	call lookupUniformSpline1d(erfcSpline, rij, erfcVal, derfcVal)
714	+	c1 = erfcVal*riji
715	+	c2 = (-derfcVal + c1)*riji
716	+	else
717	+	c1 = riji
718	+	c2 = c1*riji
719	+	endif
720
721	<	if (summationMethod .eq. 1) then
614	<	vterm = pre11 * q_i * q_j * (riji - rcuti)
721	>	preVal = pre11 * q_i * q_j
722
723	<	vpair = vpair + vterm
724	<	epot = epot + sw * vterm
723	>	if (summationMethod .eq. SHIFTED_POTENTIAL) then
724	>	vterm = preVal * (c1 - c1c)
725
726	<	dudr  = - sw * pre11 * q_i * q_j * (riji*riji*riji - rcuti2*rcuti)
726	>	dudr  = -sw * preVal * c2
727	>
728	>	elseif (summationMethod .eq. SHIFTED_FORCE) then
729	>	vterm = preVal * ( c1 - c1c + c2c*(rij - defaultCutoff) )
730
731	<	dudx = dudx + dudr * d(1)
732	<	dudy = dudy + dudr * d(2)
733	<	dudz = dudz + dudr * d(3)
734	<
731	>	dudr  = sw * preVal * (c2c - c2)
732	>
733	>	elseif (summationMethod .eq. REACTION_FIELD) then
734	>	rfVal = preRF*rij*rij
735	>	vterm = preVal * ( riji + rfVal )
736	>
737	>	dudr  = sw * preVal * ( 2.0_dp*rfVal - riji )*riji
738	>
739		else
740	<	vterm = pre11 * q_i * q_j * riji
627	<
628	<	vpair = vpair + vterm
629	<	epot = epot + sw * vterm
740	>	vterm = preVal * riji*erfcVal
741
742	<	dudr  = - sw * vterm * riji
743	<
633	<	dudx = dudx + dudr * xhat
634	<	dudy = dudy + dudr * yhat
635	<	dudz = dudz + dudr * zhat
636	<
742	>	dudr  = - sw * preVal * c2
743	>
744		endif
745
746	+	vpair = vpair + vterm
747	+	epot = epot + sw*vterm
748	+
749	+	dudx = dudx + dudr * xhat
750	+	dudy = dudy + dudr * yhat
751	+	dudz = dudz + dudr * zhat
752	+
753		endif
754
755		if (j_is_Dipole) then
756	<
757	<	pref = sw * pre12 * q_i * mu_j
756	>	! pref is used by all the possible methods
757	>	pref = pre12 * q_i * mu_j
758	>	preSw = sw*pref
759
760	<	if (summationMethod .eq. 1) then
760	>	if (summationMethod .eq. REACTION_FIELD) then
761		ri2 = riji * riji
762		ri3 = ri2 * riji
763	<
764	<	vterm = - pref * ct_j * (ri2 - rcuti2)
765	<	vpair = vpair + swi*vterm
766	<	epot = epot + vterm
763	>
764	>	vterm = - pref * ct_j * ( ri2 - preRF2*rij )
765	>	vpair = vpair + vterm
766	>	epot = epot + sw*vterm
767
768	<	!! this has a + sign in the () because the rij vector is
769	<	!! r_j - r_i and the charge-dipole potential takes the origin
770	<	!! as the point dipole, which is atom j in this case.
771	<
772	<	dudx = dudx - pref * ( ri3*( uz_j(1) - 3.0d0*ct_j*xhat) &
773	<	- rcuti3*( uz_j(1) - 3.0d0*ct_j*d(1)*rcuti ) )
774	<	dudy = dudy - pref * ( ri3*( uz_j(2) - 3.0d0*ct_j*yhat) &
775	<	- rcuti3*( uz_j(2) - 3.0d0*ct_j*d(2)*rcuti ) )
776	<	dudz = dudz - pref * ( ri3*( uz_j(3) - 3.0d0*ct_j*zhat) &
662	<	- rcuti3*( uz_j(3) - 3.0d0*ct_j*d(3)*rcuti ) )
663	<
664	<	duduz_j(1) = duduz_j(1) - pref*( ri2*xhat - d(1)*rcuti3 )
665	<	duduz_j(2) = duduz_j(2) - pref*( ri2*yhat - d(2)*rcuti3 )
666	<	duduz_j(3) = duduz_j(3) - pref*( ri2*zhat - d(3)*rcuti3 )
768	>	dudx = dudx - preSw*( ri3*(uz_j(1) - 3.0_dp*ct_j*xhat) - &
769	>	preRF2*uz_j(1) )
770	>	dudy = dudy - preSw*( ri3*(uz_j(2) - 3.0_dp*ct_j*yhat) - &
771	>	preRF2*uz_j(2) )
772	>	dudz = dudz - preSw*( ri3*(uz_j(3) - 3.0_dp*ct_j*zhat) - &
773	>	preRF2*uz_j(3) )
774	>	duduz_j(1) = duduz_j(1) - preSw * xhat * ( ri2 - preRF2*rij )
775	>	duduz_j(2) = duduz_j(2) - preSw * yhat * ( ri2 - preRF2*rij )
776	>	duduz_j(3) = duduz_j(3) - preSw * zhat * ( ri2 - preRF2*rij )
777
778		else
779	+	! determine the inverse r used if we have split dipoles
780		if (j_is_SplitDipole) then
781		BigR = sqrt(r2 + 0.25_dp * d_j * d_j)
782		ri = 1.0_dp / BigR
#	Line 674 | Line 785 | contains
785		ri = riji
786		scale = 1.0_dp
787		endif
788	<
678	<	ri2 = ri * ri
679	<	ri3 = ri2 * ri
788	>
789		sc2 = scale * scale
790	+
791	+	if (screeningMethod .eq. DAMPED) then
792	+	! assemble the damping variables
793	+	call lookupUniformSpline1d(erfcSpline, rij, erfcVal, derfcVal)
794	+	c1 = erfcVal*ri
795	+	c2 = (-derfcVal + c1)*ri
796	+	c3 = -2.0_dp*derfcVal*alpha2 + 3.0_dp*c2*ri
797	+	else
798	+	c1 = ri
799	+	c2 = c1*ri
800	+	c3 = 3.0_dp*c2*ri
801	+	endif
802
803	<	vterm = - pref * ct_j * ri2 * scale
804	<	vpair = vpair + swi * vterm
805	<	epot = epot + vterm
803	>	c2ri = c2*ri
804	>
805	>	! calculate the potential
806	>	pot_term =  scale * c2
807	>	vterm = -pref * ct_j * pot_term
808	>	vpair = vpair + vterm
809	>	epot = epot + sw*vterm
810
811	<	!! this has a + sign in the () because the rij vector is
812	<	!! r_j - r_i and the charge-dipole potential takes the origin
813	<	!! as the point dipole, which is atom j in this case.
814	<
815	<	dudx = dudx - pref * ri3 * ( uz_j(1) - 3.0d0*ct_j*xhat*sc2)
816	<	dudy = dudy - pref * ri3 * ( uz_j(2) - 3.0d0*ct_j*yhat*sc2)
817	<	dudz = dudz - pref * ri3 * ( uz_j(3) - 3.0d0*ct_j*zhat*sc2)
818	<
694	<	duduz_j(1) = duduz_j(1) - pref * ri2 * xhat * scale
695	<	duduz_j(2) = duduz_j(2) - pref * ri2 * yhat * scale
696	<	duduz_j(3) = duduz_j(3) - pref * ri2 * zhat * scale
811	>	! calculate derivatives for forces and torques
812	>	dudx = dudx - preSw*( uz_j(1)*c2ri - ct_j*xhat*sc2*c3 )
813	>	dudy = dudy - preSw*( uz_j(2)*c2ri - ct_j*yhat*sc2*c3 )
814	>	dudz = dudz - preSw*( uz_j(3)*c2ri - ct_j*zhat*sc2*c3 )
815	>
816	>	duduz_j(1) = duduz_j(1) - preSw * pot_term * xhat
817	>	duduz_j(2) = duduz_j(2) - preSw * pot_term * yhat
818	>	duduz_j(3) = duduz_j(3) - preSw * pot_term * zhat
819
820		endif
821		endif
822
823		if (j_is_Quadrupole) then
824	<	ri2 = riji * riji
703	<	ri3 = ri2 * riji
704	<	ri4 = ri2 * ri2
824	>	! first precalculate some necessary variables
825		cx2 = cx_j * cx_j
826		cy2 = cy_j * cy_j
827		cz2 = cz_j * cz_j
828	+	pref =  pre14 * q_i * one_third
829	+
830	+	if (screeningMethod .eq. DAMPED) then
831	+	! assemble the damping variables
832	+	call lookupUniformSpline1d(erfcSpline, rij, erfcVal, derfcVal)
833	+	c1 = erfcVal*riji
834	+	c2 = (-derfcVal + c1)*riji
835	+	c3 = -2.0_dp*derfcVal*alpha2 + 3.0_dp*c2*riji
836	+	c4 = -4.0_dp*derfcVal*alpha4 + 5.0_dp*c3*riji*riji
837	+	else
838	+	c1 = riji
839	+	c2 = c1*riji
840	+	c3 = 3.0_dp*c2*riji
841	+	c4 = 5.0_dp*c3*riji*riji
842	+	endif
843
844	+	! precompute variables for convenience
845	+	preSw = sw*pref
846	+	c2ri = c2*riji
847	+	c3ri = c3*riji
848	+	c4rij = c4*rij
849	+	xhatdot2 = 2.0_dp*xhat*c3
850	+	yhatdot2 = 2.0_dp*yhat*c3
851	+	zhatdot2 = 2.0_dp*zhat*c3
852	+	xhatc4 = xhat*c4rij
853	+	yhatc4 = yhat*c4rij
854	+	zhatc4 = zhat*c4rij
855
856	<	pref =  sw * pre14 * q_i / 3.0_dp
856	>	! calculate the potential
857	>	pot_term = ( qxx_j*(cx2*c3 - c2ri) + qyy_j*(cy2*c3 - c2ri) + &
858	>	qzz_j*(cz2*c3 - c2ri) )
859	>	vterm = pref * pot_term
860	>	vpair = vpair + vterm
861	>	epot = epot + sw*vterm
862
863	<	if (summationMethod .eq. 1) then
864	<	vterm1 = pref * ri3*( qxx_j * (3.0_dp*cx2 - 1.0_dp) + &
865	<	qyy_j * (3.0_dp*cy2 - 1.0_dp) + &
866	<	qzz_j * (3.0_dp*cz2 - 1.0_dp) )
867	<	vterm2 = pref * rcuti3*( qxx_j * (3.0_dp*cx2 - 1.0_dp) + &
868	<	qyy_j * (3.0_dp*cy2 - 1.0_dp) + &
869	<	qzz_j * (3.0_dp*cz2 - 1.0_dp) )
870	<	vpair = vpair + swi*( vterm1 - vterm2 )
871	<	epot = epot + ( vterm1 - vterm2 )
872	<
873	<	dudx = dudx - (5.0_dp * &
874	<	(vterm1*riji*xhat - vterm2*rcuti2*d(1))) + pref * ( &
875	<	(ri4 - rcuti4)*(qxx_j*(6.0_dp*cx_j*ux_j(1)) - &
725	<	qxx_j*2.0_dp*(xhat - rcuti*d(1))) + &
726	<	(ri4 - rcuti4)*(qyy_j*(6.0_dp*cy_j*uy_j(1)) - &
727	<	qyy_j*2.0_dp*(xhat - rcuti*d(1))) + &
728	<	(ri4 - rcuti4)*(qzz_j*(6.0_dp*cz_j*uz_j(1)) - &
729	<	qzz_j*2.0_dp*(xhat - rcuti*d(1))) )
730	<	dudy = dudy - (5.0_dp * &
731	<	(vterm1*riji*yhat - vterm2*rcuti2*d(2))) + pref * ( &
732	<	(ri4 - rcuti4)*(qxx_j*(6.0_dp*cx_j*ux_j(2)) - &
733	<	qxx_j*2.0_dp*(yhat - rcuti*d(2))) + &
734	<	(ri4 - rcuti4)*(qyy_j*(6.0_dp*cy_j*uy_j(2)) - &
735	<	qyy_j*2.0_dp*(yhat - rcuti*d(2))) + &
736	<	(ri4 - rcuti4)*(qzz_j*(6.0_dp*cz_j*uz_j(2)) - &
737	<	qzz_j*2.0_dp*(yhat - rcuti*d(2))) )
738	<	dudz = dudz - (5.0_dp * &
739	<	(vterm1*riji*zhat - vterm2*rcuti2*d(3))) + pref * ( &
740	<	(ri4 - rcuti4)*(qxx_j*(6.0_dp*cx_j*ux_j(3)) - &
741	<	qxx_j*2.0_dp*(zhat - rcuti*d(3))) + &
742	<	(ri4 - rcuti4)*(qyy_j*(6.0_dp*cy_j*uy_j(3)) - &
743	<	qyy_j*2.0_dp*(zhat - rcuti*d(3))) + &
744	<	(ri4 - rcuti4)*(qzz_j*(6.0_dp*cz_j*uz_j(3)) - &
745	<	qzz_j*2.0_dp*(zhat - rcuti*d(3))) )
746	<
747	<	dudux_j(1) = dudux_j(1) + pref * (ri3*(qxx_j*6.0_dp*cx_j*xhat) - &
748	<	rcuti4*(qxx_j*6.0_dp*cx_j*d(1)))
749	<	dudux_j(2) = dudux_j(2) + pref * (ri3*(qxx_j*6.0_dp*cx_j*yhat) - &
750	<	rcuti4*(qxx_j*6.0_dp*cx_j*d(2)))
751	<	dudux_j(3) = dudux_j(3) + pref * (ri3*(qxx_j*6.0_dp*cx_j*zhat) - &
752	<	rcuti4*(qxx_j*6.0_dp*cx_j*d(3)))
753	<
754	<	duduy_j(1) = duduy_j(1) + pref * (ri3*(qyy_j*6.0_dp*cy_j*xhat) - &
755	<	rcuti4*(qyy_j*6.0_dp*cx_j*d(1)))
756	<	duduy_j(2) = duduy_j(2) + pref * (ri3*(qyy_j*6.0_dp*cy_j*yhat) - &
757	<	rcuti4*(qyy_j*6.0_dp*cx_j*d(2)))
758	<	duduy_j(3) = duduy_j(3) + pref * (ri3*(qyy_j*6.0_dp*cy_j*zhat) - &
759	<	rcuti4*(qyy_j*6.0_dp*cx_j*d(3)))
760	<
761	<	duduz_j(1) = duduz_j(1) + pref * (ri3*(qzz_j*6.0_dp*cz_j*xhat) - &
762	<	rcuti4*(qzz_j*6.0_dp*cx_j*d(1)))
763	<	duduz_j(2) = duduz_j(2) + pref * (ri3*(qzz_j*6.0_dp*cz_j*yhat) - &
764	<	rcuti4*(qzz_j*6.0_dp*cx_j*d(2)))
765	<	duduz_j(3) = duduz_j(3) + pref * (ri3*(qzz_j*6.0_dp*cz_j*zhat) - &
766	<	rcuti4*(qzz_j*6.0_dp*cx_j*d(3)))
767	<
768	<	else
769	<	vterm = pref * ri3 * (qxx_j * (3.0_dp*cx2 - 1.0_dp) + &
770	<	qyy_j * (3.0_dp*cy2 - 1.0_dp) + &
771	<	qzz_j * (3.0_dp*cz2 - 1.0_dp))
772	<	vpair = vpair + swi * vterm
773	<	epot = epot + vterm
774	<
775	<	dudx = dudx - 5.0_dp*vterm*riji*xhat + pref * ri4 * ( &
776	<	qxx_j*(6.0_dp*cx_j*ux_j(1) - 2.0_dp*xhat) + &
777	<	qyy_j*(6.0_dp*cy_j*uy_j(1) - 2.0_dp*xhat) + &
778	<	qzz_j*(6.0_dp*cz_j*uz_j(1) - 2.0_dp*xhat) )
779	<	dudy = dudy - 5.0_dp*vterm*riji*yhat + pref * ri4 * ( &
780	<	qxx_j*(6.0_dp*cx_j*ux_j(2) - 2.0_dp*yhat) + &
781	<	qyy_j*(6.0_dp*cy_j*uy_j(2) - 2.0_dp*yhat) + &
782	<	qzz_j*(6.0_dp*cz_j*uz_j(2) - 2.0_dp*yhat) )
783	<	dudz = dudz - 5.0_dp*vterm*riji*zhat + pref * ri4 * ( &
784	<	qxx_j*(6.0_dp*cx_j*ux_j(3) - 2.0_dp*zhat) + &
785	<	qyy_j*(6.0_dp*cy_j*uy_j(3) - 2.0_dp*zhat) + &
786	<	qzz_j*(6.0_dp*cz_j*uz_j(3) - 2.0_dp*zhat) )
787	<
788	<	dudux_j(1) = dudux_j(1) + pref * ri3*(qxx_j*6.0_dp*cx_j*xhat)
789	<	dudux_j(2) = dudux_j(2) + pref * ri3*(qxx_j*6.0_dp*cx_j*yhat)
790	<	dudux_j(3) = dudux_j(3) + pref * ri3*(qxx_j*6.0_dp*cx_j*zhat)
791	<
792	<	duduy_j(1) = duduy_j(1) + pref * ri3*(qyy_j*6.0_dp*cy_j*xhat)
793	<	duduy_j(2) = duduy_j(2) + pref * ri3*(qyy_j*6.0_dp*cy_j*yhat)
794	<	duduy_j(3) = duduy_j(3) + pref * ri3*(qyy_j*6.0_dp*cy_j*zhat)
795	<
796	<	duduz_j(1) = duduz_j(1) + pref * ri3*(qzz_j*6.0_dp*cz_j*xhat)
797	<	duduz_j(2) = duduz_j(2) + pref * ri3*(qzz_j*6.0_dp*cz_j*yhat)
798	<	duduz_j(3) = duduz_j(3) + pref * ri3*(qzz_j*6.0_dp*cz_j*zhat)
863	>	! calculate derivatives for the forces and torques
864	>	dudx = dudx - preSw * ( &
865	>	qxx_j*(cx2*xhatc4 - (2.0_dp*cx_j*ux_j(1) + xhat)*c3ri) + &
866	>	qyy_j*(cy2*xhatc4 - (2.0_dp*cy_j*uy_j(1) + xhat)*c3ri) + &
867	>	qzz_j*(cz2*xhatc4 - (2.0_dp*cz_j*uz_j(1) + xhat)*c3ri) )
868	>	dudy = dudy - preSw * ( &
869	>	qxx_j*(cx2*yhatc4 - (2.0_dp*cx_j*ux_j(2) + yhat)*c3ri) + &
870	>	qyy_j*(cy2*yhatc4 - (2.0_dp*cy_j*uy_j(2) + yhat)*c3ri) + &
871	>	qzz_j*(cz2*yhatc4 - (2.0_dp*cz_j*uz_j(2) + yhat)*c3ri) )
872	>	dudz = dudz - preSw * ( &
873	>	qxx_j*(cx2*zhatc4 - (2.0_dp*cx_j*ux_j(3) + zhat)*c3ri) + &
874	>	qyy_j*(cy2*zhatc4 - (2.0_dp*cy_j*uy_j(3) + zhat)*c3ri) + &
875	>	qzz_j*(cz2*zhatc4 - (2.0_dp*cz_j*uz_j(3) + zhat)*c3ri) )
876
877	<	endif
877	>	dudux_j(1) = dudux_j(1) + preSw*(qxx_j*cx_j*xhatdot2)
878	>	dudux_j(2) = dudux_j(2) + preSw*(qxx_j*cx_j*yhatdot2)
879	>	dudux_j(3) = dudux_j(3) + preSw*(qxx_j*cx_j*zhatdot2)
880	>
881	>	duduy_j(1) = duduy_j(1) + preSw*(qyy_j*cy_j*xhatdot2)
882	>	duduy_j(2) = duduy_j(2) + preSw*(qyy_j*cy_j*yhatdot2)
883	>	duduy_j(3) = duduy_j(3) + preSw*(qyy_j*cy_j*zhatdot2)
884	>
885	>	duduz_j(1) = duduz_j(1) + preSw*(qzz_j*cz_j*xhatdot2)
886	>	duduz_j(2) = duduz_j(2) + preSw*(qzz_j*cz_j*yhatdot2)
887	>	duduz_j(3) = duduz_j(3) + preSw*(qzz_j*cz_j*zhatdot2)
888	>
889	>
890		endif
891		endif
892	<
892	>
893		if (i_is_Dipole) then
894
895		if (j_is_Charge) then
896	+	! variables used by all the methods
897	+	pref = pre12 * q_j * mu_i
898	+	preSw = sw*pref
899
900	<	pref = sw * pre12 * q_j * mu_i
900	>	if (summationMethod .eq. REACTION_FIELD) then
901
810	–	if (summationMethod .eq. 1) then
902		ri2 = riji * riji
903		ri3 = ri2 * riji
904
905	<	vterm = pref * ct_i * (ri2 - rcuti2)
906	<	vpair = vpair + swi * vterm
907	<	epot = epot + vterm
905	>	vterm = pref * ct_i * ( ri2 - preRF2*rij )
906	>	vpair = vpair + vterm
907	>	epot = epot + sw*vterm
908
909	<	!! this has a + sign in the () because the rij vector is
910	<	!! r_j - r_i and the charge-dipole potential takes the origin
911	<	!! as the point dipole, which is atom j in this case.
909	>	dudx = dudx + preSw * ( ri3*(uz_i(1) - 3.0_dp*ct_i*xhat) - &
910	>	preRF2*uz_i(1) )
911	>	dudy = dudy + preSw * ( ri3*(uz_i(2) - 3.0_dp*ct_i*yhat) - &
912	>	preRF2*uz_i(2) )
913	>	dudz = dudz + preSw * ( ri3*(uz_i(3) - 3.0_dp*ct_i*zhat) - &
914	>	preRF2*uz_i(3) )
915
916	<	dudx = dudx + pref * ( ri3*( uz_i(1) - 3.0d0*ct_i*xhat) &
917	<	- rcuti3*( uz_i(1) - 3.0d0*ct_i*d(1)*rcuti ) )
918	<	dudy = dudy + pref * ( ri3*( uz_i(2) - 3.0d0*ct_i*yhat) &
825	<	- rcuti3*( uz_i(2) - 3.0d0*ct_i*d(2)*rcuti ) )
826	<	dudz = dudz + pref * ( ri3*( uz_i(3) - 3.0d0*ct_i*zhat) &
827	<	- rcuti3*( uz_i(3) - 3.0d0*ct_i*d(3)*rcuti ) )
828	<
829	<	duduz_i(1) = duduz_i(1) - pref*( ri2*xhat - d(1)*rcuti3 )
830	<	duduz_i(2) = duduz_i(2) - pref*( ri2*yhat - d(2)*rcuti3 )
831	<	duduz_i(3) = duduz_i(3) - pref*( ri2*zhat - d(3)*rcuti3 )
916	>	duduz_i(1) = duduz_i(1) + preSw * xhat * ( ri2 - preRF2*rij )
917	>	duduz_i(2) = duduz_i(2) + preSw * yhat * ( ri2 - preRF2*rij )
918	>	duduz_i(3) = duduz_i(3) + preSw * zhat * ( ri2 - preRF2*rij )
919
920		else
921	+	! determine inverse r if we are using split dipoles
922		if (i_is_SplitDipole) then
923		BigR = sqrt(r2 + 0.25_dp * d_i * d_i)
924		ri = 1.0_dp / BigR
#	Line 839 | Line 927 | contains
927		ri = riji
928		scale = 1.0_dp
929		endif
930	<
843	<	ri2 = ri * ri
844	<	ri3 = ri2 * ri
930	>
931		sc2 = scale * scale
932	<
933	<	vterm = pref * ct_i * ri2 * scale
934	<	vpair = vpair + swi * vterm
935	<	epot = epot + vterm
936	<
937	<	dudx = dudx + pref * ri3 * ( uz_i(1) - 3.0d0 * ct_i * xhat*sc2)
938	<	dudy = dudy + pref * ri3 * ( uz_i(2) - 3.0d0 * ct_i * yhat*sc2)
939	<	dudz = dudz + pref * ri3 * ( uz_i(3) - 3.0d0 * ct_i * zhat*sc2)
932	>
933	>	if (screeningMethod .eq. DAMPED) then
934	>	! assemble the damping variables
935	>	call lookupUniformSpline1d(erfcSpline, rij, erfcVal, derfcVal)
936	>	c1 = erfcVal*ri
937	>	c2 = (-derfcVal + c1)*ri
938	>	c3 = -2.0_dp*derfcVal*alpha2 + 3.0_dp*c2*ri
939	>	else
940	>	c1 = ri
941	>	c2 = c1*ri
942	>	c3 = 3.0_dp*c2*ri
943	>	endif
944	>
945	>	c2ri = c2*ri
946	>
947	>	! calculate the potential
948	>	pot_term = c2 * scale
949	>	vterm = pref * ct_i * pot_term
950	>	vpair = vpair + vterm
951	>	epot = epot + sw*vterm
952	>
953	>	! calculate derivatives for the forces and torques
954	>	dudx = dudx + preSw * ( uz_i(1)*c2ri - ct_i*xhat*sc2*c3 )
955	>	dudy = dudy + preSw * ( uz_i(2)*c2ri - ct_i*yhat*sc2*c3 )
956	>	dudz = dudz + preSw * ( uz_i(3)*c2ri - ct_i*zhat*sc2*c3 )
957	>
958	>	duduz_i(1) = duduz_i(1) + preSw * pot_term * xhat
959	>	duduz_i(2) = duduz_i(2) + preSw * pot_term * yhat
960	>	duduz_i(3) = duduz_i(3) + preSw * pot_term * zhat
961
855	–	duduz_i(1) = duduz_i(1) + pref * ri2 * xhat * scale
856	–	duduz_i(2) = duduz_i(2) + pref * ri2 * yhat * scale
857	–	duduz_i(3) = duduz_i(3) + pref * ri2 * zhat * scale
962		endif
963		endif
964	<
964	>
965		if (j_is_Dipole) then
966	+	! variables used by all methods
967	+	ct_ij = uz_i(1)*uz_j(1) + uz_i(2)*uz_j(2) + uz_i(3)*uz_j(3)
968	+	pref = pre22 * mu_i * mu_j
969	+	preSw = sw*pref
970
971	<	pref = sw * pre22 * mu_i * mu_j
864	<
865	<	if (summationMethod .eq. 1) then
971	>	if (summationMethod .eq. REACTION_FIELD) then
972		ri2 = riji * riji
973		ri3 = ri2 * riji
974		ri4 = ri2 * ri2
975
976	<	vterm = pref * (ri3 - rcuti3) * (ct_ij - 3.0d0 * ct_i * ct_j)
977	<	vpair = vpair + swi * vterm
978	<	epot = epot + vterm
976	>	vterm = pref*( ri3*(ct_ij - 3.0_dp * ct_i * ct_j) - &
977	>	preRF2*ct_ij )
978	>	vpair = vpair + vterm
979	>	epot = epot + sw*vterm
980
981	<	a1 = 5.0d0 * ct_i * ct_j - ct_ij
981	>	a1 = 5.0_dp * ct_i * ct_j - ct_ij
982
983	<	dudx = dudx + pref*3.0d0*ri4 &
984	<	*(a1*xhat-ct_i*uz_j(1)-ct_j*uz_i(1)) - &
985	<	pref*3.0d0*rcuti4*(a1*rcuti*d(1)-ct_i*uz_j(1)-ct_j*uz_i(1))
879	<	dudy = dudy + pref*3.0d0*ri4 &
880	<	*(a1*yhat-ct_i*uz_j(2)-ct_j*uz_i(2)) - &
881	<	pref*3.0d0*rcuti4*(a1*rcuti*d(2)-ct_i*uz_j(2)-ct_j*uz_i(2))
882	<	dudz = dudz + pref*3.0d0*ri4 &
883	<	*(a1*zhat-ct_i*uz_j(3)-ct_j*uz_i(3)) - &
884	<	pref*3.0d0*rcuti4*(a1*rcuti*d(3)-ct_i*uz_j(3)-ct_j*uz_i(3))
983	>	dudx = dudx + preSw*3.0_dp*ri4*(a1*xhat-ct_i*uz_j(1)-ct_j*uz_i(1))
984	>	dudy = dudy + preSw*3.0_dp*ri4*(a1*yhat-ct_i*uz_j(2)-ct_j*uz_i(2))
985	>	dudz = dudz + preSw*3.0_dp*ri4*(a1*zhat-ct_i*uz_j(3)-ct_j*uz_i(3))
986
987	<	duduz_i(1) = duduz_i(1) + pref*(ri3*(uz_j(1) - 3.0d0*ct_j*xhat) &
988	<	- rcuti3*(uz_j(1) - 3.0d0*ct_j*d(1)*rcuti))
989	<	duduz_i(2) = duduz_i(2) + pref*(ri3*(uz_j(2) - 3.0d0*ct_j*yhat) &
990	<	- rcuti3*(uz_j(2) - 3.0d0*ct_j*d(2)*rcuti))
991	<	duduz_i(3) = duduz_i(3) + pref*(ri3*(uz_j(3) - 3.0d0*ct_j*zhat) &
992	<	- rcuti3*(uz_j(3) - 3.0d0*ct_j*d(3)*rcuti))
993	<	duduz_j(1) = duduz_j(1) + pref*(ri3*(uz_i(1) - 3.0d0*ct_i*xhat) &
994	<	- rcuti3*(uz_i(1) - 3.0d0*ct_i*d(1)*rcuti))
995	<	duduz_j(2) = duduz_j(2) + pref*(ri3*(uz_i(2) - 3.0d0*ct_i*yhat) &
996	<	- rcuti3*(uz_i(2) - 3.0d0*ct_i*d(2)*rcuti))
997	<	duduz_j(3) = duduz_j(3) + pref*(ri3*(uz_i(3) - 3.0d0*ct_i*zhat) &
998	<	- rcuti3*(uz_i(3) - 3.0d0*ct_i*d(3)*rcuti))
987	>	duduz_i(1) = duduz_i(1) + preSw*(ri3*(uz_j(1)-3.0_dp*ct_j*xhat) &
988	>	- preRF2*uz_j(1))
989	>	duduz_i(2) = duduz_i(2) + preSw*(ri3*(uz_j(2)-3.0_dp*ct_j*yhat) &
990	>	- preRF2*uz_j(2))
991	>	duduz_i(3) = duduz_i(3) + preSw*(ri3*(uz_j(3)-3.0_dp*ct_j*zhat) &
992	>	- preRF2*uz_j(3))
993	>	duduz_j(1) = duduz_j(1) + preSw*(ri3*(uz_i(1)-3.0_dp*ct_i*xhat) &
994	>	- preRF2*uz_i(1))
995	>	duduz_j(2) = duduz_j(2) + preSw*(ri3*(uz_i(2)-3.0_dp*ct_i*yhat) &
996	>	- preRF2*uz_i(2))
997	>	duduz_j(3) = duduz_j(3) + preSw*(ri3*(uz_i(3)-3.0_dp*ct_i*zhat) &
998	>	- preRF2*uz_i(3))
999	>
1000		else
899	–
1001		if (i_is_SplitDipole) then
1002		if (j_is_SplitDipole) then
1003		BigR = sqrt(r2 + 0.25_dp * d_i * d_i + 0.25_dp * d_j * d_j)
#	Line 915 | Line 1016 | contains
1016		scale = 1.0_dp
1017		endif
1018		endif
1019	<
1020	<	ct_ij = uz_i(1)*uz_j(1) + uz_i(2)*uz_j(2) + uz_i(3)*uz_j(3)
1021	<
1022	<	ri2 = ri * ri
1023	<	ri3 = ri2 * ri
1024	<	ri4 = ri2 * ri2
1019	>
1020	>	if (screeningMethod .eq. DAMPED) then
1021	>	! assemble the damping variables
1022	>	call lookupUniformSpline1d(erfcSpline, rij, erfcVal, derfcVal)
1023	>	c1 = erfcVal*ri
1024	>	c2 = (-derfcVal + c1)*ri
1025	>	c3 = -2.0_dp*derfcVal*alpha2 + 3.0_dp*c2*ri
1026	>	c4 = -4.0_dp*derfcVal*alpha4 + 5.0_dp*c3*ri*ri
1027	>	else
1028	>	c1 = ri
1029	>	c2 = c1*ri
1030	>	c3 = 3.0_dp*c2*ri
1031	>	c4 = 5.0_dp*c3*ri*ri
1032	>	endif
1033	>
1034	>	! precompute variables for convenience
1035		sc2 = scale * scale
1036	+	cti3 = ct_i*sc2*c3
1037	+	ctj3 = ct_j*sc2*c3
1038	+	ctidotj = ct_i * ct_j * sc2
1039	+	preSwSc = preSw*scale
1040	+	c2ri = c2*ri
1041	+	c3ri = c3*ri
1042	+	c4rij = c4*rij
1043	+
1044	+
1045	+	! calculate the potential
1046	+	pot_term = (ct_ij*c2ri - ctidotj*c3)
1047	+	vterm = pref * pot_term
1048	+	vpair = vpair + vterm
1049	+	epot = epot + sw*vterm
1050	+
1051	+	! calculate derivatives for the forces and torques
1052	+	dudx = dudx + preSwSc * ( ctidotj*xhat*c4rij - &
1053	+	(ct_i*uz_j(1) + ct_j*uz_i(1) + ct_ij*xhat)*c3ri )
1054	+	dudy = dudy + preSwSc * ( ctidotj*yhat*c4rij - &
1055	+	(ct_i*uz_j(2) + ct_j*uz_i(2) + ct_ij*yhat)*c3ri )
1056	+	dudz = dudz + preSwSc * ( ctidotj*zhat*c4rij - &
1057	+	(ct_i*uz_j(3) + ct_j*uz_i(3) + ct_ij*zhat)*c3ri )
1058	+
1059	+	duduz_i(1) = duduz_i(1) + preSw * ( uz_j(1)*c2ri - ctj3*xhat )
1060	+	duduz_i(2) = duduz_i(2) + preSw * ( uz_j(2)*c2ri - ctj3*yhat )
1061	+	duduz_i(3) = duduz_i(3) + preSw * ( uz_j(3)*c2ri - ctj3*zhat )
1062
1063	<	vterm = pref * ri3 * (ct_ij - 3.0d0 * ct_i * ct_j * sc2)
1064	<	vpair = vpair + swi * vterm
1065	<	epot = epot + vterm
1066	<
930	<	a1 = 5.0d0 * ct_i * ct_j * sc2 - ct_ij
931	<
932	<	dudx = dudx + pref*3.0d0*ri4*scale &
933	<	*(a1*xhat-ct_i*uz_j(1)-ct_j*uz_i(1))
934	<	dudy = dudy + pref*3.0d0*ri4*scale &
935	<	*(a1*yhat-ct_i*uz_j(2)-ct_j*uz_i(2))
936	<	dudz = dudz + pref*3.0d0*ri4*scale &
937	<	*(a1*zhat-ct_i*uz_j(3)-ct_j*uz_i(3))
938	<
939	<	duduz_i(1) = duduz_i(1) + pref*ri3 &
940	<	*(uz_j(1) - 3.0d0*ct_j*xhat*sc2)
941	<	duduz_i(2) = duduz_i(2) + pref*ri3 &
942	<	*(uz_j(2) - 3.0d0*ct_j*yhat*sc2)
943	<	duduz_i(3) = duduz_i(3) + pref*ri3 &
944	<	*(uz_j(3) - 3.0d0*ct_j*zhat*sc2)
945	<
946	<	duduz_j(1) = duduz_j(1) + pref*ri3 &
947	<	*(uz_i(1) - 3.0d0*ct_i*xhat*sc2)
948	<	duduz_j(2) = duduz_j(2) + pref*ri3 &
949	<	*(uz_i(2) - 3.0d0*ct_i*yhat*sc2)
950	<	duduz_j(3) = duduz_j(3) + pref*ri3 &
951	<	*(uz_i(3) - 3.0d0*ct_i*zhat*sc2)
1063	>	duduz_j(1) = duduz_j(1) + preSw * ( uz_i(1)*c2ri - cti3*xhat )
1064	>	duduz_j(2) = duduz_j(2) + preSw * ( uz_i(2)*c2ri - cti3*yhat )
1065	>	duduz_j(3) = duduz_j(3) + preSw * ( uz_i(3)*c2ri - cti3*zhat )
1066	>
1067		endif
1068		endif
1069		endif
1070
1071		if (i_is_Quadrupole) then
1072		if (j_is_Charge) then
1073	<
959	<	ri2 = riji * riji
960	<	ri3 = ri2 * riji
961	<	ri4 = ri2 * ri2
1073	>	! precompute some necessary variables
1074		cx2 = cx_i * cx_i
1075		cy2 = cy_i * cy_i
1076		cz2 = cz_i * cz_i
1077	+	pref = pre14 * q_j * one_third
1078
1079	<	pref = sw * pre14 * q_j / 3.0_dp
1080	<
1081	<	if (summationMethod .eq. 1) then
1082	<	vterm1 = pref * ri3*( qxx_i * (3.0_dp*cx2 - 1.0_dp) + &
1083	<	qyy_i * (3.0_dp*cy2 - 1.0_dp) + &
1084	<	qzz_i * (3.0_dp*cz2 - 1.0_dp) )
1085	<	vterm2 = pref * rcuti3*( qxx_i * (3.0_dp*cx2 - 1.0_dp) + &
973	<	qyy_i * (3.0_dp*cy2 - 1.0_dp) + &
974	<	qzz_i * (3.0_dp*cz2 - 1.0_dp) )
975	<	vpair = vpair + swi * ( vterm1 - vterm2 )
976	<	epot = epot + ( vterm1 - vterm2 )
977	<
978	<	dudx = dudx - (5.0_dp*(vterm1*riji*xhat - vterm2*rcuti2*d(1))) + &
979	<	pref * ( (ri4 - rcuti4)*(qxx_i*(6.0_dp*cx_i*ux_i(1)) - &
980	<	qxx_i*2.0_dp*(xhat - rcuti*d(1))) + &
981	<	(ri4 - rcuti4)*(qyy_i*(6.0_dp*cy_i*uy_i(1)) - &
982	<	qyy_i*2.0_dp*(xhat - rcuti*d(1))) + &
983	<	(ri4 - rcuti4)*(qzz_i*(6.0_dp*cz_i*uz_i(1)) - &
984	<	qzz_i*2.0_dp*(xhat - rcuti*d(1))) )
985	<	dudy = dudy - (5.0_dp*(vterm1*riji*yhat - vterm2*rcuti2*d(2))) + &
986	<	pref * ( (ri4 - rcuti4)*(qxx_i*(6.0_dp*cx_i*ux_i(2)) - &
987	<	qxx_i*2.0_dp*(yhat - rcuti*d(2))) + &
988	<	(ri4 - rcuti4)*(qyy_i*(6.0_dp*cy_i*uy_i(2)) - &
989	<	qyy_i*2.0_dp*(yhat - rcuti*d(2))) + &
990	<	(ri4 - rcuti4)*(qzz_i*(6.0_dp*cz_i*uz_i(2)) - &
991	<	qzz_i*2.0_dp*(yhat - rcuti*d(2))) )
992	<	dudz = dudz - (5.0_dp*(vterm1*riji*zhat - vterm2*rcuti2*d(3))) + &
993	<	pref * ( (ri4 - rcuti4)*(qxx_i*(6.0_dp*cx_i*ux_i(3)) - &
994	<	qxx_i*2.0_dp*(zhat - rcuti*d(3))) + &
995	<	(ri4 - rcuti4)*(qyy_i*(6.0_dp*cy_i*uy_i(3)) - &
996	<	qyy_i*2.0_dp*(zhat - rcuti*d(3))) + &
997	<	(ri4 - rcuti4)*(qzz_i*(6.0_dp*cz_i*uz_i(3)) - &
998	<	qzz_i*2.0_dp*(zhat - rcuti*d(3))) )
999	<
1000	<	dudux_i(1) = dudux_i(1) + pref * (ri3*(qxx_i*6.0_dp*cx_i*xhat) - &
1001	<	rcuti4*(qxx_i*6.0_dp*cx_i*d(1)))
1002	<	dudux_i(2) = dudux_i(2) + pref * (ri3*(qxx_i*6.0_dp*cx_i*yhat) - &
1003	<	rcuti4*(qxx_i*6.0_dp*cx_i*d(2)))
1004	<	dudux_i(3) = dudux_i(3) + pref * (ri3*(qxx_i*6.0_dp*cx_i*zhat) - &
1005	<	rcuti4*(qxx_i*6.0_dp*cx_i*d(3)))
1006	<
1007	<	duduy_i(1) = duduy_i(1) + pref * (ri3*(qyy_i*6.0_dp*cy_i*xhat) - &
1008	<	rcuti4*(qyy_i*6.0_dp*cx_i*d(1)))
1009	<	duduy_i(2) = duduy_i(2) + pref * (ri3*(qyy_i*6.0_dp*cy_i*yhat) - &
1010	<	rcuti4*(qyy_i*6.0_dp*cx_i*d(2)))
1011	<	duduy_i(3) = duduy_i(3) + pref * (ri3*(qyy_i*6.0_dp*cy_i*zhat) - &
1012	<	rcuti4*(qyy_i*6.0_dp*cx_i*d(3)))
1013	<
1014	<	duduz_i(1) = duduz_i(1) + pref * (ri3*(qzz_i*6.0_dp*cz_i*xhat) - &
1015	<	rcuti4*(qzz_i*6.0_dp*cx_i*d(1)))
1016	<	duduz_i(2) = duduz_i(2) + pref * (ri3*(qzz_i*6.0_dp*cz_i*yhat) - &
1017	<	rcuti4*(qzz_i*6.0_dp*cx_i*d(2)))
1018	<	duduz_i(3) = duduz_i(3) + pref * (ri3*(qzz_i*6.0_dp*cz_i*zhat) - &
1019	<	rcuti4*(qzz_i*6.0_dp*cx_i*d(3)))
1020	<
1079	>	if (screeningMethod .eq. DAMPED) then
1080	>	! assemble the damping variables
1081	>	call lookupUniformSpline1d(erfcSpline, rij, erfcVal, derfcVal)
1082	>	c1 = erfcVal*riji
1083	>	c2 = (-derfcVal + c1)*riji
1084	>	c3 = -2.0_dp*derfcVal*alpha2 + 3.0_dp*c2*riji
1085	>	c4 = -4.0_dp*derfcVal*alpha4 + 5.0_dp*c3*riji*riji
1086		else
1087	<	vterm = pref * ri3 * (qxx_i * (3.0_dp*cx2 - 1.0_dp) + &
1088	<	qyy_i * (3.0_dp*cy2 - 1.0_dp) + &
1089	<	qzz_i * (3.0_dp*cz2 - 1.0_dp))
1090	<	vpair = vpair + swi * vterm
1026	<	epot = epot + vterm
1027	<
1028	<	dudx = dudx - 5.0_dp*vterm*riji*xhat + pref * ri4 * ( &
1029	<	qxx_i*(6.0_dp*cx_i*ux_i(1) - 2.0_dp*xhat) + &
1030	<	qyy_i*(6.0_dp*cy_i*uy_i(1) - 2.0_dp*xhat) + &
1031	<	qzz_i*(6.0_dp*cz_i*uz_i(1) - 2.0_dp*xhat) )
1032	<	dudy = dudy - 5.0_dp*vterm*riji*yhat + pref * ri4 * ( &
1033	<	qxx_i*(6.0_dp*cx_i*ux_i(2) - 2.0_dp*yhat) + &
1034	<	qyy_i*(6.0_dp*cy_i*uy_i(2) - 2.0_dp*yhat) + &
1035	<	qzz_i*(6.0_dp*cz_i*uz_i(2) - 2.0_dp*yhat) )
1036	<	dudz = dudz - 5.0_dp*vterm*riji*zhat + pref * ri4 * ( &
1037	<	qxx_i*(6.0_dp*cx_i*ux_i(3) - 2.0_dp*zhat) + &
1038	<	qyy_i*(6.0_dp*cy_i*uy_i(3) - 2.0_dp*zhat) + &
1039	<	qzz_i*(6.0_dp*cz_i*uz_i(3) - 2.0_dp*zhat) )
1040	<
1041	<	dudux_i(1) = dudux_i(1) + pref * ri3*(qxx_i*6.0_dp*cx_i*xhat)
1042	<	dudux_i(2) = dudux_i(2) + pref * ri3*(qxx_i*6.0_dp*cx_i*yhat)
1043	<	dudux_i(3) = dudux_i(3) + pref * ri3*(qxx_i*6.0_dp*cx_i*zhat)
1044	<
1045	<	duduy_i(1) = duduy_i(1) + pref * ri3*(qyy_i*6.0_dp*cy_i*xhat)
1046	<	duduy_i(2) = duduy_i(2) + pref * ri3*(qyy_i*6.0_dp*cy_i*yhat)
1047	<	duduy_i(3) = duduy_i(3) + pref * ri3*(qyy_i*6.0_dp*cy_i*zhat)
1048	<
1049	<	duduz_i(1) = duduz_i(1) + pref * ri3*(qzz_i*6.0_dp*cz_i*xhat)
1050	<	duduz_i(2) = duduz_i(2) + pref * ri3*(qzz_i*6.0_dp*cz_i*yhat)
1051	<	duduz_i(3) = duduz_i(3) + pref * ri3*(qzz_i*6.0_dp*cz_i*zhat)
1087	>	c1 = riji
1088	>	c2 = c1*riji
1089	>	c3 = 3.0_dp*c2*riji
1090	>	c4 = 5.0_dp*c3*riji*riji
1091		endif
1092	+
1093	+	! precompute some variables for convenience
1094	+	preSw = sw*pref
1095	+	c2ri = c2*riji
1096	+	c3ri = c3*riji
1097	+	c4rij = c4*rij
1098	+	xhatdot2 = 2.0_dp*xhat*c3
1099	+	yhatdot2 = 2.0_dp*yhat*c3
1100	+	zhatdot2 = 2.0_dp*zhat*c3
1101	+	xhatc4 = xhat*c4rij
1102	+	yhatc4 = yhat*c4rij
1103	+	zhatc4 = zhat*c4rij
1104	+
1105	+	! calculate the potential
1106	+	pot_term = ( qxx_i * (cx2*c3 - c2ri) + qyy_i * (cy2*c3 - c2ri) + &
1107	+	qzz_i * (cz2*c3 - c2ri) )
1108	+
1109	+	vterm = pref * pot_term
1110	+	vpair = vpair + vterm
1111	+	epot = epot + sw*vterm
1112	+
1113	+	! calculate the derivatives for the forces and torques
1114	+	dudx = dudx - preSw * ( &
1115	+	qxx_i*(cx2*xhatc4 - (2.0_dp*cx_i*ux_i(1) + xhat)*c3ri) + &
1116	+	qyy_i*(cy2*xhatc4 - (2.0_dp*cy_i*uy_i(1) + xhat)*c3ri) + &
1117	+	qzz_i*(cz2*xhatc4 - (2.0_dp*cz_i*uz_i(1) + xhat)*c3ri) )
1118	+	dudy = dudy - preSw * ( &
1119	+	qxx_i*(cx2*yhatc4 - (2.0_dp*cx_i*ux_i(2) + yhat)*c3ri) + &
1120	+	qyy_i*(cy2*yhatc4 - (2.0_dp*cy_i*uy_i(2) + yhat)*c3ri) + &
1121	+	qzz_i*(cz2*yhatc4 - (2.0_dp*cz_i*uz_i(2) + yhat)*c3ri) )
1122	+	dudz = dudz - preSw * ( &
1123	+	qxx_i*(cx2*zhatc4 - (2.0_dp*cx_i*ux_i(3) + zhat)*c3ri) + &
1124	+	qyy_i*(cy2*zhatc4 - (2.0_dp*cy_i*uy_i(3) + zhat)*c3ri) + &
1125	+	qzz_i*(cz2*zhatc4 - (2.0_dp*cz_i*uz_i(3) + zhat)*c3ri) )
1126	+
1127	+	dudux_i(1) = dudux_i(1) + preSw*(qxx_i*cx_i*xhatdot2)
1128	+	dudux_i(2) = dudux_i(2) + preSw*(qxx_i*cx_i*yhatdot2)
1129	+	dudux_i(3) = dudux_i(3) + preSw*(qxx_i*cx_i*zhatdot2)
1130	+
1131	+	duduy_i(1) = duduy_i(1) + preSw*(qyy_i*cy_i*xhatdot2)
1132	+	duduy_i(2) = duduy_i(2) + preSw*(qyy_i*cy_i*yhatdot2)
1133	+	duduy_i(3) = duduy_i(3) + preSw*(qyy_i*cy_i*zhatdot2)
1134	+
1135	+	duduz_i(1) = duduz_i(1) + preSw*(qzz_i*cz_i*xhatdot2)
1136	+	duduz_i(2) = duduz_i(2) + preSw*(qzz_i*cz_i*yhatdot2)
1137	+	duduz_i(3) = duduz_i(3) + preSw*(qzz_i*cz_i*zhatdot2)
1138		endif
1139		endif
1140
1141
1142		if (do_pot) then
1143		#ifdef IS_MPI
1144	<	pot_row(atom1) = pot_row(atom1) + 0.5d0*epot
1145	<	pot_col(atom2) = pot_col(atom2) + 0.5d0*epot
1144	>	pot_row(ELECTROSTATIC_POT,atom1) = pot_row(ELECTROSTATIC_POT,atom1) + 0.5_dp*epot
1145	>	pot_col(ELECTROSTATIC_POT,atom2) = pot_col(ELECTROSTATIC_POT,atom2) + 0.5_dp*epot
1146		#else
1147		pot = pot + epot
1148		#endif
#	Line 1161 | Line 1246 | contains
1246
1247		return
1248		end subroutine doElectrostaticPair
1164	–
1165	–	!! calculates the switching functions and their derivatives for a given
1166	–	subroutine calc_switch(r, mu, scale, dscale)
1249
1250	<	real (kind=dp), intent(in) :: r, mu
1169	<	real (kind=dp), intent(inout) :: scale, dscale
1170	<	real (kind=dp) :: rl, ru, mulow, minRatio, temp, scaleVal
1250	>	subroutine destroyElectrostaticTypes()
1251
1252	<	! distances must be in angstroms
1253	<	rl = 2.75d0
1254	<	ru = 3.75d0
1255	<	mulow = 0.0d0 !3.3856d0 ! 1.84 * 1.84
1256	<	minRatio = mulow / (mu*mu)
1257	<	scaleVal = 1.0d0 - minRatio
1252	>	if(allocated(ElectrostaticMap)) deallocate(ElectrostaticMap)
1253	>
1254	>	end subroutine destroyElectrostaticTypes
1255	>
1256	>	subroutine self_self(atom1, eFrame, mypot, t, do_pot)
1257	>	logical, intent(in) :: do_pot
1258	>	integer, intent(in) :: atom1
1259	>	integer :: atid1
1260	>	real(kind=dp), dimension(9,nLocal) :: eFrame
1261	>	real(kind=dp), dimension(3,nLocal) :: t
1262	>	real(kind=dp) :: mu1, chg1
1263	>	real(kind=dp) :: preVal, epot, mypot
1264	>	real(kind=dp) :: eix, eiy, eiz
1265	>
1266	>	! this is a local only array, so we use the local atom type id's:
1267	>	atid1 = atid(atom1)
1268	>
1269	>	if (.not.summationMethodChecked) then
1270	>	call checkSummationMethod()
1271	>	endif
1272
1273	<	if (r.lt.rl) then
1274	<	scale = minRatio
1275	<	dscale = 0.0d0
1276	<	elseif (r.gt.ru) then
1277	<	scale = 1.0d0
1278	<	dscale = 0.0d0
1279	<	else
1280	<	scale = 1.0d0 - scaleVal*((ru + 2.0d0*r - 3.0d0*rl) * (ru-r)**2) &
1281	<	/ ((ru - rl)**3)
1282	<	dscale = -scaleVal * 6.0d0 * (r-ru)*(r-rl)/((ru - rl)**3)
1273	>	if (summationMethod .eq. REACTION_FIELD) then
1274	>	if (ElectrostaticMap(atid1)%is_Dipole) then
1275	>	mu1 = getDipoleMoment(atid1)
1276	>
1277	>	preVal = pre22 * preRF2 * mu1*mu1
1278	>	mypot = mypot - 0.5_dp*preVal
1279	>
1280	>	! The self-correction term adds into the reaction field vector
1281	>
1282	>	eix = preVal * eFrame(3,atom1)
1283	>	eiy = preVal * eFrame(6,atom1)
1284	>	eiz = preVal * eFrame(9,atom1)
1285	>
1286	>	! once again, this is self-self, so only the local arrays are needed
1287	>	! even for MPI jobs:
1288	>
1289	>	t(1,atom1)=t(1,atom1) - eFrame(6,atom1)*eiz + &
1290	>	eFrame(9,atom1)*eiy
1291	>	t(2,atom1)=t(2,atom1) - eFrame(9,atom1)*eix + &
1292	>	eFrame(3,atom1)*eiz
1293	>	t(3,atom1)=t(3,atom1) - eFrame(3,atom1)*eiy + &
1294	>	eFrame(6,atom1)*eix
1295	>
1296	>	endif
1297	>
1298	>	elseif ( (summationMethod .eq. SHIFTED_FORCE) .or. &
1299	>	(summationMethod .eq. SHIFTED_POTENTIAL) ) then
1300	>	if (ElectrostaticMap(atid1)%is_Charge) then
1301	>	chg1 = getCharge(atid1)
1302	>
1303	>	if (screeningMethod .eq. DAMPED) then
1304	>	mypot = mypot - (c1c + alphaPi) * chg1 * chg1
1305	>
1306	>	else
1307	>	mypot = mypot - (rcuti * chg1 * chg1)
1308	>
1309	>	endif
1310	>	endif
1311		endif
1312	<
1312	>
1313		return
1314	<	end subroutine calc_switch
1314	>	end subroutine self_self
1315
1316	<	subroutine destroyElectrostaticTypes()
1316	>	subroutine rf_self_excludes(atom1, atom2, sw, eFrame, d, rij, vpair, myPot, &
1317	>	f, t, do_pot)
1318	>	logical, intent(in) :: do_pot
1319	>	integer, intent(in) :: atom1
1320	>	integer, intent(in) :: atom2
1321	>	logical :: i_is_Charge, j_is_Charge
1322	>	logical :: i_is_Dipole, j_is_Dipole
1323	>	integer :: atid1
1324	>	integer :: atid2
1325	>	real(kind=dp), intent(in) :: rij
1326	>	real(kind=dp), intent(in) :: sw
1327	>	real(kind=dp), intent(in), dimension(3) :: d
1328	>	real(kind=dp), intent(inout) :: vpair
1329	>	real(kind=dp), dimension(9,nLocal) :: eFrame
1330	>	real(kind=dp), dimension(3,nLocal) :: f
1331	>	real(kind=dp), dimension(3,nLocal) :: t
1332	>	real (kind = dp), dimension(3) :: duduz_i
1333	>	real (kind = dp), dimension(3) :: duduz_j
1334	>	real (kind = dp), dimension(3) :: uz_i
1335	>	real (kind = dp), dimension(3) :: uz_j
1336	>	real(kind=dp) :: q_i, q_j, mu_i, mu_j
1337	>	real(kind=dp) :: xhat, yhat, zhat
1338	>	real(kind=dp) :: ct_i, ct_j
1339	>	real(kind=dp) :: ri2, ri3, riji, vterm
1340	>	real(kind=dp) :: pref, preVal, rfVal, myPot
1341	>	real(kind=dp) :: dudx, dudy, dudz, dudr
1342
1343	<	if(allocated(ElectrostaticMap)) deallocate(ElectrostaticMap)
1343	>	if (.not.summationMethodChecked) then
1344	>	call checkSummationMethod()
1345	>	endif
1346
1347	<	end subroutine destroyElectrostaticTypes
1347	>	dudx = zero
1348	>	dudy = zero
1349	>	dudz = zero
1350
1351	+	riji = 1.0_dp/rij
1352	+
1353	+	xhat = d(1) * riji
1354	+	yhat = d(2) * riji
1355	+	zhat = d(3) * riji
1356	+
1357	+	! this is a local only array, so we use the local atom type id's:
1358	+	atid1 = atid(atom1)
1359	+	atid2 = atid(atom2)
1360	+	i_is_Charge = ElectrostaticMap(atid1)%is_Charge
1361	+	j_is_Charge = ElectrostaticMap(atid2)%is_Charge
1362	+	i_is_Dipole = ElectrostaticMap(atid1)%is_Dipole
1363	+	j_is_Dipole = ElectrostaticMap(atid2)%is_Dipole
1364	+
1365	+	if (i_is_Charge.and.j_is_Charge) then
1366	+	q_i = ElectrostaticMap(atid1)%charge
1367	+	q_j = ElectrostaticMap(atid2)%charge
1368	+
1369	+	preVal = pre11 * q_i * q_j
1370	+	rfVal = preRF*rij*rij
1371	+	vterm = preVal * rfVal
1372	+
1373	+	myPot = myPot + sw*vterm
1374	+
1375	+	dudr  = sw*preVal * 2.0_dp*rfVal*riji
1376	+
1377	+	dudx = dudx + dudr * xhat
1378	+	dudy = dudy + dudr * yhat
1379	+	dudz = dudz + dudr * zhat
1380	+
1381	+	elseif (i_is_Charge.and.j_is_Dipole) then
1382	+	q_i = ElectrostaticMap(atid1)%charge
1383	+	mu_j = ElectrostaticMap(atid2)%dipole_moment
1384	+	uz_j(1) = eFrame(3,atom2)
1385	+	uz_j(2) = eFrame(6,atom2)
1386	+	uz_j(3) = eFrame(9,atom2)
1387	+	ct_j = uz_j(1)*xhat + uz_j(2)*yhat + uz_j(3)*zhat
1388	+
1389	+	ri2 = riji * riji
1390	+	ri3 = ri2 * riji
1391	+
1392	+	pref = pre12 * q_i * mu_j
1393	+	vterm = - pref * ct_j * ( ri2 - preRF2*rij )
1394	+	myPot = myPot + sw*vterm
1395	+
1396	+	dudx = dudx - sw*pref*( ri3*(uz_j(1)-3.0_dp*ct_j*xhat) &
1397	+	- preRF2*uz_j(1) )
1398	+	dudy = dudy - sw*pref*( ri3*(uz_j(2)-3.0_dp*ct_j*yhat) &
1399	+	- preRF2*uz_j(2) )
1400	+	dudz = dudz - sw*pref*( ri3*(uz_j(3)-3.0_dp*ct_j*zhat) &
1401	+	- preRF2*uz_j(3) )
1402	+
1403	+	duduz_j(1) = duduz_j(1) - sw * pref * xhat * ( ri2 - preRF2*rij )
1404	+	duduz_j(2) = duduz_j(2) - sw * pref * yhat * ( ri2 - preRF2*rij )
1405	+	duduz_j(3) = duduz_j(3) - sw * pref * zhat * ( ri2 - preRF2*rij )
1406	+
1407	+	elseif (i_is_Dipole.and.j_is_Charge) then
1408	+	mu_i = ElectrostaticMap(atid1)%dipole_moment
1409	+	q_j = ElectrostaticMap(atid2)%charge
1410	+	uz_i(1) = eFrame(3,atom1)
1411	+	uz_i(2) = eFrame(6,atom1)
1412	+	uz_i(3) = eFrame(9,atom1)
1413	+	ct_i = uz_i(1)*xhat + uz_i(2)*yhat + uz_i(3)*zhat
1414	+
1415	+	ri2 = riji * riji
1416	+	ri3 = ri2 * riji
1417	+
1418	+	pref = pre12 * q_j * mu_i
1419	+	vterm = pref * ct_i * ( ri2 - preRF2*rij )
1420	+	myPot = myPot + sw*vterm
1421	+
1422	+	dudx = dudx + sw*pref*( ri3*(uz_i(1)-3.0_dp*ct_i*xhat) &
1423	+	- preRF2*uz_i(1) )
1424	+	dudy = dudy + sw*pref*( ri3*(uz_i(2)-3.0_dp*ct_i*yhat) &
1425	+	- preRF2*uz_i(2) )
1426	+	dudz = dudz + sw*pref*( ri3*(uz_i(3)-3.0_dp*ct_i*zhat) &
1427	+	- preRF2*uz_i(3) )
1428	+
1429	+	duduz_i(1) = duduz_i(1) + sw * pref * xhat * ( ri2 - preRF2*rij )
1430	+	duduz_i(2) = duduz_i(2) + sw * pref * yhat * ( ri2 - preRF2*rij )
1431	+	duduz_i(3) = duduz_i(3) + sw * pref * zhat * ( ri2 - preRF2*rij )
1432	+
1433	+	endif
1434	+
1435	+
1436	+	! accumulate the forces and torques resulting from the self term
1437	+	f(1,atom1) = f(1,atom1) + dudx
1438	+	f(2,atom1) = f(2,atom1) + dudy
1439	+	f(3,atom1) = f(3,atom1) + dudz
1440	+
1441	+	f(1,atom2) = f(1,atom2) - dudx
1442	+	f(2,atom2) = f(2,atom2) - dudy
1443	+	f(3,atom2) = f(3,atom2) - dudz
1444	+
1445	+	if (i_is_Dipole) then
1446	+	t(1,atom1)=t(1,atom1) - uz_i(2)*duduz_i(3) + uz_i(3)*duduz_i(2)
1447	+	t(2,atom1)=t(2,atom1) - uz_i(3)*duduz_i(1) + uz_i(1)*duduz_i(3)
1448	+	t(3,atom1)=t(3,atom1) - uz_i(1)*duduz_i(2) + uz_i(2)*duduz_i(1)
1449	+	elseif (j_is_Dipole) then
1450	+	t(1,atom2)=t(1,atom2) - uz_j(2)*duduz_j(3) + uz_j(3)*duduz_j(2)
1451	+	t(2,atom2)=t(2,atom2) - uz_j(3)*duduz_j(1) + uz_j(1)*duduz_j(3)
1452	+	t(3,atom2)=t(3,atom2) - uz_j(1)*duduz_j(2) + uz_j(2)*duduz_j(1)
1453	+	endif
1454	+
1455	+	return
1456	+	end subroutine rf_self_excludes
1457	+
1458	+	subroutine accumulate_box_dipole(atom1, eFrame, d, pChg, nChg, pChgPos, &
1459	+	nChgPos, dipVec, pChgCount, nChgCount)
1460	+	integer, intent(in) :: atom1
1461	+	logical :: i_is_Charge
1462	+	logical :: i_is_Dipole
1463	+	integer :: atid1
1464	+	integer :: pChgCount
1465	+	integer :: nChgCount
1466	+	real(kind=dp), intent(in), dimension(3) :: d
1467	+	real(kind=dp), dimension(9,nLocal) :: eFrame
1468	+	real(kind=dp) :: pChg
1469	+	real(kind=dp) :: nChg
1470	+	real(kind=dp), dimension(3) :: pChgPos
1471	+	real(kind=dp), dimension(3) :: nChgPos
1472	+	real(kind=dp), dimension(3) :: dipVec
1473	+	real(kind=dp), dimension(3) :: uz_i
1474	+	real(kind=dp), dimension(3) :: pos
1475	+	real(kind=dp) :: q_i, mu_i
1476	+	real(kind=dp) :: pref, preVal
1477	+
1478	+	if (.not.summationMethodChecked) then
1479	+	call checkSummationMethod()
1480	+	endif
1481	+
1482	+	! this is a local only array, so we use the local atom type id's:
1483	+	atid1 = atid(atom1)
1484	+	i_is_Charge = ElectrostaticMap(atid1)%is_Charge
1485	+	i_is_Dipole = ElectrostaticMap(atid1)%is_Dipole
1486	+
1487	+	if (i_is_Charge) then
1488	+	q_i = ElectrostaticMap(atid1)%charge
1489	+	! convert to the proper units
1490	+	q_i = q_i * chargeToC
1491	+	pos = d * angstromToM
1492	+
1493	+	if (q_i.le.0.0_dp) then
1494	+	nChg = nChg - q_i
1495	+	nChgPos(1) = nChgPos(1) + pos(1)
1496	+	nChgPos(2) = nChgPos(2) + pos(2)
1497	+	nChgPos(3) = nChgPos(3) + pos(3)
1498	+	nChgCount = nChgCount + 1
1499	+
1500	+	else
1501	+	pChg = pChg + q_i
1502	+	pChgPos(1) = pChgPos(1) + pos(1)
1503	+	pChgPos(2) = pChgPos(2) + pos(2)
1504	+	pChgPos(3) = pChgPos(3) + pos(3)
1505	+	pChgCount = pChgCount + 1
1506	+
1507	+	endif
1508	+
1509	+	endif
1510	+
1511	+	if (i_is_Dipole) then
1512	+	mu_i = ElectrostaticMap(atid1)%dipole_moment
1513	+	uz_i(1) = eFrame(3,atom1)
1514	+	uz_i(2) = eFrame(6,atom1)
1515	+	uz_i(3) = eFrame(9,atom1)
1516	+	! convert to the proper units
1517	+	mu_i = mu_i * debyeToCm
1518	+
1519	+	dipVec(1) = dipVec(1) + uz_i(1)*mu_i
1520	+	dipVec(2) = dipVec(2) + uz_i(2)*mu_i
1521	+	dipVec(3) = dipVec(3) + uz_i(3)*mu_i
1522	+
1523	+	endif
1524	+
1525	+	return
1526	+	end subroutine accumulate_box_dipole
1527	+
1528		end module electrostatic_module

Diff Legend

–	Removed lines
+	Added lines
<	Changed lines
>	Changed lines