1 |
!! This Module Calculates forces due to SSD potential and VDW interactions |
2 |
!! [Chandra and Ichiye, J. Chem. Phys. 111, 2701 (1999)]. |
3 |
|
4 |
!! This module contains the Public procedures: |
5 |
|
6 |
|
7 |
!! Corresponds to the force field defined in ssd_FF.cpp |
8 |
!! @author Charles F. Vardeman II |
9 |
!! @author Matthew Meineke |
10 |
!! @author Christopher Fennel |
11 |
!! @author J. Daniel Gezelter |
12 |
!! @version $Id: calc_sticky_pair.F90,v 1.14 2003-08-27 16:25:11 gezelter Exp $, $Date: 2003-08-27 16:25:11 $, $Name: not supported by cvs2svn $, $Revision: 1.14 $ |
13 |
|
14 |
module sticky_pair |
15 |
|
16 |
use force_globals |
17 |
use definitions |
18 |
use simulation |
19 |
#ifdef IS_MPI |
20 |
use mpiSimulation |
21 |
#endif |
22 |
|
23 |
implicit none |
24 |
|
25 |
PRIVATE |
26 |
|
27 |
logical, save :: sticky_initialized = .false. |
28 |
real( kind = dp ), save :: SSD_w0 = 0.0_dp |
29 |
real( kind = dp ), save :: SSD_v0 = 0.0_dp |
30 |
real( kind = dp ), save :: SSD_v0p = 0.0_dp |
31 |
real( kind = dp ), save :: SSD_rl = 0.0_dp |
32 |
real( kind = dp ), save :: SSD_ru = 0.0_dp |
33 |
real( kind = dp ), save :: SSD_rlp = 0.0_dp |
34 |
real( kind = dp ), save :: SSD_rup = 0.0_dp |
35 |
real( kind = dp ), save :: SSD_rbig = 0.0_dp |
36 |
|
37 |
public :: check_sticky_FF |
38 |
public :: set_sticky_params |
39 |
public :: do_sticky_pair |
40 |
|
41 |
contains |
42 |
|
43 |
subroutine check_sticky_FF(status) |
44 |
integer :: status |
45 |
status = -1 |
46 |
if (sticky_initialized) status = 0 |
47 |
return |
48 |
end subroutine check_sticky_FF |
49 |
|
50 |
subroutine set_sticky_params(sticky_w0, sticky_v0, sticky_v0p, & |
51 |
sticky_rl, sticky_ru, sticky_rlp, sticky_rup) |
52 |
|
53 |
real( kind = dp ), intent(in) :: sticky_w0, sticky_v0, sticky_v0p |
54 |
real( kind = dp ), intent(in) :: sticky_rl, sticky_ru |
55 |
real( kind = dp ), intent(in) :: sticky_rlp, sticky_rup |
56 |
|
57 |
! we could pass all 5 parameters if we felt like it... |
58 |
|
59 |
SSD_w0 = sticky_w0 |
60 |
SSD_v0 = sticky_v0 |
61 |
SSD_v0p = sticky_v0p |
62 |
SSD_rl = sticky_rl |
63 |
SSD_ru = sticky_ru |
64 |
SSD_rlp = sticky_rlp |
65 |
SSD_rup = sticky_rup |
66 |
|
67 |
if (SSD_ru .gt. SSD_rup) then |
68 |
SSD_rbig = SSD_ru |
69 |
else |
70 |
SSD_rbig = SSD_rup |
71 |
endif |
72 |
|
73 |
sticky_initialized = .true. |
74 |
return |
75 |
end subroutine set_sticky_params |
76 |
|
77 |
subroutine do_sticky_pair(atom1, atom2, d, rij, r2, A, pot, f, t, & |
78 |
do_pot, do_stress) |
79 |
|
80 |
!! This routine does only the sticky portion of the SSD potential |
81 |
!! [Chandra and Ichiye, J. Chem. Phys. 111, 2701 (1999)]. |
82 |
!! The Lennard-Jones and dipolar interaction must be handled separately. |
83 |
|
84 |
!! We assume that the rotation matrices have already been calculated |
85 |
!! and placed in the A array. |
86 |
|
87 |
!! i and j are pointers to the two SSD atoms |
88 |
|
89 |
integer, intent(in) :: atom1, atom2 |
90 |
real (kind=dp), intent(inout) :: rij, r2 |
91 |
real (kind=dp), dimension(3), intent(in) :: d |
92 |
real (kind=dp) :: pot |
93 |
real (kind=dp), dimension(9,getNlocal()) :: A |
94 |
real (kind=dp), dimension(3,getNlocal()) :: f |
95 |
real (kind=dp), dimension(3,getNlocal()) :: t |
96 |
logical, intent(in) :: do_pot, do_stress |
97 |
|
98 |
real (kind=dp) :: xi, yi, zi, xj, yj, zj, xi2, yi2, zi2, xj2, yj2, zj2 |
99 |
real (kind=dp) :: r3, r5, r6, s, sp, dsdr, dspdr |
100 |
real (kind=dp) :: wi, wj, w, wip, wjp, wp |
101 |
real (kind=dp) :: dwidx, dwidy, dwidz, dwjdx, dwjdy, dwjdz |
102 |
real (kind=dp) :: dwipdx, dwipdy, dwipdz, dwjpdx, dwjpdy, dwjpdz |
103 |
real (kind=dp) :: dwidux, dwiduy, dwiduz, dwjdux, dwjduy, dwjduz |
104 |
real (kind=dp) :: dwipdux, dwipduy, dwipduz, dwjpdux, dwjpduy, dwjpduz |
105 |
real (kind=dp) :: zif, zis, zjf, zjs, uglyi, uglyj |
106 |
real (kind=dp) :: drdx, drdy, drdz |
107 |
real (kind=dp) :: txi, tyi, tzi, txj, tyj, tzj |
108 |
real (kind=dp) :: fxii, fyii, fzii, fxjj, fyjj, fzjj |
109 |
real (kind=dp) :: fxij, fyij, fzij, fxji, fyji, fzji |
110 |
real (kind=dp) :: fxradial, fyradial, fzradial |
111 |
real (kind=dp) :: rijtest, rjitest |
112 |
real (kind=dp) :: radcomxi, radcomyi, radcomzi |
113 |
real (kind=dp) :: radcomxj, radcomyj, radcomzj |
114 |
integer :: id1, id2 |
115 |
|
116 |
if (.not.sticky_initialized) then |
117 |
write(*,*) 'Sticky forces not initialized!' |
118 |
return |
119 |
endif |
120 |
|
121 |
if ( rij .LE. SSD_rbig ) then |
122 |
|
123 |
r3 = r2*rij |
124 |
r5 = r3*r2 |
125 |
|
126 |
drdx = d(1) / rij |
127 |
drdy = d(2) / rij |
128 |
drdz = d(3) / rij |
129 |
|
130 |
#ifdef IS_MPI |
131 |
! rotate the inter-particle separation into the two different |
132 |
! body-fixed coordinate systems: |
133 |
|
134 |
xi = A_row(1,atom1)*d(1) + A_row(2,atom1)*d(2) + A_row(3,atom1)*d(3) |
135 |
yi = A_row(4,atom1)*d(1) + A_row(5,atom1)*d(2) + A_row(6,atom1)*d(3) |
136 |
zi = A_row(7,atom1)*d(1) + A_row(8,atom1)*d(2) + A_row(9,atom1)*d(3) |
137 |
|
138 |
! negative sign because this is the vector from j to i: |
139 |
|
140 |
xj = -(A_Col(1,atom2)*d(1) + A_Col(2,atom2)*d(2) + A_Col(3,atom2)*d(3)) |
141 |
yj = -(A_Col(4,atom2)*d(1) + A_Col(5,atom2)*d(2) + A_Col(6,atom2)*d(3)) |
142 |
zj = -(A_Col(7,atom2)*d(1) + A_Col(8,atom2)*d(2) + A_Col(9,atom2)*d(3)) |
143 |
#else |
144 |
! rotate the inter-particle separation into the two different |
145 |
! body-fixed coordinate systems: |
146 |
|
147 |
xi = a(1,atom1)*d(1) + a(2,atom1)*d(2) + a(3,atom1)*d(3) |
148 |
yi = a(4,atom1)*d(1) + a(5,atom1)*d(2) + a(6,atom1)*d(3) |
149 |
zi = a(7,atom1)*d(1) + a(8,atom1)*d(2) + a(9,atom1)*d(3) |
150 |
|
151 |
! negative sign because this is the vector from j to i: |
152 |
|
153 |
xj = -(a(1,atom2)*d(1) + a(2,atom2)*d(2) + a(3,atom2)*d(3)) |
154 |
yj = -(a(4,atom2)*d(1) + a(5,atom2)*d(2) + a(6,atom2)*d(3)) |
155 |
zj = -(a(7,atom2)*d(1) + a(8,atom2)*d(2) + a(9,atom2)*d(3)) |
156 |
#endif |
157 |
|
158 |
xi2 = xi*xi |
159 |
yi2 = yi*yi |
160 |
zi2 = zi*zi |
161 |
|
162 |
xj2 = xj*xj |
163 |
yj2 = yj*yj |
164 |
zj2 = zj*zj |
165 |
|
166 |
call calc_sw_fnc(rij, s, sp, dsdr, dspdr) |
167 |
|
168 |
wi = 2.0d0*(xi2-yi2)*zi / r3 |
169 |
wj = 2.0d0*(xj2-yj2)*zj / r3 |
170 |
w = wi+wj |
171 |
|
172 |
zif = zi/rij - 0.6d0 |
173 |
zis = zi/rij + 0.8d0 |
174 |
|
175 |
zjf = zj/rij - 0.6d0 |
176 |
zjs = zj/rij + 0.8d0 |
177 |
|
178 |
wip = zif*zif*zis*zis - SSD_w0 |
179 |
wjp = zjf*zjf*zjs*zjs - SSD_w0 |
180 |
wp = wip + wjp |
181 |
|
182 |
if (do_pot) then |
183 |
#ifdef IS_MPI |
184 |
pot_row(atom1) = pot_row(atom1) + 0.25d0*(SSD_v0*s*w + SSD_v0p*sp*wp) |
185 |
pot_col(atom2) = pot_col(atom2) + 0.25d0*(SSD_v0*s*w + SSD_v0p*sp*wp) |
186 |
#else |
187 |
pot = pot + 0.5d0*(SSD_v0*s*w + SSD_v0p*sp*wp) |
188 |
#endif |
189 |
endif |
190 |
|
191 |
dwidx = 4.0d0*xi*zi/r3 - 6.0d0*xi*zi*(xi2-yi2)/r5 |
192 |
dwidy = - 4.0d0*yi*zi/r3 - 6.0d0*yi*zi*(xi2-yi2)/r5 |
193 |
dwidz = 2.0d0*(xi2-yi2)/r3 - 6.0d0*zi2*(xi2-yi2)/r5 |
194 |
|
195 |
dwjdx = 4.0d0*xj*zj/r3 - 6.0d0*xj*zj*(xj2-yj2)/r5 |
196 |
dwjdy = - 4.0d0*yj*zj/r3 - 6.0d0*yj*zj*(xj2-yj2)/r5 |
197 |
dwjdz = 2.0d0*(xj2-yj2)/r3 - 6.0d0*zj2*(xj2-yj2)/r5 |
198 |
|
199 |
uglyi = zif*zif*zis + zif*zis*zis |
200 |
uglyj = zjf*zjf*zjs + zjf*zjs*zjs |
201 |
|
202 |
dwipdx = -2.0d0*xi*zi*uglyi/r3 |
203 |
dwipdy = -2.0d0*yi*zi*uglyi/r3 |
204 |
dwipdz = 2.0d0*(1.0d0/rij - zi2/r3)*uglyi |
205 |
|
206 |
dwjpdx = -2.0d0*xj*zj*uglyj/r3 |
207 |
dwjpdy = -2.0d0*yj*zj*uglyj/r3 |
208 |
dwjpdz = 2.0d0*(1.0d0/rij - zj2/r3)*uglyj |
209 |
|
210 |
dwidux = 4.0d0*(yi*zi2 + 0.5d0*yi*(xi2-yi2))/r3 |
211 |
dwiduy = 4.0d0*(xi*zi2 - 0.5d0*xi*(xi2-yi2))/r3 |
212 |
dwiduz = - 8.0d0*xi*yi*zi/r3 |
213 |
|
214 |
dwjdux = 4.0d0*(yj*zj2 + 0.5d0*yj*(xj2-yj2))/r3 |
215 |
dwjduy = 4.0d0*(xj*zj2 - 0.5d0*xj*(xj2-yj2))/r3 |
216 |
dwjduz = - 8.0d0*xj*yj*zj/r3 |
217 |
|
218 |
dwipdux = 2.0d0*yi*uglyi/rij |
219 |
dwipduy = -2.0d0*xi*uglyi/rij |
220 |
dwipduz = 0.0d0 |
221 |
|
222 |
dwjpdux = 2.0d0*yj*uglyj/rij |
223 |
dwjpduy = -2.0d0*xj*uglyj/rij |
224 |
dwjpduz = 0.0d0 |
225 |
|
226 |
! do the torques first since they are easy: |
227 |
! remember that these are still in the body fixed axes |
228 |
|
229 |
txi = 0.5d0*(SSD_v0*s*dwidux + SSD_v0p*sp*dwipdux) |
230 |
tyi = 0.5d0*(SSD_v0*s*dwiduy + SSD_v0p*sp*dwipduy) |
231 |
tzi = 0.5d0*(SSD_v0*s*dwiduz + SSD_v0p*sp*dwipduz) |
232 |
|
233 |
txj = 0.5d0*(SSD_v0*s*dwjdux + SSD_v0p*sp*dwjpdux) |
234 |
tyj = 0.5d0*(SSD_v0*s*dwjduy + SSD_v0p*sp*dwjpduy) |
235 |
tzj = 0.5d0*(SSD_v0*s*dwjduz + SSD_v0p*sp*dwjpduz) |
236 |
|
237 |
! go back to lab frame using transpose of rotation matrix: |
238 |
|
239 |
#ifdef IS_MPI |
240 |
t_Row(1,atom1) = t_Row(1,atom1) + a_Row(1,atom1)*txi + & |
241 |
a_Row(4,atom1)*tyi + a_Row(7,atom1)*tzi |
242 |
t_Row(2,atom1) = t_Row(2,atom1) + a_Row(2,atom1)*txi + & |
243 |
a_Row(5,atom1)*tyi + a_Row(8,atom1)*tzi |
244 |
t_Row(3,atom1) = t_Row(3,atom1) + a_Row(3,atom1)*txi + & |
245 |
a_Row(6,atom1)*tyi + a_Row(9,atom1)*tzi |
246 |
|
247 |
t_Col(1,atom2) = t_Col(1,atom2) + a_Col(1,atom2)*txj + & |
248 |
a_Col(4,atom2)*tyj + a_Col(7,atom2)*tzj |
249 |
t_Col(2,atom2) = t_Col(2,atom2) + a_Col(2,atom2)*txj + & |
250 |
a_Col(5,atom2)*tyj + a_Col(8,atom2)*tzj |
251 |
t_Col(3,atom2) = t_Col(3,atom2) + a_Col(3,atom2)*txj + & |
252 |
a_Col(6,atom2)*tyj + a_Col(9,atom2)*tzj |
253 |
#else |
254 |
t(1,atom1) = t(1,atom1) + a(1,atom1)*txi + a(4,atom1)*tyi + a(7,atom1)*tzi |
255 |
t(2,atom1) = t(2,atom1) + a(2,atom1)*txi + a(5,atom1)*tyi + a(8,atom1)*tzi |
256 |
t(3,atom1) = t(3,atom1) + a(3,atom1)*txi + a(6,atom1)*tyi + a(9,atom1)*tzi |
257 |
|
258 |
t(1,atom2) = t(1,atom2) + a(1,atom2)*txj + a(4,atom2)*tyj + a(7,atom2)*tzj |
259 |
t(2,atom2) = t(2,atom2) + a(2,atom2)*txj + a(5,atom2)*tyj + a(8,atom2)*tzj |
260 |
t(3,atom2) = t(3,atom2) + a(3,atom2)*txj + a(6,atom2)*tyj + a(9,atom2)*tzj |
261 |
#endif |
262 |
! Now, on to the forces: |
263 |
|
264 |
! first rotate the i terms back into the lab frame: |
265 |
|
266 |
radcomxi = SSD_v0*s*dwidx+SSD_v0p*sp*dwipdx |
267 |
radcomyi = SSD_v0*s*dwidy+SSD_v0p*sp*dwipdy |
268 |
radcomzi = SSD_v0*s*dwidz+SSD_v0p*sp*dwipdz |
269 |
|
270 |
radcomxj = SSD_v0*s*dwjdx+SSD_v0p*sp*dwjpdx |
271 |
radcomyj = SSD_v0*s*dwjdy+SSD_v0p*sp*dwjpdy |
272 |
radcomzj = SSD_v0*s*dwjdz+SSD_v0p*sp*dwjpdz |
273 |
|
274 |
#ifdef IS_MPI |
275 |
fxii = a_Row(1,atom1)*(radcomxi) + & |
276 |
a_Row(4,atom1)*(radcomyi) + & |
277 |
a_Row(7,atom1)*(radcomzi) |
278 |
fyii = a_Row(2,atom1)*(radcomxi) + & |
279 |
a_Row(5,atom1)*(radcomyi) + & |
280 |
a_Row(8,atom1)*(radcomzi) |
281 |
fzii = a_Row(3,atom1)*(radcomxi) + & |
282 |
a_Row(6,atom1)*(radcomyi) + & |
283 |
a_Row(9,atom1)*(radcomzi) |
284 |
|
285 |
fxjj = a_Col(1,atom2)*(radcomxj) + & |
286 |
a_Col(4,atom2)*(radcomyj) + & |
287 |
a_Col(7,atom2)*(radcomzj) |
288 |
fyjj = a_Col(2,atom2)*(radcomxj) + & |
289 |
a_Col(5,atom2)*(radcomyj) + & |
290 |
a_Col(8,atom2)*(radcomzj) |
291 |
fzjj = a_Col(3,atom2)*(radcomxj)+ & |
292 |
a_Col(6,atom2)*(radcomyj) + & |
293 |
a_Col(9,atom2)*(radcomzj) |
294 |
#else |
295 |
fxii = a(1,atom1)*(radcomxi) + & |
296 |
a(4,atom1)*(radcomyi) + & |
297 |
a(7,atom1)*(radcomzi) |
298 |
fyii = a(2,atom1)*(radcomxi) + & |
299 |
a(5,atom1)*(radcomyi) + & |
300 |
a(8,atom1)*(radcomzi) |
301 |
fzii = a(3,atom1)*(radcomxi) + & |
302 |
a(6,atom1)*(radcomyi) + & |
303 |
a(9,atom1)*(radcomzi) |
304 |
|
305 |
fxjj = a(1,atom2)*(radcomxj) + & |
306 |
a(4,atom2)*(radcomyj) + & |
307 |
a(7,atom2)*(radcomzj) |
308 |
fyjj = a(2,atom2)*(radcomxj) + & |
309 |
a(5,atom2)*(radcomyj) + & |
310 |
a(8,atom2)*(radcomzj) |
311 |
fzjj = a(3,atom2)*(radcomxj)+ & |
312 |
a(6,atom2)*(radcomyj) + & |
313 |
a(9,atom2)*(radcomzj) |
314 |
#endif |
315 |
|
316 |
fxij = -fxii |
317 |
fyij = -fyii |
318 |
fzij = -fzii |
319 |
|
320 |
fxji = -fxjj |
321 |
fyji = -fyjj |
322 |
fzji = -fzjj |
323 |
|
324 |
! now assemble these with the radial-only terms: |
325 |
|
326 |
fxradial = 0.5d0*(SSD_v0*dsdr*drdx*w + SSD_v0p*dspdr*drdx*wp + fxii + fxji) |
327 |
fyradial = 0.5d0*(SSD_v0*dsdr*drdy*w + SSD_v0p*dspdr*drdy*wp + fyii + fyji) |
328 |
fzradial = 0.5d0*(SSD_v0*dsdr*drdz*w + SSD_v0p*dspdr*drdz*wp + fzii + fzji) |
329 |
|
330 |
#ifdef IS_MPI |
331 |
f_Row(1,atom1) = f_Row(1,atom1) + fxradial |
332 |
f_Row(2,atom1) = f_Row(2,atom1) + fyradial |
333 |
f_Row(3,atom1) = f_Row(3,atom1) + fzradial |
334 |
|
335 |
f_Col(1,atom2) = f_Col(1,atom2) - fxradial |
336 |
f_Col(2,atom2) = f_Col(2,atom2) - fyradial |
337 |
f_Col(3,atom2) = f_Col(3,atom2) - fzradial |
338 |
#else |
339 |
f(1,atom1) = f(1,atom1) + fxradial |
340 |
f(2,atom1) = f(2,atom1) + fyradial |
341 |
f(3,atom1) = f(3,atom1) + fzradial |
342 |
|
343 |
f(1,atom2) = f(1,atom2) - fxradial |
344 |
f(2,atom2) = f(2,atom2) - fyradial |
345 |
f(3,atom2) = f(3,atom2) - fzradial |
346 |
#endif |
347 |
|
348 |
if (do_stress) then |
349 |
|
350 |
#ifdef IS_MPI |
351 |
id1 = tagRow(atom1) |
352 |
id2 = tagColumn(atom2) |
353 |
#else |
354 |
id1 = atom1 |
355 |
id2 = atom2 |
356 |
#endif |
357 |
|
358 |
if (molMembershipList(id1) .ne. molMembershipList(id2)) then |
359 |
|
360 |
! because the d vector is the rj - ri vector, and |
361 |
! because fxradial, fyradial, and fzradial are the |
362 |
! (positive) force on atom i (negative on atom j) we need |
363 |
! a negative sign here: |
364 |
|
365 |
tau_Temp(1) = tau_Temp(1) - d(1) * fxradial |
366 |
tau_Temp(2) = tau_Temp(2) - d(1) * fyradial |
367 |
tau_Temp(3) = tau_Temp(3) - d(1) * fzradial |
368 |
tau_Temp(4) = tau_Temp(4) - d(2) * fxradial |
369 |
tau_Temp(5) = tau_Temp(5) - d(2) * fyradial |
370 |
tau_Temp(6) = tau_Temp(6) - d(2) * fzradial |
371 |
tau_Temp(7) = tau_Temp(7) - d(3) * fxradial |
372 |
tau_Temp(8) = tau_Temp(8) - d(3) * fyradial |
373 |
tau_Temp(9) = tau_Temp(9) - d(3) * fzradial |
374 |
|
375 |
virial_Temp = virial_Temp + (tau_Temp(1) + tau_Temp(5) + tau_Temp(9)) |
376 |
endif |
377 |
endif |
378 |
endif |
379 |
|
380 |
end subroutine do_sticky_pair |
381 |
|
382 |
!! calculates the switching functions and their derivatives for a given |
383 |
subroutine calc_sw_fnc(r, s, sp, dsdr, dspdr) |
384 |
|
385 |
real (kind=dp), intent(in) :: r |
386 |
real (kind=dp), intent(inout) :: s, sp, dsdr, dspdr |
387 |
|
388 |
! distances must be in angstroms |
389 |
|
390 |
if (r.lt.SSD_rl) then |
391 |
s = 1.0d0 |
392 |
dsdr = 0.0d0 |
393 |
elseif (r.gt.SSD_ru) then |
394 |
s = 0.0d0 |
395 |
dsdr = 0.0d0 |
396 |
else |
397 |
s = ((SSD_ru + 2.0d0*r - 3.0d0*SSD_rl) * (SSD_ru-r)**2) / & |
398 |
((SSD_ru - SSD_rl)**3) |
399 |
dsdr = 6.0d0*(r-SSD_ru)*(r-SSD_rl)/((SSD_ru - SSD_rl)**3) |
400 |
endif |
401 |
|
402 |
if (r.lt.SSD_rlp) then |
403 |
sp = 1.0d0 |
404 |
dspdr = 0.0d0 |
405 |
elseif (r.gt.SSD_rup) then |
406 |
sp = 0.0d0 |
407 |
dspdr = 0.0d0 |
408 |
else |
409 |
sp = ((SSD_rup + 2.0d0*r - 3.0d0*SSD_rlp) * (SSD_rup-r)**2) / & |
410 |
((SSD_rup - SSD_rlp)**3) |
411 |
dspdr = 6.0d0*(r-SSD_rup)*(r-SSD_rlp)/((SSD_rup - SSD_rlp)**3) |
412 |
endif |
413 |
|
414 |
return |
415 |
end subroutine calc_sw_fnc |
416 |
end module sticky_pair |