1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
|
.. index::
single: Program; Surfacehop
single: Surfacehop
.. _UG\:sec\:surfacehop:
:program:`surfacehop`
=====================
.. only:: html
.. contents::
:local:
:backlinks: none
.. xmldoc:: <MODULE NAME="SURFACEHOP">
%%Description:
<HELP>
This module deals with surface hop semiclassical molecular dynamics (SHMD) and has to be used together with module DYNAMIX.
Its purpose is the calculation of the relax root for the next step of the SHMD.
</HELP>
This module deals with surface hop semiclassical molecular dynamics (SHMD) and has to be used together with module DYNAMIX. Its purpose is the calculation of the relax root for the next step of the SHMD. The implemented algorithm under this module is the Tully's fewest switches :cite:`Tully1990`, using the Hammes-Schiffer/Tully scheme :cite:`Hammes-Schiffer1994` and the decoherence correction proposed by Granucci and Persico :cite:`Granucci2007`.
Under the Hammes-Schiffer/Tully scheme, the non-adiabatic population transfer between states of the same multiplicity is determined using the wavefunction overlap between the current timestep and the two previous timesteps, in an interpolation-extrapolation scheme. This is done in lieu of calculating explicitly the non-adiabatic coupling, and thus allows for surface-hopping when explicit non-adiabatic coupling is not available or is too expensive.
There are two methods to calculate the wavefunction overlap available through the :program:`SURFACEHOP` module. The default implementation calls the :program:`RASSI` module to obtain the overlap matrix between all states at the current and previous timestep. The alternative method (previously default) can be requested using the keyword :kword:`NORASSI` and uses instead a dot product of the CI vectors to approximate the overlap matrix.
.. _UG\:sec\:surfacehop_output_files:
Output files
............
.. class:: filelist
:file:`RUNFILE`
Surface hop information such as Amatrix and CI coefficients for previous steps are stored in this file.
:file:`$Project.md.xyz`
Contains the geometry of every timestep in the dynamics, in standard xyz coordinates.
:file:`$Project.md.energies`
Contains the Potential energy of the current active state, Kinetic energy, and Total energy of the system throughout the simulation, followed by the potential energies of all states in the dynamics.
.. _UG\:sec\:surfacehop_inp:
Input
-----
::
&Gateway
coord=$Project.xyz
basis=6-31G*
group=nosym
>> EXPORT MOLCAS_MAXITER=400
>> DOWHILE
&Seward
&rasscf
jobiph
cirestart
nactel = 6 0 0
inactive = 23
ras2 = 6
ciroot = 2 2 1
prwf = 0.0
mdrlxroot = 2
&Surfacehop
tully
decoherence = 0.1
psub
&alaska
&Dynamix
velver
dt = 41.3
velo = 1
thermo = 0
>>> End Do
General keywords
................
.. class:: keywordlist
:kword:`TULLY`
This keyword enables the Tully--Hammes-Schiffer integration of the TDSE for the Tully Surface Hop Algorithm. If you use this keyword you should not use the :kword:`HOP` keyword in :program:`DYNAMIX`.
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="TULLY" APPEAR="Tully surface hop" KIND="SINGLE" LEVEL="BASIC">
%%Keyword: TULLy <basic>
<HELP>
This keyword enables the Tully--Hammes-Schiffer integration of the TDSE for the Tully Surface Hop Algorithm.
</HELP>
</KEYWORD>
:kword:`NORASSI`
This keyword must be used after the :kword:`TULLY` keyword. It disables the use of :program:`RASSI` to calculate wavefunction overlaps, instead using the dot product of CI vectors (previous default option).
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="NORASSI" APPEAR="Wavefunction Overlap by CI vectors" KIND="SINGLE" LEVEL="BASIC" REQUIRE="TULLY">
%%Keyword: NORAssi <basic>
This keyword must be used after the TULLY keyword.
<HELP>
It disables the use of RASSI to calculate wavefunction overlaps.
</HELP>
</KEYWORD>
:kword:`DECOHERENCE`
This keyword must be used after the :kword:`TULLY` keyword. It enables the decoherence correction in the population density matrix as reported by Persico and Granucci. The value is called decay factor and it is usually 0.1 hartree. It can be seen as how strongly this correction is applied. It is recommendable to leave it to 0.1, unless you really know what you're doing.
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="DECOHERENCE" APPEAR="Decoherence correction" KIND="REAL" LEVEL="ADVANCED" DEFAULT_VALUE="0.1" REQUIRE="TULLY">
%%Keyword: DECOherence <advanced>
This keyword must be used after the TULLY keyword.
<HELP>
It enables the decoherence correction in the population density matrix as reported by Persico-Granucci.
</HELP>
</KEYWORD>
:kword:`SUBSTEP`
This keyword must be used after the :kword:`TULLY` keyword. This keyword specifies how many steps of integration we use to interpolate/extrapolate between two Newton's consecutive steps. The default is usually a good compromise between quickness and precision (200 substeps each femtoseconds of MD).
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="SUBSTEP" APPEAR="Electronic integration substeps" KIND="INT" LEVEL="ADVANCED" DEFAULT_VALUE="200" MIN_VALUE="0" REQUIRE="TULLY">
%%Keyword: SUBStep <advanced>
This keyword must be used after the TULLY keyword.
<HELP>
It specifies how many steps of integration we use to interpolate/extrapolate between two Newton's consecutive steps.
</HELP>
</KEYWORD>
:kword:`PSUB`
This keyword must be used after the :kword:`TULLY` keyword. To print in |molcas| output :math:`\mat{D}` matrix, :math:`\mat{A}` matrix, :math:`\mat{B}` matrix, probabilities, randoms, population and energies at each substep (quite verbose, but gives you a lot of useful information).
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="PSUB" APPEAR="Verbose output for each substep" KIND="SINGLE" LEVEL="BASIC" REQUIRE="TULLY">
%%Keyword: PSUB <basic>
This keyword must be used after the TULLY keyword.
<HELP>
To print in molcas output D matrix, A matrix, B matrix, probabilities, randoms, population and energies at each substep.
</HELP>
</KEYWORD>
:kword:`DMTX`
This keyword must be used after the :kword:`TULLY` keyword. With this keyword you can start your calculation with an initial :math:`\mat{A}` matrix (population density matrix). It is a complex matrix. In the first line after the keyword you must specify its dimension :math:`N`. Then :math:`N` lines (:math:`N` values each line) with the real part of the matrix followed by :math:`N` more lines with the imaginary part.
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="DMTX" APPEAR="Initial population density matrix" KIND="UNKNOWN" LEVEL="ADVANCED" REQUIRE="TULLY">
%%Keyword: DMTX <advanced>
This keyword must be used after the TULLY keyword.
<HELP>
Initial A matrix (population density matrix). It is a complex matrix.
In the first line after the keyword you must specify its dimension N. Then N lines (with N values each line) with the REAL part of the matrix followed by N more lines with the imaginary part.
</HELP>
</KEYWORD>
:kword:`FRANDOM`
This keyword must be used after the :kword:`TULLY` keyword. It fixes the random number to one provided by the user, in case a deterministic trajectory is needed
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="FRANDOM" APPEAR="Random number constant (deterministic MD)" KIND="REAL" LEVEL="ADVANCED" REQUIRE="TULLY">
%%Keyword: FRANdom <advanced>
This keyword must be used after the TULLY keyword.
<HELP>
It fixes the random number to one provided by the user, in case a deterministic trajectory is needed.
</HELP>
</KEYWORD>
:kword:`ISEED`
This keyword must be used after the :kword:`TULLY` keyword. The initial seed number is read from the input file. Then, seed numbers are modified (in a deterministic way), saved in the :file:`RunFile` and read in the next call to the module. This way, MD simulations are reproducible.
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="ISEED" APPEAR="Initial seed number (reproducible MD)" KIND="INT" LEVEL="ADVANCED" REQUIRE="TULLY">
%%Keyword: ISEEd <advanced>
This keyword must be used after the TULLY keyword.
<HELP>
The initial seed number is read from the input file.
Then, seed numbers are modified (in a deterministic way), saved in the RunFile and read in the next call to the module.
This way, MD simulations are reproducible.
</HELP>
</KEYWORD>
:kword:`MAXHOP`
This keyword must be used after the :kword:`TULLY` keyword. It specifies how many non-adiabatic transitions are allowed between electronic states.
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="MAXHOP" APPEAR="Maximum number of hops allowed" KIND="INT" LEVEL="ADVANCED" REQUIRE="TULLY">
%%Keyword: MAXHop <advanced>
This keyword must be used after the TULLY keyword.
<HELP>
It specifies how many non-adiabatic transitions are allowed between electronic states.
</HELP>
</KEYWORD>
:kword:`H5RESTART`
This keyword allows to restart a surface hopping trajectory calculation from an HDF5 file.
The name of the restart file is given on the next line.
.. xmldoc:: <KEYWORD MODULE="SURFACEHOP" NAME="H5RESTART" APPEAR="Restart the surface hopping trajectory from an H5 file" KIND="SINGLE" LEVEL="ADVANCED">
%%Keyword: H5REstart <advanced>
<HELP>
Restarts a surface hopping trajectory calculation from an HDF5 file, whose name is given on the next line.
</HELP>
</KEYWORD>
Input examples
..............
This example shows an excited state CASSCF MD simulation
of a methaniminium cation using the Tully Surface Hop algorithm.
Within the :program:`Surfacehop` module The keyword :kword:`TULLY` enables the TDSE integration. The options used in this case are:
(:kword:`SUBSTEP`\=200) to specify 200 substep of electronic integration between Newton's,
(:kword:`DECOHERENCE`\=1) to deal with the decoherence using a decay constant of 0.1 hartree and
(:kword:`PSUB`) to print the substeps matrices verbosely into the |molcas| log.
.. extractfile:: ug/surfacehopTULLY.input
&GATEWAY
COORD
6
angstrom
C 0.00031448 0.00000000 0.04334060
N 0.00062994 0.00000000 1.32317716
H 0.92882820 0.00000000 -0.49115611
H -0.92846597 0.00000000 -0.49069213
H -0.85725321 0.00000000 1.86103989
H 0.85877656 0.00000000 1.86062860
BASIS= 3-21G
GROUP= nosym
>> EXPORT MOLCAS_MAXITER=1000
>> DOWHILE
&SEWARD
>> IF ( ITER = 1 )
&RASSCF
LUMORB
FileOrb= $Project.GssOrb
Symmetry= 1
Spin= 1
nActEl= 2 0 0
Inactive= 7
RAS2= 2
CIroot= 3 3 1
>> COPY $Project.JobIph $Project.JobOld
>> ENDIF
&RASSCF
JOBIPH; CIRESTART
Symmetry= 1
Spin= 1
nActEl= 2 0 0
Inactive= 7
RAS2= 2
CIroot= 3 3 1
MDRLXR= 2
>> COPY $Project.JobIph $Project.JobOld
&surfacehop
TULLY
SUBSTEP = 200
DECOHERENCE = 0.1
PSUB
&ALASKA
&Dynamix
VELVer
DT= 10.0
VELO= 3
THER= 2
TEMP=300
>> END DO
.. xmldoc:: </MODULE>
|
