In the present section we consider execution of GAMESS–UK using the UNIX shell script rungamess. The jobs perform the same function as those in the previous Chapter, and the

GAMESS–UK directives themselves are identical. However, the input files contain only GAMESS–

UK directives, all reference to pathnames for executable, library, and scratch files is handled by the script.

The principle differences associated with this mode of usage of the program are as follows;

1. The users data files do not contain reference to specific pathnames for executables, scratch directories, etc. These are taken from user–set environment variables, or are supplied by the local support staff.

2. The job executes in a specially created directory, which will usually be on a scratch disk.

Files are routed back to the directory of submission, or to longer–term temporary filestore as required.

3. A variety of batch queuing systems are provided (local configuration is required).

4. The decision regarding which files to retain beyond the end of the job may be set automatically, given the type of job

5. The job may execute on a remote networked machine, without the need to explicitly move input and output, or files (binary datasets are retained on the execution machine)

Note that this documentation describes version 3.x of rungamess, these versions differs substantially from the earlier versions. From a user point of view the main differences are

• only the short forms of the arguments are now accepted (–keep must be replaced by –k, etc.).

• the job name must be the last argument.

The next section summarises the syntax and the function of the arguments to the rungamess script, and subsequent sections contain the following illustrative examples.

1. Single point SCF calculation of water using the default basis.

2. Sequence of calculations on water, including starting and restarting geometry optimisation of the neutral molecule, an RHF calculation on the

1 cation, boys localisation and 2-pair

GVB calculation of the neutral molecule.

3. Closed-shell SCF calculations on C

, C

2 and C

(NO

)

4. STO-3G calculation of Na

, together with an RHF and UHF calculation of the triplet state.

1 INTRODUCTION 3

5. Extended basis set calculation on Na

, with SCF calculation of the singlet state preceding an RHF calculation of the triplet state.

6. ECP calculation on Na

+ in a double-zeta valence basis, with SCF calculation of the singlet state preceding an RHF calculation of the triplet state.

7. 2-pair GVB calculation on 1-imino-2,4-pentadiene, CH

(CH)

8. Graphical analysis of the X

1 state of Ni(CO)

9. STO-3G direct-SCF calculations on C

NPH

Be(C

)

and the geometry optimisation of

10. Determination of the transition state for the HCN/HNC isomerisation reaction. This example demonstrates usage of the three saddle point algorithms available, the default trust-region method plus the synchronous transit and Simons-Jorgensen algorithms.

11. Location of the transition state for the HPSi/HSiP isomerisation and subsequent force constant evaluation.

12. Specification of bond-centred functions, located at the midpoint of the C-N bond in the

HCN/HNC transition state.

13. SCF geometry optimisation and analytic force constants of C

14. MP2 geometry optimisation and analytic force constants of C

15. MP2 geometry optimisation and polarisability of C

16. Direct-MP2 calculation of C

N in a 6-31G* basis.

17. CASSCF geometry optimisation of the X

1 state of H

18. CASSCF + 2nd-order CI calculations of the X

+ state of BeO.

19. MCSCF + 2nd-order CI calculations of the X

+ state of BeO.

20. Table-CI calculation of the X

1 and 1

1 states of the ammonia cation, NH

3 cation.

21. ECP calculation on NiCCH

2 state wavefunction.

with CASSCF and Direct-CI calculations of the lowest triplet

22. Table-CI calculations typical of those performed in the calculation of electronic spectra.

In this case we are studying the disposition of the lowest

1 and 1

2 states of pyridine.

The sequence of calculations involve the determination of the lowest 10 states of each category, performed in a DZ plus rydberg basis.

23. Full-CI calculations of the X

1 state of H

2 USAGE OF THE RUNGAMESS SCRIPT 4

2 Usage of the Rungamess Script

2.1 Installation

Please see the README file in the GAMESS-UK/rungamess directory for the installation and configuration instructions.

2.2 Running a Job

In its simplest usage, the script is invoked by typing rungamess, followed by a job name, for example: rungamess myjob

This will cause GAMESS–UK to read the file myjob.in, and generate the listing file myjob.out, and (when appropriate) punch file myjob.pun in the directory of submission.

The exact pathnames used for the executable, and scratch files are determined from values of environment variables, as follows.

• The executable name is taken from the environment variable GAMESS EXE.

• The environment variable GAMESS SCR must be set to the name of a directory in which the user has permission to create files. A directory will be created under this directory for each job that is run, and all binary datasets that are not routed elsewhere will be written here for the duration of the job. The directory is deleted at the end of the job.

If GAMESS SCR does not exist, an attempt will be made to create it. The environment variable GAMESS TMP should be set to a directory in which storage of files between jobs is possible (see below), although it is not used in this simple example.

• The environment variable GAMESS LIB must be set to the name of a directory containing the GAMESS–UK Library files. (see Accessing Library Files below).

The values of these variables will usually be set as part of the login procedure. On a central computer facility, the values should be provided by the support staff, in other cases the user should edit values into the appropriate login scripts. For users of the C-Shell (/bin/csh) the

AXP 433AU at Daresbury.

# setenv GAMESS_EXE /scr1/wab/GAMESS-UK/bin/gamess setenv GAMESS_LIB /scr1/wab/GAMESS-UK/libs setenv GAMESS_TMP /scr1/xyz setenv GAMESS_SCR /scr1/xyz

Users of the Borne shell (/bin/sh) and Korn shell (/bin/ksh) should make the corresponding definitions in the file .profile in their home directory.

2 USAGE OF THE RUNGAMESS SCRIPT

GAMESS_EXE=/scr1/wab/GAMESS-UK/bin/gamess export GAMESS_EXE

GAMESS_LIB=/scr1/wab/GAMESS-UK/libs export GAMESS_LIB

GAMESS_TMP=/scr1/xyz export GAMESS_TMP

GAMESS_SCR=/scr1/xyz export GAMESS_SCR

2.3 Saving Files

Command line options are used to request that datasets are to be retained beyond the end of the job. The option –k, followed by a GAMESS–UK logical filename (presented in lowercase) will cause the file to be retained in the directory from which the job is run, with a suffix derived from the logical filename. For example; rungamess -k ed3 myjob will save the dumpfile, with filename myjob.ed3.

If control of the filename is required, it may be provided, using the = symbol. The following will associate the file on LFN ed3 with the file oldjob.dump, again in the directory of job submission.

rungamess -k ed3=oldjob.dump myjob

If the filename supplied in this way is a UNIX relative pathname (ie it does not start with /), it is taken to be relative to the current directory at the time the job is submitted. Otherwise an absolute pathname is assumed, as in the following example.

rungamess -k ed3=/scr1/xyz/oldjob.dump myjob

This provides a completely general way of assigning the GAMESS–UK datasets, equivalent to the explicit setenv commands in the previous chapter. However, it becomes cumbersome if many datasets have to be named in this way. The remaining file options (–t, –r, –n and –l) serve to provide a more concise syntax.

The –t option is identical to –k, except that the default directory is the temporary directory

$GAMESS TMP. Since $GAMESS TMP will usually refer to a scratch disk, this means that large files may be retained beyond the end of the job, without being written to the users own filestore, where space will usually be limited. On most installations, the variables GAMESS TMP and

GAMESS SCR will actually be set to the same directory. The distinction is that GAMESS TMP should point to a directory which will exist beyond the end of the job (since it will contain files required to restart calculations) whereas that referenced by GAMESS SCR (a subdirectory of

$GAMESS SCR named after the job name) will be deleted after the program has finished.

As an example, consider an installation where GAMESS SCR has been set to point to a workdirectory /scr2/xyz, and GAMESS TMP to /scr1/xyz. The rungamess invocation

2 USAGE OF THE RUNGAMESS SCRIPT 6

Table 1: Keywords accepted by the –r argument keyword function hf casscf mcscf ci mrdci fullci save ed2 and ed3 save ed1, ed2, ed3, ed4, ed6, ed9, ed10, ed11 save ed2, ed3, ed4, ed6, ed13 save ed2, ed3, ed4, ed5, ed6 (for direct CI calculations) save ed2, ed3, ftn031, ftn033, ftn034, ftn035, ftn036 save ed2, ed3, ed6, ftn008 rungamess -k ed3 -t ed2 myjob will cause a temporary directory /scr2/xyz/myjob to be created, and will contain all datasets used by GAMESS–UK except that associated with LFN ed3, which will have pathname myjob.ed3, and ed2, which will have pathname /scr1/xyz/myjob.ed2.

The way in which saved files are reused is dependent on the preference of the user. If a job needs to be restarted, one option is simply to edit the input file (say to add a restart directive), and re–issue the command given above. This is the simplest method, but obviously results in the input and output files being overwritten.

If it is preferred that the restart job has a different name, it is necessary to ensure that the old files are correctly associated with the new job name. This is the model adopted in this Chapter, because of the need to include all the input files, with unique names, in the examples directory.

This can simply be achieved by giving the filename as part of the argument to –k or –t, the following command will re-use the dumpfile ed3 (but not ed2) from the above job.

rungamess -k ed3=myjob.ed3 myjob_new

If many files are to be carried over between jobs of different names, it is simpler to rename them all using the –name (or –n) option. This sets the first part of the filename for all direct access and Fortran datasets, but leaves the names of the names input, output, punch, and NQS job files unchanged. The following sequence uses “water” as the root part of the filename for both jobs.

rungamess -t ed2 -t ed3 -n water water_scf rungamess -t ed2 -t ed3 -n water water_opt

To simplify the file arguments required to allow restarting of a job as a result of time–out the –r argument has been provided. It is followed by a character string specifying the type of calculation for which the restart may be required. Valid key strings, and their effects, are given in Table

1 . Clearly, the –r argument must be given on the start–up job, as well as the restart

job.

2 USAGE OF THE RUNGAMESS SCRIPT 7

The –r option is equivalent to using a series of –t options, one for each relevant file as given in

Table

1 . Files saved in this way thus reside on the directory GAMESS TMP, with names of the

form myjob.ed11. Since rungamess processes file arguments in the order they are presented, any of the file assignments implied by –r may be overridden if the appropriate option is placed after –r.

2.4 Accessing Library Files

The –library argument (short form –l) can be used to associate a logical file name with the corresponding library file. The two LFNs which are accepted are ed0 (for the non-local pseudopotential library file, and table (for the Table–CI database used by the MRDCI module). The directory in which the library files are to be found is obtained from the environment variable

GAMESS LIB.

2.5 Batch Job Submission

Provided the local setup has been performed, the rungamess script will submit a job to a batch queuing system if the –q option is provided. A job file with suffix .script is generated and submitted. The exact behaviour of the script and any additional arguments that can be presented will depend on the queuing system in use. Template script files for submission to

NQS, PBS and LoadLeveler are provided as part of the distribution. You will need to set the environment GAMESS SUBMODE to determine which of the submission scripts are chosen.

See the file README in the rungamess directory for up-to-date details of the setup required.

2.6 Parallel Job Execution

The –p argument specifies the number of processors to be used to run the job. It will only have an effect on a parallel machine, and relies on the appropriate environment variables having been made to rungamess to invoke the parallel code. Execution scripts for execution using

TCGMSG/GA, IBM SP (under POE), LAM MPI, and the Cray T3E series are included in the distribution. You will need to set the environment variable GAMESS PAREXE to the name of the parallel executable, and GAMESS PARMODE to specify which of the execution scripts to use on the current platform. For interactive use on cluster systems GAMESS HOSTS should be set to a list of hostnames to be use. See the file README in the rungamess directory for up-to-date details of the setup required.

2.7 Summary of Command Line Arguments

The command line arguments accepted by rungamess are tabulated in Table

2 . A listing of this

information may be obtained on–line by issuing the command rungamess -i which will indicate which options are supported on your local installation.

3 GAMESS–UK PRE-DIRECTIVES 8

Table 2: Arguments to rungamess

Argument Keyword Function

-p

-s

-i

-T

-q

-Q

-k

-t

-l

-n

-r file file file name jobtype time

(none)

# nodes

(none)

(none) save file in home directory of job save file in temporary directory access library file (file must be ed0 or table) provide root part of name for files (default is the jobname) request save of all files required for restart time for job queue place job into an NQS queue queue args place job into an NQS queue (non-default job settings) specify number of nodes for parallel execution

Provide listing on stdout print information

3 GAMESS–UK Pre-Directives

The program is capable of processing a set of ‘pre-directives’, each such directive extending over one data line, and appearing as input before the program specific data. These pre-directives allow the user to define or modify, through data input, certain characteristics of the job environment e.g., time allocations, routing of output, file allocations, memory requirements etc.

3.1 The memory Pre-directive

This pre-directive provides a mechanism for specifying the dynamic core to be associated with the present run of the code. Memory requirements are in general a function of the RUNTYPE requested in the data input: allocation can be modified through MEMORY data specification.

The pre-directive consists of a single data line, the first data field being set to the character string memory or core, the second to an integer defining the number of words of memory.

Presenting the data line memory 8000000 will yield an allocation of 64 MBytes. The default allocation of 4,000,000 words will prove adequate for most runs involving both SCF and CI wavefunctions.

3.2 The time Pre-directive

GAMESS–UK monitors the CPU time available at intervals, and if it is found that insufficient time remains to usefully continue, will send restart control information to the Dumpfile, and terminate execution. The time pre-directive is used to specify the time limit for the job in CPU minutes, e.g.,

4 SINGLE POINT SCF CALCULATION 9 time 120 will allocate 2 hours of CPU time to the job. In the absence of the time pre-directive, a default allocation of 600 minutes will be in effect.

4 Single Point SCF Calculation

Let us consider a simple SCF run for H

O in a 3-21G basis, and describe submission using the script rungamess.

Assume that the following lines reside in the file myjob.in

title\h2o 3-21g scf zmat angstrom\o\h 1 roh\h 1 roh 2 theta variables\roh 0.956 hess 0.7\theta 104.5 hess 0.2 \end enter

The command rungamess may then be used to run the job in the background rungamess myjob &

If batch submission were required, the option -q should be added.

rungamess myjob -q >& myjob.err &

This will result in the files used by GAMESS–UK, in this case ED2, ED3 and ED7 being created in the scratch directory for the job, run, and deleted on job termination. The output from the job will be routed to the file myjob.out. System diagnostics will appear at the terminal stderr, unless redirected; rungamess myjob >& myjob.err &

To retain the ed2 and ed3 files in the scratch directory, rungamess myjob -t ed2 -t ed3 >& myjob.err &

5 Sequence of Calculations on H

O

In this example we include a sequence of input files and for performing various calculations based on the H

O example given above. To simplify the commands required, “water” is used as the root of the filename of all binary datasets, using the –n argument.

1. The first job (as above) generates the starting vectors. In this and subsequent jobs, the mainfile (ed2) is saved in case a restart due to lack of time occurs.

Contents of file water scf.in:

5 SEQUENCE OF CALCULATIONS ON H

O 10 title\h2o 3-21g scf zmat angstrom\o\h 1 roh\h 1 roh 2 theta variables\roh 0.956 hess 0.7\theta 104.5 hess 0.2 \end enter to run: rungamess -t ed2 -t ed3 -n water water_scf

2. This run utilises the vectors made above. Both bond length and bond angle are to be optimised.

Contents of file water opt.in: restart new title water optimisation at scf level 3-21g basis set zmat angstrom o h 1 oh h 1 oh 2 hoh variables oh 0.956

hoh 104.5

end runtype optimise enter to run: rungamess -t ed2 -t ed3 -n water water_opt

3. This example would be used to complete, if necessary, the run started in 1.

Contents of file water restopt.in: restart optimise title water optimisation at scf level 3-21g basis set zmat angstrom o h 1 oh h 1 oh 2 hoh variables oh 0.956

hoh 104.5

end runtype optimise enter to run: rungamess -t ed2 -t ed3 -n water water_restopt

5 SEQUENCE OF CALCULATIONS ON H

O 11

4. This example performs an open-shell RHF calculation at the optimised geometry from 2.

(note the RESTART usage).

Contents of file water orhf.in: restart title h2o+ doublet b1 state - ground state geometry charge 1 mult 2 zmat angs o h 1 oh h 1 oh 2 hoh variables oh 0.956

hoh 104.5

end enter to run: rungamess -t ed2 -t ed3 -n water water_orhf

5. The valence shell scf mos are to be localised. For the purpose of the example below it is assumed that orbitals 2 and 3 are the bond orbitals. The CHARGE and MULT directives are required to override the values set in the preceding job on the doublet cation.

Contents of file water loc.in: restart title lmos for h2o-ground state geometry charge 0 mult 1 zmat angs o h 1 oh h 1 oh 2 hoh variables oh 0.956

hoh 104.5

end runtype analyse local

2 to 5 end vectors 1 enter 5 to run: rungamess -t ed2 -t ed3 -n water water_loc

6 SCF CALCULATIONS ON C

, C

AND C

(NO

)

6. The localised orbitals are restored from section 5 and the NOGEN facility used to generate the virtual pairs of the two bond orbitals. The SWAP directive has been used to move the localised bond orbitals to the top of the occupied orbital list. The NOGEN facility reorders the orbitals so that the GVB pairs occur together. Note the ADAPT OFF specification, now required when using localised orbitals as the input orbital set.

Contents of file water gvb.in: restart title water gvb calculation using localised orbitals adapt off zmat ang o h 1 oh h 1 oh 2 hoh variables oh 0.956

hoh 104.5

end scftype gvb 2 vectors nogen 5 swap

2 5

3 4 end enter to run: rungamess -t ed2 -t ed3 -n water water_gvb

6 SCF Calculations on C

F

, C

H

NO

and C

H

(NO

)

CH

We show below the input files and job submission commands for three straightforward closedshell SCF calculations, providing more examples of z-matrix specification and reliance on the default options in such calculations. In each case the direct access files will be deleted on job completion.

Closed shell SCF Job for C

Contents of file cubane.in: title

**** c4f4 3/21g **** zmat angs x c 1 r1 c 1 r2 2 90.

c 1 r1 3 90.

2 180.

6 SCF CALCULATIONS ON C

, C

AND C

(NO

)

3 c 1 r2 4 90.

3 180.

x 2 1.

1 90.

3 0.

f 2 r3 6 90.

x 4 1.

1 90.

3 180.

3 0.

f 4 r3 8 90.

x 3 1.

1 90.

3 180.

4 0.

f 3 r3 10 90.

4 180.

x 5 1.

1 90.

4 0.

f 5 r3 12 90.

4 180.

variables r1 1.2

r2 1.3

r3 1.313

end enter to run: rungamess cubane

Closed shell SCF Job for C

Contents of file nitrobenzene.in: title c6h5.no2

3-21g accuracy 20 7 noprint zmat angstrom c n 1 rcn x 2 1.0 1 90.0

c 1 rcc1 2 t1 3 p1 c 1 rcc1 2 t1 3 -p1 c 4 rcc2 1 t2 2 p2 c 5 rcc2 1 t2 2 p2 c 7 rcc3 5 t3 1 p3 o 2 rno1 1 t5 3 -90.0

o 2 rno1 1 t5 3 90.0

h 4 rch1 1 t6 2 p5 h 5 rch1 1 t6 2 p5 h 6 rch2 4 t7 11 p6 h 7 rch2 5 t7 12 p6 h 8 rch3 7 t8 14 p7 variables rcn 1.49

rcc1 1.37

rcc2 1.43

rcc3 1.37

rno1 1.21

rch1 1.084

rch2 1.084

rch3 1.084

t1 120.0

6 SCF CALCULATIONS ON C

, C

AND C

(NO

)

3 t2 120.0

t3 120.0

t5 120.0

t6 120.0

t7 120.0

t8 120.0

p1 90.0

p2 180.0

p3 0.0

p5 0.0

p6 0.0

p7 0.0

end maxcyc 20 enter to run: rungamess nitrobenzene

Closed shell SCF Job for C

(NO

)

Contents of file tnt.in: o o n c n c x c c x c c title

2,4,6 tri-nitro-toluene. 3-21g basis.

noprint distance basis vectors hessian zmat angstrom

4 r1 r2 r2 r3 r3

r4 r5 r6 r7 r5

2 a1 a1 a2 a2 3

90.

a3 a4 a5 a3

180.

x n o 12 o 12

8 o 16 o 16 x 2 r6 r7

r8 r9 r9

1 a4 a5

90.

a6 a6

90.

180.

h h c 2 r10 19 h 20 r11 2 h 20 r12 h 20 r12

6 variables r13 r13

90.

1 a7 19

180.

90.

a8 21 120.

a8 21 -120.

a9 a9

180.

7 STO-3G CALCULATIONS ON NA

15 r1 1.431\r2 1.367\r3 1.397\r4 1.395

a1 60.38\a2 120.67

r5 1.521\r8 1.505\r6 1.277\r7 1.278\r9 1.277

a3 123.52\a4 121.19\a5 116.17\a6 117.53

r10 1.529\r11 1.087\r12 1.082

a7 110.07\a8 109.7

r13 1.086\a9 121.26

end enter to run: rungamess tnt

7 STO-3G Calculations on Na

Mg

In this calculation we perform an STO-3G calculation on Na

, followed by an RHF and then

UHF calculation on the triplet state. Note the use of the SUPER directive to ensure an integral file format compatible with the use of BYPASS in the subsequent RHF and UHF calculations.

Closed shell SCF Job

Contents of file na7mg rhf.in: title

* na7mg+ * sto-3g * closed shell * scf-energy= -1314.828516

mult 1 super force charge 1 zmat angs mg na 1 r1 na 1 r2 2 90.

na 1 r2 2 90. 3 72.

na 1 r2 2 90. 4 72.

na 1 r2 2 90. 5 72.

na 1 r2 2 90. 6 72.

na 1 r1 3 90. 2 180.

variables r1 3.0286740

r2 3.194799

end basis sto3g level 1.5

10 1.0

maxcyc 40 enter to run: rungamess -t ed2 -t ed3 -n na7mg na7mg_rhf

7 STO-3G CALCULATIONS ON NA

Open shell RHF Job

Contents of file na7mg orhf.in

restart title

* na7mg+ * sto-3g * triplet * scf energy=-1314.900829

super force bypass mult 3 charge 1 zmat angs mg na 1 r1 na 1 r2 2 90.

na 1 r2 2 90. 3 72.

na 1 r2 2 90. 4 72.

na 1 r2 2 90. 5 72.

na 1 r2 2 90. 6 72.

na 1 r1 3 90. 2 180.

variables r1 3.0286740

r2 3.194799

end basis sto3g maxcyc 40 enter to run: rungamess -t ed2 -t ed3 -n na7mg na7mg_orhf

UHF Job

Contents of file na7mg uhf.in: restart super force title

* na7mg+ * sto-3g * triplet * uhf energy=-1314.901919

mult 3 charge 1 bypass zmat angs mg na 1 r1 na 1 r2 2 90.

na 1 r2 2 90. 3 72.

na 1 r2 2 90. 4 72.

na 1 r2 2 90. 5 72.

na 1 r2 2 90. 6 72.

na 1 r1 3 90. 2 180.

variables

8 EXTENDED BASIS SET CALCULATIONS OF NA

17 r1 3.0286740

r2 3.194799

end basis sto3g scftype uhf vectors 5 enter to run: rungamess -t ed2 -t ed3 -n na7mg na7mg_uhf

Note that we are using the energy ordered open-shell RHF eigenvectors to initiate the UHF calculation (as written to the default section 5 by the Open shell RHF Job). Had this section not been specified using the VECTORS directive, then the closed-shell SCF MOs would been used in default.

8 Extended Basis Set Calculations of Na

Mg

In this example we use the STO-3G calculation on Na

, performed above as a starting point for a more extensive basis set calculation. In particular the set of closed-shell vectors is restored under control of GETQ, with the STO-3G Dumpfile used as a ‘foreign’ Dumpfile. We then perform an RHF calculation on the singlet state of Na

, using the integrals calculated in the closed-shell case.

Closed shell SCF Job

Contents of file na7mg ext.in: dumpfile ed3 500 title

* na7mg+ (4s3p//4s3p1d) scf energy=-1330.808718

mult 1 charge 1 super off noprint vectors zmat angs mg na 1 r1 na 1 r2 2 90.

na 1 r2 2 90. 3 72.

na 1 r2 2 90. 4 72.

na 1 r2 2 90. 5 72.

na 1 r2 2 90. 6 72.

na 1 r1 3 90. 2 180.

variables r1 3.0286740

r2 3.194799

end

8 EXTENDED BASIS SET CALCULATIONS OF NA

+ basis s mg

.005004

5609.67

.037083

841.969

.171495

191.263

.444597

53.2621

.480060

16.6003

s mg

.352170

2.97082

.692921

1.00728

s mg

1.00000

.113641

s mg

1.000000

.044678

p mg

.039884

50.9665

.223321

11.4364

.514536

3.21935

p mg

1.00000

0.914433

p mg

1.00000

0.16

d mg

1.00000

0.175

s na

.003064

6902.67

.022198

1059.04

.095576

255.445

.280448

77.3172

s na

.452587 26.8224

.29313

10.0718

s na

1.000000

2.17902

s na

1.000000 .689482

1.0

p na

.040274

.042422

38.9438

.229433

8.71012

.509774

2.42053

p na

1.00000

.661896

p na

1.0

.065

end maxcyc 40 vectors getq ed3 1 1 enter to run: rungamess -t ed2 -t ed3 -n na7mg na7mg_ext

Open shell RHF Job

8 EXTENDED BASIS SET CALCULATIONS OF NA

Contents of file na7mg ext orhf.in: dumpfile ed3 500 restart

# open shell scf using closed shell vectors from above *

# bypass integral evaluation * title

* na7mg+ (4s3p//4s3p1d) triplet rhf * energy=-1330.766950

super off bypass mult 1 charge 1 zmat angs mg na 1 r1 na 1 r2 2 90.

na 1 r2 2 90. 3 72.

na 1 r2 2 90. 4 72.

na 1 r2 2 90. 5 72.

na 1 r2 2 90. 6 72.

na 1 r1 3 90. 2 180.

variables r1 3.0286740

r2 3.194799

end basis s mg

.005004

5609.67

.037083

841.969

.171495

191.263

.444597

53.2621

.480060

16.6003

s mg

.352170

2.97082

.692921

1.00728

s mg

1.00000

.113641

s mg

1.000000

.044678

p mg

.039884

50.9665

.223321

11.4364

.514536

3.21935

p mg

1.00000

0.914433

p mg

1.00000

0.16

d mg

1.00000

0.175

s na

.003064

6902.67

.022198

1059.04

.095576

255.445

.280448

77.3172

9 ECP CALCULATIONS OF NA

.452587 26.8224

.29313

10.0718

s na

1.000000

2.17902

s na

1.000000 .689482

s na

1.0

p na

.040274

.042422

38.9438

.229433

8.71012

.509774

2.42053

p na

1.00000

.661896

p na

1.0

.065

end open 1 1 1 1 maxcyc 40 enter to run: rungamess -t ed2 -t ed3 -n na7mg na7mg_ext_orhf

9 ECP calculations of Na

Mg

In this example on Na

, we perform a local ECP calculation, using the Hay-Wadt ECP’s, together with the associated double zeta basis sets, augmented by a d-function on Mg. Having carried out the closed-shell SCF calculation, we perform an RHF calculation on the singlet state of Na

, using the integrals calculated in the closed-shell case. Note that we are overwriting the files from the previous example.

Closed shell SCF Job

Contents of file na7mg ecp.in: title na7mg+ LANL ecp /closed shell singlet charge 1 super off zmat angs mg na 1 r1 na 1 r2 2 90.

na 1 r2 2 90. 3 72.

na 1 r2 2 90. 4 72.

na 1 r2 2 90. 5 72.

na 1 r2 2 90. 6 72.

na 1 r1 3 90. 2 180.

9 ECP CALCULATIONS OF NA

+ variables r1 3.0286740

r2 3.194799

end basis ecpdz na ecpdz mg d mg

1.0 0.175

end ecp na na mg mg level 1.0

enter to run: rungamess -t ed2 -t ed3 -n na7mg na7mg_ecp

Open shell RHF Job

Contents of file na7mg ecp orhf.in: restart title na7mg+ ecp open-shell singlet rhf mult 1 charge 1 super off bypass zmat angs mg na 1 r1 na 1 r2 2 90.

na 1 r2 2 90. 3 72.

na 1 r2 2 90. 4 72.

na 1 r2 2 90. 5 72.

na 1 r2 2 90. 6 72.

na 1 r1 3 90. 2 180.

variables r1 3.0286740

r2 3.194799

end basis ecpdz na ecpdz mg d mg

1.0 0.175

end ecp na na mg mg runtype scf

10 GRAPHICAL ANALYSIS OF NI(CO)

4 open 1 1 1 1 level 0.3 1.0

enter to run: rungamess -t ed2 -t ed3 -n na7mg na7mg_ecp_orhf

10 Graphical analysis of Ni(CO)

The following example illustrates features of the Graphical Analysis module, in analysing the ground state SCF wavefunction of Ni(CO)

. Let us assume the following job has been used in constructing this wavefunction.

Contents of file nico4 scf.in: title\ni(co)4 .. 3-21g / SCF total energy -1947.864822 au zmat angstrom ni c 1 nic c 1 nic 2 109.471

c 1 nic 2 109.471 3 120.0

c 1 nic 2 109.471 4 120.0

x 2 1.0 1 90.0 3 180.0

o 2 co 6 90.0 1 180.0

x 3 1.0 1 90.0 2 180.0

o 3 co 8 90.0 1 180.0

x 4 1.0 1 90.0 5 180.0

o 4 co 10 90.0 1 180.0

x 5 1.0 1 90.0 4 180.0

o 5 co 12 90.0 1 180.0

variables nic 1.831

co 1.131

end level 1.5

enter to run: rungamess -t ed3=nico4.ed3 nico4_scf

Examination of the output reveals the following symmetry designation:

******************

MOLECULAR SYMMETRY

******************

MOLECULAR POINT GROUP TD

ORDER OF PRINCIPAL AXIS 0

SYMMETRY POINTS :

10 GRAPHICAL ANALYSIS OF NI(CO)

POINT 1 :

POINT 2 :

POINT 3 :

0.0000000

1.0000000

0.0000000

1.0000000

0.0000000

and the following atomic coordinates:

0.0000000

-1.9976836

1.9976836

-1.9976836

1.9976836

-3.2316433

3.2316433

-3.2316433

3.2316433

0.0000000

1.9976836

-1.9976836

1.9976836

3.2316433

-3.2316433

3.2316433

0.0000000

-1.9976836

1.9976836

-3.2316433

3.2316433

The following job may be used to construct a total density plot of the SCF wavefunction in a plane containing the Ni atom and two carbonyl groups, with the Ni at the centre of the plot: a contour plot will be generated on line printer output.

Contents of file nico4 grid.in: restart punch grid 151 title\ni(co)4 .. 3-21g / SCF total energy -1947.864822 au zmat angstrom ni c 1 nic c 1 nic 2 109.471

c 1 nic 2 109.471 3 120.0

c 1 nic 2 109.471 4 120.0

x 2 1.0 1 90.0 3 180.0

o 2 co 6 90.0 1 180.0

x 3 1.0 1 90.0 2 180.0

o 3 co 8 90.0 1 180.0

x 4 1.0 1 90.0 5 180.0

o 4 co 10 90.0 1 180.0

x 5 1.0 1 90.0 4 180.0

o 5 co 12 90.0 1 180.0

variables nic 1.831

co 1.131

end runtype analyse graphics gdef type 2d title square 2d grid ni(co)4 - total density calc type dens title ni(co)4 - total density section 151

11 2-PAIR GVB CALCULATION ON CH

(CH)

2 plot type line title ni(co)4 - total density vectors 1 enter to run: rungamess -t ed3=nico4.ed3 nico4_grid

The grid data will appear in the punchfile nico4 grid.pun

11 2-Pair GVB Calculation on CH

(CH)

NH

Closed shell SCF Job

Contents of file imino rhf.in:

# open shell rhf sto3g

# check super force title

1-imino-2,4-pentadiene * energy= -245.073475

super force nosym mult 1 accuracy 20 7 charge 1 zmat angs c c 1 r1 c 2 r2 1 a1 c 3 r3 2 a2 1 cx c 4 r4 3 a3 2 c2 n 1 r5 2 a4 3 c2 h 1 r6 2 a5 3 c1 h 2 r7 3 a6 7 c2 h 3 r8 4 a7 8 c2 h 4 r9 5 a8 9 c2 r3 r4 r5 r6 r7 r8 r9 h 5 r10 4 a9 10 c1 h 5 r11 4 a10 11 c2 h 6 r12 1 a11 7 c1 h 6 r13 1 a12 13 c2 variables r1 1.3463261

r2 1.4632685

1.3961480

1.3568062

1.3416773

1.0748794

1.0669518

1.0811641

1.0707072

11 2-PAIR GVB CALCULATION ON CH

(CH)

2 r10 r11 r12 r13

1.0729928

1.0747118

0.9971261

0.9985437

a1 118.2539432

a2 122.8419092

a3 119.3229511

a4 125.9857684

a5 119.9766875

a6 122.1120978

a7 116.3317717

a8 120.9014063

a9 121.5606571

a10 121.8124933

a11 120.9303349

a12 121.8324395

constants c1 0.

c2 180.

cx 90.

end basis sto3g runtype scf scftype rhf open 1 1 1 1 enter to run: rungamess -t ed2 -t ed3 -n imino imino_rhf

Two-Pair GVB Job

Contents of file imino gvb.in:

# gvb-scf with two pairs

# bypass integral evaluation restart bypass title

1-imino-2,4-pentadiene * energy=-245.114661

mult 1 super force nosym accuracy 20 7 charge 1 zmat angs c c 1 r1 c 2 r2 1 a1 c 3 r3 2 a2 1 cx c 4 r4 3 a3 2 c2 n 1 r5 2 a4 3 c2 h 1 r6 2 a5 3 c1 h 2 r7 3 a6 7 c2

12 DIRECT-SCF CALCULATIONS r3 r4 r5 r6 r7 r8 r9 h 3 r8 4 a7 8 c2 h 4 r9 5 a8 9 c2 h 5 r10 4 a9 10 c1 h 5 r11 4 a10 11 c2 h 6 r12 1 a11 7 c1 h 6 r13 1 a12 13 c2 variables r1 1.3463261

r2 1.4632685

1.3961480

1.3568062

1.3416773

1.0748794

1.0669518

1.0811641

1.0707072

r10 r11 r12 r13

1.0729928

1.0747118

0.9971261

0.9985437

a1 118.2539432

a2 122.8419092

a3 119.3229511

a4 125.9857684

a5 119.9766875

a6 122.1120978

a7 116.3317717

a8 120.9014063

a9 121.5606571

a10 121.8124933

a11 120.9303349

a12 121.8324395

constants c1 0.

c2 180.

cx 90.

end basis sto3g runtype scf scftype gvb 2 enter to run: rungamess -t ed2 -t ed3 -n imino imino_gvb

12 Direct-SCF Calculations

In the first example below we show a direct-SCF calculation in which the input geometry in cartesian coordinates is converted to z–matrix representation; the second calculation features geometry optimisation of Be(C

)

2 using the direct-SCF module.

Contents of file direct scf.in:

12 DIRECT-SCF CALCULATIONS

0.0532604

-0.3292201

-1.4442971

-2.9721287

-3.3279629

-5.1750163

-4.8159917

-6.8815215

-5.4889849

-7.5612331

-7.4278090

5.7357642

5.5713092

7.8017770

7.0224823

8.3530583

# direct scf generate zmatrix sto3g-basis title

* test2a * energy=-1550.28679356

charge -2 geometry distance angles torsions

-2.9512196

-0.1547624

-2.3287565

-1.0815728

-1.9700376

-1.1603816

-3.5363478

1.6367703

-0.8475562

-1.5385361

-2.8467927

-1.5624978

all

8.0

6.0

1.0

6.0

2.1566895

-0.4797763

3.5011251

-2.6669663

2.9474874

-3.6433853

1.7359097

1.7242063

1.3150929

4.1957289

-0.3417469

1.4125382

2.9182102

3.5022051

-3.4976155

-0.4827273

1.0782554

-0.1090391

1.8702785

-0.4054688

-1.1895255

1.0

7.0

1.0

6.0

1.0

8.0

6.0

3.8272799

5.9011224

7.1126500

2.1997957

1.8049798

3.9122805

4.4971112

2.7942026

6.1460386

6.0739323

4.3945431

3.1285958

1.7094915

5.1679189

6.8328410

5.6784964

-3.1681491

0.0783133

-0.6802213

-0.9534701

1.0197182

-1.8751357

-3.8035825

-1.8646838

-0.3379417

1.4363625

2.5896385

1.6502323

3.4875662

1.7463198

2.6479587

-0.1498644

1.0

8.0

1.0

6.0

1.0

6.0

1.0

8.0

6.0

1.0

6.0

8.0

6.0

1.0

5.9321304

-2.9668945

6.7897779

-1.7121829

7.7021385

-4.8380024

7.1302965

-6.7576698

7.6050313

-4.5640081

-1.5715523

-2.9124352

-3.9184431

-4.8124121

-3.1659952

-7.1913356

-6.0342521

-3.6456571

-4.4900093

-5.1460285

9.4719556

-4.5440252

9.3293820

-9.1872029

4.5891096

6.0381561

-7.5522104

end basis sto3g scftype direct enter

7.5798978

-1.2970693

-3.0617741

0.0564767

-0.3499893

6.0

8.0

6.0

1.0

2.0812894

0.9658559

1.0

8.0

0.9101853

15.0

-0.3438675

8.0

-0.4943459

8.0

3.8664887

-0.5381892

2.6957586

0.4734880

2.5757500

8.0

1.0

to run: rungamess direct_scf

Direct-SCF Geometry Optimisation of Be(C

)

2 h h o h c h c h o c o c h c h h p o o h o c h c o h h o h h h c h n h o c o c h c

12 DIRECT-SCF CALCULATIONS

Contents of file berylocene opt.in: time 60 title be(c5h5)2 sto3g optimised total energy = -394.279009 au zmatrix angstrom x x 1 fxa c 2 xc 1 xxc c 2 xc 1 xxc 3 cxc c 2 xc 1 xxc 4 cxc c 2 xc 1 xxc 3 -cxc c 2 xc 1 xxc 6 -cxc x 2 xx 3 xxc 4 -xxc h 2 hx 8 hxx 3 hcx h 2 hx 8 hxx 4 hcx h 2 hx 8 hxx 5 hcx h 2 hx 8 hxx 6 hcx h 2 hx 8 hxx 7 hcx x 2 fxt 3 xxc 4 xxc c 14 xc 1 xxc 3 cxxc c 14 xc 1 xxc 15 cxc c 14 xc 1 xxc 16 cxc c 14 xc 1 xxc 15 -cxc c 14 xc 1 xxc 18 -cxc x 14 xx 15 xxc 16 -xxc h 14 hx 20 hxx 15 hcx h 14 hx 20 hxx 16 hcx h 14 hx 20 hxx 17 hcx h 14 hx 20 hxx 18 hcx h 14 hx 20 hxx 19 hcx be 2 fxa 3 xxc 4 xxc variables fxa 1.47

fxt 3.37

xc hx

1.22

2.12

hxx 88.4

constants cxc 72.0

cxxc 36.0

xxc 90.0

hcx 0.0

xx 1.0

end basis sto3g runtype optimize scftype direct level 2.0 10 1.4

enter to run: rungamess berylocene_opt

13 HCN/HNC TRANSITION STATE LOCATION 29

13 HCN/HNC Transition State Location

Transition state calculation for the HCN,HNC isomerisation process. The first job uses the default trust region algorithm, the second the synchronous transit algorithm, and the third the

Jorgensen-Simons algorithm.

Contents of file hcn tr.in: title hcn 4-31G saddle point zmat angs c x 1 1.0

n 1 cn 2 90.0

h 1 ch 2 90.0 3 hcn variables cn 1.1484 type 3 ch 1.5960 type 3 hcn 90.0 type 3 end basis 4-31g runtype saddle enter to run: rungamess hcn_tr

Saddle point for HCN using the synchronous transit algorithm - note the definition of the minima required (on the variable definition lines) and the LSEARCH directive. The default saddle point method does nor require minima definition (see above)

Contents of file hcn st.in: title hcn saddle point - synchronous transit zmat angs c x 1 1.0

n 1 cn 2 90.0

h 1 ch 2 90.0 3 hcn variables cn 1.1484 minima 1.1371 1.1597

ch 1.5960 minima 1.0502 2.1429

hcn 90.0

minima 180.0

0.0

end basis 4-31g runtype saddle lsearch 0 4 enter to run: rungamess hcn_st

14 HSIP/HPSI TRANSITION STATE LOCATION

Saddle point for HCN using the Jorgensen-Simons algorithm.

Contents of file hcn js.in: title hcn/hnc ts search . jorgensen-simons zmat angs c x 1 1.0

n 1 cn 2 90.0

h 1 ch 2 90.0 3 hcn variables cn 1.1484 type 3 ch 1.5960 type 3 hcn 90.0

type 3 end basis 4-31g runtype saddle jorgensen powell maxjor 55 recalc off rfo off cutoffs optprint on xtol 0.0018

enter to run: rungamess hcn_js

14 HSiP/HPSi Transition State Location

This example is concerned with locating the transition state in the HPSi, HSiP isomerisation process, and calculating the associated vibrational frequencies. We provide sample jobs using both numerical and analytical techniques in the transition state location and subsequent force constant evaluation. Note that latter example is computationally the most efficient, and should certainly be adopted for small–medium sized molecules.

14.1 Numerical Force Constants

In the first step we perform an initial SCF for subsequent use in the saddle point calculation.

Closed shell SCF Job

Contents of file hsip scf.in: title psih saddle point

14 HSIP/HPSI TRANSITION STATE LOCATION 31 zmat ang p x 1 1.0

si 1 psi 2 90.0

h 1 ph 2 90.0 3 hpsi variables psi 2.053 type 3 ph 2.44

type 3 hpsi 51.02 type 3 end enter to run: rungamess -t ed2 -t ed3 -n hsip hsip_scf

In the subsequent location of the transition state, note the use of TYPE 3 which causes the program to calculate the complete force constant matrix numerically before commencing the search for the saddle point, and the use of XTOL to provide more stringent optimisation criteria in view of the subsequent force constant evaluation. LOCK is used to retain the initial SCF configuration throughout the search.

Transition State Job

Contents of file hsip ts.in: restart new title psih <-> hpsi saddle point bypass zmat angs p x 1 1.0

si 1 psi 2 90.0

h 1 ph 2 90.0 3 hpsi variables psi 2.053 type 3 ph 2.44

type 3 hpsi 51.02 type 3 end runtype saddle lock xtol 0.0005

enter 2 to run: rungamess -t ed2 -t ed3 -n hsip hsip_ts

Finally we present the job for numerical evaluation of the force constants at the optimised geometry. Note the use of restart in requesting usage of the geometry from the Dumpfile, rather

14 HSIP/HPSI TRANSITION STATE LOCATION 32 than from the data file.

Numerical Force Constant Job

Contents of file hsip fc.in: restart title psih <-> hpsi saddle point numerical fcm zmat angs p x 1 1.0

si 1 psi 2 90.0

h 1 ph 2 90.0 3 hpsi variables psi 2.053 type 3 ph 2.44

type 3 hpsi 51.02 type 3 end runtype force vectors 2 lock enter 3 to run: rungamess -t ed2 -t ed3 -n hsip hsip_fc

14.2 Analytic Force Constants

In the first step we perform the computation of the trial hessian under RUNTYPE HESSIAN control for subsequent use in the saddle point calculation.

Computing the trial Hessian

Contents of file hsip fcm1.in: title psih trial hessian / SCF zmat ang p x 1 1.0

si 1 psi 2 90.0

h 1 ph 2 90.0 3 hpsi variables psi 2.053

ph 2.44

hpsi 51.02

end runtype hessian enter

14 HSIP/HPSI TRANSITION STATE LOCATION 33 to run: rungamess -t ed2 -t ed3 -n hsip hsip_fcm1

In the subsequent location of the transition state, note the use of the FCM keyword on the

RUNTYPE data line to restore the trial hessian computed in the first job and the use of XTOL to provide more stringent optimisation criteria in view of the subsequent force constant evaluation. LOCK is used to retain the initial SCF configuration throughout the search.

Transition State Job

Contents of file hsip tsfcm.in: restart new title psih <-> hpsi saddle point location / using trial hessian zmat angs p x 1 1.0

si 1 psi 2 90.0

h 1 ph 2 90.0 3 hpsi variables psi 2.053

ph 2.44

hpsi 51.02

end runtype saddle fcm xtol 0.0005

vectors 1 lock enter 2 to run: rungamess -t ed2 -t ed3 -n hsip hsip_tsfcm

Finally we present the job for analytic computation of the force constants at the optimised geometry under control of runtype hessian. Note the use of restart in requesting usage of the geometry from the Dumpfile, rather than from the data file.

Analytic Force Constant Job

Contents of file hsip fcm2.in: restart title psih <-> hpsi saddle point / force constants zmat angs p

15 USE OF BOND-CENTRED FUNCTIONS 34 to run: x 1 1.0

si 1 psi 2 90.0

h 1 ph 2 90.0 3 hpsi variables psi 2.053

ph 2.44

hpsi 51.02

end runtype hessian vectors 2 lock enter 2 rungamess -t ed2 -t ed3 -n hsip hsip_fcm2

15 Use of Bond-centred Functions

In this example we demonstrate the use of bond-centred functions (s,p), cited at the mid-point of the C-N bond in the HCN, HNC transition state.

Contents of file hcn bf.in: title hcn-hnc basis - dunning (9s5p-3s2p) bond(s,p) + p(h) zmat angstrom c bq 1 rcn2 x 2 1.0 1 90.0

n 2 rcn2 3 90.0 1 180.0

x 1 1.0 2 90.0 3 0.0

h 1 rch 5 90.0 4 phi variables rcn2 0.5991

rch 1.2128

phi 71.2

end basis sv h p h

1.0 0.7

s bq

1.0 1.0

p bq

1.0 0.7

sv c sv n end enter to run: rungamess hcn_bf

16 SCF ANALYTIC FORCE CONSTANTS FOR C

16 SCF Analytic Force Constants for C

H

We consider below computing the analytic force constants for C

, initially optimising the molecule at the SCF level, followed by the force constant calculation. Note the use of the

XTOL directive in the optimisation job to ensure a higher degree of optimisation than that derived using the default XTOL.

Geometry Optimisation

Contents of file ethene opt.in: title ethylene 6-31g** geometry optimisation zmatrix angstrom c c 1 cc h 1 ch 2 hcc h 1 ch 2 hcc 3 180.0

h 2 ch 1 hcc 3 0.0

h 2 ch 1 hcc 3 180.0

variables cc 1.40

ch 1.10

hcc 118.0

end basis 6-31g** runtype optimize xtol 0.0001

enter to run: rungamess -t ed3=ethene.ed3 ethene_opt

Analytic Force Constants

Contents of file ethene fcm.in: restart title ethylene 6-31g** ground state vibrational frequencies zmatrix angstrom c c 1 cc h 1 ch 2 hcc h 1 ch 2 hcc 3 180.0

h 2 ch 1 hcc 3 0.0

h 2 ch 1 hcc 3 180.0

variables cc 1.40

ch 1.10

17 MP2 ANALYTIC FORCE CONSTANTS FOR C

4 hcc 118.0

end basis 6-31g** runtype hessian enter to run: rungamess -t ed3=ethene.ed3 ethene_fcm

17 MP2 Analytic Force Constants for C

H

We consider below computing the analytic force constants for C

, initially optimising the molecule at the MP2 level, followed by the force constant calculation. Note again the use of the XTOL directive in the optimisation job to ensure a higher degree of optimisation than that derived using the default XTOL.

MP2 Geometry Optimisation

Contents of file ethene mp2opt.in: title ethylene 6-31g** MP2/ optimised total energy = -78.3272309

zmatrix angstrom c c 1 cc h 1 ch 2 hcc h 1 ch 2 hcc 3 180.0

h 2 ch 1 hcc 3 0.0

h 2 ch 1 hcc 3 180.0

variables cc 1.40

ch 1.10

hcc 118.0

end basis 6-31g** runtype optimize scftype mp2 xtol 0.0001

enter to run: rungamess -t ed3=ethene.ed3 ethene_mp2opt

MP2 Analytic Force Constants

Contents of file ethene mp2fcm.in:

18 MP2 POLARISABILITY FOR C

4 restart title ethylene MP2/6-31g** ground state vibrational frequencies

#freq 847.7, 941.6, 994.1, 1091.6, 1267.4, 1414.5,

#freq 1525.2, 1729.7, 3241.3, 3259.2, 3336.6, 3359.8

zmatrix angstrom c c 1 cc h 1 ch 2 hcc h 1 ch 2 hcc 3 180.0

h 2 ch 1 hcc 3 0.0

h 2 ch 1 hcc 3 180.0

variables cc 1.40

ch 1.10

hcc 118.0

end basis 6-31g** runtype hessian scftype mp2 enter to run: rungamess -t ed3=ethene.ed3 ethene_mp2fcm

18 MP2 Polarisability for C

H

We consider below computing the molecular polarisability of C

, initially optimising the molecule at the MP2 level, followed by the property calculation. Note again the use of the

XTOL directive in the optimisation job to ensure a higher degree of optimisation than that derived using the default XTOL.

MP2 Geometry Optimisation

Contents of file ethene mp2opt.in: title ethylene 6-31g** MP2/ optimised total energy = -78.3272309

zmatrix angstrom c c 1 cc h 1 ch 2 hcc h 1 ch 2 hcc 3 180.0

h 2 ch 1 hcc 3 0.0

h 2 ch 1 hcc 3 180.0

variables cc 1.40

ch 1.10

hcc 118.0

19 DIRECT-MP2 CALCULATION OF C

N end basis 6-31g** runtype optimize scftype mp2 xtol 0.0001

enter to run: rungamess -t ed3=ethene.ed3 ethene_mp2opt

MP2 Polarisability

Contents of file ethene mp2pol.in: restart title ethylene MP2/6-31g** ground state polarisability zmatrix angstrom c c 1 cc h 1 ch 2 hcc h 1 ch 2 hcc 3 180.0

h 2 ch 1 hcc 3 0.0

h 2 ch 1 hcc 3 180.0

variables cc 1.40

ch 1.10

hcc 118.0

end basis 6-31g** runtype polarisability scftype mp2 enter to run: rungamess -t ed3=ethene.ed3 ethene_mp2pol

19 Direct-MP2 Calculation of C

H

N

We show below the data for performing a direct–MP2 calculation on the C

N molecule, conducted in a 6–31G* basis set. Note the use of the MEMORY pre-directive in requesting a memory allocation of 4 MWords.

Contents of file pyridine.in: title pyridine 6-31g* direct-mp2 zmat angstrom n

20 CASSCF GEOMETRY OPTIMISATIONS x 1 1.0

x 1 1.0 2 90.

x 1 1.0 2 90. 3 90.

c 1 c4n 3 90. 2 180.

x 5 1.0 1 90. 3 0.0

x 5 1.0 1 90. 4 0.0

h 5 ch4 6 90. 1 180.

c 1 c2n 2 c2nz 3 180.

c 1 c2n 2 c2nz 3 0.0

c 9 c2c3 1 ccn 2 180.

c 10 c2c3 1 ccn 2 180.

h 9 c2h6 1 nch2 2 0.0

h 10 c2h6 1 nch2 2 0.0

h 11 c3h5 9 c2c3h 1 180.

h 12 c3h5 10 c2c3h 1 180.

variables c4n 2.7845546

ch4 1.0823078

c2n 1.3372389

c2nz 120.641858

c2c3 1.3944571

ccn 122.662269

c2h6 1.0814291

c3h5 1.0809550

nch2 116.400433

c2c3h 120.158516

end basis 6-31g* scftype direct mp2 enter to run: rungamess -t ed3 pyridine

20 CASSCF Geometry Optimisations

We consider below a CASSCF calculation on the X

1 state of H

O, using a full valence criterion in specifying the active space so that the formally vacant SCF virtual MOs, 4a

1 and

, are permitted variable occupancy. This example utilises the vectors from the closed shell

SCF calculation of Example 1.

Contents of file water cas.in

restart new title water at casscf level 3-21g basis set zmat angstrom o h 1 oh h 1 oh 2 hoh variables oh 0.956

20 CASSCF GEOMETRY OPTIMISATIONS 40 hoh 104.5

end scftype casscf config print doc 1 to 5 uoc 6 7 end superci 1 to 8 newton 9 to 20 hessian 9 to 20 simul 9 to 20 enter to run: rungamess -t ed3 -n water water_cas

The following points should be noted:

• It is not possible to use BYPASS in the above, given the data for the SCF job of Example

1. This would have resulted in generation of a P-supermatrix which is not usable in a

CASSCF run (see Part 2, Table 1).

• CASSCF calculations require two scratch FORTRAN data sets, FT01 and FT02.

In the above we have assumed that the CASSCF calculation completes in the time allocated, with the associated direct-access files allocated in default scratch status. The following rungamess invocation is typical of that required if restarts of the CASSCF step are envisaged, here we are using the –r argument, which will cause all datasets required for a casscf (scf or optimise) restart calculation (specifically ed1 ed2 ed4 ed6 ed9 ed10, and ed11) to be saved.

rungamess -r casscf -n water water_cas

Assuming the above job terminated prior to convergence, the calculation might be restarted as follows:

Contents of file water cas rest.in: restart scf title water at casscf level 3-21g basis set zmat angstrom o h 1 oh h 1 oh 2 hoh variables oh 0.956

hoh 104.5

end scftype casscf config bypass

21 CASSCF + 2ND-ORDER CI CALCULATIONS ON BEO 41 doc 1 to 5 uoc 6 7 end superci 1 to 2 newton 3 to 20 hessian 3 to 20 simul 3 to 20 vectors 6 7 enter 6 7 to run: rungamess -r casscf -n water water_cas_rest where the default sections housing the CASSCF vectors and ci coefficients (sections 6 and 7 respectively) created in the startup job are explicitly declared above, and CONFIG processing is bypass’ed. Having completed the single point calculation, the following might be used to perform a geometry optimisation at the CASSCF level.

Contents of file water cas opt.in: restart new title water - geometry optimisation at casscf level zmat angstrom o h 1 oh h 1 oh 2 hoh variables oh 0.956

hoh 104.5

end runtype optimise scftype casscf config bypass doc 1 to 5 uoc 6 7 end superci 1 to 5 newton 6 to 20 hessian 6 to 20 simul 6 to 20 enter to run: rungamess -r casscf -n water water_cas_opt

21 CASSCF + 2nd-order CI Calculations on BeO

First, we consider below a CASSCF calculation on the X

+ configuration state of BeO, characterised by the

1σ

2σ

3σ

4σ

1π

(1)

21 CASSCF + 2ND-ORDER CI CALCULATIONS ON BEO 42

The initial closed–shell SCF using a DZP basis was conducted with the following job:

Contents of file beo rhf.in: super off nosym title\beo .. dzp zmat angstrom\be\o 1 beo\ variables\beo 1.300 hessian 0.7\end basis dzp enter to run: rungamess -t ed2 -t ed3 -n beo beo_rhf

An examination of the closed-shell SCF output reveals the following symmetry adapted basis information

=============================

IRREP NO. OF SYMMETRY ADAPTED

BASIS FUNCTIONS

=============================

1 17

============================= and the SCF MO ordering shown below:

===============================================

M.O.

IRREP ORBITAL ENERGY ORBITAL OCCUPANCY

===============================================

1 1 -20.45769692

2.0000000

-4.72825831

-1.15792230

-0.46629250

-0.39257378

-0.05704423

0.09936415

2.0000000

0.0000000

0.09936415

0.15573442

0.25156032

0.29518660

0.57290305

0.66957865

0.0000000

0.66957865

0.84348942

1.01733643

1.05237116

1.24301610

0.0000000

21 CASSCF + 2ND-ORDER CI CALCULATIONS ON BEO 43

1.45883277

1.80402496

2.36534068

0.0000000

2.62307878

3.16009549

4.37857805

45.42979631

0.0000000

===============================================

We wish to perform a CASSCF calculation in which the inner shell and O2s orbitals (the 1σ–3σ) remain doubly occupied, with the active space including the formally vacant SCF virtual MOs, the 5σ and 2π. This example utilises the vectors from the closed shell SCF calculation. We wish to perform the CASSCF calculation under RUNTYPE CI specification where, having performed the 6 electrons in 6 orbital CASSCF, we use the natural orbitals in carrying out a second-order

CI using the Direct-CI module. Specifically we aim to use a reference space in the CI consisting of all CSFs which can be generated by distributing 6 electrons in 6 MOs i.e. the CASSCF space.

This may be achieved in a single run through the following job specification:

Contents of file beo casscf ci.in: restart new title\beo .. dzp casscf+ci (6 electrons in 6 mos) bypass zmat angstrom\be\o 1 beo\ variables\beo 1.300 hessian 0.7\end basis dzp runtype ci active\4 to 31\end core\1 to 3\end scftype casscf thresh 4 config print fzc 1 to 3 doc 4 to 6 uoc 7 to 9 end superci 1 to 8 newton 9 to 20 hessian 9 to 20 simul 9 to 20 direct 6 6 22 conf

2 2 2 0 0 0 refgen

1 4 1 5 1 6 2 4 2 5 2 6 3 4 3 5 3 6 refgen

22 MCSCF + 2ND-ORDER CI CALCULATIONS ON BEO

1 4 1 5 1 6 2 4 2 5 2 6 3 4 3 5 3 6 refgen

1 4 1 5 1 6 2 4 2 5 2 6 3 4 3 5 3 6 enter to run: rungamess -t ed2 -t ed3 -n beo beo_casscf_ci

If restarts are envisaged, the following invocation would be suitable: rungamess -r ci -r casscf -n beo beo_casscf_ci

22 MCSCF + 2nd-order CI Calculations on BeO

First, we consider below a MCSCF calculation on the X

+ configuration state of BeO, characterised by the

1σ

2σ

3σ

4σ

1π

(2)

The initial closed–shell SCF using a DZP basis was conducted with the following job:

Contents of file beo rhf.in: super off nosym title\beo .. dzp zmat angstrom\be\o 1 beo\ variables\beo 1.300 hessian 0.7\end basis dzp enter to run: rungamess -t ed2 -t ed3 -n beo beo_rhf

An examination of the closed-shell SCF output reveals the following symmetry adapted basis information

=============================

IRREP NO. OF SYMMETRY ADAPTED

BASIS FUNCTIONS

=============================

1 17

============================= and the SCF MO ordering shown below:

22 MCSCF + 2ND-ORDER CI CALCULATIONS ON BEO 45

===============================================

M.O.

IRREP ORBITAL ENERGY ORBITAL OCCUPANCY

===============================================

1 1 -20.45769692

2.0000000

-4.72825831

-1.15792230

-0.46629250

-0.39257378

2.0000000

-0.05704423

0.09936415

0.15573442

0.25156032

0.29518660

0.0000000

0.57290305

0.66957865

0.84348942

1.01733643

1.05237116

1.24301610

0.0000000

1.45883277

1.80402496

2.36534068

2.62307878

3.16009549

0.0000000

4.37857805

45.42979631

0.0000000

===============================================

We wish to perform a CASSCF calculation in which the inner shell and O2s orbitals (the 1σ–3σ) remain doubly occupied, with the active space including the the formally vacant SCF virtual

MOs, the 5σ and 2π. This example utilises the vectors from the closed shell SCF calculation.

Contents of file beo mcscf.in: restart new super off nosym title\beo .. dzp bypass zmat angstrom\be\o 1 beo\ variables\beo 1.300 hessian 0.7\end basis dzp scftype mcscf thresh 4 mcscf orbital\3cor1 doc1 doc3 doc2 uoc1 uoc2 uoc3 \end enter to run:

22 MCSCF + 2ND-ORDER CI CALCULATIONS ON BEO 46 rungamess -t ed2 -t ed3 -n beo beo_mcscf

The following points should be noted:

• The MCSCF Natural orbitals are routed to section 10 of the Dumpfile on convergence, the default section used for NO output.

• Integral evaluation has been bypassed as the initial SCF job specified the necessary integral format for the subsequent SCF.

In the above we have assumed that the MCSCF calculation completes in the time allocated, with the associated direct-access files allocated in default scratch status. The following rungamess invocation is typical of that required if restarts of the MCSCF step are envisaged: rungamess -r mcscf -n beo beo_mcscf

Note that the symmetry adapted list of integrals are sorted at the outset of processing to ED13, and this file should be preserved across restart jobs, given that the DONT SORT data line is presented to the restart job. Assuming the above job terminated prior to convergence, the calculation might be restarted as follows:

Contents of file beo mcscf rest.in

restart scf super off nosym title\beo .. dzp zmat angstrom\be\o 1 beo\ variables\beo 1.300 hessian 0.7\end basis dzp scftype mcscf thresh 4 mcscf orbital\3cor1 doc1 doc3 doc2 uoc1 uoc2 uoc3 \end dont sort enter to run: rungamess -r mcscf -n beo beo_mcscf_rest

Now let us consider performing the MCSCF calculation under RUNTYPE CI specification where, having performed the 6 electrons in 6 orbital CASSCF, we use the natural orbitals in carrying out a second-order CI using the Direct-CI module. Specifically we aim to use a reference space in the CI consisting of all CSFs which can be generated by distributing 6 electrons in 6 MOs i.e.

the CASSCF space. This may be achieved in a single run through the following job specification:

Contents of file beo mcscf ci.in

restart new super off nosym

23 TABLE-CI CALCULATIONS ON THE AMMONIA CATION 47 title\beo .. dzp mcscf+2nd-order ci (6 electrons in 6 mos) zmat angstrom\be\o 1 beo\ variables\beo 1.300 hessian 0.7\end basis dzp runtype ci active\4 to 31\end core\1 to 3\end scftype mcscf thresh 4 mcscf orbital\3cor1 doc1 doc3 doc2 uoc1 uoc2 uoc3 \end direct 6 6 22 conf

2 2 2 0 0 0 refgen

1 4 1 5 1 6 2 4 2 5 2 6 3 4 3 5 3 6 refgen

1 4 1 5 1 6 2 4 2 5 2 6 3 4 3 5 3 6 vectors 1\enter 20 21 to run: rungamess -t ed3 -n beo beo_mcscf_ci

Note that the vectors specification is requesting that the closed-shell SCF eigenvectors be used to initiate the MCSCF calculation. The MCSCF natural orbitals, routed to section 20 of the

Dumpfile, will be used as the orbitals for the Direct-CI calculation.

23 Table-CI calculations on the Ammonia Cation

We consider below a Table-CI calculation of the X

1 state and 1

1 state of the ammonia cation. In the first instance we consider performing the calculation in two steps, initially the open-shell SCF calculation followed by the MRD-CI 2-reference state calculation. We then subdivide the CI calculation into 5 separate steps, performing the symmetry adaptation + integral transformation, followed by configuration selection, construction of the CI Hamiltonian, the diagonalisation and, finally, the subsequent analysis of the CI wavefunctions.

Open-shell SCF Job

Contents of file nh3 rohf: title

23 TABLE-CI CALCULATIONS ON THE AMMONIA CATION nh3+ * 3-21g * scf-energy=-55.53325817 hartree super off nosym charge 1 mult 2 zmat angstrom n h 1 roh h 1 roh 2 theta h 1 roh 2 theta 3 theta 1 variables roh 1.03

hessian 0.7

theta 104.2

hessian 0.2

end enter to run: rungamess -t ed2 -t ed3 -n nh3 nh3_rohf

Table-CI Job

Contents of file nh3 mrdci.in: restart title

* nh3+ * 3-21g * mrdci-energies 1r -55.6393336 2r -55.4116210

bypass scf charge 1 mult 2 zmat angstrom n h 1 roh h 1 roh 2 theta h 1 roh 2 theta 3 theta 1 variables roh 1.03

hessian 0.7

theta 104.2

hessian 0.2

end runtype ci mrdci adapt tran table select symmetry 1 spin 2 cntrl 9 conf

1 4 1 2 3 12

1 3 1 2 4 12 roots 2 thresh 5 5 ci diag

23 TABLE-CI CALCULATIONS ON THE AMMONIA CATION 49 extrap 3 dthr 0.0001 0.0001

natorb cive 1 2 prop cive 1 2

1 4 1 2 3 12

1 3 1 2 4 12 moment

36 1 36 2 1 enter to run: rungamess -t ed2 -t ed3 -t table -n nh3 nh3_mrdci

The orbitals employed in the CI calculation will be taken from the default section associated with the open-shell RHF module, section 5, that containing the energy-ordered canonicalised open-shell vectors written on termination of the SCF process. Note that the table keyword will activate generation of the Table-CI data base, to be used in the subsequent steps below. In this case the data base will be written to the file nh3.table in the temporary directory for subsequent use. Note that on most installations a copy of the data base will already be available and some time is saved if recomputation is avoided. It is accessed with the –library ( or –l ) option on the rungamess command line. In this case the table keyword must be removed.

rungamess -t ed2 -t ed3 -l table -n nh3 nh3_mrdci

Now let us consider dividing the above CI calculation. The following points should be noted in this division:

• the “-r mrdci” option has been used to save all files that must be retained between separate runs of the program. The option “-t ed2” is now implicit and so may be omitted. We assume below that the Table-CI data-base is available from the previous job, thus omitting the TABLE step.

• Note the use of the BYPASS keyword on the various steps comprising the Table-CI procedure. Such a keyword is required on both those steps already completed and those steps to be handled in a subsequent run of the program.

Table-CI Data I. Symmetry Adaption and Integral Transformation

Contents of file nh3 sa tran.in: restart title

* nh3+ * 3-21g * mrdci-energies 1r -55.6393336 2r -55.4116210

bypass scf charge 1 mult 2

23 TABLE-CI CALCULATIONS ON THE AMMONIA CATION zmat angstrom n h 1 roh h 1 roh 2 theta h 1 roh 2 theta 3 theta 1 variables roh 1.03

hessian 0.7

theta 104.2

hessian 0.2

end runtype ci mrdci adapt tran select bypass symmetry 1 spin 2 cntrl 9 conf

1 4 1 2 3 12

1 3 1 2 4 12 roots 2 thresh 5 5 ci bypass diag bypass extrap 3 dthr 0.0001 0.0001

enter to run: rungamess -r mrdci -t table -n nh3 nh3_sa_tran

Table-CI Job II. Configuration Selection

Contents of file nh3 select.in: restart ci title

* nh3+ * 3-21g * mrdci-energies 1r -55.6393336 2r -55.4116210

bypass scf charge 1 mult 2 zmat angstrom n h 1 roh h 1 roh 2 theta h 1 roh 2 theta 3 theta 1 variables roh 1.03

hessian 0.7

theta 104.2

hessian 0.2

end runtype ci mrdci adapt bypass

23 TABLE-CI CALCULATIONS ON THE AMMONIA CATION tran bypass select symmetry 1 spin 2 cntrl 9 conf

1 4 1 2 3 12

1 3 1 2 4 12 roots 2 thresh 5 5 ci bypass diag bypass extrap 3 dthr 0.0001 0.0001

enter to run: rungamess -r mrdci -t table -n nh3 nh3_select

Table-CI Job III. CI Hamiltonian Construction

Contents of file nh3 hamil.in: restart ci title

* nh3+ * 3-21g * mrdci-energies 1r -55.6393336 2r -55.4116210

bypass scf charge 1 mult 2 zmat angstrom n h 1 roh h 1 roh 2 theta h 1 roh 2 theta 3 theta 1 variables roh 1.03

hessian 0.7

theta 104.2

hessian 0.2

end runtype ci mrdci adapt bypass tran bypass select bypass symmetry 1 spin 2 cntrl 9 conf

1 4 1 2 3 12

1 3 1 2 4 12 roots 2 thresh 5 5 ci diag bypass

23 TABLE-CI CALCULATIONS ON THE AMMONIA CATION extrap 3 dthr 0.0001 0.0001

enter to run: rungamess -r mrdci -t table -n nh3 nh3_hamil

Table-CI Job IV. Diagonalisation

Contents of file nh3 diag.in: restart ci title

* nh3+ * 3-21g * mrdci-energies 1r -55.6393336 2r -55.4116210

bypass scf charge 1 mult 2 zmat angstrom n h 1 roh h 1 roh 2 theta h 1 roh 2 theta 3 theta 1 variables roh 1.03

hessian 0.7

theta 104.2

hessian 0.2

end runtype ci mrdci adapt bypass tran bypass select bypass symmetry 1 spin 2 cntrl 9 conf

1 4 1 2 3 12

1 3 1 2 4 12 roots 2 thresh 5 5 ci bypass diag extrap 3 dthr 0.0001 0.0001

enter to run: rungamess -r mrdci -t table -n nh3 nh3_diag

Table-CI Job V. CI Wavefunction Analysis

Contents of file nh3 analy.in:

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

2 restart ci title

* nh3+ * 3-21g * mrdci-energies 1r -55.6393336 2r -55.4116210

bypass scf charge 1 mult 2 zmat angstrom n h 1 roh h 1 roh 2 theta h 1 roh 2 theta 3 theta 1 variables roh 1.03

hessian 0.7

theta 104.2

hessian 0.2

end runtype ci mrdci adapt bypass tran bypass select bypass symmetry 1 spin 2 cntrl 9 conf

1 4 1 2 3 12

1 3 1 2 4 12 roots 2 thresh 5 5 ci bypass diag bypass extrap 3 dthr 0.0001 0.0001

natorb cive 1 2 prop cive 1 2

1 4 1 2 3 12

1 3 1 2 4 12 moment

36 1 36 2 1 enter to run: rungamess -r mrdci -t table -n nh3 nh3_analy

24 ECP, CASSCF and Direct-CI Calculations on NiCCH

This example illustrates the use of CASSCF and Direct-CI calculations in the framework of

ECP studies. The molecular system under investigation is NiCCH

, with a 5-reference direct-CI calculation performed using a CASSCF wavefunction for the lowest triplet state. Five data files are presented below:

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

1. Start–up closed–shell SCF calculation for the

1 state. Note the SUPER directive for compatibility with the subsequent open–shell calculation. The ECP library file /scr1/gamess/GAMESS-

UK/libs/ecplib (on the HP735 at Daresbury) is allocated to LFN ed0, with the NIHAY and C non-local ECPs requested under control of the PSEUDO directive.

2. Restart SCF job, with appropriate use of the SWAP directive to converge the closed–shell

SCF.

3. RHF calculation for the

1 state.

4. CASSCF calculation for the

1 state.

5. 5-reference Direct-CI calculation for the

1 state.

Closed-shell SCF Start-up Job

Contents of file nicch2 rhf.in

title ni(cch2) 1a1 rhf ,hay’s ni, bar’s nm ecp’s mult 1 super force zmat angstrom ni c 1 nica x 2 1.0 1 90.0

c 2 cacb 3 90.0 1 180.0

x 4 1.0 2 90.0 3 0.0

h 4 hcb 2 hcc 5 90.0

h 4 hcb 2 hcc 5 -90.0

variables nica 2.0895

cacb 1.3604

hcb 1.1047

hcc 122.646

end basis s h

0.032828

0.231208

0.817238

s h

1.000000

s c

13.3615

2.0133

0.4538

0.1233

1.000000

s c

1.000000

p c

0.018534

0.115442

0.386206

0.640089

p c

0.4962

0.1533

18.1557

3.9864

1.1429

0.3594

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

1.000000

s ni

0.1146

-0.4372528 0.6778

1.1889453 0.1116

s ni

1.0000000 0.0387

p ni

1.0000000 0.0840

p ni

1.0000000 0.0240

d ni

0.0360414 42.7200

0.1938645 11.7600

0.4596238 3.8170

0.5599305 1.1690

d ni

1.0000000 0.2836

end pseudo nonlocal nihay ni c c runtype scf maxcyc 30 level 2.5 15 1.0

vectors hcore enter to run: rungamess -t ed2 -t ed3 -l ed0 -n nicch2 nicch2_rhf

Closed-shell SCF Restart Data

Contents of file nicch2 swap.in

restart new title ni(cch2) 1a1 rhf restart ,hay’s ni, bar’s nm ecp’s mult 1 bypass super force zmat angstrom ni c 1 nica x 2 1.0 1 90.0

c 2 cacb 3 90.0 1 180.0

x 4 1.0 2 90.0 3 0.0

h 4 hcb 2 hcc 5 90.0

h 4 hcb 2 hcc 5 -90.0

variables nica 2.0895

cacb 1.3604

hcb 1.1047

hcc 122.646

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

2 end basis s h

0.032828

0.231208

0.817238

s h

1.000000

s c

13.3615

2.0133

0.4538

0.1233

1.000000

s c

1.000000

p c

0.018534

0.115442

0.386206

0.4962

0.1533

18.1557

3.9864

1.1429

0.3594

0.640089

p c

1.000000

s ni

0.1146

-0.4372528 0.6778

1.1889453 0.1116

s ni

1.0000000 0.0387

p ni

1.0000000 0.0840

p ni

1.0000000 0.0240

d ni

0.0360414 42.7200

0.1938645 11.7600

0.4596238 3.8170

0.5599305 1.1690

d ni

1.0000000 0.2836

end pseudo nonlocal nihay ni c c runtype scf maxcyc 40 level 3.0 10 1.0

swap

4 5

6 8

8 10

10 11 end enter to run: rungamess -t ed2 -t ed3 -l ed0 -n nicch2 nicch2_swap

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

Open-shell SCF Job

Contents of file nicch2 orhf.in

restart new title ni(cch2) 3a1 rhf ,hay’s ni, bar’s nm ecp’s mult 3 super force bypass zmat angstrom ni c 1 nica x 2 1.0 1 90.0

c 2 cacb 3 90.0 1 180.0

x 4 1.0 2 90.0 3 0.0

h 4 hcb 2 hcc 5 90.0

h 4 hcb 2 hcc 5 -90.0

variables nica 2.0895

cacb 1.3604

hcb 1.1047

hcc 122.646

end basis s h

0.032828

0.231208

0.817238

s h

1.000000

s c

1.000000

s c

1.000000

p c

13.3615

2.0133

0.4538

0.1233

0.4962

0.1533

0.018534

0.115442

0.386206

0.640089

p c

18.1557

3.9864

1.1429

0.3594

1.000000

s ni

0.1146

-0.4372528 0.6778

1.1889453 0.1116

s ni

1.0000000 0.0387

p ni

1.0000000 0.0840

p ni

1.0000000 0.0240

d ni

0.0360414 42.7200

0.1938645 11.7600

0.4596238 3.8170

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

0.5599305 1.1690

d ni

1.0000000 0.2836

end pseudo nonlocal nihay ni c c runtype scf scftype gvb maxcyc 50 level 2.0 3.0 15 1.0 1.0

open 2 2 swap

11 12

14 15 end enter to run: rungamess -t ed2 -t ed3 -l ed0 -n nicch2 nicch2_orhf

Eigenvector utilisation in the above job will drive off the default sections of the Dumpfile, with the open-shell SCF module using the closed-shell SCF vectors from section 1 to initiate the SCF process, and writing the SCF open-shell orbitals to sections 4 and 5 (the energy-ordered SCF

MOs). These latter orbitals will be used below to instigate the CASSCF processing.

CASSCF Job

Contents of file nicch2 cas.in

restart new title ni(cch2) 3a1 c2v cas at opt (3 2 2 0), hay’s ni, bar’s nm ecp’s mult 3 super off nosym zmat angstrom ni c 1 nica x 2 1.0 1 90.0

c 2 cacb 3 90.0 1 180.0

x 4 1.0 2 90.0 3 0.0

h 4 hcb 2 hcc 5 90.0

h 4 hcb 2 hcc 5 -90.0

variables nica 2.0895

cacb 1.3604

hcb 1.1047

hcc 122.646

end basis s h

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

0.032828

0.231208

0.817238

s h

13.3615

2.0133

0.4538

1.000000

s c

1.000000

s c

1.000000

0.1233

0.4962

0.1533

p c

0.018534

0.115442

0.386206

18.1557

3.9864

1.1429

0.3594

0.640089

p c

1.000000

0.1146

s ni

-0.4372528 0.6778

1.1889453 0.1116

s ni

1.0000000 0.0387

p ni

1.0000000 0.0840

p ni

1.0000000 0.0240

d ni

0.0360414 42.7200

0.1938645 11.7600

0.4596238 3.8170

0.5599305 1.1690

d ni

1.0000000 0.2836

end pseudo nonlocal nihay ni c c runtype scf scftype casscf config print fzc 1 to 7 doc 8 to 9 alp 10 to 11 uoc 12 to 14 end superci 1 to 12 newton 13 to 20 hessian 13 to 20 simul 15 to 20 vectors 5 swap

14 15 end enter to run:

24 ECP, CASSCF AND DIRECT-CI CALCULATIONS ON NICCH

2 rungamess -t ed2 -t ed3 -t ed4 -t ed6 -l ed0 -n nicch2 nicch2_cas

Direct-CI Job

Contents of file nicch2 ci.in

restart new title ni(cch2) 3a1 mrsdci (3 2 2 0) ,hay’s ni, bar’s nm ecp’s mult 3 super off nosym bypass scf zmat angstrom ni c 1 nica x 2 1.0 1 90.0

c 2 cacb 3 90.0 1 180.0

x 4 1.0 2 90.0 3 0.0

h 4 hcb 2 hcc 5 90.0

h 4 hcb 2 hcc 5 -90.0

variables nica 2.0895

cacb 1.3604

hcb 1.1047

hcc 122.646

end basis s h

0.032828

0.231208

0.817238

s h

1.000000

s c

1.000000

13.3615

2.0133

0.4538

0.1233

0.4962

s c

1.000000

p c

0.018534

0.115442

0.386206

0.640089

0.1533

18.1557

3.9864

1.1429

0.3594

p c

1.000000

0.1146

s ni

-0.4372528 0.6778

1.1889453 0.1116

s ni

1.0000000 0.0387

p ni

1.0000000 0.0840

p ni

1.0000000 0.0240

d ni

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE61

0.0360414 42.7200

0.1938645 11.7600

0.4596238 3.8170

0.5599305 1.1690

d ni

1.0000000 0.2836

end pseudo nonlocal nihay ni c c runtype ci core end active

1 to 40 end direct 20 14 26 spin triplet conf

2 2 2 2 2 2 2 2 2 1 1 0 0 0

2 2 2 2 2 2 2 2 1 2 0 0 1 0

2 2 2 2 2 2 2 2 1 1 1 0 1 0

2 2 2 2 2 2 2 2 0 1 1 0 2 0

2 2 2 2 2 2 2 1 2 2 0 1 0 0

2 2 2 2 2 2 2 1 2 1 1 1 0 0 vprint 100 0.01

maxcyc 10 enter to run: rungamess -t ed2 -t ed3 -t ed5 -t ed6 -l ed0 -n nicch2 nicch2_ci

25 Table-CI Calculations of the Electronic Spectra of Pyridine

This example demonstrates the use of the Table-CI module in the calculation of the low-lying states of Pyridine. Specifically, we are involved in determining the disposition of the first ten

1 and

2 states, using a common set of orbitals (the X

SCF-MOs) in a DZ plus Rydberg basis set of 91 functions. Five job files are presented below:

1. Start–up closed–shell SCF calculation for the X

1 state. Note the SUPER directive for compatibility with the subsequent CI calculation. Note also the particular syntax for siting the DZ basis on H: the third and fourth data fields are to provide an unscaled hydrogen basis, since the default specification will scale the two components by 1.2 (the more contracted) and 1.15 (the more diffuse component)

2. 1M/1R Table-CI calculation of the X

1 state.

3. 6M/1R Table-CI calculation of the X

1 state.

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE62

4. 21M/10R Table-CI calculation of the ten lowest

1 states.

5. 19M/10R Table-CI calculation of the ten lowest

2 states.

1. Closed-shell SCF Job

Contents of file pyridine2 rhf.in

title pyridine dz+bond-centred functions super off nosym zmat angstrom n x 1 1.0

x 1 1.0 2 90.

x 1 1.0 2 90. 3 90.

c 1 c4n 3 90. 2 180.

x 5 1.0 1 90. 3 0.0

x 5 1.0 1 90. 4 0.0

h 5 ch4 6 90. 1 180.

c 1 c2n 2 c2nz 3 180.

c 1 c2n 2 c2nz 3 0.0

c 9 c2c3 1 ccn 2 180.

c 10 c2c3 1 ccn 2 180.

h 9 c2h6 1 nch2 2 0.0

h 10 c2h6 1 nch2 2 0.0

h 11 c3h5 9 c2c3h 1 180.

h 12 c3h5 10 c2c3h 1 180.

bq 1 1.39 3 90. 2 180.

variables c4n 2.7845546

ch4 1.0823078

c2n 1.3372389

c2nz 120.641858

c2c3 1.3944571

ccn 122.662269

c2h6 1.0814291

c3h5 1.0809550

nch2 116.400433

c2c3h 120.158516

end basis dz h 1.0 1.0

dz n dz c s bq

1.0 0.021

s bq

1.0 0.008

s bq

1.0 0.0025

p bq

1.0 0.017

p bq

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE63

1.0 0.009

d bq

1.0 0.015

d bq

1.0 0.008

end enter to run: rungamess -t ed2 -t ed3 -n pyridine2 pyridine2_rhf

2. 1M/1R Table-CI Job for the X

State

An examination of the SCF output reveals the following symmetry adapted basis functions, given the C

2v geometry and DZ plus Rydberg basis set:

=============================

IRREP NO. OF SYMMETRY ADAPTED

BASIS FUNCTIONS

=============================

1 45

4 6

=============================

Thus the orbital reordering performed by the Table-CI module will yield the sequence numbers

1–45 for the a

MOs, 46–57 for the b

MOs, 58–85 for the b

MOs and 86–91 for the a

MOs.

We are both freezing and discarding orbitals in the subsequent CI calculations. Confining this to both a

1 and b

MOs, the first six orbitals of a

1 symmetry and first four orbitals of b

2 symmetry are to be frozen, and the eight highest energy orbitals of a

1 symmetry and six highest of b

2 are to be discarded. Thus the final sequence numbers for the active orbitals in the CI are 1–31 for the a

MOs, 32–43 for the b

MOs, 44–61 for the b

MOs and 62–68 for the a

MOs.

The CONF data line specifying the reference configuration is based on the associated sequence numbers of these active orbitals.

Contents of file pyridine2 1m1r.in

restart title pyridine dz+bond-centred functions super off nosym bypass scf zmat angstrom n x 1 1.0

x 1 1.0 2 90.

x 1 1.0 2 90. 3 90.

c 1 c4n 3 90. 2 180.

x 5 1.0 1 90. 3 0.0

x 5 1.0 1 90. 4 0.0

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE64 h 5 ch4 6 90. 1 180.

c 1 c2n 2 c2nz 3 180.

c 1 c2n 2 c2nz 3 0.0

c 9 c2c3 1 ccn 2 180.

c 10 c2c3 1 ccn 2 180.

h 9 c2h6 1 nch2 2 0.0

h 10 c2h6 1 nch2 2 0.0

h 11 c3h5 9 c2c3h 1 180.

h 12 c3h5 10 c2c3h 1 180.

bq 1 1.39 3 90. 2 180.

variables c4n 2.7845546

ch4 1.0823078

c2n 1.3372389

c2nz 120.641858

c2c3 1.3944571

ccn 122.662269

c2h6 1.0814291

c3h5 1.0809550

nch2 116.400433

c2c3h 120.158516

end basis dz h 1.0 1.0

dz n dz c s bq

1.0 0.021

s bq

1.0 0.008

s bq

1.0 0.0025

p bq

1.0 0.017

p bq

1.0 0.009

d bq

1.0 0.015

d bq

1.0 0.008

end runtype ci mrdci adapt tran freeze discard

6 0 4 0

1 to 6 1 to 4

8 0 6 0

38 to 45 23 to 28 select cntrl 22 spin singlet symmetry 1 conf

0 1 2 3 4 5 32 33 44 45 46 62

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE65 roots 1 1 thresh 30 10 ci diag title pyridine 1m1r ground state dz + 3s2p2d rydberg basis extrap 3 natorb civec 1 putq aos 2 enter to run: rungamess -r mrdci -l table -n pyridine2 pyridine2_1m1r

3. 6M/1R Table-CI Job for the X

State

Contents of file pyridine2 6m1r.in

restart title pyridine dz+bond-centred functions super off nosym bypass scf zmat angstrom n x 1 1.0

x 1 1.0 2 90.

x 1 1.0 2 90. 3 90.

c 1 c4n 3 90. 2 180.

x 5 1.0 1 90. 3 0.0

x 5 1.0 1 90. 4 0.0

h 5 ch4 6 90. 1 180.

c 1 c2n 2 c2nz 3 180.

c 1 c2n 2 c2nz 3 0.0

c 9 c2c3 1 ccn 2 180.

c 10 c2c3 1 ccn 2 180.

h 9 c2h6 1 nch2 2 0.0

h 10 c2h6 1 nch2 2 0.0

h 11 c3h5 9 c2c3h 1 180.

h 12 c3h5 10 c2c3h 1 180.

bq 1 1.39 3 90. 2 180.

variables c4n 2.7845546

ch4 1.0823078

c2n 1.3372389

c2nz 120.641858

c2c3 1.3944571

ccn 122.662269

c2h6 1.0814291

c3h5 1.0809550

nch2 116.400433

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE66 c2c3h 120.158516

end basis dz h 1.0 1.0

dz n dz c s bq

1.0 0.021

s bq

1.0 0.008

s bq

1.0 0.0025

p bq

1.0 0.017

p bq

1.0 0.009

d bq

1.0 0.015

d bq

1.0 0.008

end runtype ci mrdci adapt bypass tran freeze discard bypass

6 0 4 0

1 to 6 1 to 4

8 0 6 0

38 to 45 23 to 28 select cntrl 22 spin singlet symmetry 1 conf

0 1 2 3 4 5 32 33 44 45 46 62

0 1 2 3 4 5 32 38 44 45 46 62

0 1 2 3 4 5 32 33 44 45 46 65

4 32 33 38 39 1 2 3 4 5 44 45 46 62

4 32 39 62 65 1 2 3 4 5 33 44 45 46

4 33 38 62 65 1 2 3 4 5 32 44 45 46 roots 1 1 thresh 30 10 ci diag title pyridine 6m1r ground state dz + 3s2p2d rydberg basis extrap 3 natorb civec 1 putq aos 2 enter to run: rungamess -r mrdci -l table -n pyridine2 pyridine2_6m1r

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE67

4. 21M/10R Table-CI Job for the

States

Contents of file pyridine2 21m10r.in

restart title pyridine dz+rydberg basis super off nosym bypass scf zmat angstrom n x 1 1.0

x 1 1.0 2 90.

x 1 1.0 2 90. 3 90.

c 1 c4n 3 90. 2 180.

x 5 1.0 1 90. 3 0.0

x 5 1.0 1 90. 4 0.0

h 5 ch4 6 90. 1 180.

c 1 c2n 2 c2nz 3 180.

c 1 c2n 2 c2nz 3 0.0

c 9 c2c3 1 ccn 2 180.

c 10 c2c3 1 ccn 2 180.

h 9 c2h6 1 nch2 2 0.0

h 10 c2h6 1 nch2 2 0.0

h 11 c3h5 9 c2c3h 1 180.

h 12 c3h5 10 c2c3h 1 180.

bq 1 1.39 3 90. 2 180.

variables c4n 2.7845546

ch4 1.0823078

c2n 1.3372389

c2nz 120.641858

c2c3 1.3944571

ccn 122.662269

c2h6 1.0814291

c3h5 1.0809550

nch2 116.400433

c2c3h 120.158516

end basis dz h 1.0 1.0

dz n dz c s bq

1.0 0.021

s bq

1.0 0.008

s bq

1.0 0.0025

p bq

1.0 0.017

p bq

1.0 0.009

d bq

1.0 0.015

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE68 d bq

1.0 0.008

end runtype ci mrdci adapt bypass tran freeze discard bypass

6 0 4 0

1 to 6 1 to 4

8 0 6 0

38 to 45 23 to 28 select cntrl 22 spin singlet symmetry 1 conf

2 62 65 1 2 3 4 5 32 33 44 45 46

2 33 34 1 2 3 4 5 32 44 45 46 62

2 5 8 1 2 3 4 32 33 44 45 46 62

2 62 63 1 2 3 4 5 32 33 44 45 46

2 33 38 1 2 3 4 5 32 44 45 46 62

2 33 35 1 2 3 4 5 32 44 45 46 62

2 5 7 1 2 3 4 32 33 44 45 46 62

2 33 37 1 2 3 4 5 32 44 45 46 62

2 5 9 1 2 3 4 32 33 44 45 46 62

2 62 64 1 2 3 4 5 32 33 44 45 46

2 33 36 1 2 3 4 5 32 44 45 46 62

2 5 11 1 2 3 4 32 33 44 45 46 62

2 33 39 1 2 3 4 5 32 44 45 46 62

2 32 34 1 2 3 4 5 33 44 45 46 62

2 32 38 1 2 3 4 5 33 44 45 46 62

4 5 8 33 38 1 2 3 4 32 44 45 46 62

4 33 38 62 65 1 2 3 4 5 32 44 45 46

4 32 33 34 38 1 2 3 4 5 44 45 46 62

4 33 34 62 65 1 2 3 4 5 32 44 45 46

4 32 38 62 63 1 2 3 4 5 33 44 45 46

4 5 7 33 38 1 2 3 4 32 44 45 46 62 roots 10 1 2 3 4 5 6 7 8 9 10 thresh 30 10 ci diag title pyridine 21m10r 1a1 extrap 3 enter dz + 3s2p2d rydberg basis to run: rungamess -r mrdci -l table -n pyridine2 pyridine2_21m10r

5. 19M/10R Table-CI Job for the

States

Contents of file pyridine2 19m10r.in

restart

25 TABLE-CI CALCULATIONS OF THE ELECTRONIC SPECTRA OF PYRIDINE69 title pyridine dz+rydberg basis super off nosym bypass scf zmat angstrom n x 1 1.0

x 1 1.0 2 90.

x 1 1.0 2 90. 3 90.

c 1 c4n 3 90. 2 180.

x 5 1.0 1 90. 3 0.0

x 5 1.0 1 90. 4 0.0

h 5 ch4 6 90. 1 180.

c 1 c2n 2 c2nz 3 180.

c 1 c2n 2 c2nz 3 0.0

c 9 c2c3 1 ccn 2 180.

c 10 c2c3 1 ccn 2 180.

h 9 c2h6 1 nch2 2 0.0

h 10 c2h6 1 nch2 2 0.0

h 11 c3h5 9 c2c3h 1 180.

h 12 c3h5 10 c2c3h 1 180.

bq 1 1.39 3 90. 2 180.

variables c4n 2.7845546

ch4 1.0823078

c2n 1.3372389

c2nz 120.641858

c2c3 1.3944571

ccn 122.662269

c2h6 1.0814291

c3h5 1.0809550

nch2 116.400433

c2c3h 120.158516

end basis dz h 1.0 1.0

dz n dz c s bq

1.0 0.021

s bq

1.0 0.008

s bq

1.0 0.0025

p bq

1.0 0.017

p bq

1.0 0.009

d bq

1.0 0.015

d bq

1.0 0.008

end runtype ci mrdci

26 FULL-CI CALCULATIONS 70 adapt bypass tran freeze discard bypass

6 0 4 0

1 to 6 1 to 4

8 0 6 0

38 to 45 23 to 28 select cntrl 22 spin singlet symmetry 4 conf

2 5 65 1 2 3 4 32 33 44 45 46 62

2 33 47 1 2 3 4 5 32 44 45 46 62

2 5 63 1 2 3 4 32 33 44 45 46 62

2 33 48 1 2 3 4 5 32 44 45 46 62

2 8 62 1 2 3 4 5 32 33 44 45 46

2 7 62 1 2 3 4 5 32 33 44 45 46

2 33 50 1 2 3 4 5 32 44 45 46 62

2 5 64 1 2 3 4 32 33 44 45 46 62

2 33 49 1 2 3 4 5 32 44 45 46 62

2 6 62 1 2 3 4 5 32 33 44 45 46

4 5 33 38 65 1 2 3 4 32 44 45 46 62

4 5 33 38 63 1 2 3 4 32 44 45 46 62

4 33 47 62 65 1 2 3 4 5 32 44 45 46

4 5 33 34 65 1 2 3 4 32 44 45 46 62

4 7 33 38 62 1 2 3 4 5 32 44 45 46

4 33 47 62 63 1 2 3 4 5 32 44 45 46

4 8 33 38 62 1 2 3 4 5 32 44 45 46

4 5 33 34 63 1 2 3 4 32 44 45 46 62

4 33 48 62 63 1 2 3 4 5 32 44 45 46 roots 10 1 2 3 4 5 6 7 8 9 10 thresh 30 10 ci diag title pyridine 19m10r 1a2 dz + 3s2p2d rydberg basis extrap 3 enter to run: rungamess -r mrdci -l table -n pyridine2 pyridine2_19m10r

26 Full-CI calculations

We consider below Full-CI calculations of the X

1 state of the H

O molecule. In the first instance we consider correlating all electrons. We then perform a valence-only calculation, freezing the O1s orbital through the ACTIVE and CORE directives, specifying a total of 8 electrons on the FULLCI data line.

All-electron Job

26 FULL-CI CALCULATIONS 71

Contents of file fullci all.in

core 8000000 title h2o - DZ basis - full-ci super off nosym zmat angstrom\o\h 1 roh\h 1 roh 2 theta variables\roh 0.956 hess 0.7\theta 104.5 hess 0.2 \end basis dz runtype ci\fullci 14 5 5 enter to run: rungamess -r fullci -n h2o_fullci fullci_all

Assuming the above job did not complete in the time allocated, and dumped to disk in a controlled fashion, the following job would act to continue the processing, assuming that the

FORTRAN file from stream ftn008 had been saved, along with the Mainfile (ed2) Dumpfile

(ed3) and Transformed integral file (ed6). All these files should be stored on temporary disk space, given the –r fullci argument.

Restarting the Full-CI job

Contents of file fullci rest.in: core 8000000 restart ci title h2o - DZ basis - restart full-ci super off nosym zmat angstrom\o\h 1 roh\h 1 roh 2 theta variables\roh 0.956 hess 0.7\theta 104.5 hess 0.2 \end basis dz runtype ci\fullci 14 5 5 enter to run: rungamess -r fullci -n h2o_fullci fullci_rest

Valence-electron Job

Contents of file fullci val.in: core 8000000 restart new title h2o - DZ basis - valence full-ci super off nosym

26 FULL-CI CALCULATIONS bypass zmat angstrom\o\h 1 roh\h 1 roh 2 theta variables\roh 0.956 hess 0.7\theta 104.5 hess 0.2 \end basis dz runtype ci active\2 to 14 end\core\1\end fullci 13 4 4 enter to run: rungamess -r fullci -n h2o_fullci fullci_val

The following points should be noted:

• the use of –r, which causes the FORTRAN stream ftn008 to be saved;

• the use of the core pre-directive to specify memory requirements.

안녕하세요! 저는 Computing for Science Ltd. GAMESS-UK Rungamess Computational Chemistry Software User's Guide을(를) 지원하도록 특별히 훈련된 AI 챗봇입니다. 문서를 철저히 검토했으며 필요한 정확한 정보를 찾거나 내용을 명확하고 간단하게 설명하는 데 도움을 드릴 수 있습니다. 특정 기능에 대한 지침, 문제 해결 단계 또는 일반적인 사용법을 찾고 계시든 언제든지 질문하십시오. 귀하의 우려 사항이나 필요 사항에 대해 더 자세히 알려주실수록 더 정확하고 효과적으로 도와드릴 수 있습니다!

어시스턴트 입력 중

어시스턴트가 키보드에서 벗어났습니다. 나중에 다시 시도하십시오.

Generalised Atomic and Molecular Electronic Structure System

PART 13. GAMESS-UK under UNIX - Rungamess

M.F. Guest, J. Kendrick, J.H. van Lenthe and P. Sherwood

Contents

1 Introduction

2 Usage of the Rungamess Script

2.1

Installation

2.2

Running a Job

2.3

Saving Files

2.4

Accessing Library Files

2.5

Batch Job Submission

2.6

Parallel Job Execution

2.7

Summary of Command Line Arguments

3 GAMESS–UK Pre-Directives

3.1

The memory Pre-directive

3.2

The time Pre-directive

4 Single Point SCF Calculation

5 Sequence of Calculations on H

O

6 SCF Calculations on C

F

, C

H

NO

and C

H

(NO

)

CH

7 STO-3G Calculations on Na

Mg

8 Extended Basis Set Calculations of Na

Mg

9 ECP calculations of Na

Mg

10 Graphical analysis of Ni(CO)

11 2-Pair GVB Calculation on CH

(CH)

NH

12 Direct-SCF Calculations

13 HCN/HNC Transition State Location

14 HSiP/HPSi Transition State Location

14.1

Numerical Force Constants

14.2

Analytic Force Constants

15 Use of Bond-centred Functions

16 SCF Analytic Force Constants for C

H

17 MP2 Analytic Force Constants for C

H

18 MP2 Polarisability for C

H

19 Direct-MP2 Calculation of C

H

N

20 CASSCF Geometry Optimisations

21 CASSCF + 2nd-order CI Calculations on BeO

22 MCSCF + 2nd-order CI Calculations on BeO

23 Table-CI calculations on the Ammonia Cation

24 ECP, CASSCF and Direct-CI Calculations on NiCCH

25 Table-CI Calculations of the Electronic Spectra of Pyridine

26 Full-CI calculations

관련 항목

CFS GAMESS-UK User Manual: AI Chat & PDF Download

GAMESS-UK Version 8.0 User's Guide | AI Chat & PDF Download

GAMESS-UK v8.0 Parallel Implementation Manual

SAPT Software User's Guide: AI Chat & PDF Download

GAMESS-UK Electronic Structure System AI & PDF Access

Electronic Structure System GAMESS-UK User's Guide and Reference Manual

GAMESS-UK Version 8.0 User Manual | AI Chat & PDF Download