Home

GAMESSPLUS: - Chemical Theory Center

Contents

1. ELTAGW AND DELTAGSOIL based on GAMESSPLUS 2008 Tue Apr 8 19 58 36 2008 KOC CALCULATION FOR BENZENE USING CALCULATED VALUES FOR D El C 6 0 000000 1 208141 697522 C 6 0 000000 1 208141 209 522 C 6 0 000000 000000 1 395043 C 6 0 000000 1 208141 697522 C 6 0 000000 1 208141 697522 C 6 0 000000 000000 1 395043 H 1 0 000000 2 160477 1 247356 H Tyo 000000 2 160477 1 247356 H 1 0 000000 000000 2 494707 H T0 000000 2 160477 247356 H 1 0 000000 2 160477 1 247356 H 13 0 000000 000000 2 494707 SSCF AM1 SSOILDENSITY 0 11 g mL All values are for a temperature of 298 K Bre nergies are calculated using a 1M gt 1M standard state log log basel0 Free Energy of Transfer air gt water 1 51 kcal mol Free Energy of Transfer air gt soil 4 32 kcal mol Free Energy of Transfer water gt soil 2 81 kcal mol log Koc 3 02 124 125 End of output test2 log GAMESSPLUS Soil sorption utility calculation based on GAMESSPLUS 2008 Tue Apr 8 20 00 14 2008 KOC CALCULATION FOR BENZENE USING AN EXPERIMENTAL VALUE FOR DELTAGW AND A CALCULATED VALUE FOR DELTAGSOIL Ck C 6 0 000000 1 208141 697522 C 6 0 000000 1 208141 697522 6 0 000000 000000 1 395043 C 6 0 000000 1 208141 697522 C 6 0 000000 1 208141 697522 C 6 0 000000 000000
2. of gt Sp 6j a o coefficient type 1000 J 2000 J 3000 J 4000 J atomic coefficient for atoms 7 1 100 J 1 100 J 1 100 J 1 100 with nuclear charge i 1100 j 2100 j 3100 J 4100 j H X where j is the nuclear jJ 1 100 1 100 j 1 100 j 1 100 charge on atom X 1201 2201 3201 4201 C C 1 1202 2202 3202 4202 C C 2 1203 2203 3203 4203 O C 1204 2204 3204 4204 O O 1205 2205 3205 4205 N C 1 1206 2206 3206 4206 O N 1207 2207 3207 4207 S S 1210 2210 3210 4210 C N 1211 2211 3211 4211 N C 2 1212 2212 3212 4212 H N 2 1213 2213 3213 4213 H O 2 1214 2214 3214 4214 O P 1215 2215 3215 4215 S P 1216 2216 3216 4216 N C 3 molecular surface tension coefficients 5001 6 5002 68 5003 gi 5004 gv Example 51 The following input defines the values of the following four coefficients 6 40 80 oe 24 51 B 346 42 and 6 0 3136 All other coefficients are set to zero Icoeff 2 4 2 2001 40 80D0 3006 24 51D0 4108 346 42D0 5001 0 3136D0 Namelist CM2SRP The current version of GAMESSPLUS provides an option for entering external parameters for the CM2 charge model and this is called CM2 SRP This option is required when the electronic wave function used for a calculation does not have a matched CM2 CM3 or CM4 parameter set for example HF STO 3G or when it is desired to adjust the parameters in the charge model to obtain a better estimation of the
3. 0 02 for a H and C pair and D 9 11 for a C and Cl pair one sets DCMSRP 1 0 02 and DCMSRP 13 0 11 The default is that all the DCMSRP and CCMSRP are zero OQ DR Defines the displacement size in bohr of the coordinate used in computing aR EETYP DQDR by numerical differentiation of the charges Default 0 01 OQ DPHI Defines the displacement size in a u of the electrostatic potential used in computing am EETYP DQDPHI by numerical differentiation of the charges Default 0 01 IRDMM Determines how the electrostatic potential distribution is obtained See Section Electrostatically Embedded QM Calculation with a Site Site Representation of the QM MM Electrostatic Interaction IRDMM 0 is obtained from the input data PHI which is explained above default IRDMM 1 is calculated from MM charges and coordinates which are read from MM namelist In this case PHI is ignored and namelist MM is required ke Parameters below are effective only if IRDMM 1 IADDGP Specifies whether is regarded as the function of R and R or not See Section Electrostatically Embedded QM Calculation with a Site Site Representation of the QM MM Electrostatic Interaction IADDGP 0 is regarded as an independent variable not dependent on R default Vv HEM dv EEM IADDGP 1 is regarded as a function of R and R and are dR dR a A outputted IUCMM Determines the units of the MM coordinates which a
4. 90 aqueous and organic solvent by AM1 and PM3 y 11 and 12 L wdin and CM2 charges by HF and B3LYP testB9 02y y 1 2 3 6 8 9 SM5 42 energy evaluation in aqueous and organic testBw g2y where w 1 8 q 1 and 2 solvent by HF and B3LYP y 1 6 8 9 SM5 42 energy and analytical gradient evaluation in testBw q2y where w 1 8 q 1 and 2 aqueous and organic solvent by HF y 2 and 3 Eight of these nine molecules are used to test most of the CM3 and CM4 parameters and all of the SMS 43 parameters CM3 and SM5 43 do not contain parameters for I Because CM3 and CM4 are also parameterized for Li and for molecules that containing Si O Si F and Si Cl bonds three additional compounds orthosilicic acid HSi OH FCI1 and LigCyN OFCISH 3 w 10 11 and 12 respectively have been added to the test suite These test calculations are named testS1 gy testS12 gy where S A or B q 0 1 or 2 and y is the value of ICMD used in the calculation Note that in the portion of the test suite that tests the CM2 parameters a value of 2 for CM2 was included in the name before the value of ICMD while an analogous value of 3 for CM3 and 4 for CM4 is already included in the values of ICMD that request a calculation of CM3 or CM4 charges The table below summarizes this portion of the test suite Calculation type L wdin and CM3 charges by AM1 and PM3 L wdin and CM3 charges by HF BLYP and
5. 13 SM5 42 SM5 43 SM6 SM8 SM8AD and SM8T Solvation Models SMS 42 our earliest ab initio solvation model implemented in GAMESSPLUS is a universal solvation model based on SM5 functional forms for atomic surface tensions hence the first three characters in the name of the method are SM5 built on class IV point charges hence 4 comes next of the CM2 type hence 2 A more recent model called SM5 43 uses the same functional forms for atomic surface tensions as does SM5 42 but SM5 43 uses CM3 charges hence the 3 in the name The SM6 model is based on SM6 functional forms for atomic surface tensions and uses class IV CM4 point charges The SM6 model has only been parametrized for aqueous solvent The SM8 and SM8AD are the most recent universal continuum solvation models where universal denotes applicable to all solvents see MOO7 for more details With universal models if desired one can calculate solvation free energies for two different solvents e g water and 1 octanol and use the results to calculate log P where P is the partition coefficient SM8 SM8AD is applicable to any charged or uncharged solute composed of H C N O F Si P S Cl and or Br in any solvent or liquid medium for which a few key descriptors are known in particular dielectric constant refractive index bulk surface tension and acidity and basicity parameters It may be used with any level of electronic structure theory as long as accurate partial char
6. I rea seb In GAMESSPLUS one can calculate V and its first and second derivatives with respect to R and for given R and The first derivative of V with respect to a component of R can be obtained in a similar way in Ref ZL99 In addition to terms that appear in the gas phase calculation 22 T one has to calculate with P fixed at the converged value P and add it to the first derivative When Q is the LPA charge the first derivative with respect to R is given by aQ LPA x as P 2492 po S i 18 oR i R OR Cc rea c c rr 1 2 The method to calculate is given in Ref ZL99 When Q is the RLPA charge the first derivative Cc with respect to R is given by 6Q RLPA _ 0Q LPA OR po p oR p C Z E exp a Ra 20 VO exp a R Ro c Jp bza oY 2 2 5 Z exp a Ri 2Z Y a exp a R2 Ra 19 b a oR po where 1 X OY _ S OS po gtystpo S 20 oR p rea oR oR When Q is the CMx charge the first derivative with respect to R is given by 0 Q _ oQ pi 5 B D ea 2C Bap i 21 oR po OR po oR Pp c c b a e where Bay a OS a 1 ay 3S oR l g rr OR P S j P S P oR l 22 rea seb c s In GAMESSPLUS the second derivative of V with respect to R is obtained by numerical differentiations of the first derivatives EEQM The first derivative of V with respect to a comp
7. 201350 0 172260 073030 725570 455040 147070 291310 863270 201350 0 172260 073030 725570 455040 147070 4 251310 863270 201350 0 501800 681444 151398 0 501800 681444 151398 0 501800 681444 151398 0 501800 00000 t SD KD C2 EA gD Ee aa ae gt 00000 CO 00000 CO 00000 j ae a E 00000 l1 0000000000 198348 0707400000 3953800000 6633110000 0813800000 5748530000 0000000000 1099310000 4627130000 0000000000 000000 i 0707400000 3953800000 6633110000 0813800000 5748530000 0000000000 1099310000 4627130000 0000000000 130455 1 0704520000 4078260000 0000000000 130455 1 0704520000 4078260000 0000000000 000000 S2 0704520000 4078260000 0000000000 130455 1 0704520000 691865 383730 230017 230017 460034 230017 123 2 S 1 1 H 1 S 2 1 2 S 1 1 H 1 S 2 1 2 S 1 1 SEND Output testl log 6814440 1513980 5018000 6814440 1513980 501800 681444 1513980 00000 00000 com 0 0 4078260000 1 0000000000 2 130455 0 0704520000 0 4078260000 1 0000000000 1 230017 000000 2 460034 0 0704520000 0 4078260000 1 0000000000 GAMESSPLUS Soil sorption utility calculation
8. where R and R stand for the collection of the coordinates R a 1 2 N e and R A 1 2 N a of atoms in the QM and MM subsystems respectively The first term V2 is the electronic energy of the QM region and the last term V is the MM potential energy The middle term V is the QM MM interaction energy and can be separated into three terms y QM MM R R y QMMM R R aye R R 4 VQM MM R R l 31 ele val where V2W MM yONMM and VOM M are the electrostatic van der Waals and valence interaction energies respectively In GAMESSPLUS VOMM is represented by a site site representation ele VEM RRB Y 0 Y 32 where is the electronic wave function of the QM region is the population operator that generates the partial charge Q on QM atomic site a is the electrostatic potential at atom a from the MM region In GAMESSPLUS the user can choose the operator according to L wdin population analysis LPA redistributed L wdin population analysis RLPA Charge Model 2 CM2 Charge Model 3 CM3 Charge Model 4 CM4 or Charge Model 4M CM4M For the details see the section entitled Electrostatically Embedded QM Calculation with a Site Site Representation of the QM MM Electrostatic Interaction For QM MM calculations with site site electrostatics GAMESSPLUS uses the AMBER force field as the MM potential energy function The AMBER force field ref CC95 i
9. 2DHC loadPdb 2DHC_wat20 pdb set 2DHC cap 0 0 0 0 0 0 20 0 87 set default OldPrmtopFormat on savePdb 2DHC 2DHC_wat20_tleap pdb saveAmberParm 2DHC 2DHC_wat20 top 2DHC_wat20 crd quit The first line source leaprc gaff means that the GAFF parameters will be loaded The second line mods loadAmberParams DCE_resp frcmod means that additional parameters will be loaded from the DCE_resp frcmod file In the example case there is no additional parameter in the DCE_resp fremod file However this line was added specifically for this tutorial The third line DCE loadMol2 DCE_resp mo12 means that the parameters for DCE will be loaded from the DCE_resp mol2 file and assigned to residues named DCE The fourth line 2DHC loadPdb 2DHC_wat20 pdb means that atoms and their coordinates will be loaded from 2DHC_wat20 pdb file and assigned to variable 2DHC which becomes the name of the target system in the program If some atoms are missing in the PDB file usually the PDB file does not contain hydrogen atoms they are automatically generated by LEaP The fifth line set 2DHC cap 0 0 0 0 0 0 20 0 is part of the control of the equilibrium MD simulation see below and is ignored in QM MM calcuations by the current version of GAMESSPLUS This line requests that a water cap be added to the 2DHC system A half harmonic potential will be added at 20 A from the center of the system in order to
10. 69 SCM2 ISCRF 1 IGAS 0 ICMD 8 IAQU 0 SolN 1 3288 SolA 0 43 SolB 0 47 SolG 31 77 Dielec 32 613 SEND SDATA SM5 42R HF 6 31 G d calculation of water in methanol Gil O 8 0 0 000000 0 000000 0 000000 H1 1 0 0 967300 0 000000 0 000000 H1 1 0 0 210300 0 944200 0 000000 SEND Example 3 SM5 42 HF AM1 geometry optimization of water in aqueous solution SCONTRL SCFTYP RHF RUNTYP OPTIMIZE COORD UNIQUE SEND SBASIS GBASIS AM1 SEND SSTATPT NSTEP 50 SEND SGMSOL ISCRF 1 ICMD 11 IAQU 1 SEND SDATA optimization of water in water C1 O 8 0 0 000000 0 000000 0 000000 HI 1 0 0 967300 0 000000 0 000000 HI O 0 210300 0 944200 0 000000 SEND Example 4 Numerical frequency evaluation using central differences and a step size of 0 0005 bohr for water in liquid acetone using SM5 42 HF PM3 SCONTRL SCFTYP RHF RUNTYP HESSIAN COORD UNIQUE SEND SFORCE METHOD NUMERIC NVIB 2 VIBSIZ 0 0005 VIBANL TRUE SEND SCM2 ISCRF 1 ICMD 12 IAQU 0 Dielec 20 493 SolN 1 3588 SolA 0 04 SolB 0 49 SolG 33 77 SolC 0 000 SolH 0 000 SEND SDATA frequency calculation of water in acetone C1 O 8 0 0 000000 0 000000 0 000000 HI 1 0 0 967300 0 000000 0 000000 HI 1 0 0 210300 0 944200 0 000000 END Example 5 Transition state optimization for the reaction
11. 1 3 and a 0 25 428 606 or 999 and t 5 gas phase or 10 embedded electrostatic potential There are two test calculations test4 0 inp and test4 1 inp in the EEQM test suit with IRDMM 1 located in EEQMTests The QM MM system consists of 1 2 dichloroethane and haloalkane dehalogenase The QM subsystem is 1 2 dichloroethane and the side chain of Asp124 and the MM subsystem is the rest of the protein and water ADDGP 0 in test4 0 inp and IADDGP 1 in test4 1 inp Description of Test Suite for INTFRZ There are 4 test calculation in the INTFRZ test suite located in QMMM Tests The test1 inp calls for geometry optimization of a water molecule with two O H bonds fixed at 1 0 A The test2 inp calls for geometry optimization of a water molecule with the H O H angle fixed at 120 0 degrees The test3 inp calls for geometry optimization of 1 2 dichloroethane with the Cl C C Cl torsional angle fixed at 60 0 84 degrees The test4 inp calls for geometry optimization of the reaction Cl CH F CICH F with the difference between C Cl and C F bonds fixed at 0 03 A which corresponds to the transition state In all cases calculations were performed using the M06 2X density functional and the 6 31G d basis set Description of Test Suite for QM MM There is one test calculation in the QM MM test suite located in QMMM Tests The test input testl inp calls for QM MM geometry optimization of a system consisting of 1 2 dichloroethane and ha
12. 88 June Converting N terminal residue name to PDB format Converting C terminal residue name to PDB format Checking Unit WARNING T gt ignori Bu Bu Bu Bu ldi ldi ldi lding Building Building total Buildi ng topology Ng t a ng b a H H H H H H Marking per residue Residues lacking co these don t have res CGLU MOL NMET WAT 1 1 a 317 no restraints iMaxAoms 2 24 Quit a Cc total affec he unperturbed charge of the unit ng the warning tom parameters ond parameters ngle parame proper torsio improper torsion parameters 1035 improper ng H Bond parame Not Marking per residue atom chain types ters n parameters torsions applied ters tom chain types nnect0 connectl hain types marked ted templates NMET gt MET CGLU gt GLU 16 999999 is not zero One can use the AMBER parameter topology and coordinate files created in this way as the input for GAMESSPLUS QM MM calculations by pasting them with AMBTOP or AMBCRD before the first line and END after the last line into a GAMESSPLUS input file 89 4 Before the QM MM geometry optimization is performed an equilibrium MD simulation and a minimization with a pure MM force field should be carried out by the AMBER program because the structure created from the PDB file in particular the automatically generated hydrogen ato
13. B3LYP test cases testAw Oy where w 1 2 4 12 y 312 testAw 0y where w 1 2 4 11 y 311 testBw 0y where w 1 2 4 8 y 301 testBw 0y where w 1 2 4 12 y 302 320 321 testBw 0y where w 9 10 11 12 y 303 313 L wdin RLPA and CM3 charges by B3LYP testBw Oy where w 1 2 4 12 y 314 L wdin and CM3 charges by MPWX testBw x 0y where w 9 12 x 0 517 999 y 315 316 317 RLPA and CM3 charges by MPWX testBw x 0y where w 9 12 x 0 517 999 y 318 and 319 L wdin and CM3 1 charges by HF MIDI testBw 0322 where w 1 4 5 6 7 SM5 43 energy and analytical gradient evaluation in aqueous and organic solvent by HF 6 31G d and B3LYP 6 31G d testBw qgy where w 1 2 4 8 q 1 or2 y 303 and 313 SM5 43 energy evaluation in aqueous and organic solvent by MPWX testBw x gy where w 1 2 4 8 x 0 517 999 qg 1or2 y 315 316 317 318 319 L wdin and CM4 charges by MPWX testBw x 0y where w 9 12 x 0 517 999 y 416 and 417 RLPA and CM4 charges by MPWX testBw x 0y where w 9 12 x 0 517 999 y 418 and 419 The remaining test jobs test the remaining keywords available in GAMESSPLUS Two test cases in subset A test the SRP models gas phase CM2 SRP AM1 SRP evaluation for methanol using Form A input testA20a gas phase AM1 SRP SCF evaluation for methanol using Form B in
14. Chem A 1998 102 1820 1831 The 6 31G d basis set was not available for I so we used the MIDI 6D instead The 6 31 G d basis set was also not available for I so we used the MIDI 6D basis set augmented by diffuse s and p shells with an exponent of 0 03 For the user s convenience the 6 31G d and 6 31 G d basis sets are included in the files 6 31GS bas and 6 31PGS bas respectively These files are located in the gmsplus v4 6 Basis directory Special Notes on SCF Schemes If ISCRF 0 only a gas phase calculation is performed and this is the default If ISCRF equals either 1 or 2 then the code does a liquid phase SCF calculation Two different SCF schemes have been implemented The standard scheme called scheme I uses the current solution phase bond order matrix values in the Fock or Kohn Sham operator at every step of the iteration Scheme I is chosen by setting ISCRF 1 Scheme II uses the gas phase bond order matrix values at the current geometry to calculate CM2 CM3 or CM4 corrections of L wdin or RLPA charges in solution Scheme II and its physical meaning may be justified as follows The modification to the L wdin or RLPA charges is parameterized to fit to experimental gas phase dipole moments So one can assume that the same modification applies to the solvated molecule with the same geometry in the liquid phase as in the gas phase The change of atomic partial charges due to the solvation effects is reflected by the ch
15. D 80 NTHE 16 NPHI 32 SEND UNC 1 END SEEQM EETYP DQDPH PHI 1 1 2873 0 0734 0 0734 END DFT DFTTYP MPWX HFE 0 428 NRA BASIS GBASIS N31 NGAUSS 6 NDF DATA ter MPW1K 6 31G d 0 8 0 0000000000 1 0 7633156080 1 0 7633156080 H END 0000000000 0931046833 0000000000 5148742072 0000000000 5148742072 71 Density Functionals Recommended for Use with CM4 CM4M and SM6 SM8 The CM4 and CM4M charge models and the SM6 and SM8 solvation models can be used with any density functional as long as the density functional gives a reasonable electronic distribution for the molecule of interest Shown in the table below is a list of density functional methods that are recommended for use with CM4 CM4M and SM6 SM8 in GAMESSPLUS For the hybrid density functional theory methods recommended for use in GAMESSPLUS the percent Hartree Fock exchange for each functional is also given Pure DFT functionals recommended for use in GAMESSPLUS Note that the keyword HFE 0 00 should be specified in the CM2 or MNGSM namelist when these or any other pure DFT functionals are used in conjunction with the CM4 or SM6 SM8 models Method GAMESSPLUS Keyword Reference s BLYP DFTTYP BLYP Becke A D Phys Rev A 1988 38 3098 HFE 0 0 must be Lee C Yang W Parr R G Phys Rev B 1988 37 785 specified in the CM2 or Miehlich B Savin A Stoll H Preuss H Chem Phys Lett GMSO
16. M J VWN in the CM2 or GMSOL J Phys Chem 1994 98 11623 namelist only mPW1PW91 0 250 DFTTYP MPWX Adamo C Barone V J Chem Phys HFE 0 250 must be specified 1998 708 664 12 in both the DFT namelist and the CM2 or GMSOL namelist MPWIS 0 060 DFTTYP MPWX Lynch B J Zhao Y Truhlar D G HFE 0 060 must be specified J Phys Chem A 2003 107 1384 in both the DFT namelist and the CM2 or GMSOL namelist MPWIN 0 406 DFTTYP MPWX Kormos B L Cramer C J HFE 0 406 must be specified J Phys Org Chem 2002 15 712 in both the DFT namelist and the CM2 or GMSOL namelist MPWIK 0 428 DFTTYP MPWX Lynch B J Fast P L Harris M HFE 0 428 must be specified Truhlar D G J Phys Chem A 2000 in both the DFT namelist and 104 4811 the CM2 or GMSOL namelist MPWX 0 000 0 999 DFTTYP MPWX Winget P Thompson J D Xidos J HFE X must be specified in D Cramer C J Truhlar D G both the DFT namelist and the J Phys Chem A 2002 106 10707 CM2 or GMSOL namelist Minnesota density functionals recommended for use in GAMESSPLUS Note that the HFE keyword should be specified in the CM2 or GMSOL namelist when these or any other hybrid DFT functionals are used in conjunction with the CM4M or SM6 SM8 models Method Keyword Fraction HFE Reference DFTTYP M05 0 280 Zhao Y Schultz N E Tru
17. Parameters below are effective only if IQMMM 1 IQMATM Integer array used to specify which atoms in AMBTOP input correspond to QM atom in DATA IQMATM i m means that i th QM atom given in DATA corresponds to m th atom in AMBTOP If IQMATM is negative and IQMATM j n j th QM atom in DATA is a link atom placed on a Q1 M1 bond and n th atom in AMBTOP is the M1 atom Q1 atom corresponding to the M1 atom is automatically detected from AMBTOP input Note that in that case j th coordinate in DATA is regarded not as that of the link atom but rather as that of the M1 atom The coordinate of the link atom is generated according to LNKTYP option RCUT Defines a cutoff distance r in A for the QM MM electrostatic interaction The default is cut 15 0 A ITAPER Specifies whether the TINKER tapering function is used for the QM MM electrostatic interactions or not See Section The TINKER tapering function for long range electrostatic interactions ITAPER 0 The TINKER tapering function is not used default ITAPER 1 The TINKER tapering function is used CTAPER LNKTYP LNKTYP 0 LNKTYP 1 RLINK SCLINK IBNDRY IBNDRY 0 IBNDRY 1 IBNDRY 2 IBNDRY 3 IBNDRY 4 CONVMM MXMSTP IPRIMM 62 Defines the fraction of 7 function for long range electrostatic interactions 7a CTAPER x value is 0 65 which is the same value as that in TINKER with respect to r cut See Section The
18. SRP Models GAMESSPLUS can use specific reaction parameters i e nonstandard parameters optimized for a specific system or reaction or limited range of systems or reactions for the NDDO Hamiltonians of the AM1 and PM3 models in the gas phase for the CM2 AM1 and CM2 PM3 methods and in the liquid phase for the CM2 AM1 CM2 PM3 SM5 42 AM1 and SM5 42 PM3 methods AM1 and PM3 calculations in either the gas phase or liquid phase may be performed without using the arithmetic mean rule for the resonance parameters In standard AM1 and PM3 calculations the 19 resonance parameter Jx for interaction of an orbital with angular momentum on an atom of element x and an orbital with angular momentum on an atom of element y is given by Prey Bix Bry 2 7 where fj and y are standard parameters The user can now override eq 7 by inputting specific values of the resonance parameter for one or more sets of l x l and y A reference for this general procedure is reference CE95 in the Literature References section Solubility Calculations The solubility of a given solute A in a liquid solvent B are calculated using a thermodynamic relationship between the solubility free energy of solvation and pure substance vapor pressure of solute A which is given by _ Px z acs 2 v s RT 8 In this equation S is the solubility of solute A in solvent B Px is the equilibrium vapor pressure of solute A over a pure solution of A P
19. Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions April 11 2008 R1 plus others see the section on Makepatch Method for Updating and Compiling GAMESSPLUS New version by M Higashi and D G Truhlar The major modifications of the code in this version are as follows e The capability to carry out QM MM geometry optimization with a site site representation of the QM MM electrostatic interaction using link atoms if the QM MM boundary passes through a covalent bond has been added The new namelists AMBTOP AMBCRD and QMMM were added for this purpose e The capability to carry out constrained geometry optimization in Cartesian coordinates by a projection operator method has been added The new namelist INTFRZ was added for this purpose GAMESSPLUS Version 2010 2 September 2010 Authors M Higashi A V Marenich R M Olson A C Chamberlin J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions April 11 2008 R1 plus others see the section on Makepatch Method for Updating and Compiling GAMESSPLUS The major modifications of the code in this version are as follows e The SM8AD solvation model was added 110 APPENDIX I GAMESSPLUS Solubi
20. hence 2 SM5 42 provides a way to calculate electronic wave functions in different liquid phases and in soil and free energies of transfer between the gas phase and various condensed phases The SM5 42 solvation model is universal i e it is parameterized for water and any medium that can be characterized by a set of solvent descriptors In the case of bulk soil a set of solvent descriptors have been empirically optimized using a data base of experimental Ko values see Winget P Cramer C J Truhlar D G Environ Sci Technol 2000 34 4733 In the GAMESSPLUS Soil sorption utility program standard state free energies of transfer are calculated using rigid gas phase geometries i e no geometry optimization in solution is performed Our experience has shown that in most cases there is very little error associated with using gas phase geometries to calculate free energies of solvation Thus one may use as input gas phase geometries optimized at any reliable level of theory However it is important to point out that in some cases the geometry of a given solute might undergo significant relaxation between the gas phase and solution or soil As a result the transfer free energy calculated using a rigid gas phase geometry might differ significantly from the transfer free energy calculated using a relaxed geometry In cases where significant geometric relaxation is expected to occur upon transfer from the gas phase to water or soil a g
21. page http ambermd org Questions resp html When one wants to use AM1 BCC charges one can directly obtain an AMBER parameter file DCE_bcc mol2 from the PDB file by antechamber antechamber i DCE pdb fi pdb o DCE_bcc mol2 fo mol2 c bec Note that the charges as determined above are not used in the QM MM calculation because the QM charges are replaced by the ones derived from the QM electronic structure calculations The charge determination procedure is nevertheless required if the charges are to be used in equilibrium MM MD simulations see below 86 After AMBER parameter mol2 files are created by antechamber one should check if there are missing parameters for the molecules in the GAFF parameters file gaff dat which is in AMBERHOME dat leap parm directory by parmchk program Note that antechamber can find atom types e g c3 or h1 in the above case and bonds for the molecules but does not check if all the parameters for the molecules are available or not The usage of parmchk is as follows parmchk i DCE_resp mol2 f mol2 o DCE_resp frcmod If there are missing parameters they will be written with parameters for more general atom types which will be used instead in a force field modification frcmod file The fremod file looks like this remark goes here MASS BOND ANGLE Ca e3 el 64 784 110 735 Calculated with empirical approach cl cl cx 56 400 LTT 29 90 same as cl cl
22. src int2a src mpcint src mpcmol src mthlib src rhfuhf src and statpt src The subroutine tdgrad was removed from mhtlib src because in the latest version of GAMESS this subroutine is located in the module gamess src The size of pnrm 35 in the shinrm common block in the GAMESSPLUS patch of the grd2c srce module was not compatible with the size of this block in other modules The correct size is 84 GAMESSPLUS Version 2009 April 2009 109 Authors M Higashi A V Marenich R M Olson A C Chamberlin J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions April 11 2008 R1 plus others see the section on Makepatch Method for Updating and Compiling GAMESSPLUS New version by M Higashi and D G Truhlar The major modifications of the code in this version are as follows e The electrostatic potential distribution can be calculated from the MM charges and coordinates in the EEQM calculation Some options were added in namelist EEQM Namelist MM was added e The previous versions of GAMESSPLUS do not support ICMD 420 new CM4 and CM4M charge models added in GAMESSPLUS v2008 2 in the EEQM calculations This version supports ICMD gt 420 in the EEQM calculations GAMESSPLUS Version 2010 February 2010 Authors M Higashi A V Marenich R M Olson A C Chamberlin J Pu J D
23. 0 000000 1 198348 691865 S 3 Q PPNHPNRPENENWNHE W PRPENOPNFPEPENENWNHE fon fon oO fo PPNHPNHNPENENWNHE W OY O K HH R A NLA DO WN A WW 155333 23 923290 153 23 923290 T53 23 923290 153 23 923290 172260 073030 725570 455040 147070 251310 863270 201350 0 172260 073030 725570 455040 147070 251310 863270 201350 0 172260 073030 725570 455040 147070 251310 863270 201350 0 172260 073030 7255970 455040 147070 251310 863270 Hoey Soe Rae oe fae a Ea ce 00000 OO CO le Ba A a a 00000 C lt 3 a CO Eaa O 1 0707400000 3953800000 6633110000 0813800000 5748530000 0000000000 1099310000 4627130000 0000000000 198348 0707400000 3953800000 6633110000 0813800000 5748530000 0000000000 1099310000 4627130000 0000000000 000000 0707400000 3953800000 6633110000 0813800000 5748530000 0000000000 1099310000 4627130000 0000000000 198348 0707400000 3953800000 6633110000 0813800000 5748530000 0000000000 1099310000 4627130000 691865 1 383730 691865 122 6 PRPFNHRFNRFRPRFNFNWDNDNE W 6 PRFNHNHRFRFNRPRPNFNWDNEN EF W PRPNMNEN PRPNOEFE ND PRPNOFR ND 1533 23 923290 T53 23 923290
24. 1 395043 H 1 0 000000 2 160477 1 247356 H 15 0 000000 2 160477 1 247356 H 1 0 000000 000000 2 494707 H 1 0 000000 2 160477 1 247356 H 1 0 000000 2 160477 1 247356 H 1 0 000000 000000 2 494707 SSCF AM1 SSOILDENSITY 0 11 g mL SDELTAGW 0 87 kcal mol All values are for a temperature of 298 K Fr nergies are calculated using a 1M gt 1M standard state log log basel0 Free Energy of Transfer air gt water 0 87 kcal mol Free Energy of Transfer air gt soil 4 32 kcal mol Free Energy of Transfer water gt soil 3 45 kcal mol log Koc 34 00 End of output test3 log GAMESSPLUS Soil sorption utility calculation based on GAMESSPLUS 2008 Tue Apr 8 20 00 43 2008 KOC CALCULATION FOR BENZENE USING CALCULATED VALUES FOR DELTAGW AND DELTAGSOIL C1 SSCF ABINITIO SSOILDENSITY S 0 11 g mL All values are for a temperature of 298 K EX nergies are calculated using a 1M gt 1M standard state log log basel0 126 Free Energy of Transfer air gt water 1 34 kcal mol Free Energy of Transfer air gt soil 3 77 kcal mol Free Energy of Transfer water gt soil 2 43 kcal mol log Koc 2 74 End of output Installing and running the soil sorption utility program A working version of GAMESSPLUS must be installed first see the sections entitled Notes on Running GAMESSPLUS and Updating and Compiling GAMESSPLUS in this m
25. 103 4893 Solubility TCO3 Thompson J D Cramer C J Truhlar D G Predicting Aqueous Solubilities From Aqueous Free Energies of Solvation and Experimental or Calculated Vapor Pressures of Pure Substances J Chem Phys 2003 119 1661 1670 QM MM with GHO AIHF boundary treatment PG04 Pu J Gao J Truhlar D G Generalized Hybrid Orbital GHO Method for Combining Ab Initio Hartree Fock Wave Functions with Molecular Mechanics J Phys Chem A 2004 108 632 650 Charge response kernel MK97 Morita A Kato S Ab Initio Molecular Orbital Theory on Intramolecular Charge Polarization Effect of Hydrogen Abstraction on the Charge Sensitivity of Aromatic and Nonaromatic Species J Am Chem Soc 1997 119 4021 4032 LY04 Lu Z Yang W Reaction path potential for complex systems derived from combined ab initio quantum mechanical and molecular mechanical calculations J Chem Phys 2004 121 89 100 HTO8 Higashi M Truhlar D G Electrostatically Embedded Multiconfiguration Molecular Mechanics Based on the Combined Density Functional and Molecular Mechanical Method J Chem Theory Comput 2008 4 790 803 TINKER tapering fucntion 35 PJ97 Ponder J W TINKER version 3 5 Washington University St Louis MO 1997 SB93 Steinbach P J Brooks B R New Spherical Cutoff Methods for Long Range Forces in Macromolecular Simulation J Comput Chem 1993 15 667 683 QM
26. 2003 R2 The generalized hybrid orbital GHO module is added into the GAMESSPLUS program to combine ab initio HF wave functions with molecular mechanics Four orbital orthogonalization schemes for this purpose were implemented with energy and analytical gradients available for all four Using these methods requires building the CHARMM GAMESSPLUS combination program which is described in the CGPLUS user manual Twelve modules bassto gamess grdl grd2a grd2b grd2c inputa inputb intl int2a rhfuhf symorb of GAMESS have been modified to incorporate the GHO functionality in GAMESSPLUS The GHO related modifications can be located by the string QLINK in the corresponding src files for these modules listed above A new gho module is added to GAMESSPLUS in a file called gho src to accommodate routines for defining GHO data structure constructing the basis transformations and performing gradient calculations 102 e A file called ghodum src has been added to compile GAMESSPLUS based on GAMESS in the absence of CHARMM The gho module in GAMESSPLUS is written largely following the CHARMM programming style which utilizes include statements and conditional compilations through a pre processor in CHARMM For compiling the CHARMM GAMESSPLUS as an integrated executable locating the CHARMM pre processor will be properly handled by the utility package CGPLUS when the gho module is compiled therefore it is not a problem However for a user who
27. 303 SM5 43 HF 6 31G d 313 SM5 43 B3LYP 6 31G d 315 SM5 43 MPWX MIDI 316 SM5 43 MPWX MIDI 6D 317 SM5 43 MPWX 6 31G d 318 SM5 43 MPWX 6 31 G d 319 SM5 43 MPWX 6 31 G d p ICDS 0 All atomic surface tension coefficients equal zero Gcpg 0 By default ICDS is set equal to 1 Note that the B3LYP method in the table above corresponds to the one requested by the DFTTYP B3LYP3 keyword in the DFT data group see the section entitled Notes on GAMESSPLUS input above 114 IAQU Determines the solvent type IAQU 0 Organic solvent additional solvent data must be input see below TAQU 1 Aqueous solvent default no additional solvent data is required Solvent Descriptors If IAQU 0 then the solvent properties are specified by a series of solvent property descriptor values Dielec dielectric constant of solvent default is the value for water 78 3 SolN index of refraction at optical frequencies at 293 K ny default is 1 0 SolA Abraham s hydrogen bond acidity 2 default is 0 0 SolB Abraham s hydrogen bond basicity rp default is 0 0 SolG Y YmlY default is 0 0 where Ym is the macroscopic surface tension at air solvent interface at 298 K and y is 1 cal mol A note that 1 dyne cm 1 43932 cal mol A SolC aromaticity the fraction of non hydrogenic solvent atoms that are aromatic carbon atoms default is 0 0 SolH electronegative halogenicity y the fraction of non hydrogenic s
28. 40 c d SM8 M06 DZVP 6 508 801 6 0 40 c d SM8 M06 6 31B d 6 509 801 6 0 40 c d SM8 M06 6 31B d p 6 510 801 6 0 40 cd In the above table X corresponds to the percentage of Hartree Fock exchange used in the mPW hybrid density functional see the HFE keyword for more details For SM6 DFT refers to any good density functional A list of recommended density functionals that are supported in GAMESSPLUS is given in the section entitled Density Functionals Recommended for Use with CM4 CM4M and SM6 SM6 in GAMESSPLUS Also given in this section are the HFE values required by all of the recommended density functionals This keyword is specified in the CONTRL namelist b Analytic gradients not available for this method When using any of the CM4 or SM6 SM8 SM8AD methods the HFE keyword must always be specified HFE 0 00 should be entered when using pure DFT functionals dM06 stands for the M06 suite of density functionals M06 L M06 M06 2X or M06 HF Namelist EEQM Namelist EEQM controls EEQM calculations with a site site representation of the QMW MM electrostatic interaction It is required that RUNTYP EEQM in namelist CONTRL to carry out an EEQM calculation Currently GAMESSPLUS doesn t support restricted open shell Hartree Fock and restricted open shell Kohn Sham calculations when the EEQM option is selected EETYP Determines the type of EEQM calculation VF R WY H 0 4Q D EETYP ENERGY Calculate en
29. 6D MPWX 6 31G d MPW4 6 31 G d and MPWX 6 31 G d p TCO5 Thompson J D Cramer C J Truhlar D G Density Functional Theory and Hybrid Density Functional Theory Continuum Solvation Models for Aqueous and Organic Solvents Universal SM5 43 and SM5 43R Solvation Models for Any Fraction of Hartree Fock Exchange Theor Chem Acc 2005 113 107 SM6 aqueous model KCOSb Kelly C P Cramer C J Truhlar D G SM6 A Density Functional Theory Continuum Solvation Model for Calculating Aqueous Solvation Free Energies of Neutrals Ions and Solute Water Clusters J Chem Theory Comput 2005 1 1133 SM8 universal model MO7 Marenich A V Olson R M Kelly C P Cramer C J Truhlar D G Self consistent reaction field model for aqueous and nonaqueous solutions based on accurate polarized partial charges J Chem Theory Comput 2007 3 2011 33 SM8AD universal model MCO09 Marenich A V Cramer C J Truhlar D G Universal solvation model based on the generalized Born approximation with asymmetric descreening J Chem Theory Comput 2009 5 2447 Temperature dependent aqueous models AC06 Chamberlin A C Cramer C J Truhlar D G Predicting Free Energies of Solvation as Functions of Temperature J Phys Chem B 2006 110 5665 ACO8 Chamberlin A C Cramer C J Truhlar D G Extension of a temperature dependent aqueous solvation model to compounds containin
30. 80 1 98 2 22 2 12 Cl 1 75 2 13 2 28 2 02 Ar 1 88 K 2 75 Ni 1 63 Cu 1 40 Zn 1 39 Ga 2 40 Ge 2 10 As 1 85 Se 1 90 Br 1 85 2 31 2 38 2 60 Kr 2 02 Pd 1 63 Ag 1 72 Cd 1 58 In 2 50 Sn 2 20 Sb 2 10 Te 2 06 I 1 98 2 66 Xe 2 60 For atoms not listed in this table a radius of 2 0 A is used Bondi s values for the van der Waals radius Bondi A J Phys Chem 1964 68 441 In cases where the atomic radius is not defined Bondi s values are used In cases where the atomic radius has not been defined by Bondi 2 0 A is used for the atomic radius These radii have been optimized for use with the SM5 42 model These radii have been optimized for use with the SM5 43 model These radii have been optimized for use with the SM6 model ISTS ISTS 5 ISTS 6 Determines the type of surface tension functionals Use SM5 functional forms to evaluate Gcps This is the default when the SM5 42 and SMS5 43 models are invoked Use SM6 functional forms to evaluate Gcps This is the default when the SM6 and SM8 models are invoked 47 Note The ISTS option is deprecated in the current version of GAMESSPLUS SolvRd IAQU IAQU 0 IAQU 1 Value of the solvent radius in angstroms for the calcul
31. CM3 and CM4 yield very accurate charges both in the gas phase and in liquid phase solutions and this is useful for a qualitative understanding of solvent induced changes in the solute We should note here that partial atomic charges are not physical observables but they can still be considered accurate within a given model context if they vary physically with molecular geometry and environment and can be used to accurately reproduce observables such as dipole moments or if they can be derived consistently and realistically from accurate experimental data for instance from the dipole moment of a diatomic molecule e SM5 SM6 SM8 and SM8AD parameterizations included an extremely broad range of solute functional groups including molecules containing phosphorus which are very hard to treat e SMx do not need to be corrected for outlying charge error which can be large in some other methods Furthermore our most recent models SM8 and SM8AD have several advantages compared to earlier solvent models e g SM5 42 SM5 43 or SM6 developed within our group e SM8 can be used with any of the density functional methods supported in GAMESSPLUS e SM8 significantly outperforms SM5 42 SM5 43 and all other competing continuum solvation models against which it has been tested prior to SMVLE for predicting aqueous solvation free energies of ions This is important because aqueous solvation free energies of ions can be used in various thermodynami
32. Calculate RLPA charges 301 CM3 HE MIDI 302 CM3 HEF MIDI 6D 303 CM3 HF 6 3 1G d 311 CM3 AM1 312 CM3 PM3 313 CM3 B3LY P 6 31G d 44 314 CM3 B3LY P 6 31 G d 315 CM3 MPWX MIDI 316 CM3 MPWX MIDI 6D 317 CM3 MPWX 6 31G d 318 CM3 MPWX 6 31 G d uses RLPA charges 319 CM3 MPWX 6 31 G d p uses RLPA charges 320 CM3 B3LYP MIDI 6D 321 CM3 BLYP 6 31G d 322 CM3 1 HF MIDI 415 CM4 DFT MIDI 416 CM4 DFT MIDI 6D 417 CM4 DFT 6 31G d 418 CM4 DFT 6 31 G d uses RLPA charges 419 CM4 DFT 6 31 G d p uses RLPA charges 420 CM4 DFT 6 31G d p 422 CM4 DFT cc pVDZ 423 CM4 DFT DZVP 424 CM4 DFT 6 31B d 425 CM4 DFT 6 31B d p 500 CM4M M06 MIDI 501 CM4M M06 MIDI 6D 502 CM4M M06 6 31G d 503 CM4M M06 6 31 G d 504 CM4M M06 6 31 G d p 505 CM4M M06 6 31G d p 506 CM4M M06 cc pVDZ 508 CM4M M06 DZVP 509 CM4M M06 6 31B d 510 CM4M M06 6 31B d p In the above tables X corresponds to the percentage of Hartree Fock exchange see the HFE keyword for more details For SM6 DFT refers to any good density functional A list of density functionals that are available in GAMESS and GAMESSPLUS is given in the section entitled Density Functional Methods Available in GAMESS and GAMESSPLUS In cases where the default value of ICDS 0 a solvation model does not exist for that particular set of charges For ICMD 300 all re
33. Cramer and D G Truhlar GAMESS version October 31 1996 and January 6 1998 e This version can be used with either the Oct 31 1996 version or the Jan 6 1998 version of GAMESS The same CM2 and SM5 42R methods are supported as in version 1 0 GAMESOL Version 1 1 1 September 1998 95 Authors J Li G D Hawkins D A Liotard C J Cramer and D G Truhlar GAMESS version October 31 1996 and January 6 1998 e This version fixed two bugs in subroutine SMS5CDS for the surface tension functional and a bug in subroutine GBSCRF for option ISCRF 1 A few typos in the text file of solvent properties have also been corrected GAMESOL Version 2 0 September 1998 Authors J Li T Zhu G D Hawkins D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version January 6 1998 e This version added analytical gradients and the capability to optimize solute geometries in liquid solutions by the SM5 42 HF solvation model GAMESOL Version 2 0 1 January 1999 Authors J Li T Zhu G D Hawkins D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version January 6 1998 e Bugs in the subroutine for the CDS term were fixed The bugs were for the terms involving CC triple bonds and amides e Some errors in parameters for SM5 42R AM1 and SM5 42R PM3 models have been corrected e The code now uses the correct Coulomb radii for SM5 42R AM1 SM5 42R PM3 SM5 42 AM1 and SM5 42 PM3 when these methods are invo
34. DK99 Link atom method fixed length WC07 AMBER redistributed charge WC07 RC and RCD LTOS BRC and BRCD WT10 38 39 Usage This section highlights important GAMESS input options and describes the input options in the namelists CM2 GMSOL CM2SRP NDDOSRP EEQM MM AMBTOP AMBCRD QMMM and INTFRZ Namelists CM2 and GMSOL are used to define the specifics of calculations involving CM2 CM3 CM4 CM4M SM5 42 SM5 43 SM6 and SM8 Note that GMSOL and CM2 define the same input and only one of them should be defined in a given input file If they are both defined in the same input file the input parameters defined in the second instance will be ignored Namelist CM2SRP is used for reading the external parameters for CM2 SRP methods and namelist NDDOSRP is used for reading the external parameters for NDDO SRP methods Namelist EEQM is used for reading some parameters for the EEQM method Namelist MM is used for reading MM charges and coordinates when IRDMM 1 in EEQM AMBTOP and AMBCRD is used for reading AMBER parameter topology and coordinate inputs for QM MM calculation Namelist QMMVM is used for reading some parameters for QM MM method Namelist INTFRZ is used to define internal coordinates to be constrained Note GHO AIHF calculations are carried by the CHARMM GAMESSPLUS combination package and the CHARMM input file controls the GHO options A detailed description is available in documentation for the CGPLUS
35. Extracting Accurate Partial Charges from Wave Functions J Phys Chem A 1998 102 1820 Parameterization of the CM2 model for HF cc pVDZ LX99 Li J Xing J Cramer C J Truhlar D G Accurate Dipole Moments from Hartree Fock Calculations by Means of Class IV Charges J Chem Phys 1999 111 885 Original parameterizations of the CM3 Charge Model WTO02 Winget P Thompson J D Xidos J D Cramer C J Truhlar D G Charge Model 3 A Class IV Charge Model based on Hybrid Density Functional Theory with Variable Exchange J Phys Chem A 2002 106 10707 Extension of CM3 to AM1 PM3 BLYP and B3LYP TCO3 Thompson J D Cramer C J Truhlar D G Parameterization of Charge Model 3 For AM1 PM3 BLYP and B3LYP J Comput Chem 2003 24 1291 Extension of CM3 and CM3 1 to HF MIDI KCO0Sa Kelly C P Cramer C J Truhlar D G Accurate Partial Atomic Charges for High Energy Molecules with the MIDI Basis Set Theor Chem Acc 2005 113 133 Parameterizations of the CM4 Charge Model KCOSb Kelly C P Cramer C J Truhlar D G SM6 A Density Functional Theory Continuum Solvation Model for Calculating Aqueous Solvation Free Energies of Neutrals Ions and Solute Water Clusters J Chem Theory Comput 2005 1 1133 SM5 42 DFT method and ISCRF 1 option ZL98 Zhu T Li J Hawkins G D Cramer C J Truhlar D G Density Functional Solvation Model Bas
36. For example for SGI computers the optimization level has been set to O3 in the compilation script comp for most of the source files For this setting the gradient evaluations in testw x2y where w 1 8 x 1 or 2 and y 2 3 11 and 12 produces erroneous results and the geometry optimizations in test cases test13 and test14 both fail Reducing the optimization level for grd1 src to O1 solves this problem The optimization level is set in the compilation script comp The following is the portion of the script comp that sets compile options for SGI machines modified to lower the optimization level for grdl src if STARGET sgi32 TARGET sgi64 then set OPT 03 if SMODULE grd2c set OPT 02 if SMODULE rxncrd set OPT 02 if SMODULE grdl set OPT Ol1 set ARCH r12000 set INTS if STARGET sgi64 set INTS i8 set FLAGS 64 mips4 automatic G0 woff 2290 OPT Olimit 0 set echo 77 c SOPT SARCH SFLAGS S INTS SMODULE f unset echo endif 79 6 Type compall and GAMESS should start compiling After the GAMESS compilation type ked An executable named gamess 00 x will be created Optionally typing ked name version where name is a filename and version is a number string produces a GAMESS executable named name version x e g typing ked gamessplus 01 gives an executable named gamessplus 01 x Note that modifying name will require appropriate modifications to
37. GAMESSPLUS combination package for calculations that combine ab initio HF wave functions with molecular mechanics For the QMW MM partition along a covalent bond the generalized hybrid orbital GHO method is used to provide a smooth connection between the QM subsystem and the MM subsystem In the GHO treatment sp carbons are often chosen as GHO boundary atoms denoted by B Such a B atom is both a QM atom and an MM atom The QM atom bonded to B is called a QM frontier atom denoted by A The other three MM atoms directly bonded to B are denoted by X Y and Z A set of generalized hybrid orbitals 78 gt ho 1y 1z is placed on the GHO boundary atom B where the hybridization scheme is completely determined by the local geometry of the QM MM boundary atoms Q B X Y and Z Among the four hybrid orbitals one approximately pointing toward A denoted by 7p will participate in the SCF procedure with other QM basis functions and is therefore called an active hybrid orbital The remaining three hybrid orbitals 7 7y 1z are called auxiliary orbitals and they are excluded from the SCF procedure With this restriction on one hand the active molecular orbitals MOs in GHO are only allowed to be expanded over the active basis functions including 7g On the other hand each auxiliary hybrid orbital forms an auxiliary MO by itself and it is occupied by a fixed auxiliary charge density To distribute the MM point charge on B over the three auxiliary or
38. How well does a restrained electrostatic potential RESP model perform in calculating conformational energies of organic and biological molecules J Comput Chem 2000 27 1049 1074 HAO6 Hornak V Abel R Okur A Strockbine B Roitberg A Simmerling C Comparison of multiple AMBER force fields and development of improved protein backbone parameters Proteins Struct Funct Genet 2006 65 712 725 TIP3P water model JC83 Jorgensen W L Chandrasekhar J Madura J Klein M L Comparison of simple potential functions for simulating liquid water J Chem Phys 1983 79 926 935 36 General AMBER force field GAFF WW04 Wang J Wolf R M Caldwell J W Kollman P A Case D A Development and Testing of a General Amber Force Field J Comput Chem 2004 25 1157 1174 Link atom method with the Q1 QL bond length held to a fixed fraction of the Q1 M1 bond length DK99_ Dapprich S Kom romi I Byun K Morokuma K Frisch M A new ONIOM implementation in Gaussian98 Part I The calculation of energies gradients vibrational frequencies and electric field derivatives THEOCHEM 1999 461 462 1 21 Link atom method with fixed Q1 QL bond lengths and with the AMBER redistributed charge scheme WCO07 Walker R C Crowley M F Case D A The implementation of a fast and accurate QM MM potential method in Amber J Comput Chem 2007 29 1019 1031 Redistributed charge RC
39. METHOD M M is GRID to request grid based DFT calculation default Note All SM5 42 and SMS 43 parameterizations that are based on the B3LYP method use version III of the VWN correlation functional so these types of calculations must be carried out using DFTTYP B3LYP3 not DFTTYP B3LYP35 In addition all SM5 42 and SM5 43 methods that are based on the BLYP method and or the B3LYP B3LYP3 method were parameterized using grid based DFT so METHOD GRID the default for METHOD should always be used Namelist BASIS input of available standard basis sets GBASIS N31 Pople s N 31G basis sets AMI AM1 model Hamiltonian PM3 PM3 model Hamiltonian NGAUSS N the N of N31 e g N 6 for 6 31G NDFUNC N N is the number of polarizing d subshells on heavy atoms N 1 for 6 31G d default 0 DIFFSP TRUE adds a diffuse sp shell to the basis set default FALSE Note The d subshells have 5 functions if ISPHER 1 and they have 6 functions if ISPHER 1 where ISPHER is defined in the CONTRL data group Examples 6 31G d SBASIS GBASIS N31 NGAUSS 6 NDFUNC 1 SEND 6 31 G d SBASIS GBASIS N31 NGAUSS 6 NDFUNC 1 DIFFSP Namelist DATA input molecule information Input consists of the following information line 1 title line TRUE END 113 line 2 symmetry group always enter C1 for GAMESSPLUS calculations line 3 molecular coordinates for Cartesian input each line consists
40. MM method with a site site representation of the QM MM electrostatic interaction HOOO Hayashi S Ohmine I Proton Transfer in Bacteriorhodopsin Structure Excitation IR Spectra and Potential Energy Surface Analyses by an ab Initio QM MM Method J Phys Chem B 2000 104 10678 10691 HTO8 Higashi M Truhlar D G Electrostatically Embedded Multiconfiguration Molecular Mechanics Based on the Combined Density Functional and Molecular Mechanical Method J Chem Theory Comput 2008 4 790 803 HTO9 Higashi M Truhlar D G Efficient Approach to Reactive Molecular Dynamics with Accurate Forces J Chem Theory Comput 2009 5 2925 2929 AMBER ff94 force field this provides the functional form that we call the AMBER force field CC95 Cornell W D Cieplak P Bayly C I Gould I R Merz Jr K M Ferguson D M Spellmeyer D C Fox T Caldwell J W Kollman P A A Second Generation Force Field for the Simulation of Proteins Nucleic Acids and Organic Molecules J Am Chem Soc 1995 117 5179 5197 AMBER ff03 force field DW0O3 Duan Y Wu C Chowdhury S Lee M C Xiong G Zhang W Yang R Cieplak P Luo R Lee T A point charge force field for molecular mechanics simulations of proteins based on condensed phase quantum mechanical calculations J Comput Chem 2003 24 1999 2012 AMBER ff99SB force field WC00 Wang J Cieplak P Kollman P A
41. Power 3 processors nodes running AIX version 5 1 compiled with the XL Fortran compiler version 7 1 12 in the comp compall and lked compilation scripts TARGET was set to ibm64 IBM SP with NightHawk Power3 processors running AIX version 5 1 compiled with the XL Fortran compiler version 7 1 12 in the comp compall and lked compilation scripts TARGET was set to ibm64 IBM Regatta Power 4 processors that is pSeries 690 and pSeries 655 nodes running AIX version 5 1 compiled with the XL Fortran compiler version 7 1 12 in the comp compall and lked compilation scripts TARGET was set to ibm64 Netfinity Linux cluster running RedHat Linux version 7 2 and kernel version 2 4 9 compiled with the g77 compiler version 3 2 in the comp compall and lked compilation scripts TARGET was set to linux pc Sun Blade 2000 with UltraSparc III processors running Solaris 8 compiled with Forte Developer 7 Fortran version 7 0 compiler in the comp compall and lked compilation scripts TARGET was set to sun64 SGI Altix 3000 with Madison processors running RedHat Linux and compiled with the Intel Fortran compiler version 8 0 in the comp compall and lked compilation scripts TARGET was set to linux ia64 Versions previous to version 3 9 have been successfully tested on SGI Origin 3800 with R14000 CPUs running IRIX 6 5 12f compiled with MIPSpro compiler version 7 3 1 2m in th
42. TINKER tapering The default Nout r Specifies how the position of the link atom is determined See the section entitled QM MM Potential Energy Calculation and Geometry Optimization with a Site Site Representation of the QM MM Electrostatic Interaction The length of Q1 QL bond is fixed Eq 35 default The fraction of Q1 QL bond length to QI M1 one is fixed Eq 34 Defines the Q1 QL bond length in A when LNKTYP 0 The default value is 1 090 which is the same as the CT HC bond length in the AMBER force field Defines fractions of Q1 QL bond length to Q1 M1 one when LNKTYP 1 The default value is 0 714 which is the fraction of CT HC bond length 1 090 A to CT CT one 1 526 A in the AMBER force field Specifies how the QM MM electrostatic interaction near the QM MM boundary is treated AMBER default scheme corresponding to adjust_g 2 in AMBER is used default The redistributed charge RC scheme is used The redistributed charge and dipole RCD scheme is used The balanced redistributed charge BRC scheme is used The balanced redistributed charge dipole BRCD scheme is used Specifies the gradient convergence criterion in hartree bohr in MM geometry optimization MM geometry optimization will be finished when the maximum MM gradient component becomes below CONVMM The default value is 1 0x 10 4 Defines the maximum number of MM geometry optimization steps The default is 10000 The MM energy and gradie
43. and redistributed charge and dipole RCD schemes LTOS Lin H Truhlar D G Redistributed Charge and Dipole Schemes for Combined Quantum Mechanical and Molecular Mechanical Calculations J Phys Chem A 2005 109 3991 4004 Balanced redistributed charge BRC and balanced redistributed charge and dipole BRCD schemes WT10 Wang B Truhlar D G Combined Quantum Mechanical and Molecular Mechanical Methods for Calculating Potential Energy Surfaces Tuned and Balanced Redistributed Charge Algorithm J Chem Theory Comput 2010 6 in press Projection Operator Method for Geometry Optimization with Constraints LZ91 D h Lu M Zhao and D G Truhlar Projection Operator Method for Geometry Optimization with Constraints Journal of Computational Chemistry 12 376 384 1991 Quick index to literature The references for the methods supported by GAMESSPLUS are as follows Method Reference Lowdin charges LD50 RLPA charges TX02 mPW hybrid functional AB98 PTO2 and LZ03 MPWIK LF00 MPWX PTO2 CM2 AM1 LZ98 CM2 PM3 LZ98 CM2 HF MIDI LZ98 CM2 HEF MIDI 6D LZ98 CM2 HF 6 31G d LZ98 CM2 BPW91 MIDI LZ98 CM2 BPW91 MIDI 6D LZ98 CM2 B3LYP MIDI LZ98 CM2 BPW91 6 31G d LZ98 CM2 HF 6 31 G d LZ98 CM2 HF cc pVDZ LX99 CM2 BPW91 DZVP LZ98 CM3 AM1 TCO
44. build the molecular cavity to be used in the SCRF calculation This is the default for ICDS 800 Use the atomic radii optimized for predicting solvation free energies with the SM8AD solvation model to build the molecular cavity to be used in the SCRF calculation This is the default for ICDS 801 The four sets of radii described above are listed below Note that Coulomb radii for solutes in water for the case of IRADII 5 will be equal to those for the case of IRADII 4 The definition of nonaqueous radii for the case of IRADII 5 is given in the reference MO07 which describes the SM8 model In cases where the atomic radius of a given atom is not defined the default value is Bondi s value for the radius For elements whose atomic radii are not listed in Bondi s paper a value of 2 0 A is used for the radius Bondi A J Phys Chem 1964 68 441 Thus Generalized Born calculations may be carried out for a molecule containing any element on the periodic table Element IRADII 1 IRADI 24 IRADII 3 IRADII 4 H 1 20 0 91 0 79 1 02 Li 1 82 1 32 C 1 70 1 78 1 81 1 57 46 N 1 55 1 92 1 66 1 61 O 1 52 1 60 1 63 1 52 F 1 47 1 50 1 58 1 47 Ne 1 54 Na 2 27 Mg 1 73 Al 2 50 Si 2 10 P 1 80 2 27 2 01 1 80 S 1
45. c3 cl cx he 48 300 109 750 same as cl c3 he cl cx cx 64 200 TL 590 same as cl c3 c3 DIHE IMPROPER ca ca ca ha Lal 180 0 220 General improper torsional angle 2 general atom types n O c OSs 105 180 0 2 0 General improper torsional angle 2 general atom types c ca n NN Tl 180 0 23 0 General improper torsional angle 2 general atom types ca ca ca n Lal 180 0 2 0 Using default value NONBON This is fremod save file in AMBERHOME test antechamber sustiva directory In that case for example parameters for cl cl cx angles where cl means general sp carbon and cx means sp carbon in triangle systems are unavailable and parameters for cl cl c3 angles where c3 means general sp carbon will be used for cl cl cx angles instead In the example case all the parameters required to describe DCE are available in the GAFF parameter file Therefore no additional parameter appears in the frcmod file DCE_resp frcmod Make AMBER topology parameter and coordinate files by LEaP Note that before running LEaP one has to specify what version of the AMBER force field is used for proteins In the example case the AMBER ff03 force field is chosen by using the following commands cd SAMBERHOME dat leap cmd ln s leaprce ff 03 r1l leapre First a LEaP input file leap inp which consists of a series of commands is prepared like this source leaprc gaff mods loadAmberParams DCE_resp frcmod DCE loadMol2 DCE_resp mol2
46. charges with minimal computational cost PhysChemComm 2001 4 72 Corrected modified Perdew Wang hybrid DFT method LZ03 Lynch B J Zhao Y Truhlar D G The Effectiveness of Diffuse Basis Functions for Calculating Relative Energies by Density Functional Theory J Phys Chem A 2003 107 1384 MPWIK LFOO Lynch B J Fast P L Harris M Truhlar D G Adiabatic Connection for Kinetics J Phys Chem A 2000 104 4811 MPWPW91 mPW1PW91 and MPWX AB98 Adamo C Barone V Exchange functionals with Improved Long Range Behavior and Adiabatic Connection Methods without Adjustable Parameters The mPW and mPW1PW Models J Chem Phys 1998 108 664 PTO2 Pu J Trublar D G Parameterized Direct Dynamics Study of Rate Constants of H with CH4 from 250 to 2400K J Chem Phys 2002 116 1468 L wdin population analysis LD50 L wdin P O J Chem Phys 1950 18 365 b Del Re G In Quantum Science Calais J L Goscinski O Linderberg J Ohrn Y Eds Plenum Press New York 1976 p 53 31 Redistributed L wdin population analysis TX02 Thompson J D Xidos J D Sonbuchner T M Cramer C J Truhlar D G More Reliable Partial Atomic Charges when using Diffuse Basis Sets PhysChemComm 2002 5 117 Original parameterizations of the CM2 Charge Model LZ98 Li J Zhu T Cramer C J Truhlar D G A New Class IV Charge Model for
47. eE ESEO SEEPS EEE Trae REEERE SESS 118 S lyent d scriptors fort B lk solli seitas aare aea a eae aa eaaa E a DEES eS E EEE EAER E EEEa 119 USA GC ss E EERE E EE NE 120 IRES REII FLAKO aT EE ATEA Reece T Eo ea eR 120 Executive Summary GAMESSPLUS is a module that currently incorporates the following methods into GAMESS e L wdin population analysis redistributed L wdin population analysis RLPA CM2 CM3 CM4 and CM4M charge models SMx x 5 42 5 43 6 8 8AD solvation models SM8 with temperature dependence SM8T electrostatically embedded quantum mechanical EEQM energy and its first and second derivatives with respect to coordinates and electrostatic potentials with a site site representation of the QM MM electrostatic interaction e QM MM geometry optimization with a site site representation of the QM MM electrostatic interaction e internal coordinate constrained geometry optimization in Cartesian coordinates by projection operator method e combined quantum mechanics and molecular mechanics QM MM with the generalized hybrid orbital GHO boundary treatment The current version of GAMESSPLUS version 2010 2 has been developed to work with the latest R1 revision of GAMESS version of April 11 2008 The SMx solvation models are based on the generalized Born method for electrostatics augmented with semiempirical surface tensions for non bulk electrostatics These models can calculate free energies of solvation
48. either the GMSOL or the CM2 namelists are required To use the CM2 SRP method the CM2SRP namelist is required and to use the NDDO SRP capability the NDDOSRP namelist is required Below is a set of input examples for a variety of typical GAMESSPLUS calculations Example 1 Gas phase CM2 dipole moment calculation for water using HF MIDI where MIDI is inputted as a general basis SCONTRL SCFTYP RHF RUNTYP ENERGY COORD UNIQUE ISPHER 1 SEND SGMSOL ISCRF 0 ICMD 1 SEND SDATA water gas phase CM2 dipole moment Cl O 8 0 0 000000 0 000000 0 000000 S 3 1 281 8665800 0 0690600000 2 42 4160000 0 3931590000 3 9 0956200 0 6656690000 S 2 11 4660300 0 0808200000 2 0 8878600 0 5820900000 S 0 2788000 1 0000000000 P 2 1 838 0472400 0 1242710000 2 1 6684200 0 4765940000 P 1 1 0 3725100 0000000000 D 1 1 0 8000000 1 0000000000 H1 1 0 0 967300 0 000000 0 000000 S 2 1 4 5018000 0 0704520000 2 0 6814440 0 4078260000 s 0 1513980 1 0000000000 H1 1 0 0 210300 0 944200 0 000000 S 2 1 4 5018000 0 0704520000 2 0 6814440 0 4078260000 3 l 1 0 1513980 1 0000000000 SEND Example 2 SM5 42R HF 6 31 G d calculation of a water molecule in methanol solution SCONTRL SCFTYP RHF RUNTYP ENERGY COORD UNIQUE SEND SBASIS GBASIS N31 NGAUSS 6 NDFUNC 1 POLAR POPLE DIFFSP TRUE SEND
49. interpreted with some degree of caution Another important issue is the difference between using SCF Scheme I and SCF Scheme II for liquid phase calculations involving diffuse basis functions This is an important issue because past experience has shown that in some cases when SCF Scheme I is used with diffuse basis functions the liquid phase SCF calculation is unable to reach convergence in particular drastic fluctuations in the Mayer bond orders occur between the gas phase and the liquid phase when using SCF Scheme I leading to poor SCF convergence In cases where this occurs using SCF Scheme II usually resolves this problem In contrast this issue rarely arises when non diffuse basis sets are used as demonstrated by the results from above Because of the small difference between the aqueous solvation free energies computed using SCF Scheme I and those computed using SCF Scheme II and because for the majority of the test cases where diffuse basis functions were used SCF Scheme I led to poor convergence SCF Scheme I is no longer available for calculations that use diffuse basis functions Subset D Subset D contains 22 input files for testing the temperature dependent solvation model Five molecules were selected to test the SM8T atomic surface tension parameters In all cases calculations were performed using MPWX where X is the percentage of Hartree Fock exchange The five molecules are benzene testD1 a b furfural testD2 a b w
50. is the pressure of an ideal gas for a given standard state a 1 molar standard state at 298 K is used in this calculation for all phases therefore P is 24 45 atm AGg is the standard state free energy of solvation of solute A in solvent B R is the universal gas constant and T is temperature Full details are given in the Appendix I of this manual entitled GAMESSPLUS Solubility Utility Soil Sorption Calculations For a given solute the soil sorption coefficient Koc is defined as Koc Coit Coil 9 Cyl Cw where Coi is the concentration of solute per gram of carbon in standard soil Cy is the concentration of solute per volume of aqueous solution and C and Cy are the standard state concentrations of organic carbon for soil and aqueous solution respectively Typically a standard state of 1 wg of solute g of organic carbon is used for C and 1 mol L is used for Cy Koc may be calculated according to Koc Psoil lacs Kee 10 where p o is the density of soil in g mL AG is the standard state free energy associated with transferring a solute from the gas phase to aqueous solution and AG is the standard state free 20 energy associated with transferring a solute from the gas phase to soil Full details are given in the Appendix II of this manual entitled GAMESSPLUS Soil Sorption Utility QM MM Calculations at the Ab Initio HF Level with the GHO Boundary Treatment GAMESSPLUS can be compiled into a CHARMM
51. modeling soil as a homogenous medium see Winget P Cramer C J Truhlar D G Prediction of Soil Sorption Coefficients Using a Universal Solvation Model Environ Sci Technol 2000 34 4733 The GAMESSPLUS Soil sorption utility program also allows Ko values to be calculated with user supplied data for AG Thus in cases where an experimental value for AG is known it can be used in eq 2 to calculate Koc In fact a AG value obtained from any reliable method e g SM5 43 or SM6 can be used in lieu of the AG value calculated using SM5 42 in eq 2 This section and several subsections below of this manual provide a stand alone introduction to the GAMESSPLUS Soil sorption utility program for users who just want to calculate soil sorption coefficients Such users should first install GAMESSPLUS according to the instructions given in the sections entitled Notes on Running GAMESSPLUS and Updating 119 and Compiling GAMESSPLUS of this manual Users of this utility program should give the appropriate references described in the section entitled GAMESSPLUS Reference Then all of the information required to run a standard calculation with the GAMESSPLUS Soil sorption utility program is in the following few sections The SM5 42 continuum solvation model uses SM5 functional forms for atomic surface tensions hence the first three characters in the name of the method are SM5 class IV point charges hence 4 comes next of the CM2 type
52. new subroutines GPDER1 SMXPUN and DOGEOM were added Subroutine GPDERI calculates the analytical gradient for SM5 42 AM1 and SM5 42 PM3 and is called by subroutine MPCGRD in mpcgrd src Subroutine SMXPUN prints out the energy components from solvation calculations in required by POLYRATE and is called by subroutine DOSOLV Subroutine DOGEOM initiates liquid phase geometry optimization and is called from subroutine DISPLC in Statpt src The capability of inputting specific reaction parameters SRP for AM1 PM3 and or CM2 is added The former capability is called NDDO SRP and the latter is called CM2 SRP A new subroutine RCM2SRP has been added to the smx src and a new module nddosrp has been implemented in file nddosrp src Namelist input is now implemented by calling the GAMESS subroutine NAMEIO This enhances the portability of the code since namelist input is machine dependent ICDS is set equal to ICMD by default Thus a user does not need to specify ICDS The GAMESOL version number is printed in the summary of solvation calculations GAMESOL now works with both the Jan 6 1998 and May 6 1998 versions of GAMESS The Users Manual has been re organized and improved GAMESOL Version 2 2 April 1999 Authors J Li T Zhu G D Hawkins Y Y Chuang D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS versions January 6 1998 May 6 1998 and December 1 1998 GAMESOL now supports the Mar 15 1999 version
53. of GAMESS as well as with the Jan 6 1998 and May 6 1998 version of GAMESS New NDDO SRP options BETSS and BETSP are added with a new test run 11 inp as an example The DATA statements in smx src are moved after the declaration statements of the variables according to the standard FORTRAN syntax GAMESOL Version 2 2 1 April 1999 Authors J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS versions January 6 1998 May 6 1998 and March 15 1999 The name of directory gamesol 2 2 mod Dec98 is changed to gamesol2 2 1 mod Mar99 to reflect that the version of GAMESS we received from Iowa State had changes later than the date printed in the box 97 GAMESOL Version 2 2 2 May 1999 Authors J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS versions January 6 1998 May 6 1998 and March 15 1999 e The parameter NUMATM has been assigned to be the same as MXATM which is 500 in the subroutines BORNRD CALCDS CALSTN DAREAL GBMOD GPDER GPDER1 and SMSCDS in smx src to allow for solvation calculation on molecules containing more than 100 atoms GAMESOL Version 2 2 3 July 1999 Authors J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS versions January 6 1998 May 6 1998 and March 15 1999 e The paramet
54. of data available for other solvents Solvent parameters for common organic solvents are tabulated in the Minnesota Solvent Descriptor Database The latest version of this database is available at http comp chem umn edu solvation Solvent Temperature If AQU 1 then the solvent temperature may be specified using one of the SOLK following keywords Specifies that the free energy of solvation for the solute be computed at a single temperature The current model is only applicable to H C and O containing compounds for the range of temperatures in which water is liquid 273 to 373 K Note the model uses a temperature dependent dielectric constant the equation for it was obtained from the CRC Handbook of Chemistry and Physics 76 edition ed Lide D R 1995 CRC Press New York The default value is SOLK 298 READK AVGK IGEOM IGEOM 0 IGEOM 1 IGAS IGAS 0 IGAS 1 IGAS 2 IGAS 3 48 Causes GamessPlus to read in a list of temperatures at which to compute free energies of solvation from a file labeled KELT The default value is READK FALSE KELT must be located in the scratch directory Example KELT file 3 273 298 373 The first line in the input file must list the number of temperatures at which the free energy of solvation is to be computed Then each succeeding line should list only one temperature This option is similar to READK however instead of computing the electronic energy of the solute in solution at eac
55. package http comp chem umn edu cgplus see CGPLUS v1 0 User Manual Notes on GAMESSPLUS Input Users of GAMESSPLUS should consult at least the first two sections of the GAMESS Users Manual that is distributed with the GAMESS package These sections outline the specifics for all aspects of GAMESS input For QM MM calculations with site site interactions and link atoms users should also consult AmberTools Users Manual because such QM MM calculations require parameter topology and coordinate inputs generated by AmberTools Most GAMESS input is entered in a pseudo namelist format where a namelist is a particular group of keywords The title of this group the namelist name is prefixed by a that must begin in the second column of an input file e g SCF A namelist accepts as arguments a number of possible keywords that are entered after the namelist name in a free format style that can span multiple lines Each keyword accepts either character or numeric values A namelist is terminated with a END Only input between a namelist name and its corresponding END will be read from a GAMESS input file GAMESSPLUS introduces several new namelists to GAMESS CM2 GMSOL CM2 and GMSOL define the same input options CM2SRP NDDOSRP EEQM MM AMBTOP A4MBCRD QMMM and INTFRZ these new namelists are discussed at length in separate subsections below Below is a brief summary of GAMESS namelists and their keywords that are most i
56. solvents used in the calculation i e the solvent descriptors for the pure solution of the solute required for the calculation of the free energy of self solvation and the solvent descriptors of the solvent used in the calculation of AG For example for the solubility calculation of n pentane in methanol the VAPOR namelist would be used to specify the solvent descriptors of n pentane and the DGS namelist would be used to specify the solvent descriptors of methanol Below are descriptions of the keywords used in both VAPOR and DGS ISCRF Controls the type of calculation to be performed ISCRF 1 Calculates the free energy of solvation of a solute using the SM5 42 or SMS 43 solvation model and SCF Scheme I see SCF Schemes This is the default value for ISCRF SCF Scheme I is not available for methods that employ diffuse basis functions This is the default for all other methods ISCRF 2 Calculates the free energy of solvation of a solute using the SM5 42 or SM5 43 solvation model and SCF Scheme II see SCF Schemes ICDS Selects the set of coefficients to use for the SM5 42 or SM5 43 solvation model Coefficients have been optimized for specific wave functions as listed below Value of ICDS Wave function available methods 1 SMS5 42 HF MIDI 2 SMS5 42 HF MIDI 6D 3 SMS5 42 HF 6 31G d 6 SMS5 42 B3LYP MIDI 8 9 SM5 42 HF 6 31 G d SM5 42 HF cc pVDZ 11 SM5 42 AM1 12 SM5 42 PM3
57. the file midi bang bas which contains the coefficients necessary to run calculations that use the MIDI basis set You need to edit rungmsplus to set up a variable called GMSPATH that indicates the path to the location of gamessplus v2010 2 x and ddikick xto You may need to make other corrections of rungmsplus and solubility pl to make them run on your platform To execute the utility one should run the following command solubility pl input where input is the name of the input file and x is the number of the given version of GAMESSPLUS The solubility utility program creates two input files vapor input log and dgs input log corresponding to the calculation of P and AG respectively When the PRESSURE keyword is used a vapor pressure calculation is not required so there will be no 118 vapor input log file in this case The program also creates a file named input log which contains a summary of the calculations of P AGS and the solubility of the solute APPENDIX II GAMESSPLUS Soil Sorption Utility Executive summary The GAMESSPLUS soil sorption utility program is a utility program for calculating soil sorption coefficients For a given solute the soil sorption coefficient Koc is defined as Csoit Cool 1 Cure where Coi is the concentration of solute per gram of carbon in standard soil C is the concentration of solute per volume o of aqueous solution and Chi and C are the standard state concent
58. the molecular charge default 0 112 MULT N N is the multiplicity of the electronic state default 1 COORD UNIQUE input symmetry unique Cartesian coordinates default CART input all Cartesian coordinates ZMT input Gaussian style Z matrix internal coordinates ISPHER 1 use Cartesian basis functions e g 6D 10F default 1 use spherical harmonic basis functions e g 5D 7F Note Symmetry should not be used with GAMESSPLUS calculations The best way to input Cartesian coordinates is to set COORD UNIQUE which prevents coordinate rotation and to set the symmetry group of the molecule to C1 in DATA see below Namelist SYSTEM information for controlling the computer s operation MEMORY N N is the maximum memory the job can use in words default 1 000 000 Namelist DFT density functional theory DFT input DFTTYP N N is BLYP which requests a BLYP calculation or N is B3LYP3 which requests a B3LYP as it is implemented in Gaussian and HONDOPLUS i e using version III of the VWN correlation functional calculation or N is B3LYP5 which requests a B3LYP calculation but using version V of the VWN correlation functional HFE N Defines the fraction of Hartree Fock exchange N to be used when the MPWX functional is used This keyword must be specified when MPWX is used For CM3 and SM5 43 calculations that use MPWX ICMD 315 319 the HFE N must also be specified in the GMSOL or CM2 namelist
59. the rungms script modifying version requires the user to type sh rungms version to run GAMESS Note The compilation of CHARMM with GAMESSPLUS as a combination package with an integrated executable is supported by a utility package called CGPLUS The step by step instructions for modifying GAMESS GAMESSPLUS and CHARMM to create the CHARMM GAMESSPLUS combination package to perform QM MM calculations at the HF ab initio level with the GHO boundary treatment can be found in the chapter Compiling CHARMM with GAMESSPLUS of the CGPLUS v2008 User Manual http comp chem umn edu cgplus 80 Platforms The current version of GAMESSPLUS has been tested with the April 11 2008 R1 version of GAMESS on the following platforms the hardware specifications in more detail can be found at http www msi umn edu hardware SGI Altix Intel Itanium 2 running SUSE Linux 2 6 5 compiled with the Intel Fortran compiler version 8 1 in the comp compall and lked compilation scripts TARGET was set to linux ia64 SGI Altix Intel Itanium 2 running SUSE Linux 2 6 5 compiled with the GNU Fortran compiler version gcc 4 2 4 gfortran in the comp compall and lked compilation scripts TARGET was set to linux64 SGI Altix XE 1300 with Intel Xeon processors running SUSE Linux 2 6 16 compiled with the Intel Fortran compiler version 11 0 in the comp compall and ked compilation scripts TARGET was set to linux ia64 SGI A
60. then uses a scaling factor to compute the electronic energy at a given temperature Note this option is considerably faster than the READK option An additional test suite for SM6T has been added GAMESSPLUS Version 2008 April 2008 Authors M Higashi A C Chamberlin J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions March 24 2007 R6 plus others see the section on Makepatch Method for Updating and Compiling GAMESSPLUS This version works with the March 24 2007 R6 version of the GAMESS program Some modules of GAMESS have been modified This version also works with the CGPLUS v2008 package The capability to carry out electrostatically embedded quantum mechanical EEQM calculations with a site site representation of the QM MM electrostatic interaction has been added RUNTYP EEQM in the CONTRL namelist carries out the EEQM calculation The new namelist EEQM is available Two new parameters MXSATM and MXSBAS have been added in smx src to reduce the memory requirement MXSATM and MXSBAS control the maximum number of atoms and basis functions in the smx module Now these parameters have been set at MXSATM 100 and MXSBAS 512 The default parameter SolvRd in Namelists GMSOL and CM2 has been changed Now SolvRd 0 A for SM5 42 calculation and SolvRd 0 40 A for SM5 43 or later models
61. to calculate AGS and a user defined pure solute vapor pressure with the PRESSURE keyword Below are the input files and output files for these two calculations Input pentane inp SCONTRL SCFTYP RHF RUNTYP GRADIENT COORD UNIQUE SEND SBASIS GBASIS N31 NGAUSS 6 NDFUNC 1 POLAR POPLE SEND SEND SDATA Solubility calculation of n pentane El C 6 0 2 551305 329567 000000 C 6 0 1 279490 531652 000000 C 6 0 000000 318908 000000 H 1 0 000000 976801 884631 H 1 0 1 277493 1 188738 883613 1 0 1 277493 1 188738 883613 H 1 0 2 584647 975537 889230 H 1 0 2 584647 975537 889230 H 1 0 3 455107 294693 000000 C 6 0 1 279495 531646 000000 H 1 0 1 277498 1 188730 883614 H 1 0 1 277498 1 188730 883614 C 6 0 2 551302 329578 000000 H 1 0 2 584636 975547 889231 H 1 0 2 584636 975547 889231 H 1 0 3 455111 294675 000000 H 1 0 000000 976801 884631 SEND Svapor ICDS 3 ISCRF 1 IAQU 0 SolN 1 357 SolA 0 0 SolB 0 0 SolC 0 0 SolH 0 0 Dielec 1 837 SolG 22 2951 DENSITY 8 607067 SEND SDGS ICDS 3 ISCRF 1 SEND pentane2 inp SCONTRL SCFTYP RHF RUNTYP GRADI SBASIS GBASIS N31 NGAUSS 6 NDF ENT COORD UNIQUE SEND UNC 1 POLAR POPLE END 000000 000000 000000 884631 8 883613 8 883613 889230 889230 000000 000000 883614 883614 000000 889231 889231 000000 884631 272 Units Bar Density 8 607067 S
62. with HF MIDI and is described in Accurate Partial Atomic Charges for High Energy Molecules with the MIDI Basis Set by Kelly C P Cramer C J Truhlar D G Theor Chem Acc 2005 113 133 e Gas phase and liquid phase CM4 class IV charges can be determined for the following combinations of electronic structure theory and basis set using either a restricted or an unrestricted formalism BLYP MIDI 6D BLYP 6 31 G d BLYP 6 31G d BLYP 6 31 G d p G96LYP MIDI 6D G96LYP 6 31 G d G96LYP 6 31G d G96LYP 6 31 G d p B3LYP MIDI 6D B3LYP 6 31 G d B3LYP 6 31G d B3LYP 6 31 G d p MPWX MIDI MPWX MIDI 6D MPWX 6 31G d MPWX 6 31 G d MPWX 6 31G d p MPWX 6 31 G d p MPWX cc pVDZ MPWX DZVP MPWX 6 31B d MPWX 6 31B d p e The CM4M charge model is an extension of the earlier CM4 model The CM4M model was individually optimized for the M06 suite of density functionals namely M06 L M06 M06 2X and M06 HF for eleven basis sets which are MIDI MIDI 6D 6 31G d 6 31 G d 6 31 G d p 6 31G d p cc pVDZ DZVP 6 31B d and 6 31B d p e Calculation of the solvent accessible surface areas SASAs of the atoms of a given solute The SASA is that defined by Lee and Richards see Lee B Richards F M Mol Biol 1971 55 379 and Hermann see Hermann R B J Phys Chem 1972 76 2754 In this definition the solvent is taken to be a sphere of radius r and the solute is represented by a set of atom centered spheres of a gi
63. work for hydrogen atom GAMESSPLUS Version 4 6 February 2005 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions November 22 2004 R1 May 19 2004 R3 e This version works with two different versions of GAMESS e More helpful error messages are produced when the installation fails GAMESSPLUS Version 4 7 August 2005 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions November 22 2004 R1 plus others see the section on Makepatch Method for Updating and Compiling GAMESSPLUS An alternate way to compile GAMESSPLUS is now available This way called the Makepatch Method for Compiling GAMESSPLUS is explained in the manual A script called makepatch pl is located in the gmsplus4 7 Code directory and a directory called Patches has been added under the gmsplus4 7 directory This alternate method of compiling GAMESSPLUS was designed with making GAMESSPLUS compatible with multiple previous versions of GAMESS and future versions of GAMESS in mind The keyword MPWxX is now an option for DFTTYP in the DFT namelist Note that using this keyword requires specifying the percentage of Hartree Fock exchange X in MPWX wi
64. 00000D 01 0 1379770000D 00 0 4446000000D 00 0 4781480000D 00 S 1 1 00 0 1220000000D 00 0 1000000000D 01 P 1 1 00 0 7270000000D 00 0 1000000000D 01 These two ways of expressing the basis set give identical wave functions and energies because the basis functions span the same space Furthermore they lead to identical Mulliken charges However the wave function coefficients and the L wdin charges are different Our CM2 charge model was parameterized using the Form 1 expression Therefore when you want to use the cc pVDZ basis set use the form that is provided in the file ccpVDZ bas in the Basis directory Whenever we refer to the cc pVDZ basis sets in conjunction with CM2 and or SM5 42 we refer to the cc pVDZ basis set as extended in Accurate Dipole Moments from Hartree Fock Calculations by Means of Class IV Charges J Li J Xing C J Cramer and D G Truhlar J Chem Phys 1999 111 885 892 This extended basis set is identical to the cc pVDZ basis set for all elements except Br and I where MIDI is used instead 66 6 31G d and 6 31 G d basis sets in CMx x 2 or 3 and SMx x 5 42 5 43 6 or 8 Whenever we refer to the 6 31G d and 6 31 G d basis sets in conjunction with CM2 SM5 42R or SMS 42 calculations we refer to these basis sets as extended in A New Class IV Charge Model for Extracting Accurate Partial Charges from Wave Functions J Li T Zhu C J Cramer and D G Truhlar J Phys
65. 16075 8837967 6150000 2166616 4830000 5831452 15 16 f 3 14 4 8830131 4842584 0449078 1156195 8408663 7973495 15 16 2789673 14 alas 6393860 4005517 9564566 1398508 2140845 14 17 1844928 8405406 16 Ll 16 1580741 1912484 7073872 8397302 61 QM MM energy calculation and geometry optimization QM gradient and numerical Hessian calculations with MM atoms fixed are also available Example of namelist 0MMM SQMMM IQMMM 1 IQMATM 1 1974 1972 1975 1976 1977 1978 1979 4859 4860 4861 4862 4863 4864 4865 4866 ITAPER 1 RCUT 15 0D 00 IBNDRY 3 IMMFIX 1 1 490 495 780 787 790 938 941 988 1000 1008 1038 1047 1231 1296 1299 1342 1361 1373 1382 1425 1434 1444 1446 1626 1633 1675 1903 2116 2118 2122 2292 2303 2304 2315 2321 2473 2492 2886 2887 2894 2899 2934 2942 2988 2990 3046 3050 3155 3197 3203 3208 3225 3229 3277 3280 3321 3340 3361 3365 3376 3386 3641 3651 3698 3703 3742 3745 3748 3764 3777 3778 3781 3784 3792 3831 3842 3845 3847 3989 4074 4077 4225 4234 4277 4308 4319 4330 4338 4448 4451 4452 4468 4487 4606 4609 4617 4654 4665 4858 SEND IQMMM Determines whether a QM MM calculation will be performed or not IQMMM 0 A QM MM calculation will not be performed default IQMMM 1 AQM MM calculation will be performed k
66. 3 CM3 PM3 TCO3 CM3 HE MIDI KCOSa CM3 HE MIDI 6D WT02 CM3 HF 6 31G d WT02 CM3 MPWX MIDI WT02 CM3 MPWX MIDI 6D WT02 CM3 MPWX 6 31G d WT02 CM3 MPWX 6 31 G d WT02 CM3 MPWX 6 31 G d p WT02 CM3 BLYP 6 31G d TCO3 CM3 B3LYP MIDI 6D TCO3 CM3 B3LYP 6 31G d TCO3 CM3 B3LY P 6 31 G d TCO3 CM3 1 HF MIDI KCOSa CM4 DFT MIDI 6D KCO0S5b CM4 DFT 6 31G d KCO0S5b CM4 DFT 6 31 G d KCO05b CM4 DFT 6 31 G d p KCO0S5b SMS 42 HF MIDI LZ99 SMS 42 HF MIDI 6D LH98 SMS5 42 HF 6 31G d LZ99 SMS 42 BPW91 MIDI LZ99 SMS 42 BPW91 MIDI 6D ZL98 SM5 42 B3LYP MIDI LZ99 SM5 42 BPW91 6 31G d ZL98 SM5 42 HF 6 31 G d LZ99 SM5 42 HF cc pV DZ LZ99 SM5 42 BPW91 DZVP ZL98 SM5 43 HF 6 31G d TC04 SM5 43 B3LYP 6 31G d TC04 SM5 43 MPWX MIDI TCO5 SMS 43 MPWX MIDI 6D TCO5 SMS5 43 MPWX 6 3 1G d TCO5 SMS5 43 MPWX 6 3 1 G d TCO5 SMS5 43 MPWX 6 31 G d p TCOS5 SM6 DFT MIDI 6D KCO5b 37 SM6 DFT 6 31G d KCO5b SM6 DFT 6 31 G d KCO5b SM6 DFT 6 31 G d p KCOS5b SM8 M007 SM8AD MC09 SM8T AC06 ACOS SMx liquid phase optimizations ZL99 and CR99 GHO AIHF PG04 CRK MK97 LY04 and HTO8 TINKER tapering function PJ97 and SB93 Site site QM MM HOOO0 HTO8 HTO9 and references for the force field AMBER ff94 force field CC95 AMBER ff03 force field DW03 AMBER ff99SB force field WCO00 and HA06 TIP3P JC83 General AMBER force field Wwwo4 Link atom method fixed fraction
67. 6 6 529000 5 543000 5 615000 H1 0 199059 5 883013 5 639386 4 840541 H2 0 199059 7 456819 5 447792 5 216662 59 H3 0 199059 6 484258 6 400552 6 158074 Note Three namelists below AMBTOP AMBCRD and QMMM are used for QM MM calculations with site site electrostatics and if the QM MM boundary passes through a covalent bond with link atoms In these QM MM calculations QM atoms and coordinates are read by namelist DATA as usual for GAMESSPLUS calculations while MM atoms and coordinates are read by namelists AMBTOP and AMBCRD In addition to these three namelists namelist EEQM is required for QM MM calculations with link atoms in order to specify which charge model is used to describe QM MM electrostatic interactions Note For how to make AMBER topology parameter and coordinate files see the section entitled Short Tutorial for Making AMBER Parameter Topology and Coordinate Files and the AmberTools manual Namelist AMBTOP Namelist AMBTOP is used for reading AMBER parameter topology input generated by LEaP which is one of programs included in AmberTools Details of input format are described on the AMBER home page http ambermd org formats html Note that all QM and atoms need to be included in the AMBER parameter topology file because evaluation of QM MM interactions such as van der Waals interactions requires an MM force field for both QM and MM regions Some parameters such as charges and bond
68. 6 31G d HF 6 31G d HF 6 31 G d BPW91 MIDI HF cc pVDZ BPW91 MIDI 6D BPW91 DZVP e Liquid phase calculations based on gas phase geometries can be performed with SM5 43 for the following restricted and unrestricted Hartree Fock DFT and adiabatic connection method wave functions i e hybrid DFT wave functions that employ spherical harmonic or Cartesian d functions HF 6 31G d B3LYP 6 31G d MPWX MIDI MPWX MIDI 6D MPWX 6 31G d MPWX 6 31 G d MPWX 6 31 G d p e Liquid phase calculations based on gas phase geometries can be performed with SM6 for the following restricted and unrestricted DFT and adiabatic connection method wave functions the four basis sets for which SM6 is parameterized use Cartesian d functions BLYP MIDI 6D BLYP 6 31 G d BLYP 6 31G d BLYP 6 31 G d p G96LYP MIDI 6D G96LYP 6 31 G d G96LYP 6 31G d G96LYP 6 31 G d p B3LYP MIDI 6D B3LYP 6 31 G d B3LYP 6 31G d B3LYP 6 31 G d p MPWX MIDI 6D MPWX 6 31 G d MPWX 6 31G d MPWX 6 31 G d p e Liquid phase calculations based on gas phase geometries can be performed with SM8 or SM8AD and any choice of electronic structure method and basis set combination for which CM4 or CM4M charges can be calculated The CM4M charge model is recommended for use with the M06 suite of density functionals M06 M06 HF M06 L M06 2X e Liquid phase analytical gradients for SM6 SM8 and SM8AD are available for basis sets that use Cartesian d shells e Note that the B3LYP options
69. 7 0 885 ATOM 8 CL1 DCE 1 1 679 1 347 0 000 For partial charges on atoms in molecules treated by the charges in the GAFF force field ref WW04 recommends using restrained electrostatic potential RESP charges calculated by the HF 6 31G d method or AM1 BCC bond charge correction charges When one wants to use RESP charges one can generate a Gaussian input file from the PDB file by antechamber antechamber i DCE pdb fi pdb o DCE ginp fo gert Assuming that DCE gout is the Gaussian output file calculated by Gaussian using the input file DCE ginp created by the above command one can get an AMBER parameter file DCE_resp mol2 by antechamber antechamber i DCE gout fi gout o DCE_resp mol2 fo mol2 c resp The output file DCE_resp mol2 looks like this lt TRIPOS gt MOLECULE MOL 8 7 1 0 0 SMALL resp lt TRIPOS gt ATOM 1 cl 0 4730 0 5920 0 0000 c3 MOL 0 065447 2 H1 1 0950 0 5990 0 8800 h MOL 0 131379 3 H2 1 0950 0 5990 0 8800 hl MOL 0 131379 ACU 0 4730 2 1120 0 0000 cl MOL 0 197311 5 C2 0 4730 0 5920 0 0000 c3 MOL 0 065447 6 H3 1 0950 0 5990 0 8800 h MOL 0 131379 7 H4 1 0950 0 5990 0 8800 h MOL 0 131379 8 C12 0 4730 2 1120 0 0000 c MOL 0 197311 lt TRIPOS gt BOND 1 1 2 2 1 3 3 I 4 4 1 IL 5 5 6 1 6 5 7 7 5 8 1 lt TRIPOS gt SUBSTRUCTURE 1 MOL 1 TEMP Q KKK AREK 0 ROOT If one wants to calculate RESP charges by GAMESSPLUS see the AMBER RESP FAQ web
70. 88atm 89083 23Pa 668 18torr 116 117 0 89bar Solubility of solute 0 0025 mol L Logarithm basel10 of solubility of solute 2261 pentane2 log GAMESSPLUS solubility utility calculation based on GAMESSPLUS 2008 Wed Apr 16 21 05 56 2008 ee Summary of Solubility Calculation A value of the pure solute vapor pressure was supplied by the user so no information regarding the free energy of self solvation is available Standard state fr nergy of solvation of solute DeltaE EN liq elect nuc reorganization 012 kcal mol G P liq polarization free energy of solvation 192 kcal mol G CDS liq cavity dispersion solvent structure 1 866 kcal mol Free energy of self solvation 1 687 kcal mol Density of Solute 8 607067 mol L Vapor pressure of solute 0 88atm 89083 23Pa 668 18torr 0 89bar Solubility of solute 0 0025 mol L Logarithm basel10 of solubility of solute 2 61 Installing and running the solubility utility program A working version of GAMESSPLUS must be installed first see the sections entitled Notes on Running GAMESSPLUS and Updating and Compiling GAMESSPLUS in this manual The GAMESSPLUS solubility utility program comes with the GAMESSPLUS distribution It is located in the gmsplus_solubility directory In this directory you will find a PERL script named solubility pl a csh script named rungmsplus two test cases pentane 1 inp and pentane2 inp and
71. April 11 2008 dftxca src April 11 2008 gamess src April 11 2008 grdl src August 20 2007 grd2a src April 11 2008 grd2b src December 22 2006 grd2c src December 22 2006 inputa src April 21 2008 inputb src April 11 2008 intl sre April 11 2008 int2a src April 11 2008 mpcgrd src November 6 2006 mpcint src March 12 2008 mpcmol src March 12 2008 mthlib sre April 11 2008 rhfuhf src April 14 2008 statpt srce April 11 2008 symorb src August 20 2007 For each of these files in the GAMESS distribution whose latest modification date matches the date above simply replace the whole file in the GAMESS distribution with the corresponding file in the gmssplus x Code directory where x is the number of the given version of GAMESSPLUS for instance x v2009 If the latest modification date for one of the above files in the GAMESS distribution is later than the corresponding date above then the modifications should be carried out line by line Note that all modifications to the GAMESS code start with CGMSPLUSSTR and end with CGMSPLUSEND Thus the corresponding file in the Code directory can be used as a template for modifying the file in the GAMESS distribution If any problems are encountered in this 78 procedure contact the GAMESSPLUS developers see http comp chem umn edu gamessplus for contact information 2 Copy gmsplus x Code smx src gmsplus x Code nddosrp src gmsplus x Code ghodum src and gmsplus x Code eeqm s
72. B force field refs WC00 and HA06 for proteins If the user wants to use the AMBER ff03 or ff99SB force field for proteins and the TIP3P model for water the user should enter or where AMBERHOME is the AmberTools install directory before one runs the LEaP cd AMBERHOME dat leap cmd ln s leaprc ff03 r1i leapre cd AMBERHOME dat leap cmd in s leaprc ff99sS B leapre program For the details of the LEaP program see the AmberTools Manual Namelist AMBCRD 60 Namelist AMBTOP is used for reading AMBER coordinate input generated by LEaP which is one of programs included in AmberTools Details of input format are described in AMBER home page http ambermd org formats html As in the case of namelist AMBTOP all coordinates of QM and MM atoms are required However coordinates corresponding to QM atoms and M1 atoms when there are bonds in QM MM boundary are overwritten by ones specified in namelist DATA Example of namelist AMBCRD SAMBCRD 5817 16 5290000 15 5430000 17 4568187 15 4477921 16 2240000 14 4120000 14 9410000 14 6040000 4 0821117 4 8053907 6 1114530 2 4676554 6 5097623 3 3345304 SEND Namelist QMMM Namelist QMMM controls QM MM calculations with a site site representation of the QM MM electrostatic interaction and possibly with link atoms The current version of GAMESSPLUS supports I5 LS 16 3290000 17 16 10
73. END SDATA Solubility calculation of n pentane GL C 6 0 2 551305 329567 C 6 0 1 279490 531652 C 6 0 000000 318908 H 1 0 000000 976801 H 1 0 1 277493 1 18873 H 1 0 1 277493 1 18873 H 1 0 2 584647 975537 H 1 0 2 584647 975537 H 1 0 3 455107 294693 C 6 0 1 279495 531646 H 1 0 1 277498 1 188730 H 1 0 1 277498 1 188730 C 6 0 2 551302 329578 H 1 0 2 584636 975547 H 1 0 2 584636 975547 H 1 0 3 455111 294675 H 1 0 000000 976801 SEND Svapor Pressure 0 890832 SEND SDGS ICDS 3 ISCRF 1 SEND Output pentanel log GAMESSPLUS solubility uti lity calculation based on GAMESSPLUS 2008 Wed Apr 16 21 03 51 2008 Hees Summary of Solubility Calculation Standard state fr DeltaE EN liq elect nuc nergy of self solvation of solute reorganization 002 kcal mol G P liq polarization free energy of solvation 082 kcal mol G CDS liq cavity disper sion solvent structure 3 164 kcal mol Free energy of self solvation Standard state fr DeltaE EN liq elect nuc 3 244 kcal mol nergy of solvation of solute reorganization 012 kcal mol G P liq polarization free energy of solvation 192 kcal mol G CDS liq cavity dispersion solvent structure 1 766 kcal mol Free energy of self solvation 1 586 kcal mol Density of Solute Vapor pressure of solute 8 607067 mol L 0
74. GAMESS intl src GAMESS smx src GAMESSPLUS gho src GAMESSPLUS int2a src GAMESS statpt src GAMESS ghodum src GAMESSPLUS _ intfrz src GAMESSPLUS symorb src GAMESS grdl src GAMESS mpcegrd src GAMESS makepatch EEQM Tests contains two subdirectories called Tests and Output The subdirectory Tests contains the EEQM test suite input inp files and the subdirectory Output contains the corresponding output samples log There are additional files in EEQMTests Tests a script used to verify correct installation of GAMESSPLUS called compare pl and other scripts used for running the EEQM test suite run bat rungmsplus and run pbs INTFRZtTests contains two subdirectories called Tests and Output The subdirectory Tests contains the INTFRZ test suite input inp files and the subdirectory Output contains the corresponding output samples log gmsplus_soil contains files required to run calculations with the soil sorption utility program midi bang bas rungmsplus soil pl testl inp test2 inp test3 inp and the subdirectory Output with output samples test1 log test2 log and test3 log gmsplus_solubility contains files required to run calculations with the solubility utility program midi bang bas pentanel inp pentane2 inp rungmsplus solubility pl and the subdirectory Output with output samples pentane1 log and pentane2 log Patch contains patch files for GAMESS version April 11 2008 R1 and execution script to genera
75. GAMESSPLUS A Module Incorporating Electrostatic Potential Hessians for Site Site Electrostatic Embedding QM MM Geometry Optimization Internal Coordinate Constrained Cartesian Geometry Optimization Generalized Hybrid Orbital QM MM Methods the SM5 42 SM5 43 SM6 SM8 SM8AD and SM8T Solvation Models the L wdin and Redistributed L wdin Population Analysis Methods and the CM2 CM3 CM4 and CM4M Charge Models into GAMESS Users Manual Version 2010 2 Date of finalization of this version of the software Sep 30 2010 Date of most recent change in this document Sep 30 2010 Masahiro Higashi Aleksandr V Marenich Ryan M Olson Adam Chamberlin 4 Jingzhi Pu Casey P Kelly Jason D Thompson James D Xidos Jiabo Li Tianhai Zhu Gregory D Hawkins Yao Yuan Chuang Patton L Fast Benjamin J Lynch Daniel A Liotard Daniel Rinaldi Jiali Gao Christopher J Cramer and Donald G Truhlar a Department of Chemistry and Supercomputer Institute University of Minnesota Minneapolis MN 55455 0431 U S A b Laboratoire de Physico Chimie Theorique Universite de Bordeaux 1 351 Cours de la Liberation 33405 Talence Cedex France c Laboratoire de Chimie Theorique Universite de Nancy I Vandoeuvre Nancy 54506 France Distribution site http comp chem umn edu gamessplus The code and manual are copyrighted 1998 2010 Contents Executive Summary soises essessentecbvnaustscasatsce
76. ICMD 322 respectively The CM4 parameter sets for DFT MIDI 6D ICMD 416 DFT 6 31G d CMD 417 DFT 6 31 G d CMD 418 and DFT 6 31 G d p CMD 419 were added and tested The SM5 43 parameter sets for aqueous and organic solvents for MPWX MIDI ICDS 315 MPWX MIDI 6D ICDS 316 MPWX 6 31G d ICDS 317 MPWX 6 31 G d ACDS 318 and MPWX 6 31 G d p CDS 319 were added and tested The SM6 parameter sets for aqueous solution for DFT MIDI 6D ICDS 416 DFT 6 31G d CDS 417 DFT 6 31 G d ICDS 418 and DFT 6 31 G d p ICDS 419 were added and tested The SM6 test suite has been added This test suite tests the ability of GAMESSPLUS to perform liquid phase geometry optimizations and liquid phase single point calculations with methods that use diffuse functions SCF Scheme Iis no longer available for liquid phase calculations that involve the use of diffuse basis functions due to convergence problems If ISCRF 1 SCF Scheme I is specified for a method that uses diffuse functions the code prints out a message and automatically switches to ISCRF 2 SCF Scheme II A new method for installing the GAMESSPLUS solubility utility is used in this version of GAMESSPLUS This new method uses the script called install pl The GAMESSPLUS Soil sorption utility has been added This program uses the same installation procedure as the solubility utility GAMESSPLUS Version 4 8 May 2006 Authors A C Chamb
77. In GAMESSPLUS v4 7 and v4 8 SM5 42 calculations without specifying SolvRd 0 A had errors because the default parameter SolvRd was set at 0 40 A for all the solvation model calculations All the test input files have been rechecked All the test output files have been replaced with those calculated by GAMESSPLUS v2008 In addition in many cases of the SM6 geometry optimization calculations better initial guesses were provided for the test run geometries and the natural coordinate option was turned on for some of the geometry optimizations GAMESSPLUS Version 2008 2 August 2008 108 Authors M Higashi A V Marenich R M Olson A C Chamberlin J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions April 11 2008 R1 plus others see the section on Makepatch Method for Updating and Compiling GAMESSPLUS The major modifications of the code in this version are as follows The SM8 model was added The temperature dependent SM8T model was added The CM4M model was added The CM4 model parameters for additional basis sets were added The maximum number of atoms in the solvation module has been extended to 250 instead of 100 The maximum number of basis functions in the solvation module has been extended to 2500 instead of 512 The corresponding limits in the EEQM module have not been
78. L namelist only 1989 157 200 GI6LYP DFTTYP GLYP Gill P M W Mol Phys 1996 89 433 HFE 0 0 must be Adamo C Barone V J Comp Chem 1998 19 419 specified in the CM2 or Lee C Yang W Parr R G Phys Rev B 1988 37 785 GMSOL namelist only Miehlich B Savin A Stoll H Preuss H Chem Phys Lett 1989 157 200 mPWPW91 DFTTYP MPWX Adamo C Barone V J Chem Phys 1998 108 664 HFE 0 0 must be specified in both the DFT namelist and in the CM2 or GMSOL namelist Burke K Perdew J P Wang Y in Electronic Density Functional Theory Recent Progress and New Directions Dobson J F Vignale G Das M P Eds Plenum New York 1998 Perdew J P in Electronic Structure of Solids 91 Ziesche P Eschrig H Eds Akademie Verlag Berlin 1991 Perdew J P Chevary J A Vosko S H Jackson K A Pederson M R Singh D J Fiolhais C Phys Rev B 1992 46 6671 Perdew J P Burke K Wang Y Phys Rev B 1996 54 16533 Hybrid DFT functionals recommended for use in GAMESSPLUS Note that the HFE keyword should be specified in the CM2 or GMSOL namelist when these or any other hybrid DFT functionals are used in conjunction with the CM4 or SM6 SM8 models Method Fraction HFE GAMESSPLUS Keyword s Reference s B3LYP 0 200 DFTTYP B3LYP3 Stephens P J Devlin F J version III HFE 0 200 must be specified Chabalowski C F Frisch
79. M06 2X 6 31 G d geometry optimization using numerical gradients for water in water Note that testE14 fails to run in parallel Subset F Subset F contains 5 input files for testing the SM8AD solvation model ICDS 801 as well as the CM4 and CM4M charge models Test file description testF1 HF STO 3G analytical gradients of an arbitrary molecule containing 47 atoms in water testF2 HF 6 31G d analytical gradients for 1 1 1 3 3 3 hexafluoropropan 2 ol water cluster testF3 HF MIDIX 6D analytical gradients for 5 fluorouracil in fluorobenzene testF4 HF MIDIX 6D numerical gradients for 5 fluorouracil in fluorobenzene testF5 HF MIDIX 6D seminumerical frequencies for 5 fluorouracil in fluorobenzene 94 Note that testF14 fails to run in parallel Verifying Installation of GAMESSPLUS Using Test Suite Results A PERL script named compare pl is provided in the GAMESSPLUS distribution it is located in the Tests directories This script compares important output from test suite output files to corresponding output files in a subdirectory called Output To use this script after running some or all of the test suite calculations use the command compare pl output file s where output file is a test suite output file Note that you can select more than one output file to compare e g you can use the command compare pl log gt compare out to compare all the test calculations that you have completed A di
80. MPW1IKK For all of the MPWX methods listed above CM3 has been parameterized for five specific values of X namely 0 25 42 8 60 6 and 99 9 and these parameter sets are available in MN GSM Every CM3 and CM4 parameter is a linear or a quadratic function of the percentage of HF exchange used in the mPW exchange functional So in addition to the specific CM3 and CM4 parameter sets i e when X in MPWX is 0 25 42 8 60 6 and 99 the CM3 and CM4 Charge Models are available for any value of X in MPWX between 0 0 and 100 0 Note that the CM3 and CM4 parameters were optimized using a corrected version of the modified Perdew Wang density functional as implemented in Gaussian The details of this correction are described fully in The Effectiveness of Diffuse Basis Functions for Calculating Relative Energies by Density Functional Theory by Lynch B J Zhao Y Truhlar D G J Phys Chem A 2003 107 1384 The CM3 model for the BLYP and B3LYP methods uses a slightly modified mapping scheme for compounds that contain N and O For more information see Parameterization of Charge Model 3 For AM1 PM3 BLYP and B3LYP by Thompson J D Cramer C J Truhlar D G J Comput Chem 2003 24 1291 We have also developed a special CM3 model for assigning partial atomic charges to high energy materials This model is called CM3 1 and it uses the same mapping scheme as the CM3 model for BLYP and B3LYP This model has been parameterized for use
81. P wat Polariz amp LP incl 02 04 99 Loadi Readi parameters force field ng ng amber10 dat leap parm fremod ff03 m odification type file Readi title Duan Loadi Loadi Loadi Loadi Loading ng et ng ng ng ng library library library library library Loading library Sourcing leap Source in iog file Loadi Readi AMBER 2003 Loadi Readi ng parameters ng title ng parameters a a a a a m m m m m m al ff03 phi psi torsions a frcmod berl0 dat leap lib ions94 1lib berl10 dat leap lib solvents lib berl0 dat leap lib all_nucleic94 lib berl0 dat leap lib all_aminoct03 lib berl0 dat leap lib all_aminont03 1lib ber10 dat leap lib all_amino03 1lib amber10 dat leap cmd leaprce gaff Source of amberl10 dat leap cmd leapre gaff done leap log amberl10 dat leap parm gaff dat General Force Field for organic mol DCE_resp frcmod Readi remar oadi ng title k goes here ng Mol2 file Reading MOLECULE Loading PDB file starting new total atoms i ng force field modification type fil i DCE_resp mol2 named 2DHC_wat20 pdb molecule for chain X n file MOL 3334 add in e frcmod Leap added 2483 missing atoms according to residu 2483 H lone pairs Writing pdb file 2DHC_wat20_tleap pdb fo at the end
82. S distribution no longer contains the file mpcdat src GAMESSPLUS Version 4 3 1 September 2004 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS version May 19 2004 R3 e A bug was fixed in the program This bug was due to duplicate lines appearing in the file inputb src This bug was introduced in version 4 3 GAMESSPLUS Version 4 4 December 2004 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS version May 19 2004 R3 e A script to compile GAMESSPLUS was added The script modgms will automatically make all the changes described in the installation section GAMESSPLUS Version 4 5 January 2005 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS version May 19 2004 R3 105 e The mPW exchange functional was added for restricted and unrestricted calculations e The PW91 correlation functional was added for restricted and unrestricted calculations e The keywords MPWIK and mPW1PW91 are now options for DFTTYP in the DFT input section e A numerical issue was resolved in the VWN3 and VWNS5 functionals so that they now both
83. SM5 functional forms for atomic surface tensions hence the first three characters in the name of the method are SM5 class IV point charges hence 4 comes next of the CM2 or CM3 type hence 2 or 3 and are parameterized for rigid hence R gas phase geometries These solvation models provide a way to calculate electronic wave functions in the liquid phase and free energies of solvation The SMx x 5 0 5 2 5 4 5 42 and 5 43 solvation models are universal i e they are parameterized for water and any organic solvent With universal models one can calculate solvation free energies of a solute in two different 111 solvents e g water and 1 octanol and use the results to calculate the partition coefficient In this part of the program the standard state free energy of solvation AG R is calculated at the gas phase geometry and is given by two components AG5 R AGgp Gcps 0 where AGgEp AER Gp 2 In this equation AGEp is the bulk electrostatic component of the solvation free energy it is the sum of the polarization energy Gp representing favorable solute solvent interactions and the associated solvent rearrangement cost and the distortion energy AER the cost of distorting the solute electronic charge distribution to be self consistent with the solvent electric polarization The quantity GCps accounts for first solvation shell effects The quantity AGEp is determined by a self consistent reaction field SCRF calc
84. a a e EE E eb cs Rao e SAAD ee a a eh es a A e 36 URC REE E REEE E A E EE EE E E sig mentees 39 Notes oh GAMESSPLUS Input secsvccsecsecedcccccescseccccscssccuss cose cessvesdeutesecdcvesedecusetedccesscesccesstesdcestesdcodedeccceccvedcoessce 39 Namelists GMSOL and CM2 s osesesesesesssessosososossssososososesessososososeseseosososesesesesessssososososossosososososesessosososese 42 Namelist BCTV IDS BRP oc ckve och scecccticwnczuck suas ath cwsinsstatoukacaleceted avon ENEE EEEO SEEE AASE ECEE EEEN Se 51 Nae list SNIP OSE mE E toast scien sean tcaupvounstavsasash EEEE EEA T EEE 52 GAMESSPLUS REY WORGS scisiesciscssssosgussasessusddvoguedious sista ceiousdabecnousnoxbabunusesducvadooss donsusiuepnwieasessuseassnaead teneassanens 53 Namelist SEE QM ess tescsccasssccsssdacsceccccisceessieuscscascsacesscesncesttcesecevecesecdieseascedabesccvstecosbiosecs dodesceessious cocdevsaaseucsesessuee 55 Nam list MM a eeror esorare kroost soere oo Eeee eaen EErEE oarre onoo aSo SES EE rai EErEE O PEE thant EENE EEEN Eor iE 58 Namelist AMBTOP veao ee as scene eae cedaacesnctceces oxcuastacsucsanctes oaeuntva vxciaune succes EAEE ve vost eceeadessivenctocaeccssvecss 59 Namielist AIVIRE RD ANTA E E E EEEE E 60 Namelist OMMM secsseiecctinbssivcsieesescsedinsecescteacccencesnuesscdissaes a a a a i aaea eerie 60 DETTATI AH D N e VAE E E EE T SATE EE AA OTE EREE 63 Special Notes on Basis Sets isc cciscissecsccsssescecessonsvesnnsscntsssacseswesssnassnecesuweeseadss
85. able for the QM MM electrostatic interactions in the EEQM calculation with IRDMM 1 If the charge charge electrostatic interaction energy V between atom a and b is sharply truncated at a cutoff distance r namely Q Q ab lt Tout Vab r Fab 5 29 0 hae where r is the distance between atoms a and b and Q and Q are the atomic charges on atom a u In order to make V a continuous and and b V is not a continuous function at rp differentiable function many shifted or switched functions have been developed see Ref B93 In the TINKER tapering method the charge charge electrostatic potential is given by 2 0 2 0 Tap Nap Top R 5 Tap Da 22 aQ 22 0 OY fur Tab Tiap lt Tab lt Tout 30 k 0 Tap K 0 Tout lt Tap and c tap cut where rap lt To is a tapering distance the beginning of the tapering window r F lt r r and determined to connect V at r fap and Tout tap c and f are coefficients calculated from 7 and y r cut smoothly This potential energy inches is continuous and differentiable in the entire range i rp and it has continuous second derivatives QM MM Potential Energy Calculation and Geometry Optimization with a Site Site Representation of the QM MM Electrostatic Interaction In QM MM methods the total potential energy V of a QM MM system is described as the sum of three terms pital R R you R R y OMM R R y R 31 25
86. adients for UHF wave functions is os Density force W 1 y Wav T 1 uv where W is the total energy weighted density matrix UHF expression SA Ebb Peb Wu a y4 no Cis Civct DF No imi ve 2 i l i l In GAMESSPLUS v4 0 the UHF energy weighted matrix is evaluated in an alternative way 1 P7F P P F P v W 42 m 3 P F P PPEPPP wey where P and F are the density matrix and the Fock matrix for the alpha and the beta set respectively The expression in Eq 3 is identical to Eq 2 except that the diagonal elements of Eq 3 are one half of those in Eq 2 However they give identical results when the density force in Eq 1 is evaluated because all gradients of the diagonal overlap elements vanish dS u AR 0 Note that the above equivalence only holds if one uses the density matrix and the Fock matrix consistently For example in gas phase calculations both P and F are obtained without any 103 reaction fields included In solvation calculations the solvation effect for GB model the mutual polarization of the solute and solvent through the Gp term is included as a correction term to the Fock matrix during the SCF procedure HO OGp F pF PaP 4 Unfortunately the Fock matrix in the conventional storage space is still the gas phase Fock matrix F When GAMESSPLUS uses Eq 3 for gradient in any SCRF calculations the density matrix P is the solvated one but the gas phase Fock mat
87. analysis calculations are available for the CM2 CM3 and CM4 CM4M methods for which analytical gradients of the generalized Born solvation energy are available The necessary modification of NDDO Hamiltonians to carry out AM1 SRP and PM3 SRP calculations has been implemented GAMESSPLUS includes the GAMESSPLUS solubility utility for calculating the solubility of a given solute A in a given solvent B This utility is described in a self contained section of this manual Therefore users who only want to calculate solubilities do not need to be familiar with the entire GAMESSPLUS manual GAMESSPLUS includes the GAMESSPLUS soil sorption utility for calculating the soil sorption coefficients This utility is described in a self contained section of this manual Therefore users who only want to calculate soil sorption coefficients do not need to be familiar with the entire GAMESSPLUS manual GAMESSPLUS can now be used for GHO QM MM calculations with the CHARMM package through the CHARMM GAMESSPLUS interface for QM MM calculations GHO QM MM calculations are combined QM MM calculations with the QM MM boundary treated by the generalized hybrid orbital GHO method at the ab initio HF level GHO AIHF A parametrized version of GHO AIHF is available for the MIDI basis set The QM energy can be calculated in the presence of an external electrostatic potential with a site site representation of the QM MM electrostatic interaction energy The first a
88. ange of L wdin or RLPA charges and all gh are constants see references ZL98 LH98 and LZ99 for more detail This leads to a very simple expression for solvated Fock matrix elements ul i FS FS p E Ven CDi Bn oe n where Vk is the reaction field acting on atom k the indices i j and n refer to basis functions and kj is the atom on which basis function n is centered Usually scheme I and scheme II give very similar results see the SM6 and SM8 test suite for examples In principle and in practice scheme II converges to slightly different results and thus it could be considered to be a different model However in our experience the difference is usually small and scheme II sometimes has better convergence properties particularly for larger basis sets and basis sets containing diffuse functions See the references ZL98 LH98 and LZ99 for more detail on the SCF schemes SCF scheme III ISCRF 3 performs a liquid phase SCRF based on user provided atomic charges ICREAD 1 These atomic charges remain constant during the SCRF calculation Analytical gradients have not been implemented for ISCRF 3 SMS 42 SM5 43 SM6 and SM8 calculations cannot use restricted open shell wave functions but unrestricted wave functions can be used With the corrected implementation of UHF gradients 67 beginning with GAMESSPLUS v4 2 SM5 42 SM5 43 SM6 and SM8 are available for both restricted and unrestricted wave functions but they
89. anual The GAMESSPLUS soil sorption utility program comes with the GAMESSPLUS distribution It is located in the gmsplus_soil directory In this directory you will find a PERL script named soil pl and a csh script named rungmsplus three test cases test1 inp test2 inp and test3 inp and the file midi bang bas which contains the coefficients necessary to run calculations that use the MIDI basis set You need to edit rungmsplus to set up a variable called GMSPATH that indicates the path to the location of gamessplus v2010 2 x and ddikick xto You may need to make other corrections of rungmsplus and soil pl to make them run on your platform To execute the utility one should run the following command soil pl input where input is the name of the input file The program creates two input files water input inp and soil input inp corresponding to the calculation of AGS and AG respectively When the DELTAGW keyword is used a free soil energy calculation in water is not required so there will be no water input inp file in this case The program also creates a file named input log which contains a summary of the calculations of AGS AG and the logarithm base 10 of Koc
90. are still unavailable for restricted open shell wave functions The SCF procedures used by GHO AIHF algorithm for combining HF and molecular mechanics are different from the SCF scheme described above The GHO wave function is only optimized over an active space consisting of fully QM basis functions and active hybrid basis functions Therefore the conventional SCF procedure is modified to prevent the auxiliary basis functions on GHO boundary atom from mixing with all active basis functions For convenience in terms of the implementation the total Fock matrix and density matrix are still constructed in the atomic orbital basis in the usual way The active Fock matrix and density matrix for solving Roothaan s equations in the GHO active space are then obtained by a certain basis transformation followed by dropping the corresponding auxiliary entries The specific transformation between atomic basis functions and the orthogonalized hybrid basis functions are also dependent on the specific orthogonalization scheme used by the GHO AIHF method The detailed recipe of the modified SCF procedure for GHO AIHF based different orthogonalization procedures can be found in the reference PG04 68 Input Examples GAMESSPLUS is transparent if one performs standard GAMESS calculations that is the program may be used in exactly the same way that one uses GAMESS However if one would like to use the extra capabilities provided by GAMESSPLUS then at the very least
91. as follows e His289 was protonated at the delta position according to the previous study J Am Chem Soc 1998 120 5611 while other histidines were protonated at the epsilon position The name of the 289th residue in the PDB file was changed from HIS to HID The default setting of AmberTools assumes that HIS is an epsilon protonated histidine HIE e The center of the Cartesian coordinate system was moved to the C1 atom of DCE where the side chain of Asp129 attacks and the reaction occurs e Water molecules beyond 20 A from the C1 atom of DCE in the PDB file were deleted whereas water molecules were added within 20 A of the C1 atom of DCE to solvate the reaction center The name of the modified PDB file is 2DHC_20wat pdb 2 If there are some non protein molecules that are not included in the AMBER parameter library the user should make parameter files for those molecules using antechamber and parmchk which are included in AmberTools The general AMBER force field GAFF ref WW04 is used for determining such parameters In the example case one has to make a parameter file for DCE First a PDB file including only DCE DCE pdb was created as follows ATOM 1 Cl DCE 1 0 000 0 754 0 000 ATOM 2 H11 DCE 1 0 487 1 147 0 885 ATOM 3 H12 DCE 1 0 487 1 147 0 885 ATOM 4 CLI DCE 1 1 679 1 347 0 000 85 ATOM 5 Cl DCE 1 0 000 0 754 0 000 ATOM 6 H11 DCE 1 0 487 1 147 0 885 ATOM 7 H12 DCE 1 0 487 1 14
92. ate constraints in Cartesian coordinates Projection operator is used to project out forces along the constraints ref LZ9 The projection operator P is described as P ee 36 where m is the number of constraints and e has 3n components in the Cartesian coordinate for an n atom system and is the orthonormalized vector constructed from the row vector e of Wilson B matrix corresponding to the constrained internal coordinate For example if one wants to constraint the distance between atom A and B the nonzero elements of the row vector e are 27 6 R R Ry Rep i a a 7 i Aa AB OR AB IR R R gt 37 and _ a R R N Rea Rag 38 e H i Ba AB AB OR ga IR R where N is the normalized constant R Rea A X y z is the Cartesian coordinate of atom K K A B When there is only one constraint e e If there are more than one constrains e is obtained by Gram Schmidt orthogonalization e N i 5 aoe 39 j l The constrained geometry optimization is performed with the projected gradient 1 P g g is force vector in Cartesian coordinates Even if the geometry optimization 1s carried out with the projected gradient the values of the constrained internal coordinates may deviate from the initial ones due to the nonlinear character of the constraints In GAMESSPLUS SHAKE method is used to maintain the constraints every geometry optimization step The current v
93. ater testD3 a b dimethyl sulfoxide testD4 a b and urea testD5 a b In the names below a is the type of temperature calculation b is temperature All of the computations were carried out using the 6 31 G d p basis set using ISCRF 2 ICMD 419 ICDS 419 for testD1 testD2 and testD3 or ICDS 800 for testD4 and testD5 The KELT file used to run test files testDx 2 and testDx 3 should be kept in the scratch directory Calculation type test cases Single temperature calculations using SOLK k testDx 1 k where x 1 5 k 273 298 348 373 Multiple temperature calculations using testDx 2 where x 1 5 individually computed electronic terms at each temperature using the READK keyword Multiple temperature calculations using scaled testDx 3 where x 1 5 electronic terms at each temperature using the AVGK keyword Subset E 93 Subset E contains 20 input files for testing the SM8 solvation model ICDS 800 as well as the CM4 and CM4M charge models Test file description testE1 B3LYP 6 31G d energy of 2 2 dichloroethenyl dimethyl phosphate in water testE1 M06 M06 6 31G d energy of 2 2 dichloroethenyl dimethyl phosphate in water testE1 M06 2X M06 2X 6 31G d energy of 2 2 dichloroethenyl dimethyl phosphate in water testE1 M06 HF M06 HF 6 31G d energy of 2 2 dichloroethenyl dimethyl phosphate in water testE1 M06 L M06 L 6 31G d energy of 2 2 dichloroethenyl dimet
94. ation of the solvent accessible surface areas of the atoms of the solute the default is O for SM5 42 and 0 40 for SM5 43 SM6 and SM8 For predicting solvation free energies with SM5 42 SolvRd should be set to 0 For predicting solvation free energies with SM5 43 or SM6 or SM8 or SM8AD the default value of 0 40 should be used Determines the solvent type Organic solvent additional solvent data must be input see below Aqueous solvent default no additional solvent data is required Solvent Descriptors If IAQU 0 then the solvent properties are specified by a series of solvent Dielec SoIN SolA SolB SolG SolC SolH property descriptor values dielectric constant of solvent default is the value for water 78 3 index of refraction at optical frequencies at 293 K ni default is 1 0 Abraham s hydrogen bond acidity X default is 0 0 Abraham s hydrogen bond basicity spe default is 0 0 Ym V default is 0 0 where y is the macroscopic surface tension at air solvent interface at 298 K and y is 1 cal mol A note that 1 dyne cm 1 43932 cal mol A aromaticity o the fraction of non hydrogenic solvent atoms that are aromatic carbon atoms default is 0 0 electronegative halogenicity the fraction of non hydrogenic solvent atoms that are F Cl or Br default is 0 0 For a desired solvent these values can be derived from experiment or from interpolation or extrapolation
95. be used to specify this namelist If they are both defined in the same input file the parameters defined in the second instance will be ignored Below are descriptions of all input variables defined in GMSOL or CM2 ISCRF Controls the type of calculation to be performed ISCRF 0 Invokes a gas phase calculation of L wdin RLPA CM2 CM3 or CM4 charges default ISCRF 1 Invokes a gas and liquid phase solution calculation of L wdin RLPA CM2 CM3 or CM4 charges and the calculation of solvation free energy using SCF Scheme I see SCF Schemes This method is only available for methods that use nondiffuse basis functions ISCRF 2 Invokes a gas and liquid phase solution calculation of L wdin or RLPA charges and CM2 CM3 or CM4 charges and the calculation of solvation free energy using SCF Scheme II see SCF Schemes This is the default method for methods that use diffuse basis sets Note that when a calculation of RLPA charges is requested L wdin charges are also calculated However for liquid phase calculations that use RLPA charges the L wdin charges correspond to the converged SCRF achieved with RLPA charges not with L wdin charges Also note that RLPA charges should only be calculated by methods that use the 6 31 G d or 6 31 G d p basis sets while L wdin charges should be used for all other methods ICDS Selects the set of coefficients that will be used to evaluate the atomic surface tensions The table above matches the val
96. bitals the charge density for each auxiliary orbital is determined as 1 gp 3 0 where gp denotes the MM point charge on B In GAMESSPLUS the implementation of GHO at the ab initio HF level GHO AIHF is based on algorithms described in the paper of J Pu J Gao and D G Truhlar see Ref PGO4 The major features of this extension include 1 The basis set on the GHO boundary B is chosen as an STO 3Gv set the 1s core electrons are not explicitly present ii The active basis functions are orthogonalized to the auxiliary orbitals to maintain the global MO orthonormal constraints Four orthogonalization schemes are proposed and implemented iii The GHO gradients are calculated analytically by incorporating additional forces due to the basis transformations of the GHO scheme Further details of the CHARMM GAMESSPLUS combination package are given in the CGPLUS user manual Electrostatically Embedded QM Calculation with a Site Site Representation of the QM MM Electrostatic Interaction In the electrostatically embedded QM calculations with a site site representation of the QM MM electrostatic interaction the sum of the QM electronic energy and the QM MM electrostatic interaction energy is given by VIEM R P H Q D Y 1 where R stands for the collection of the coordinates R a 1 2 N a of atoms in the QM region is the electronic wave function H o is the electronic Hamiltonian including nuclear 21 repulsion of th
97. by the Gauss Legendre quadrature method of reference LH95 in subroutine BORNRD This change in method improves dramatically the accuracy of the analytical gradient leading to more successful optimizations e Subroutine SETSOL was modified to allow for transition state optimizations and for Hessian evaluations in solution i e RUNTYP SADPOINT and RUNTYP HESSIAN are now recognized e The new CM2 keyword ISREAD has been added to allow users to input surface tension coefficients in a file called CDS_Param 99 GAMESOL now produces more informative output consistent with other codes that implement the SMS5S 42R and or the SM5 42 solvation models In addition the printing of charges and dipole moments has been changed for option ICREAD 1 A check for whether the Coulomb radii for all of the atoms in the input are available in the code has been introduced If there are any missing radii and a liquid phase calculation is being performed then the program stops If there are any missing radii and only gas phase CM2 charges are being calculated then the program continues but the printing of NOPOL information is suppressed A new comprehensive test suite has been added GAMESOL Version 3 0 1 August 2002 Authors J D Xidos J Li T Zhu G D Hawkins J D Thompson Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version June 11 2000 In previous versions of GAMESOL the
98. c cycles to calculate pK e SM8 and SM8AD use an improved set of surface tension functionals using this new set of surface tension functionals significantly improves the performance of the model for molecules containing peroxide functional groups e SM8 and SM8AD use class IV CM4 or CM4M charges which give more realistic partial atomic charges for aliphatic groups than our previous class IV models this is important for accurately modeling hydrophobic effects Analytical gradients and geometry optimization in liquid phase solutions Analytical gradients for liquid phase calculations have been implemented in GAMESSPLUS beginning with version 2 0 In particular GAMESSPLUS contains analytical gradients for restricted and unrestricted wave functions for basis sets with Cartesian d shells However analytical gradients are not available for basis sets with spherical harmonic d functions e g for HF MIDI HF cc pVDZ and methods using basis sets containing functions higher in angular momentum than f Analytical gradients are also available when the AM1 and PM3 or method is used The availability of gradients allows for efficient geometry optimization in liquid phase solution This is necessary in many cases For example the transition state geometry for the Sy2 reaction of ammonia and chloromethane the Menschutkin reaction depends strongly on solvent Other applications include 18 the study of phase dependent reaction mechanisms and solvent
99. changed Errors in the output of CDS energy contributions for the temperature dependent model formerly SM6T now SM8T were fixed A problem with wrong numerical gradients in the SMx solvation module was detected and fixed The problem did not affect any other functionality of GAMESSPLUS or GAMESS except calculations using the SMx solvation models A compiler gfortran gcc 4 1 2 error related to the misplaced initialization of maxqmlink in the GAMESSPLUS patch of int2a src was fixed The previous versions of GAMESSPLUS failed to run correctly EXAMO3 in the GAMESS test suite gas phase ROHF gradients due to a bug in the GAMESSPLUS patch for a GAMESS module called grd1 src This problem was fixed All tabs were removed because they caused run time I O errors unexpected element in format while running the code compiled by gfortran gcc 4 1 2 This version of GAMESSPLUS was modified to be run in parallel Running on up to 32 CPUs was tested A problem with analytical gradients in the SMx solvation module for running in parallel was detected and fixed There was no such problem in the EEQM module To make this version of GAMESSPLUS to be compatible with the latest version of GAMESS April 11 2008 the following modules located in the directory Code were modified line by line to reflect the change from the old version of GAMESS to the new one dftxca sre the old name is dftexc src dft src gamess src grd2a src inputa src inputb src intl
100. d IBM Regatta computers Previous versions of GAMESSPLUS all parts except the GHO ATHF module have been tested on the following additional platforms IBM Blade Center AMD Opteron Linux Cluster running SUSE Linux 2 6 5 compiled with the version 6 2 Portland Group Fortran in the comp compall and lked compilation scripts TARGET was set to linux64 SGI Altix XE 1300 running SUSE Linux 2 6 16 compiled with the Intel Fortran compiler version 10 1 in the comp compall and lked compilation scripts TARGET was set to linux ia64 IBM pSeries 690 and pSeries 655 Nodes Power 4 processors running AIX version 5 3 compiled with the XL Fortran compiler version 10 1 in the comp compall and lked compilation scripts TARGET was set to ibm64 81 IBM pSeries 690 and pSeries 655 Nodes Power 4 processors running AIX version 5 2 compiled with the XL Fortran compiler version 9 1 in the comp compall and lked compilation scripts TARGET was set to ibm64 Netfinitiy Linux cluster running Red Hat Linux kernel version 2 4 21 compiled with the g77 compiler version 3 2 3 in the comp compall and lked compilation scripts TARGET was set to linux pc SGI Altix with Itanium 2 processors running Red Hat Linux kernel 2 4 21 compiled with the Intel Fortran compiler version 8 0 in the comp compall and lked compilation scripts TARGET was set to linux ia64 IBM SP with WinterHawk
101. d angles and torsional angles To use GAMESSPLUS the user needs to obtain the GAMESS package from Iowa State University April 11 2008 R1 version of GAMESS and GAMESSPLWUS version 2010 2 from the University of Minnesota For QM MM calculations with a site site representation of the electrostatic potential the user also needs to obtain AmberTools we used version 1 3 from the Amber Home Page http ambermd org in order to make parameter topology and coordinate files of the total QM MM system This is done in a separate run and the output is then used to make input for GAMESSPLUS The GHO QM MM method is available by means of a CHARMM GAMESSPLUS combination package for treating the QM subsystem at the ab initio Hartree Fock level The GHO analytical gradients are also available for QM MM geometry optimizations The compilation of the CHARMM GAMESSPLUS combination package as an integrated executable is supported by a utility package called CGPLUS which is available at http comp chem umn edu cgplus The usage of the CHARMM GAMESSPLUS combination package for carrying out GHO AIHF calculations is covered in the CGPLUS manual see the CGPLUS v2008 User Manual CGPLUS also provides a separate test suite for testing the GHO ATHF functionality of the CHARMM GAMESSPLUS combination package To perform GHO QM MM calculations the user needs to obtain GAMESS from Iowa State University April 11 2008 R1 version of GAMESS GAMESSPLUS from the Univers
102. d p 6 419 419 4 0 40 c SM8 DFT MIDI 6 415 800 5 0 40 c SM8 DFT MIDI 6D 3 6 416 800 5 0 40 c SM8 DFT 6 31G d 6 417 800 5 0 40 c SM8 DFT 6 31 G d 6 418 800 5 0 40 c SM8 DFT 6 31 G d p 6 419 800 5 0 40 c SM8 DFT 6 31G d p 6 420 800 5 0 40 c SM8 DFT cc pVDZ 6 422 800 5 0 40 c SM8 DFT DZVP 6 423 800 5 0 40 c SM8 DFT 6 31B d 6 424 800 5 0 40 c SM8 DFT 6 31B d p 6 425 800 5 0 40 c SM8 M06 MIDI 6 500 800 5 0 40 c d SM8 M06 MIDI 6D 2 6 501 800 5 0 40 cd SM8 M06 6 31G d 6 502 800 5 0 40 cd SM8 M06 6 31 G d 6 503 800 5 0 40 cd SM8 M06 6 31 G d p 6 504 800 5 0 40 c d SM8 M06 6 31G d p 6 505 800 5 0 40 cd SM8 MO06 cc pVDZ 6 506 800 5 0 40 cd SM8 M06 DZ VP 6 508 800 5 0 40 cd SM8 M06 6 31B d 6 509 800 5 0 40 c d SM8 M06 6 31B d p 6 510 800 5 0 40 c d SM8AD DFT MIDI 6 415 801 6 0 40 c SM8AD DFT MIDI 6D R 6 416 801 6 0 40 c SM8AD DFT 6 31G d 6 417 801 6 0 40 c SM8AD DFT 6 31 G d 6 418 801 6 0 40 c SM8AD DFT 6 31 G d p 6 419 801 6 0 40 c SM8AD DFT 6 31G d p 6 420 801 6 0 40 c SM8AD DFT cc pVDZ 6 422 801 6 0 40 c SM8AD DFT DZVP 6 423 801 6 0 40 c SM8AD DFT 6 31B d 6 424 801 6 0 40 c SM8AD DFT 6 31B d p 6 425 801 6 0 40 c SM8AD M06 MIDI 6 500 801 6 0 40 cd SM8AD M06 MIDI 6D 6 501 801 6 0 40 cd SM8AD M06 6 31G d 6 502 801 6 0 40 cd SM8AD M06 6 31 G d 6 503 801 6 0 40 c d SM8AD M06 6 31 G d p 6 504 801 6 0 40 cd SM8 M06 6 31G d p 6 505 801 6 0 40 cd 55 SM8 M06 cc pVDZ 6 506 801 6 0
103. del portion of GAMESSPLUS Thus keywords used by the rest of the GAMESSPLUS program such as the level of theory basis set SCF and geometry optimization options etc must also be included in the route section of the input file For convenience the ISPHER keyword which is specified in the CONTRL namelist is also included in the table below Method ISPHER2 ISTS ICMD ICDS IRADII SolvRd HFE SM5 42 HF MIDI 2 1 5 1 1 2 0 00 N A SM5 42 HF MIDI 6D 1 5 2 2 2 0 00 N A SM5 42 HF 6 31G d 5 3 3 2 0 00 N A SM5 42 HF 6 31 G d 5 8 8 2 0 00 N A SM5 42 HF cc pVDZ 5 9 9 2 0 00 N A SMS 42 BPW91 MIDI 2 1 5 4 4 2 0 00 N A SM5 42 BPW91 MIDI 6D 1 5 5 5 2 0 00 N A SM5 42 BPW91 6 31G d 5 7 7 2 0 00 N A SM5 42 BPW91 DZVP 5 10 10 2 0 00 N A SM5 42 B3LYP MIDI 2 1 5 6 6 2 0 00 N A SM5 43 HF 6 31G d 5 303 303 3 0 40 N A SM5 43 B3LYP 6 31G d 5 313 313 3 0 40 N A 54 SMS5 43 MPWX MIDI 2 1 5 315 315 3 0 40 X SM5 43 MPWX MIDI 6D 1 5 316 316 3 0 40 X SM5 43 MPWX 6 31G d 5 317 317 3 0 40 X SM5 43 MPWX 6 31 G d 5 318 318 3 0 40 X SM5 43 MPWX 6 31 G d p 5 319 319 3 0 40 X SM6 DFT MIDI 6D 1 6 416 416 4 0 40 c SM6 DFT 6 31G d 6 417 417 4 0 40 c SM6 DFT 6 31 G d 6 418 418 4 0 40 c SM6 DFT 6 31 G
104. dependent molecular conformational preferences A full derivation of the analytical gradient is presented in the paper by T Zhu et al entitled Analytical Gradients of a Self Consistent Reaction Field Solvation Model Based on CM2 Atomic Charges J Chem Phys 1999 110 5503 5513 Notation for Solvation Models l Geometry optimized at level X Y in the gas phase followed by a single point SMx solvation calculation at level W Z where W Z is one of the choices supported by ICMD SMx W Z X Y 2 If X Y is the same as W Z then X Y may be substituted by g where g denotes gas phase SMx W Z g Previously solvation calculations carried out using gas phase geometries were denoted by including an R suffix after the name of the SMx model Here this older notation has been replaced with the notation above 3 For a liquid phase geometry optimization the X Y is dropped and this calculation is denoted as follows SMx W Z Previously solvation calculations carried out using liquid phase geometries were denoted by dropping the R suffix after the name of the SMx model Here we drop this suffix for all solvation calculations and use the notation described above Solvent Parameters Solvent parameters for common organic solvents are tabulated in the Minnesota Solvent Descriptor Database The latest version of this database is available at http comp chem umn edu solvation NDDO and CM2 Specific Reaction Parameters
105. dipole moment To carry out a CM2 SRP CM3 SRP or CM4 SRP calculation the namelist CM2SRP is required In this namelist there is only one variable NAME this variable is used to indicate the name of a CM2 SRP file which has a maximum of 8 characters Note the CM2 SRP file must be located in the scratch directory used throughout a given calculation The scratch directory used for a particular calculation is defined with the SCR variable in the rungms script see the section entitled Notes on Running GAMESSPLUS below In order to describe the syntax for the input file defined in CM2SRP it will be useful to describe the CM2 charge mapping scheme The CM2 CM3 or CM4 charge qg on an atom k is given by qk q D Bi Dix Br Crx 16 k where dy is the L wdin or Redistributed L wdin charge on atom k Bgg is the Mayer bond order between atoms k and k and Dg and Cpg are the CM2 parameters It is the D and Cz that may be defined in the CM2 SRP file Thus the syntax of a CM2 SRP file is lt Variable type gt lt Atomic pair no gt lt Value gt lt Variable type gt indicates if the parameter is a C parameter or a D parameter acceptable input is C or D according to equation 16 above and lt Atomic pair no gt corresponds to a particular pair of atoms k and k in equation 16 Allowed values of lt Atomic pair no gt and the corresponding pair of atoms defined for the value of lt Atomic pair no gt are given be
106. ditionally the covariance between the two terms in the above equation and the relatively small number of points for each compound on average 10 points were used means that the actual numerical values of these two terms may vary significantly from experimental entropies and heat capacities of solvation while still reproducing experimental values with high accuracy Note that the model has only been developed for solutes in aqueous solutions for the temperature 273 15 to 373 15 K A comment on using gas phase geometries to calculate solvation free energies For SM1 4 and SM5 4 geometry optimization in solution was an essential part of the parameterization SM5 42 SM5 43 SM6 SM8 and SM8AD are parameterized in such a way that one 16 fixes the geometry at a reasonable value any reasonably accurate gas phase geometry should be acceptable and calculates the solvation energy without changing the geometry Thus geometry optimization in the presence of solvent is not required to obtain accurate solvation free energies This method of obtaining solvation parameters based on gas phase geometries was adopted for several reasons First previous experience has shown that the difference one gets from re optimizing the geometry in the presence of solvent in almost all cases is small less than the average uncertainty in the method or in any competing method Second for many solutes less expensive e g semiempirical or molecular mechanics methods can
107. drogenic solvent atoms that are aromatic carbon atoms aromaticity and WV square of the fraction y of nonhydrogenic solvent atoms that are F Cl or Br electronegative halogenicity For bulk soil gand ware both zero The remaining five solvent descriptors have been empirically optimized against experimental values Note that the solvent descriptors for soil which are listed below should only be used with the levels of theory for which they are optimized AM1 and HF MIDI SM5 42 Solvent Descriptors for Bulk Soil AM1 HF MIDI Descriptor GAMESSPLUS Keyword SCF AM1 SCF ABINITIO E DIELEC 15 0 15 0 n SOLN 1 541 1 379 a SOLA 0 36 0 61 B SOLB 0 34 0 60 y SOLG 63 3 46 0 SOLC 0 0 0 0 Y SOLH 0 0 0 0 aNot required input for the GAMESSPLUS Soil sorption utility program 120 Usage A description of the input that is required to run this utility program is given below For more detailed information for setting up GAMESS or other GAMESSPLUS calculations see the documentation that comes with the GAMESS distribution and the main part of the GAMESSPLUS manual All of the input required to run a calculation with the GAMESSPLUS Soil Sorption program is entered in a pseudo namelist format a namelist can be thought of as a particular group of keywords The title of this group the namelist name is prefixed by a that must be entered in column 2 of an input file e g SCF A namelist accep
108. e QM region is the population operator that generates the partial charge Q on QM atomic site a Q 0 12 and is the electrostatic potential at atom a from the MM region In GAMESSPLUS one can choose the operator according to L wdin population analysis LPA redistributed L wdin population analysis RLPA Charge Model 2 CM2 Charge Model 3 CM3 Charge Model 4 CM4 or Charge Model 4M CM4M The LPA charge Q LPA is given by Qn LPA S PS 13 where r is the indices of atomic basis function S is the overlap matrix and P is the density matrix The RLPA charge Q RLPA is given by Q9 RLPA Q LPA Z Y gt exp a Ri gt ZY exp 0 R3 14 b a b a where Z is an empirical parameter is the diffuse orbital exponent on atom a and Y is the L wdin population that is associated with the diffuse basis functions on atom a diffuse r gt S Ps 15 a rea The CMx x 2 3 4 charge model is determined from wave function dependent charges the Mayer bond order and empirical parameters that are determined to reproduce experimental or converged theoretical charge dependent observables Q 0 gt Bus Das CaBa gt 16 b a where Q is the partial atomic charge from either a LPA for nondiffuse basis sets or a RLPA for diffuse basis sets D p and C are empirical parameters B a and b a p is the Mayer bond order between atom Bu D PS PS
109. e compilation scripts comp compall and lked the variable TARGET was set to sgi64 Notes on Running GAMESSPLUS It is recommended that GAMESS be run using a modified version of the script called rungms that is provided in the top level directory of the GAMESS distribution An example of such a script called rungmsplus modified to run GAMESSPLUS comes with the GAMESSPLUS distribution see the directories EEQMTests Tests and SMxTests Tests The user needs to edit rungmsplus to set up a variable called GMSPATH that indicates the path to the location of gamessplus v2010 2 x and ddikick x The user may need to make other corrections of rungmsplus to make it run on a certain platform The usage of the rungmsplus script is as follows cungmsplus input_file total_number_of_nodes total_number_of_CPUs gt amp output_file 82 For GHO AIHF calculations one needs to run the CHARMM GAMESSPLUS combination package which is compiled as an integrated executable charmm To start running a calculation with the CHARMM input file JOB inp type path charmm lt SJOB inp gt S JOB out where path is the directory path to the integrated charmm executable The user should run the full GAMESSPLUS test suite to make sure that the GAMESSPLUS installation is correct The GAMESS test suite that comes with the original GAMESS distribution should also be run Many GAMESS jobs except semi empirical and some other calculations can be run as pa
110. e energies in both aqueous and nonaqueous solutions Unlike SMS5 42 and SM5 43 the SM7 model is based on SM6 functional forms for atomic surface tensions and uses class IV CM4 point charges as well as the SM6 model However the electrostatic part of the SM7 model is based on the SM6 model s Coulomb radii which were optimized for aqueous solution only In the new model called SM8 the radii depend on the nature of a solvent This feature of the SM8 model makes it more accurate than SM7 when there is a need to calculate solvation energies in nonaqueous solutions Thus we skip the SM7 model hereafter There was also the SM6T model The SM6T model is an extension of SM6 to include the temperature dependence of the free energy of solvation relative to 298 K When the SM8 model came into existence the old temperature dependent terms from SM6T were augmented with a few new ones and the SM8T merged the SMO6T Since the SM8T model has some additional functionalities we opt to skip the more inferior SM6T model hereafter 14 The SMx solvation models provide a way to calculate electronic wave functions in liquid phase solution and free energies of solvation For solvation calculations based on gas phase geometries the standard state free energy of solvation AGg R is given by two components AGS R AG p Gcps 1 where AGgp AER Gp 2 In equations 1 and 2 AGgp is the bulk electrostatic component of the solvation free energy it is t
111. e gradients of the effective Born radii the gradients of the Coulomb integrals the gradients of the solvent accessible surface areas and the gradients of Gcps These quantities are now stored on GAMESS s direct access file i e the DICTNRY file records 267 270 and GAMESS main work array GAMESSPLUS Version 3 9 April 2003 Authors J D Xidos J Li T Zhu G D Hawkins J D Thompson Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version January 14 2003 R3 The name of the program GAMESOL has been changed to the more general name GAMESSPLUS The parameters for CM3 AM1 CM3 PM3 CM3 BLYP 6 31G d CM3 B3LYP MIDI 6D CM3 B3LYP 6 31G d and CM3 B3LYP 6 31 G d were added The new mapping scheme for compounds that contain N and O for these CM3 methods was also added For these methods this provides the capability to evaluate gas phase CM3 charges liquid phase CM3 charges and the electrostatic contribution to the free energy of solvation using the generalized Born GB model For all six methods the corresponding CM3 GB gradient may also be evaluated and used for geometry optimizations Modifications were made to routine STVDER in the source file grd1 sre and to various subroutines in the source file smx src The test suite was extended to test all of the CM3 parameters and to test the new charge mapping scheme for compounds that contain N and O The extended test suite a
112. e theory and basis set are described using keywords ICMD and ICDS see the section entitled Notes on GAMESSPLUS input below The SM5 42 and SM5 43 models have been parametrized for a few combinations of the methods and they should be applied with these combinations The SM6 model has been tested against several different density functionals and has been shown to retain its accuracy when different density functionals besides MPWX the method against which the CM4 and SM6 parameters were originally developed Thus the SM6 model is only basis set dependent and can be used with any good density functional There is a single set of the SM8 parameters radii and CDS terms that can be used with any basis set as long as accurate partial charges can be computed for that basis set The SM8T model is applicable to 15 the same combinations of theory and basis set as SM8 but it has been parametrized only for aqueous solution A list of density functionals that are available in GAMESSPLUS and that are recommended for use with SM6 and SM8 is in the section entitled Density Functionals Recommended for use with CM4 CM4M and SM6 SM8 in GAMESSPLUS Incorporating temperature dependence into the SMx models SM8T To account for the variation of the free energy of solvation as a function of temperature the temperature dependence of both the bulk electrostatic AGEP and the non bulk electrostatic AGeps contributions are included The effect of temperat
113. easestaxsas ed dacadvtadsctessistsaascousneadendessdvcclenssdes 8l Representative Performance Data on Running GAMESSPLUS in Parallel sccsscsssssssssesssssseees 82 GAMESSPLUS Test SUilE isee eneee eanan eE E Ee E oe Ee aE EN EES aoas e OSESE 83 Description of Test Suite for EEQM sesssessoessoessocesoeescsssecssecssecesecssecesccesoceoosscossoosssosssosssosssosesesssesssessseseseee 83 Description of Test Suite for INTERZ ssscccsssssssssssssssssssscsssssssscssessssenscsssssssssessssssssnscsessssssssssssessssescescesss 83 Description of Test Suite for QM MM sscssssccssssssssscsssscssssssssscessesssssecsssssssscssssesssssassssssscesssessesessescssseecs 84 Short Tutorial for Making AMBER Parameter Topology and Coordinate Files ccsccsssscssssccsesees 84 Description of Test Suite for CM2 CM3 CM4 CM4M SM5 42 SM5 43 SM6 SM8 and SMB8T 89 S bs t A and Subset Boece i cccccccveiccd eran en ie Eea placws evi salencasvhecd Kani laced eves i aar EnA E erie ers nai taiea 89 S bset C a Eara E EE e kod E E RAEE lab ee Ee EEE R A EEEE E AEA ERER REEE ERa 91 Subset D rar ei EE La E heed deeded E RE E E e raka EA A E RE a 92 Subset E anr aren a A E a R aae a EAEE AER eles Al EREA RE TER 93 S bset Eoc a E eE avai a E K E A E E OESTE TE REAS EE arie R a 93 Verifying Installation of GAMESSPLUS Using Test Suite Results sccssssssssssscssssssssesssscesecesees 94 GAMESSPLUS Revision History and Vers
114. ed as discussed in work by Kollman and Francl and their respective coworkers Class IV charge models provide a practical and stable way to obtain reasonable charges for buried atoms e Class III charges are at best as good as the basis set and wave function used whereas class IV charges represent extrapolation to full CI with a complete basis Class IV charges are useful for any purpose for which ChEIPG or Merz Kollman charges are useful but we believe that they are better Charge Model 2 CM2 Charge Model 3 CM3 and Charge Model 4 CM4 are our second third and fourth generation models of class IV charges respectively The CM4M charge model is an extension of the CM4 model Charge Model 3 has been parameterized with a larger training set than CM2 398 data vs 198 data and it is available for different combinations of electronic structure theory and basis sets than CM2 Furthermore it is parameterized for Li and for molecules that contain Si O Si F and Si Cl bonds CM2 is not Charge Model 4 has been parametrized against the same training set that CM3 was except that CM4 gives improved charges for aliphatic functional groups which is important for modeling hydrophobic effects The CM4M model was individually optimized for the M06 suite of density functionals see details in Olson R M Marenich A V Cramer C J Truhlar D G Charge Model 4 and intramolecular charge polarization J Chem Theory Comput 2007 3 2046
115. ed Notes on Running GAMESSPLUS below The syntax of the NDDO SRP file is a series of lines each of which having one of two possible forms Form A is lt Variable type gt lt Atomic no gt lt Value gt where lt Variable type gt indicates the type of the parameter and lt Atomic no gt indicates the atomic number Form A may be used for changes in the one electron one center energies USS or UPP in the monatomic parameter for one electron resonance integrals BETAS or BETAP and in orbital exponents ZS or ZP For example 53 USS 6 49 850 UPP 6 40 337 USS 8 99 181 UPP 8 80 762 BETAS 6 16 912 BETAP 6 9 190 BETAS 8 28 998 BETAP 8 29 249 Form B is used for modifying the two center resonance integrals in the NDDO SRP method without using the arithmetic mean prescription see last section of Executive summary In form B each line has the syntax BETxy lt atom which x orbital taken from gt lt atom which y orbital taken from gt lt value gt where BETxy is BETSS BETSP or BETPP i e x and y denote orbital types only S and P type are available For example BETSP 1 8 17 711 Note that BETSP 6 8 is different than BETSP 8 6 GAMESSPLUS Keywords Shown in the below table are the GAMESSPLUS keywords required for running standard SM5 42 SM5 43 SM6 SM8 and SM8AD calculations Note that the keywords entered in the GMSOL or CM2 namelist are used by the solvation and charge mo
116. ed information for setting up GAMESS calculations than is given here see the documentation that comes with the GAMESS distribution particularly the first two sections of the GAMESS user s manual These sections specify all aspects of a GAMESS input file Most GAMESS input is entered in a pseudo namelist format a namelist can be thought of as a particular group of keywords The title of this group the namelist name is prefixed by a that must be entered in column 2 of an input file e g SCF A namelist accepts as arguments a number of possible keywords that are entered after the namelist name in a free format style that can span over multiple lines Each keyword accepts either character or numeric values A namelist is terminated with a END Only input between a namelist name and its corresponding END will be read from a GAMESS input file Below is a brief summary of GAMESS namelists and their keywords that are most important to users who want to compute solubilities with the GAMESSPLUS solubility utility program In this section the namelist name is followed by a brief description and below the namelist name is a set pertinent keywords and their options Namelist CONTRL group of global settings for the calculation SCFTYP RHF restricted Hartree Fock calculation default UHF unrestricted Hartree Fock calculation RUNTYP ENERGY SCF evaluation default MAXIT N N is the maximum number of SCF cycles default 30 ICHARG N N is
117. ed on CM2 Atomic Charges J Chem Phys 1998 109 9117 SM5 42 MIDI 6D method and more efficient version of ISCRF 1 option LH98 Li J Hawkins G D Cramer C J Truhlar D G Universal Reaction Field Model Based on Ab Initio Hartree Fock Theory Chem Phys Lett 1998 288 293 32 Additional SM5 42 parameterizations and ISCRF 2 option LZ99 Li J Zhu T Hawkins G D Winget P Liotard D A Cramer C J Truhlar D G Extension of the Platform of Applicability of the SM5 42R Universal Solvation Model Theor Chem Acc 1999 103 9 SMS5 42 parameterization for Si WTO02 Winget P Thompson J D Cramer C J Truhlar D G Parameterization of a Universal Solvation Model for Molecules Containing Silicon J Phys Chem B 2002 106 5160 Parameterization of SM5 42 for soil WCOO Winget P Cramer C J Truhlar D G Prediction of Soil Sorption Coefficients Using a Universal Solvation Model Env Sci Technol 2000 34 4733 SMS5 43 parameterizations for HF 6 31G d B3LYP 6 31G d mPW1PW91 6 31G d and mPW1PW91 6 31 G d TC04 Thompson J D Cramer C J Truhlar D G New Universal Solvation Model and Comparison of the Accuracy of the SM5 42R SM5 43R C PCM D PCM and IEF PCM Continuum Solvation Models for Aqueous and Organic Solvation Free Energies and Vapor Pressures J Phys Chem A 2004 108 6532 SMS5 43 parameterizations for MPWX MIDIX MPWX MIDI
118. eometry optimized in soil should be used as input for a GAMESSPLUS Soil Sorption calculation for water a separate calculation should be carried out with the main GAMESSPLUS program to obtain the value of AG for the relaxed geometry this value should then be used as input for a soil sorption calculation The instructions for performing geometry optimizations with the main GAMESSPLUS are outlined in the main part of this manual In the main GAMESSPLUS program the solvent descriptors for soil must be explicitly defined in the input file in the GAMESSPLUS Soil sorption utility program the descriptors are automatically defined depending on the level of theory For aqueous solution solvent descriptors for bulk water do not need to be provided in either the GAMESSPLUS Soil sorption utility program or the main GAMESSPLUS program The solvent descriptors for soil are described in the section below Solvent descriptors for bulk soil For nonaqueous solvents the SM5 42 universal continuum solvent model uses a set of 7 solvent descriptors to characterize the properties of the solvent of interest These seven solvent descriptors are as follows the dielectric constant of the solvent n refractive index at the wavelength of the Na D line Abraham s hydrogen bond acidity parameter Xa P Abraham s hydrogen bond basicity parameter Xf y macroscopic molecular surface tension in units of cal mol A 2 P square of the fraction of nonhy
119. er NUMATM has been assigned to 100 in the subroutines BORNRD CALCDS CALSTN DAREAL GBMOD GPDER GPDER1 and SM5CDS in smx src due to the size of the executable files GAMESOL Version 2 2 4 August 1999 Authors J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS versions January 6 1998 May 6 1998 and March 15 1999 e The subroutines RNDDOSRP and RCM2SRP were modified to allow them to read the NDDO SRP and CM2SRP parameter files on both the IBM SP and SGI Origin2000 workstations Before compiling nddosrp src and smx src on IBM SP machines that run the AIX operating system the AIX at the beginning of lines contained in these subroutines must be removed No source modification is required for the SGI workstations 98 GAMESOL Version 2 2 5 October 2000 Authors J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS versions January 6 1998 May 6 1998 and March 15 1999 e Corrected an error in the analytical derivative of the N C O surface tension term in subroutine SMSCDS e Moved the evaluation of DIELEC after the IFTSOL LT O line in subroutine DOSOLV The previous implementation would crash during gas phase calculations on some machines e Uncommented the following line in subroutine DISPLC in statpt src IF CVGED GO TO 700 1454x0700 The absence of this
120. ergy V and partial charges Q in the presence of a given electrostatic potential distribution default EEQM in EETYP GRADIENT Calculate energy VEM and partial charges Q plus the gradient the presence of a given electrostatic potential distribution EEQM EETYP HESSIAN Calculate energy Vand partial charges Q plus the gradient 21 EEQM plus the Hessian AR in the presence of a given electrostatic potential distribution EETYP DQDR Calculate energy VF and partial charges Q plus the derivatives of the OQ charges with respect to coordinates R in the presence of a given electrostatic potential distribution See eq 22 56 EETYP DQDPHI Calculate energy VF and partial charges Q plus the derivatives of the gy p 8 p OQ charges with respect to electrostatic potentials gt in the presence of a given electrostatic potential distribution See eq 23 PHI Electrostatic potential distribution that is array of electrostatic potentials on the atoms k The units used for PHI are controlled by IUPHI The default is 0 which corresponds to a gas phase calculation IUPHI Determines the units of the electrostatic potentials PHI IUPHI 0 PHI is in a u IUPHI 1 PHI is in eV default ICMD Select the set of coefficients to use to evaluate the CM2 CM3 CM4 or CM4M charges This option is the same as ICMD in namelists GMSOL and CM2 Two additional options are allowed ICMD 998 Use the se
121. erlin J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS versions November 22 2004 R1 plus others see the section on Makepatch Method for Updating and Compiling GAMESSPLUS Minor corrections to the code were implemented The atomic surface tensions used to compute the temperature dependence of the free energy of solvation have been added 107 A new keyword SOLK to read in the temperature of liquid aqueous solution and to compute free energies of solvation at the given has been added The keyword only accepts values within the range of 273 to 373 K A new keyword READK to read in temperatures from a file named KELT and to compute free energies of solvation for a particular molecule in aqueous solution has been added Note that this approach used the solution phase electronic structure as an initial guess for successive temperatures so one may obtain identical electronic energies if the temperatures are very near one another To avoid this either use very strict convergence criteria or do not order similar temperatures very near one another The differences in electronic energies are negligible anyway Another keyword AVGK which is similar to READK computes the electrostatics by computing the electronic energy of the solute in solution at the average of all the temperatures in the file KELT and
122. ersion GAMESSPLUS supports four types of internal coordinate constraints bond lengths sums or differences of bond lengths bond angles and torsional angles GHO AIHF QM MM Calculations The GHO AIHF model in combined QM MM calculations has been tested for a series of closed shell and open shell small molecules and ions with various functional groups close to the QM MM boundary The rotation barrier around the central C C bond in n butane has been studied by using GHO AIHF The proton affinities of small alcohols amines thiols and acids computed by GHO AIHF showed that the method is also reliable for energetics In those tests various basis sets were used for the QM part namely STO 3G 6 31G d 6 314 G d 6 31 G d p 6 314 G d p and MIDI As compared to a projected basis scheme and a scheme based on neglect of diatomic differential overlap involving auxiliary orbitals using hybrid orbitals based on global L wdin orthogonalized atomic orbitals is more robust It has been shown that only the non orthogonality of atoms near the boundary is important and needs to be removed by the explicit orthogonalization scheme in GHO therefore only a local orthogonalization is necessary Considering the localization of the boundary treatment this method is more theoretically promising Therefore a local L wdin orthogonalization algorithm has also been implemented Instead of doing a L wdin orthogonalization over the entire QM system only orbitals o
123. erssossetecdecsseeecasatsssaosessenedsencseeseseee 64 MIDE GAs E E esti Sica elaleies ek ike Sh whe See RA eed Ae ee 64 Ce pV DZ basis St Ul GAUSS laity nriran ties eaea eoid Mine ieden ipea eee eE eea ie S epee Ahan eters 65 6 31G d and 6 31 G d basis sets in CMx x 2 or 3 and SMx x 5 42 5 43 6 OF 8 eeeeccceesseeceereeeesseeeeeeeeeee 66 Special Notes n SCE SCHEMES s sissccessesscsesdisecsossesseasdencnsssessssebscosensevdeacsendssecasssedcessscsesensseveconseessnsasvesesenseessese 66 TIPU DEL AEE euweb assis waa EE A AS eA E AAO EI TSE 68 Density Functionals Recommended for Use with CM4 CM4M and SM6 SMB sssssssssesseesscesees 71 Program DISIIDUU ON sssscvissseresieradevcsssavessienasecassdecwedsvad reisens ttds Eeee NAE EE E SRE EERE TEENER Ese Sn 73 A Note on GAMESS Versions iiicicsccsctessscssccccsaksnsvsiendsecesunsnenssatecuessvvea wade cedanuvenssadheessuabasvisersbanesedsoetes 75 Standard Method for Updating and Compiling GAMESSPLUS s sssssccssssccssscccsssccsssscessssceseaces 76 Makepatch Method for Updating and Compiling GAMESSPLUS u ccsssccsssssssssscsssccssesscssssscesees 76 Manually Updating and Compiling GAMESSPLUS sccsssscssssscssnsscssssccesssccsssscesssccesssscesssscesssssesees 77 PUGS OF INS scissisesssssic cd vsicas cuans dah aaeveas esi cada apes staker ad oath CeRi ASEE AENEA kA ang h SSAA aA VERNA ESA A aN Nae 80 Notes on Running GAMESSPLUS wisvccscssscecssvszsdvccacnsyccdenscastsst
124. f Minnesota 2 Place both files in the same directory e g jsmith gamessplus and gunzip and untar them The files will untar into the directories jsmith gamessplus gamess and jsmith gamessplus gmsplus x where x is the number of the given version of GAMESSPLUS for instance x v2010 2 3 Move into the gmsplus Code directory and execute makepatch makepatch 4 Move into the gmsplus directory and execute modgms nodgms This script might prompt you for a few bits of system information and then it will compile GAMESSPLUS If GAMESSPLUS compiled successfully the script should end with something like if x v2010 2 77 done with all compilations GAMESS will be linked into the executable image gamessplus v2010 2 x chdir object xlf o gamessplus v2010 2 x q64 W1 m WI1 bloadmap Iked map gamess o unport o messages but no errors from linker End of GAMESPLUSS INSTALL The executable gamessplus v2010 2 x should now be in the jsmith gamessplus gamess directory Manually Updating and Compiling GAMESSPLUS If there was a problem with either of the installations outlined above you can make all the modifications by hand following the next six steps detailed below 1 In the gamess source directory of the GAMESS distribution check the first line of each of the GAMESS files listed below for the latest modification date bassto src September 19 2005 dft sre
125. fference in the last digit is acceptable TestB 13 can produce a difference in the second digit that is acceptable too GAMESSPLUS Revision History and Version Summaries Note First and second level revisions e g 2 0 2 1 are enhancements Third level revisions e g 1 1 1 2 0 1 are bug fixes Versions 1 0 through 3 1 were called GAMESOL Beginning with version 3 9 the name is changed to GAMESSPLUS In the revision histories for each version of GAMESSPLUS GAMESOL we list the authors of that version and we also list the version of GAMESS on which it was based For recent versions we also list the names of the persons responsible for the new version GAMESOL Version 1 0 December 1997 Authors J Li G D Hawkins D A Liotard C J Cramer and D G Truhlar GAMESS version October 31 1996 e This version added the L wdin and CM2 charge models and the SM5 42R solvation model to the Oct 31 1996 version of GAMESS Parameters for following methods were included 5D wave functions and DFT models not yet available in GAMESS CM2 HF MIDI SM5 42R HF MIDI CM2 BPW91 6 31G d CM2 HF MIDI 6D SM5 42R HF MIDI 6D CM2 HF 6 31 G d SM5 42R HF 6 31 G d CM2 HF 6 31G d SM5 42R HF 6 31G d CM2 HF cc pVDZ CM2 BPW91 MIDI CM2 BPW91 DZVP SM5 42R BPW91 DZVP CM2 BPW91 MIDI 6D SMS5 42R BPW91 MIDI 6D CM2 AM1 CM2 B3LYP MIDI SM5 42R B3LYP MIDI CM2 PM3 GAMESOL Version 1 1 March 1998 Authors J Li G D Hawkins D A Liotard C J
126. g nitrogen fluorine chlorine bromine and sulfur J Phys Chem B 2008 112 3024 Analytical gradients of the SM5 42 SM5 43 SM6 and SM8 models ZL99 Zhu T Li J Liotard D A Cramer C J Truhlar D G Analytical Gradients of a Self Consistent Reaction Field Solvation Model Based on CM2 Atomic Charges J Chem Phys 1999 710 5503 Overview article including SM5 42 HZ99 Hawkins G D Zhu T Li J Chambers C C Giesen D J Liotard D A Cramer C J Truhlar D G Universal Solvation Models in Combined Quantum Mechanical and Molecular Mechanical Methods Gao J Thompson M A Eds American Chemical Society Washington DC 1998 pp 201 219 Sample applications of SM5 42 TC9S Truhlar D G Cramer C J Solvent Effects on 1 3 Dipolar Addition Reactions Faraday Discussions 1998 110 477 LC99 Li J Cramer C J Truhlar D G Application of a Universal Solvation Model to Nucleic Acid Bases Comparison of Semiempirical Molecular Orbital Theory Ab Initio Hartree Fock Theory and Density Functional Theory Biophys Chem 1999 78 147 WHOO Winget P Hawkins G D Cramer C J Truhlar D G Prediction of Vapor Pressures from Self Solvation Free Energies Calculated by the SM5 Series of Universal Solvation Models J Phys Chem B 2000 104 4726 WW00 Winget P Weber E J Cramer C J Truhlar D G Computational Electrochemistry Aqueous Oxidation P
127. ges can be computed for that level of theory we recommend using it with self consistently polarized Charge Model 4 or other self consistently polarized class IV charges in which case analytic gradients are available The cavities for the bulk electrostatics calculation are defined by superpositions of nuclear centered spheres whose sizes are determined by intrinsic atomic Coulomb radii The difference between SM8 and SM8AD is that the SM8 model uses the formula of Still et al for the Born radius used in the generalized Born approximation for bulk electrostatics while the SM8AD model utilizes the asymmetric descreening AD algorithm for the Born radius suggested by Grycuk See MC0O9 for more detail The SM8T solvation model is an extension of SM8 to include the temperature dependence of the free energy of solvation relative to 298 K The SM8T model models the temperature dependence of the solvation free energy using the same functional forms as those in SM8 but with additional terms added to account for temperature dependence thus a calculation carried out at 298 K with the SM8T model will yield the same solvation free energy as the same calculation carried out with SM8 The SM8T model has only been parametrized for aqueous solution There was also the SM7 model The SM7 model is an intermediate model between SM6 and SM8 Like in the case of SM5 42 and SMS 43 the non bulk electrostatic part of the SM7 model was parametrized to predict solvation fre
128. h temperature it computes the electronic energy of the solute at the average temperature and scales the electronic energy of the solute using the following factor ye Er i 7 E Avg where is the dielectric constant of water at the average temperature and is the dielectric constant of water at a given temperature T This corresponds to assuming that the charge on the solute does not change significantly as a function of temperature This approach can compute a large number of temperatures with nearly the same computation time as that of a single temperature This approach does deviate mildly from the individual computations performed at each temperature however the deviation is negligible The default value is AVGK FALSE Determines whether the inputted structure is already optimized in the gas phase Perform a geometry optimization in the gas phase and use the optimized gas phase structure as the initial guess for the liquid phase optimization default Use the inputted structure as the initial point for the liquid phase geometry optimization i e do not perform a gas phase optimization on the inputted structure Determines which gas phase energy is used in the evaluation of AGg Use the gas phase energy of either the inputted geometry IGEOM 1 or of the optimized gas phase structure IGEOM 0 IGAS 0 is the default Once this gas phase energy is obtained the gas phase SCF is skipped i e the liquid phase SCRF begi
129. he CM2 or GMSPLUS namelists If ICMD 315 319 or ICMD 415 419 or ICMD 500 510 and HFE is not defined by the user GAMESSPLUS will be terminated IRADII IRADII 1 IRADII 2 IRADII 3 IRADII 4 IRADII 5 IRADII 6 Determines the atomic radii that will be used during an SCRE calculation Use Bondi s values for the van der Waals radii to build the molecular cavity 2 0 A is used in cases where the atomic radii were not defined by Bondi For evaluating the SASA this is always the default set of radii For other values of IRADII these radii are used in cases where the radius of a given atom is not defined for a particular model see the table below Use the atomic radii optimized for predicting solvation free energies with the SM5 42 solvation model to build the molecular cavity to be used in the SCRF calculation This is the default for ICMD 1 to ICMD 10 Use the atomic radii optimized for predicting solvation free energies with the SM5 43 solvation model to build the molecular cavity to be used in the SCRF calculation This is the default for ICMD 301 to ICMD 322 Use the atomic radii radii optimized for predicting solvation free energies with the SM6 solvation model to build the molecular cavity to be used in the SCRF calculation This is the default for ICMD 0 ICMD 300 and ICMD 416 to ICMD 419 Use the atomic radii optimized for predicting solvation free energies with the SM8 solvation model to
130. he sum of the polarization energy Gp representing favorable solute solvent interactions and the associated solvent rearrangement cost and the distortion energy AE g the cost of distorting the solute electronic charge distribution to be self consistent with the solvent electric polarization Gcps accounts for first solvation shell effects AGP is determined by a self consistent reaction field SCRF calculation which allows the solvent induced change in the solute electronic wave function to be optimized variationally Gcps is not a self consistent term it has no effect on the solute electronic wave function In its simplest form Gcps is defined as Gcps DAK 3 where Ax is the exposed surface area of atom k this depends on the solute s 3 D geometry and is calculated by the Analytical Surface Area ASA algorithm as described in reference LH95 and as included in recent versions of AMSOL and ox is the atomic surface tension of atom k The atomic surface tension ox is itself a function of the solute s 3 D geometry and a small set of solvent descriptors References LH98 ZL98 and LZ99 present a more expanded form of Gcps than what appears in equation 3 The surface tension functional forms are the same in all SM5 42 and SM5 43 models SM6 and SM8 use a different set of functional forms The SM6 and SM8 functional forms are better for most purposes than those used in SM1 SM5 Allowed combinations of solvent model electronic structur
131. heoretical Chemistry Accounts style Higashi M Marenich AV Olson RM Chamberlin AC Pu J Kelly CP Thompson JD Xidos JD Li J Zhu T Hawkins GD Chuang Y Y Fast PL Lynch BJ Liotard DA Rinaldi D Gao J Cramer CJ Truhlar DG 2009 GAMESSPLUS version 2010 2 University of Minnesota Minneapolis 2009 based on the General Atomic and Molecular Electronic Structure System GAMESS as described in Schmidt MW Baldridge KK Boatz JA Elbert ST Gordon MS Jensen JH Koseki S Matsunaga N Nguyen KA Su SJ Windus TL Dupuis M Montgomery JA 1993 J Comput Chem 14 1347 In addition as usual the user should give literature references for any methods used A convenient collection of literature references in provided in the following section 30 Literature References ASA algorithm LH95 Liotard D A Hawkins G D Lynch G C Cramer C J Truhlar D G Improved Methods for Semiempirical Solvation Models J Comp Chem 1995 16 422 MIDI Basis set EG96 Easton R E Giesen D J Welch A Cramer C J Truhlar D G The MIDI Basis Set for Quantum Mechanical Calculations of Molecular Geometries and Partial Charges Theor Chim Acta 1996 93 281 LC9S Li J Cramer C J Truhlar D G MIDI Basis Set for Silicon Bromine and Iodine Theor Chem Acc 1998 99 192 TWO Thompson J D Winget P Truhlar D G MIDIX basis set for the lithium atom Accurate geometries and partial atomic
132. hermore AGg R depends on standard state choices the values given directly by the SM5 SM6 and SM8 models correspond to using the same molar density e g one mole per liter in the gas phase and in the liquid phase solution Furthermore the liquid solution standard state corresponds to an ideal dilute solution at that concentration However one may adjust the results to correspond to other choices of standard state by standard thermodynamic formulae Note that changing the standard state corresponds to adding a constant to WI thus the gradient of WI which is used for dynamics is not affected Why use SM5 42 SM5 43 SM6 SM8 or SM8AD e The semiempirical CDS terms make the above models more accurate than alternative models for absolute free energies of solvation of neutral solutes e SM5 42 SM5 43 SM8 and SM8AD are universal models i e the semiempirical parameters are adjusted for water and for all solvents for which a small number of required solvent descriptors are known or can be estimated this includes essentially any organic solvent 17 e SMS5 42 SM5 43 SM6 SM8 and SM8AD use class IV charges to calculate the bulk electrostatic contribution to the solvation free energy this is typically more accurate than calculating the charge distribution directly from the approximate wave function This has two consequences 1 The electrostatic contributions to the solvation free energy are estimated more realistically 2 CM2
133. hlar D G J Chem Theory Comput 2006 2 364 DFTTYP M05 2X 0 560 Zhao Y Schultz N E Truhlar D G J Chem Theory Comput 2006 2 364 DFTTYP M06 0 270 Zhao Y Truhlar D G Th Chem Acc 2008 120 215 DFTTYP M06 2X 0 540 Zhao Y Truhlar D G Th Chem Acc 2008 120 215 DFTTYP M06 L 0 000 Zhao Y Truhlar D G Th Chem Acc 2008 120 215 DFTTYP M06 HF 1 000 Zhao Y Truhlar D G Th Chem Acc 2008 120 215 73 Program Distribution The GAMESSPLUS program package consists of a compressed tar file called gmsplus x tar gz where x is the number of the given version of GAMESSPLUS The top level directory of this file system is gmsplus x that contains the file modgms and the following subdirectories Basis Code EEQMtTests gmsplus_soil gmsplus_solubility Patch SMxTests The content of these subdirectories are explained below Basis contains files with basis set information 631GS bas 631PGS bas ccpVDZ bas midi bang bas Code contains the modified GAMESS modules new GAMESSPLUS modules and a script to make new patch files bassto src GAMESS grd2a src GAMESS mpcint src GAMESS dft src GAMESS grd2b src GAMESS mpcmol src GAMESS dftxca src GAMESS grd2c src GAMESS mthlib src GAMESS eeqmmm src GAMESSPLUS inputa src GAMESS nddosrp src GAMESSPLUS eeqm src GAMESSPLUS inputb src GAMESS rhfuhf src GAMESS gamess src
134. hyl phosphate in water testE2 B3LYP 6 31G d analytical gradients of 2 6 dichlorothiobenzamide in aniline testE3 B3LYP 6 31 G d analytical gradients of 2 6 dichlorothiobenzamide in methanol testE4 B3LYP 6 31G d analytical gradients of protonated methanol CH OH in water testE5 B3LYP 6 31G d numerical gradients of protonated methanol CH OH in water testE6 HF 6 31 G d semi numerical frequencies for CHOH in methanol by numerical differentiation of analytical gradients METHOD SEMINUM NVIB 2 in FORCE testE7 HF 6 31 G d numerical frequencies for CH3O0H2 in methanol by double differentiation METHOD FULLNUM NVIB 2 VIBSIZ 0 001 in FORCE testE8 HF MIDI geometry optimization and frequency calculation for the transition state of an SN2 reaction in water testE9 HF MIDI geometry optimization and frequency calculation for the Ru II complex with 6 water molecules in water testE10 UHF MIDI analytical gradients for the Rud complex with 6 water molecules in water testE11 HF STO 3G analytical gradients for fullerene C60 in toluene testE12 HF 6 31G d analytical gradients for 2 2 dichloroethenyl dimethyl phosphate in water testE13 HF 6 31 G d p analytical gradients for 5 fluorouracil in fluorobenzene testE14 HF STO 3G analytical gradients for an arbitrary molecule containing 47 atoms in water testE15 M06 2X 6 31 G d geometry optimization using analytical gradients for water in water testE16
135. in the lists above should use the standard version HI VWN functional which is requested with the DFTTYP B3LYP3 keyword in data group DFT e L wdin population analysis partial atomic charges can be used in conjunction with the generalized Born method to calculate the electrostatic contribution to the free energy of solvation using HF DFT and hybrid DFT and basis sets containing s p d and f functions For basis sets involving 10 Cartesian d and f functions analytic gradients of the generalized Born free energy are available and they can be used for geometry optimizations and numerical Hessian and vibrational frequency calculations Redistributed L wdin population analysis charges can be used in conjunction with the generalized Born method to calculate the electrostatic contribution to the free energy of solvation using HF DFT and hybrid DFT and the 6 31G d and 6 31 G d p basis sets Analytic gradients of the generalized Born free energy are available and they can be used for geometry optimizations and numerical Hessian and vibrational frequency calculations by numerical differentiation of analytically calculated gradients CM2 CM3 and CM4 CM4M charges can be used in conjunction with the generalized Born method to calculate the electrostatic contribution to the free energy of solvation using any of the CM2 CM3 and CM4 CM4M methods detailed above Liquid phase geometry optimizations and Hessian and vibrational frequency
136. ion Summaries 11ccsssccsssscssssscsssscssssccsssscesssscsesssceseaces 94 APPENDIX I GAMESSPLUS Solubility Utility cccccccccssscssssscssssscssssscsssscccsssccssssscessscssssscesssscessaces 110 Executive Summary aese meno eae aao e ren Ea O E a S RE AE nia E e Bats va E e e ei 110 The SM5 42 and SM5 43 continuum solvation models cccccsscesscecsscesccceseceescecsseceeececsseeeseeeceseceeseecueeseseeesseeenees 110 WSA Ge e E EEA EEEE A EE E EEEE O E A T 111 Input specific to the GAMESSPLUS solubility utility eee eeecseeecsseceseeseeecsaeceeesecseesecnecuee caseecesecsesseeneseeesaeeeseeens 113 Input options specific to the SVAPOR namelist eeecseseseesceecseeseesesecseeseseeseesesecsesecsesseeaesecaeeaeeeceeeasseseeseseeeeeaeeees 114 Test calculations setes a Mie bau kth es edits haves E hess A cise Have AEP ieee Ree Rs 115 TiN e a scevens a sdescncvkecs Gates EE EEE T A sug th siocbveus Bsehsuscnsvades Sig des costenes b abaus sessed detest seauuecs aye 115 QUEL anae ara e Er a AEE E leueben tonne duct uacsen de E ade otieeseeas ches Ub tveenion seth ae E SS 116 Installing and running the solubility utility program 2 0 0 2 ccceccceescesceeececesceseceseceeeceaecaeecaeesaeceseeeeeeeeeseeeeeeeeeeseneeereeees 117 APPENDIX IT GAMESSPLUS Soil Sorption Utility ccccccscccsssscssssccssssccssssccsssccsssscesssscesssscessnees 118 Exec tive SUMIMNALY sisses orae se Epee aeaeo e Eroa E sve ptoaeastesdassssssspesies sees EEEE R Tesei
137. is not planning to use GHO AIHF in GAMESSPLUS the compilation of GAMESSPLUS with CHARMM is not necessary and the compilation of GAMESSPLUS with GAMESS itself is still needed to run GAMESSPLUS enhancements to GAMESS To meet this need the file ghodum src has been created it contains all necessary dummy routine required to compile GAMESSPLUS with GAMESS e A set of integral scaling factors is included for GHO AIHF the parameters were optimized for GHO AITHF MIDI with the local L wdin orthogonalization treatment e The comment lines to mark the GAMESSPLUS modifications have been changed from CGAMESOLSTR and CGAMESOLEND to CGMSPLUSSTR and CGMSPLUSEND to be consistent with the name of the program e In mpcgrd src the lines for reading the gradients of the CDS term and the gradients of Coulomb integrals from the direct access file records 269 and 270 are commented These calls to DAREAD occurred even for gas phase MOPAC type calculations For gas phase only MOPAC type calculations these resulted in an error because these records 269 and 270 were never created To avoid this error a separate subroutine GET_GRAD is called instead to handle both the gas phase case and the solvation case properly The following three items are related to the SM5 42 UHF analytical gradients e A bug has been fixed to evaluate the density weighted matrix correctly for SCRF calculations with UHF wave functions The density force contribution in analytical gr
138. ity of Minnesota and CHARMM from Harvard University In order to make the following description of some of the capabilities of GAMESSPLUS more clear we note that the following basis sets use Cartesian d functions MIDI 6D also known as MIDIX6D 6 31G d 6 31 G d 6 31 G d p 6 31G d p DZVP and the following basis sets use spherical harmonic d functions MIDI also known as MIDI 5D and MIDIXSD cc pVDZ GAMESSPLUS adds the following new capabilities to GAMESS e The B3LYP hybrid density functional theory method as it is implemented in Gaussian and HONDOPLUS i e using version III of the VWN correlation functional has been added This method can be used to obtain restricted and unrestricted wave functions and is requested with the DFTTYP B3LYP3 keyword in the DFT data group see the section entitled Notes on GAMESSPLUS input below The DFTTYP B3LYP5 keyword uses version V of the VWN functional which is the non standard form of the VWN functional e The MPWX where X is the percentage of Hartree Fock exchange hybrid density functional theory method This method can be used to obtain restricted and unrestricted wave functions and is requested with the DFTTYP MPW lt X keyword in the DFT data group see the section entitled Notes on GAMESSPLUS input below e For all restricted and unrestricted HF DFT and hybrid DFT methods using basis sets containing functions up to fin angular momentum gas phase and liquid phase L wdin par
139. ked Note that these Coulomb radii are different from the SM5 42R HF and SM5 42 HF Coulomb radii for P and S e In version 2 0 GAMESOL would crash after the gas phase Hartree Fock calculation was done during the gas phase geometry optimization phase of an SM5 42 calculation The Hessian matrix from the last step of gas phase geometry optimization was not being stored and thus there was no previous Hessian available for the SM5 42 calculation e Some common blocks have been rearranged so that longer variables come first i e the common blocks have been made byte aligned floating point variables before integer variables before logical variables Although this is not required on most machines on some machines running the Linux operation system a compilation error will be generated if the variables in a common block are not properly arranged e Some debug lines have been removed GAMESOL Version 2 1 January 1999 Authors J Li T Zhu G D Hawkins Y Y Chuang D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version January 6 1998 and May 6 1998 e This version added SM5 42R AM1 SM5 42R PM3 SM5 42 AM1 and SM5 42 PM3 solvation models The following models can be used in GAMESOL Rigid solvation model Solvation models with analytical gradient SMS5 42R MIDI 6D SMS5 42 MIDI 6D SMS5 42R 6 31G d SM5 42 6 31G d 96 SM5 42R 6 31 G d SM5 42 6 31 G d SMS5 42R AM1 SM5 42 AM1 SMS5 42R PM3 SM5 42 PM3 Three
140. l sorption utility program uses a standard state of 1 mol L in both the gas and aqueous phase to calculate Koc Thus user supplied values for aG should also correspond to the above standard state DATA Input molecule information Input consists of the following information line 1 title line line 2 symmetry group always enter C1 for GAMESSPLUS calculations line 3 molecular coordinates for Cartesian input each line consists of the following atom label the atom s nuclear charge which should be a floating point value i e 1 0 for H and the atom s x y and z coordinate See examples in the Input and Output Examples section and in the test suite Test calculations Three test calculations are given with the GAMESSPLUS Soil sorption utility program They are named test inp test2 inp and test3 inp and are located under the directory gmsplus_soil The first test calculation calculates jG and AG values at the AM1 SM5 42 level of theory and then uses these calculated free energies in eq 2 to compute Koc The second test calculation calculates aG at the AM1 SMS 42 level of theory and then uses this calculated value along with an experimental value for age in eq 2 to compute Koc The third test calculation repeats test calculation 1 at the HF MIDI level of theory Note that for test3 the coefficients for the MIDI basis set are entered in the data section This basis set is 121 included with the GAMESSPLUS distrib
141. l frequency evaluation numerical second order derivatives based on analytical first derivatives numerical second order derivatives based on double differentiation of the total energy forward difference for numeric Hessian evaluation default central difference for numeric Hessian evaluation N is the displacement size in Bohr default 0 01 print vibrational analysis default TRUE for RUNTYP HESSIAN FALSE otherwise Note GAMESSPLUS does not have analytical second derivatives for solvation energies Do not use METHOD ANALYTIC that is default for certain electronic structure methods in GAMESS in the group FORCE for RUNTYP HESSIAN in CONTRL and for RUNTYP OPTIMIZE or RUNTYP SADPOINT in CONTRL with HESS CALC or HSSEND TRUE in STATPT Namelist STATPT NSTEP N HESS GUESS input information for geometry optimization N is the maximum number of steps default 20 guess positive definite Hessian default for 42 RUNTYP OPTIMIZE READ read Hessian from HESS namelist default for RUNTYP SADPOINT CALC calculate the Hessian HSSEND TRUE calculate the Hessian at the end of a successful optimization default FALSE Note We recommend calculating the liquid phase Hessian separately from the geometry optimization Namelists GMSOL and CM2 GMSOL and CM2 are alternative names for a namelist that controls most of the options for the solvation and charge models Only one of these two names should
142. latlONs id cs ccscsssscssseessceccssesessoossedocesoncesssossecoscesessceuaedessssesucssnsnsccssessasessoscessnesodescssesuessessecoceseess 19 Soil SOLptiOn CalCulatiOns sass scvecccescsccsessesasccoseedeccsssonysoseveseensssnedsontecouacssesecsestesenscevencecesescssssedecsesasesesssenesevens 19 QM MM Calculations at the Ab Initio HF Level with the GHO Boundary Treatment sscssese 20 Electrostatically Embedded QM Calculation with a Site Site Representation of the QM MM Electrostatic Interaction cssccssscssssssssssscssssessscscsssensscssssccesssessscsscssssssessessssesceseesesssseneesesssseseesecssosseees 20 The TINKER tapering function for long range electrostatic interactions sssscccscscsesecssessseeseseees 24 QM MM Potential Energy Calculation and Geometry Optimization with a Site Site Representation of the QM MM Electrostatic Interaction scccssssscccsssscccssssccssssscccscssccecscsscesecssseccscssscccesssscecessseesesssseecees 24 Constrained Geometry Optimization in Cartesian Coordinates by Projection Operator Method 26 GHO ATHF QM MM Calculations siciccccciecccesceteccesstecesssuceccsscsceceosedecscessuccesseuteceesstucsesssuceceessduceessveccsevscdeceesse 27 GAMESS PLUS WAGON oxo scx scence he sasievesaveskdsaguccuusaauataas tava tanita deal ues sseueasvaagiateskcbuonvstetassvallotovscrveakest 29 Literature Referentes carrement anise aeit RE a a ananin 30 Quick index to literaire se ens
143. line causes incorrect printing of molecular information after geometry optimization has concluded for both gas phase and liquid phase geometry optimizations e Common block GBCNTL in subroutine SMXPUN was made consistent with all other instances of this common block e The absolute free energy of the solute is now passed to the routines that handle geometry optimization This should improve convergence in some cases GAMESOL Version 3 0 February 2001 Authors J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version June 11 2000 e CM2 and L wdin atomic charge and SM5 42R solvation energies can now be evaluated using wave functions that use spherical harmonic d functions As a consequence the following new methods are added to GAMESOL in this version CM2 HF MIDI SM5 42R HF MIDI CM2 HF cc pVDZ SM5 42R HF cc pVDZ e SCRF evaluations can now be performed using L wdin charges This option is available for restricted or unrestricted HF wave functions that employ Cartesian or spherical harmonic basis functions Analytical gradients of SCRF energies with respect to the nuclear coordinates are available for restricted or unrestricted HF wave functions that employ Cartesian basis functions up to f shell 1 e 6D 10F e The surface tension parameters for SM5 42R HF cc pVDZ have been added e Trapezoidal numerical integration has been replaced
144. lity Utility Executive summary The GAMESSPLUS solubility utility program is a utility program for GAMESSPLUS that predicts the solubility of a solute in a given solvent It utilizes the thermodynamic relationship see Thompson J D Cramer C J Truhlar D G J Chem Phys 2003 119 1661 that exists between the solubility free energy of solvation and the pure substance vapor pressure of a solute which is shown below for a given solute A in a liquid solvent B nee In this equation S is the solubility of solute A in solvent B Px is the equilibrium vapor pressure of solute A of a pure solution of A P is the pressure of an ideal gas for a given standard state a 1 molar standard state at 298 K is used in this calculation for all phases therefore P is 24 45 atm AG is the standard state free energy of solvation of solute A in solvent B R is the universal gas constant and T is temperature This relationship is valid on the condition that all phases in question are ideal i e the saturated solution of the solute in a given solvent the solute in the gas phase and the pure solution of the solute It has been shown to be valid for a diverse set of liquid and solid solute data in water solvent see Thompson J D Cramer C J Truhlar D G J Chem Phys 2003 119 1661 The solubility utility program uses various features in the GAMESSPLUS program to calculate solubilities In particular it uses GAMESSPLUS to calculate aG fo
145. loalkane dehalogenase in water solvent The QM subsystem is 1 2 dichloroethane and the side chain of Asp124 and the MM subsystem is the rest of the protein and water The electronic structure theory used is MPWIK and the basis set is 6 31G d p for C and H atoms and 6 31 G d p for O and Cl The MM force field used for this test run is AMBER ff03 for the protein and TIP3P for water This system is treated in reference HT09 Short Tutorial for Making AMBER Parameter Topology and Coordinate Files As written in the Namelist AMBTOP and Namelist A MBCRD sections QM MM calculations with a site site representation of the QM MM electrostatic interaction require AMBER parameter topology and coordinate files Here as an example we treat a system consisting of 1 2 dichloroethane DCE and haloalkane dehalogenase and we briefly show how to make AMBER parameter topology and coordinate files from a PDB file using AmberTools version 1 3 The files used in this tutorial are located in QMMMtTests AmberTools directory For the details of how to use AmberTools see the AmberTools manual 1 Get or make a PDB file for the target system One can get the PDB file from Protein Data Bank http www rcsb org One may have to edit the PDB file to meet one s purpose In the case of the example the crystal structure of the enzyme substrate complex 2DHC pdb Nature 1993 363 693 was downloaded from the Protein Data Bank The modification was made
146. low lt Atomic pair no gt Description 1 H C 2 H N 3 H O 4 H Si 5 H P 6 H S 7 C N 8 C O Example input vECEOne Namelist NDDOSRP 10 11 12 13 14 15 16 17 18 19 20 o0 o0 U e 52 n Hi OZ FORDE RRR Ao VYU wWONY T N 0 0200 0 0149 0 0874 0 0215 Although the AM1 and PM3 methods are parameterized using broad sets of stable molecules these methods sometimes do not perform equally well for molecules outside the AM1 or PM3 training set for radical species or for transition states Furthermore sometimes one does not want the parameters that give the best results on the average i e the standard parameters rather one wants nonstandard parameters that perform the best for a specific reaction or for a limited range of systems In order to fix the energetic information for a chemical dynamics calculation on a specific reaction or a set of calculations on a specific range of systems the specific reaction parameter NDDO SRP method can be used To carry out an NDDO SRP calculation the namelist NDDOSRP is required In this namelist there is only one variable NAME which is used to indicate the name with maximum of 8 characters of an NDDO SRP file Note that the NDDO SRP file must be located in the scratch directory defined for the given calculation The scratch directory used for a particular calculation is defined with the SCR variable in the rungms script see the section entitl
147. lso tests the B3LYP method implemented in this version of GAMESSPLUS The PM3 parameters for Li which are necessary to carry out CM3 PM3 calculations for compounds containing Li were added Modifications were made to subroutine MPCDAT in the source file mpcdat src The source file mpcdat src is now part of the GAMESSPLUS distribution These parameters are tested in the portion of the test suite that tests the new CM3 PM3 parameters The B3LYP hybrid density functional theory method that uses version III of the VWN correlation functional i e the version of B3LYP coded in Gaussian and HONDO S was added This method 101 is requested by using the keyword DFTTYP B3LYP3 in data group DFT Minor modifications were made to subroutines INPGDFT CALCEXC in the source file dftexc src In addition a new subroutine called VWN3SCEF was created This routine is located in the source file smx src The source file dftexc src is now part of the GAMESSPLUS distribution Using the keyword B3LYP now causes the program to stop and ask the user to specify B3LYP3 or B3LYPS5 In the previous version of GAMESSPLUS GAMESOL version 3 1 common block DSOLVA which stored the gradients of the effective Born radii the gradients of the Coulomb integrals the gradients of the solvent accessible surface areas and the gradients of Gcps was removed The data in these arrays were instead stored on GAMESS s direct access file i e the DICTNRY file reco
148. ltix XE 1300 with Intel Xeon processors running SUSE Linux 2 6 16 compiled with the GNU Fortran compiler version gcc 4 1 2 gfortran in the comp compall and lked compilation scripts TARGET was set to linux64 SGI Altix XE 1300 with Intel Xeon processors running SUSE Linux 2 6 16 compiled with the PathScale compiler version 3 2 in the comp compall and lked compilation scripts TARGET was set to linux64 Sun Fire X4600 Linux cluster with AMD Opteron processors running SUSE Linux 2 6 16 compiled with the Intel Fortran compiler version 11 1 in the comp compall and lked compilation scripts TARGET was set to linux ia64 Sun Fire X4600 Linux cluster with AMD Opteron processors running SUSE Linux 2 6 16 compiled with the PathScale compiler version 3 2 in the comp compall and lked compilation scripts TARGET was set to linux Sun Fire X4600 Linux cluster with AMD Opteron processors running SUSE Linux 2 6 16 compiled with the Portland Group Fortran compiler version 8 0 in the comp compall and lked compilation scripts TARGET was set to linux Sun Fire X4600 Linux cluster with AMD Opteron processors running SUSE Linux 2 6 16 compiled with the GNU Fortran compiler version gcc 4 1 2 gfortran in the comp compall and ked compilation scripts TARGET was set to linux Note that the GHO AIHF module has been tested with CGPLUS v2008 and CHARMM version c30al on the IBM SP an
149. mory the job can use in words default 1 000 000 density functional theory DFT input DFTTY P HFE METHOD M N is the density functional or hybrid density functional A full list of the density functionals available in GAMESSPLUS that are suggested for use with CM4 and SM6 are given in the section entitled Density Functionals Recommended for use with CM4 CM4M and SM6 SMS in GAMESSPLUS Defines the fraction of Hartree Fock exchange N to be used when the MPWX functional is used This keyword must be specified when MPWX is used For CM3 and SM5 43 calculations that use MPWX ICMD 315 319 the HFE N must also be specified in the G6MSOL or CM2 namelist For CM4 and SM6 ICMD 416 419 HFE N must always be specified in the GMSOL or CM2 Namelist M is GRID to request grid based DFT calculation default Note Using the charge or solvation model parameterized for B3LYP one should specify DF TTYP B3LYP3 not DFTTYP B3LYP3 because the original parametrization has been done for the B3LYP method that uses version III of the VWN correlation functional In addition the charge and solvation models that are based on DFT were parameterized using grid based DFT so METHOD GRID the default for METHOD should always be used Namelist BASIS GBASIS N31 AM1 PM3 NGAUSS N NDFUNC N DIFFSP TRUE 41 input of available standard basis sets Pople s N 31G basis sets AM1 m
150. mportant to GAMESSPLUS users In this section the namelist name is followed by a brief description and below the namelist name is a set pertinent keywords and their options Namelist CONTRL group of global settings for the calculation SCFTYP RHF restricted Hartree Fock calculation default UHF unrestricted Hartree Fock calculation RUNTYP ENERGY SCF evaluation default MAXIT ICHARG MULT COORD ISPHER 40 GRADIENT gradient evaluation HESSIAN OPTIMIZE SADPOINT EEQM N N N UNIQUE CART ZMT 1 1 Hessian and possibly vibrational frequency evaluation geometry optimization to minima geometry optimization to saddle points EEQM calculation with a site site representation of the QM MM electrostatic interaction N is the maximum number of SCF cycles default 30 N is the molecular charge default 0 N is the multiplicity of the electronic state default 1 input symmetry unique Cartesian coordinates default input all Cartesian coordinates input Gaussian style Z matrix internal coordinates use Cartesian basis functions e g 6D 10F default use spherical harmonic basis functions e g 5D 7F Note The best way to input Cartesian coordinates is to set COORD UNIQUE this prevents coordinate rotation and to set the symmetry group of the molecule to C1 in DATA see below Namelist SY STEM MEMORY Namelist DFT N information for controlling the computer s operation N is the maximum me
151. ms may be too compressed In the case of testl inp an equilibrium MM MD simulation and a minimization with only hydrogen atoms allowed to move were performed first Then a simulation was performed in which all the atoms within 20 A of the center of the system moved were allowed to move Description of Test Suite for CM2 CM3 CM4 CM4M SMS5 42 SM5 43 SM6 SM8 and SM8T There are 763 test calculations in the SMx test suite located in SMxTests This test suite is broken in five subsets A B C D E and F Subset A and Subset B Subset A contains 57 input files for testing gas phase CM2 and CM3 charges and SM5 42 solvation energies and analytical gradients in aqueous and organic solvent calculated in combination with semi empirical models AM1 and PM3 Note that all of these tests fail being run in parallel because the parallel execution cannot be carried out for semi empirical methods in GAMESS Subset B contains 564 input files for testing gas phase CM2 CM3 and CM4 charges and SM5 42 and SM5 43 solvation energies and analytical gradients in aqueous and organic solvent calculated in combination with the Hartree Fock method and DFT There are twelve molecules used in subset A and in subset B Nine molecules out of 12 have been selected for these tests such that all of the CM2 parameters and all of the SM5 42 atomic surface tension parameters are tested The nine molecules w 1 9 are nitroethyne methyl disulfide fluorochlorobrom
152. n GHO boundary atoms QM frontier atom A and QM atoms directly bonded to A these atoms are also called geminal atoms are orthogonalized to each other before the hybridization By using this local L wdin orthogonalization method the mixing of tails from other QM atoms far from the boundary is eliminated and the perturbation introduced to the QM subsystem is minimized 28 Although the unparametrized GHO AIHF method gives reasonable optimized geometries and charges one can obtain even better results by scaling the integrals involving the boundary orbitals Such a parametrized version of GHO AIHF based on local L wdin orthogonalization is available for the MIDI basis set in which the scaling factors are obtained from a small training set containing propane propanol propanoic acid n butane and 1 butene 29 GAMESSPLUS Citation Publications including work performed with GAMESSPLUS should cite the software package in the following ways Journal of Chemical Physics or World Scientific style M Higashi A V Marenich R M Olson A C Chamberlin J Pu C P Kelly J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESSPLUS version 2010 2 University of Minnesota Minneapolis 2010 based on the General Atomic and Molecular Electronic Structure System GAMESS as described in M W Schmidt K K Baldridge J A Boa
153. n made in module smx src With the bug fix mentioned above and the correct implementation of the Mayer bond order derivatives the new version of GAMESSPLUS gives the correct analytical gradients for both GB Lowdin UHF and SM5 42 UHEF calculations Therefore GAMESSPLUS v4 1 now supports analytical gradients using unrestricted wave functions Note that GAMESSPLUS v4 1 is still unable to carry out analytical gradients for SM5 42 ROHE wave functions If ROHF is specified with SCREF on the program prints an error message and stops 104 GAMESSPLUS Version 4 2 March 2004 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS version July 3 2003 R2 e The SM5 43 parameter sets for HF 6 31G d and B3LYP 6 31G d were added The test suite was updated to test these new models Modifications to subroutines SETSOL DOSOLV COULRD SM5STN and OSM5 in smx src were made GAMESSPLUS Version 4 3 July 2004 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS version May 19 2004 R3 e The GAMESSPLUS module was implemented into the May 19 2004 R3 version of GAMESS This version of GAMESS provides PM3 parameters for lithium that were previously provided by the GAMESSPLUS module Thus the GAMESSPLU
154. nalysis and Redistributed L wdin Population Analysis L wdin population analysis like Mulliken analysis provides class II atomic partial charges but the L wdin method has certain advantages It has been implemented in GAMESSPLUS because L wdin population analysis charges are used for obtaining CM2 CM3 CM4 and CM4M charges However there may be some independent interest in L wdin analysis since it can be used with any basis set whereas CM2 CM3 CM4 and CM4M are defined only for selected basis sets and L wdin analysis will usually yield more useful population analyses than Mulliken s method Note that L wdin and Mulliken charges are identical for AM1 and PM3 because overlap is neglected in these methods Partial atomic charges obtained from L wdin population analysis can however be sensitive to basis set size particularly for extended basis sets that include diffuse functions We have developed and implemented a new method called redistributed L wdin population analysis or RLPA which alleviates some of this sensitivity to basis set size For methods using diffuse basis sets 6 31 G d and 6 31 G d p RLPA charges are used for obtaining CM3 and CM4 charges Charge Models Based on Class IV Charges CM2 CM3 CM4 and CM4M Class IV charges have the following advantages over class III charge models e g ChEIPG and Merz Kollman algorithms e Class III charges are unreliable for buried charges this problem is widely recogniz
155. nd SM5 42 calculations can now be carried out for wave functions that use spherical harmonic f functions The CM3 parameter sets which allow for evaluation of CM3 charges were added to this version of GAMESOL This provides the capability to evaluate gas phase CM3 charges liquid phase CM3 charges and the electrostatic contribution to the free energy of solvation using the generalized Born model The corresponding CM3 GB gradient may also be evaluated and used for geometry optimizations 100 The redistributed L wdin population analysis RLPA method was also added to this version of GAMESOL This new method can be used to evaluate gas phase RLPA charges and liquid phase RLPA charges The RLPA charges can further be used in a calculation of the electrostatic contribution to the free energy of solvation using the generalized Born model and of the corresponding free energy gradient which can also be used for geometry optimizations The namelist has been broadened to support the new options as well as the previous ones The namelist is now called GMSOL although the old name CM2 may also be used if desired To accommodate the new methods the allowed values of the ICMD keyword have been extended to include 300 302 and 303 The test suite was extended to test all of the CM3 parameter sets and to test the use of RLPA charges For portability issues common block DSOLVA was removed in this version of GAMESOL This common block stored th
156. nd second derivatives with respect to coordinates and electrostatic potentials are available Note that when the electrostatic potential of the MM subsystem is treated with a site site representation if there is a QM MM boundary that passes through a covalent bond the link atom method is used The option for QM MM calculations with a site site interaction should not be confused with the option for GHO QM MM calculations QM MM energy calculations and geometry optimization can be performed whereby the QM MM electrostatic interaction is treated by a site site representation and the AMBER force field is used 11 as the MM potential energy function Whereas the MM potential energy terms and their derivatives are evaluated by CHARMM when one uses the GHO QM MM option and therefore one must link to CHARMM these terms are evaluated by routines in the eeqmmm src file of GAMESSPLUS when one carries out QM MM calcuations with a site site represenatation of the electrostatics Therefore one does not need to add a separate program for calculating the MM terms However this part of the code does use AmberTools to read the MM input in Amber format GAMESSPLUS can carry out the geometry optimization in Cartesian coordinates but with constraints expressed in internal coordinates The user can enforce constraints on bond lengths sums or differences of bond lengths bond angles and torsional angles 12 Extended Abstract L wdin Population A
157. ns at first SCF cycle Gas phase energy is updated for every new geometry Gas phase energy is read in using the ETGAS keyword Gas phase energy is read in using the ETGAS keyword and gas phase SCF is always skipped i e the liquid phase SCRF begins at first SCF cycle This option is not available when ISCRF 2 Note The relative free energy of solvation indicated in the string 9 DeltaG S liq free energy of solvation see an output file depends on the reference gas phase energy indicated in the string 0 E EN g gas phase 49 Note Since the SCRF calculation with ISCRF 2 cannot skip the gas phase step IGAS 3 is not available for single point energy calculations and geometry optimizations using ISCRF 2 For the same reason IGAS 0 is not available for analytical geometry optimizations using ISCRF 2 either To carry out the analytical optimization with ISCRF 2 one needs to use IGAS 2 along with the ETGAS option or IGAS 1 However in case of IGAS 1 the user needs to add the nuclear or geometry relaxation energy term to the resulting free energy of solvation indicated in the string 9 DeltaG S liq free energy of solvation see an output file The geometry relaxation term in case of IGAS 1 should be equal to the difference in kcal mol between the gas phase total energy calculated at the gas phase equilibrium structure and the gas phase total energy calculated at the liquid phase equilibrium struc
158. nt will be printed during MM geometry optimization at every IPRIMM th step The default is 10 ICYCQM IMMFIX IFIRMM 63 Defines the maximum iteration cycle between QM and MM geometry optimizations The default is 1 which means that QM MM geometry optimization will be stopped only when both QM and MM gradients become less than the convergence criteria Integer array that specifies which MM atoms are to be fixed during MM geometry optimization IF IMMFIX Z is negative MM atoms from IMMFIX i 1 to IMMFIX i will be fixed For example when IMMFIX 1 1 3 5 7 MM atoms 1 3 4 5 7 in AMBTOP file will be fixed Logical value that specifies first iteration cycle between QM and MM geometry optimization When IFIRMM is false default QM geometry optimization will be performed at first When IFIRMM is true at first QM energy calculation will be performed to determine QM charges then MM geometry optimization will be carried out with QM atoms fixed Namelist INTFRZ Namelist INTFRZ controls internal coordinate constrained geometry optimization in Cartesian coordinates This option can be used with any type of geometry optimizations such as gas phase solvation model and QM MM calculations IFZBND IFZANG IFZTOR Integer array that specifies bond lengths and sums or differences of bond lengths to be constrained 0 means separation of constraints and more than four nonzero sequential numbers means sum or diffe
159. odel Hamiltonian PM3 model Hamiltonian the N of N31 e g N 6 for 6 31G N is the number of polarizing d subshells on heavy atoms N 1 for 6 31G d default 0 adds a diffuse sp shell to the basis set default FALSE Note The d subshells have 5 functions if ISPHER 1 and they have 6 functions if ISPHER 1 where ISPHER is defined in the CONTRL data group Thus when using the MIDI basis set ISPHER 1 should be specified in the CONTRL namelist When using the MIDI 6D basis set ISPHER 1 should be specified in the CONTRL namelist Examples 6 31G d SBASIS GBASIS N31 NGAUSS 6 NDFUNC 1 SEND N31 NGAUSS 6 NDFUNC 1 DIFFSP TRUE SEND 6 31 G d S BASIS GBASIS Namelist DATA input molecule information Input consists of the following information line 1 title line line 2 symmetry group always enter C1 for GAMESSPLUS calculations line 3 molecular coordinates for Cartesian input each line consists of the following atom label the atom s nuclear charge which should be a floating point value i e 1 0 for H and the atom s x y and z coordinate If a general basis set is required then the basis set for each atom follows the atom s coordinates See examples in the Input and Output Examples section and in the test suite Namelist FORCE METHOD SEMINUM FULLNUM NVIB 1 2 VIBSIZ N VIBANL TRUE controls Hessian and vibrationa
160. of Cl and CH3Br using Z matrix input and SM5 42 HF AM1 with numerical Hessian evaluation before and after the optimization using forward differences and a step size of 0 001 bohr ICHARG 1 SEND NT COORD ZMT SCONTRL SCFTYP RHF RUNTYP SADPOI SBASIS GBASIS AM1 SEND SSTATPT NSTEP 50 HESS CALC HSSEND TRUE SEND SFORCE METHOD NUMERIC NVIB 1 VIBSIZ 0 001 SEND SGMSOL ISCRF 1 ICMD 11 IAQU 1 SEND SDATA SN2 rxn TS Cl CH3Br C1 c H2 C BI H3 CI B2 H2 Al H4 C1 B3 H2 A2 Cl C B4 H2 A3 Br GI B5 H2 A4 B1 1 09846481 B2 1 09846481 B3 1 09846481 B4 2 03270616 B5 2 35911154 Al 119 44411379 A2 19 44411379 A3 94 29818687 A4 85 70181313 D1 165 20732678 D2 97 39633661 D3 82 60366339 SEND H3 H3 H3 70 D1 D2 D3 Note The user cannot enter both a Z matrix and a general basis set in DATA i e for Z matrix input the user is limited to using only GAMESS supported basis sets that can be entered using the BASIS namelist Example 6 EEQM calculation of charge response kernel with respect to electrostatic potential for water in the presence of an electrostatic potential using MPW1K 6 31G d SCONTRL SCFTYP RHF RUNTYP EEQM SEND CMD 417 HFE 0 428 wa CA O H
161. of the following atom label the atom s nuclear charge which should be a floating point value i e 1 0 for H and the atom s x y and z coordinate If a general basis set is required then the basis set for each atom follows the atom s coordinates See examples in the Input and Output Examples section and in the test suite Input specific to the GAMESSPLUS solubility utility The GAMESSPLUS solubility utility adds two new namelists to GAMESSPLUS the VAPOR namelist and the DGS namelist Both namelists define the specifics for a particular type of SMx calculation For a given solute A in a given liquid solvent B the DGS namelist provides the specifics for calculating the standard state free energy of solvation of the solute A in solvent B AGS The VAPOR namelist provides the specifics for calculating the standard state free energy of self solvation of solute A which defines the pure solute vapor pressure of A P The VAPOR namelist can also supply a user given pure solute vapor pressure of solute A in several different units in this case a free energy of self solvation calculation of solute A is not carried out These two namelists are used to specify the SM5 42 or SM5 43 parameter set which corresponds to a particular wave function for which the SM5 42 or SM5 43 parameters were optimized the available wave functions are shown in the section entitled Executive Summary above They also specify the solvent descriptors for the two
162. oiodomethane water hydrazine acetamide hydrogen cyanide hydrogen thionophosphate and 1 P fluorophosphino 1 silylmethanesulfonic acid These calculations are named testS1 x2y testS9 x2y where S A or B x indicates if a gas phase calculation of the CM2 charges x 0 an evaluation of the SM5 42 free energy of solvation in aqueous solution x 1 or in liquid chloroform x 2 is carried out the number 2 indicates that CM2 charges are used in the calculation and the value of y corresponds to the value of ICMD see the section entitled Namelists GMSOL and CM2 for allowed values of ICMD used in the calculation For calculations corresponding to x gt 0 SCF Scheme I is used except for those calculations involving SMS5 42 HF 6 31 G d ICMD 8 testBw x28 w 1 8 many SCRF evaluations using ICMD 8 and ISCRF 1 diverge note that ISCRF 1 is no longer available for methods that use diffuse basis functions because of this property so where appropriate SCF scheme II is used ISRCF 2 In addition when x gt 0 when the basis set employed uses Cartesian d shells and when ISCRF 1 analytical gradient evaluations are performed i e for ICMD 2 3 11 and 12 The table below summarizes this portion of the test suite Calculation type test cases L wdin and CM2 charges by AM1 and PM3 testA9 02y y 11 and 12 SM5 42 energy and analytical gradient evaluation in testAw g2y where w 1 8 q 1 and 2
163. olvent atoms that are F Cl or Br default is 0 0 For a desired solvent these values can be derived from experiment or from interpolation or extrapolation of data available for other solvents Solvent parameters for common organic solvents are tabulated in the Minnesota Solvent Descriptor Database The latest version of this database is available at http comp chem umn edu solvation Note that unless IAQU is set to 1 these solvent descriptors are required input The calculation will not run if they are not specified HFE Defines the fraction of Hartree Fock HF exchange to be used when ICMD 315 to ICMD 319 For example if a calculation of CM3 charges by the mPW1PW91 MIDI method were to be carried out then ICMD would be set to 315 and HFE would be set to 0 25 The HFE keyword only defines the fraction of HF exchange for a given calculation in the charge and solvation model portion of GAMESSPLUS not in the entire electronic structure part of the GAMESSPLUS program thus when using the MPWX functional the HFE keyword must also be specified in the DFT namelist This keyword is only active when ICMD 315 319 and in these instances HFE must be explicitly specified by the user in both the DFT and the CM2 or GMSPLUS namelists If ICMD 315 319 and HFE is not defined by the user GAMESSPLUS will terminate Input options specific to the VAPOR namelist There are three other options specific to the VAPOR namelist These opti
164. on coefficients to zero Note that the CM2CHG and CDS_Param files must be located in the directory defined to be the scratch directory for the calculation i e the directory defined by SCR in the rungms script see the section entitled Notes on Running GAMESSPLUS below for a description of the rungms script User provided surface tension coefficients are entered into the input file in the following fashion the first line of file contains the total number of surface tension coefficients that the user has provided in the file followed by the value of Coeff where ICoeff 1 use the surface tension coefficients provided in the code for any surface tension coefficients that are not provided by the user in this input file ICoeff 2 set all surface tension coefficients that have not been provided to zero 50 All coefficient types have been assigned a four digit coefficient label see the table below the user provides a coefficient label followed by its corresponding value in cal A on each line following the first line until all data is entered Note that only one parameter convention can be entered either a set of solvent descriptors for water or a set of solvent descriptor dependent coefficients can be entered but not a mix of both 1 e coefficient labels must be either all less than 2000 or all greater than 2001 One two and three atom surface tension coefficients
165. ond length to the Q1 M1 bond length is fixed Ror Roi Cor Ry Ro gt 34 where Co is constant and is specified in the input file The second way is the one proposed by Walker et al ref WCO7 and is the method used in the AMBER program In this type of link atom placement the Q1 QL bond length is fixed Rm Ro Ro Roi doL 35 mi Ra where do is the length of the Q1 QL bond In the current version of GAMESSPLUS the redistributed charge redistributed charge and dipole balanced redistributed charge balanced redistributed charge and dipole and AMBER default methods are all available to treat the QM MM electrostatic interaction near the QM MM boundary GAMESSPLUS can perform QM MM geometry optimization The QM and MM portions of geometry optimization are carried out separately First the QM geometry is optimized with the MM atoms fixed by the original GAMESS routine usually the quasi Newton Raphson method Then the MM geometry is optimized with the QM atom fixed by the GAMESSPLUS conjugate gradient method This procedure is repeated until the gradients of both the QM and MM atoms are below the convergence criterion When the QM MM cuts a covalent bond and the link atoms are placed on the QI M1 bonds the M1 atoms are optimized with the QM geometry Constrained Geometry Optimization in Cartesian Coordinates by Projection Operator Method GAMESSPLUS can carry out the geometry optimization with internal coordin
166. onent of is given by 23 ay EEM op a 9 P 23 EEQM Then the second partial derivatives of V first or second order in electrostatic potential are OV FEQM GQ Ky 24 O 0R OR and OV FEQM GQ f 25 2D oD 00 i These variables y and are known as charge response kernels CRKs In GAMESSPLUS the CRKs can be obtained by numerical differentiations of the charges In GAMESSPLUS can be given directly IRDMMB 0 in namelist EEQM or calculated from the MM charges Q and coordinates R which are read from namelist MM IRDMM 1 In the latter case M is given by N MM D R R 2 ik RT 26 where N is the number of MM atoms One can use the TINKER tapering function see next section for the QM MM electrostatic interactions Furthermore can be regarded as a function of R and R IADDGP 1 In this case the first derivative of V with respect to R is given by dy SEM E gy EEQM i gy EEQM ao dR oR OR eye aw aR aR ae and that with respect to R is given by dv FEM OV FEM ob RN OD ORY ab gt O RUT 28 A a 24 v QM and its first and second derivatives can be used as input for electrostatically embedded multiconfiguration mechanics calculation The TINKER tapering function for long range electrostatic interactions In GAMESSPLUS the TINKER tapering function is avail
167. ons are given below DENSITY Specifies the density of the pure solution of the solute in units of mol L This keyword is REQUIRED for all solubility calculations PRESSURE Defines a user specified pure solute vapor pressure of the solute P in units defined by the user with the UNITS keyword described below This keyword allows the user to enter a value for the pure solute vapor pressure of the solute instead of calculating it When this keyword is given the keywords DIELEC IAQU SolA SolB SolC SolG SolH SoIN and ICDS are not 115 required in VAPOR The default units for the vapor pressure are pascals however the pressure can be given in other units specified by the UNITS keyword see below UNITS Specifies the units of the user supplied vapor pressure The allowed values of this keyword are Pa atm bar and torr for pressure in units of pascals atmospheres bars and torr respectively The default for this keyword is UNITS Pa Test calculations Two test calculations are given with the GAMESSPLUS solubility utility program They are named pentane1 inp and pentane2 inp and are located under the directory gmsplus x gmsplus_solubility where x is the number of the given version of GAMESSPLUS The first test calculation computes the solubility of n pentane in water using SM5 42 HF 6 31G d to calculate both AG and P The second test calculation computes the solubility of n pentane in water using SM5 42 HF 6 31G d
168. ons on hydrogen either Thus bang is part way between star and no star In the MIDI basis set these d shells and the p shell for Li have been carefully optimized Note that the following convention regarding MIDI basis sets the original MIDI basis set used 5 spherical harmonic components for each d subshell and MIDI by itself denotes using spherical harmonic d subshells i e 5 component d subshells However we sometimes use 6 Cartesian 65 components for d shells this is denoted MIDI 6D For emphasis sometimes we use MIDI SD to denote MIDI The MIDI basis set is not stored internally in GAMESS but it can be used as an external basis set For convenience the MIDI basis set is provided in GAMESS format in the file gmsplus v4 6 Basis midi bang bas for all 12 elements for which it is defined H Li C N O F Si P S Cl Br I MIDI and MIDIX are synonymous as are MIDI 6D and MIDIX6D cc pVDZ basis set in Gaussian There are two ways to write down the cc pVDZ basis set For example for the hydrogen atom it can be written as Form 1 H O S 4 1 00 0 1301000000D 02 0 1968500000D O01 0 1962000000D 01 0 1379770000D 00 0 4446000000D 00 0 4781480000D 00 0 1220000000D 00 0 5012400000D 00 S 1 1 00 0 1220000000D 00 0 1000000000D 01 P 1 1 00 0 7270000000D 00 0 1000000000D 01 or as Form 2 H 0 S 3 1 00 0 1301000000D 02 0 1968500000D 01 0 19620
169. otentials for Substituted Anilines Phys Chem Chem Phys 2000 2 1231 CR99 Chuang Y Y Radhakrishnan M L Fast P L Cramer C J Truhlar D G Direct Dynamics for Free Radical Kinetics in Solution Solvent Effect on the Rate Constant for the Reaction of Methanol with Atomic Hydrogen J Phys Chem A 1999 103 4893 34 STOI Sicinska D Truhlar D G Paneth P Solvent Dependent Transition states for Decarboxylations J Am Chem Soc 2001 123 7863 SP02 Sicinska D Paneth P Truhlar D G How Well Does Microsolvation Represent Macrosolvation A Test case Dynamics of Decarboxylation of 4 Pyridylacetic Acid Zwitterion J Phys Chem B 2002 106 2708 AMI SRP and PM3 SRP models GO91 Gonzalez Lafont A Truong T N Truhlar D G Direct Dynamics Calculations with Neglect of Diatomic Differential Overlap Molecular Orbital Theory with Specific Reaction Parameters J Phys Chem 1991 95 4618 CE95 Corchado J C Espinosa Garcia J Hu W P Rossi I Truhlar D G Dual Level Reaction Path Dynamics The Approach to VTST with Semiclassical Tunneling Application to OH NH3 H20 NH7 J Phys Chem 1995 99 687 CR99 Chuang Y Y Radhakrishnan M L Fast P L Cramer C J Truhlar D G Direct Dynamics for Free Radical Kinetics in Solution Solvent Effect on the Rate Constant for the Reaction of Methanol with Atomic Hydrogen J Phys Chem A 1999
170. prevent water molecules from drifting away during the MD simulation The sixth line set default OldPrmtopFormat on means that the AMBER parameter topology files will be printed with old format Note that the current version of GAMESSPLUS supports only old formats The seventh line is optional which mean the atoms and their coordinates including those generated by LEaP in the 2DHC system will be printed into a file called 2DHC_wat20_tleap pdb with PDB format The eighth line saveAmberParm 2DHC 2DHC_wat20 top 2DHC_wat20 crd means that the AMBER topology parameter and coordinate files for the 2DHC system will be saved as files called 2DHC_wat20 top and 2DHC_wat20 crd respectively The LEaP program will be stopped by the last command quit One can also enter these commands step by step on the LEaP terminal After the LEaP input file is created one can run the LEaP program as follows tleap f leap inp gt amp leap out The output file leap out will be like this SEs I I f Adding amber10 dat leap prep to search path Adding amber10 dat leap lib to search path Adding amber10 dat leap parm to search path Adding amber10 dat leap cmd to search path Source leap in Welcome to LEaP Sourcing leapre amber10 dat leap cmd leapre og file leap log Loading parameters amber10 dat leap parm parm99 dat Reading title PARM99 for DNA RNA AA organic molecules TIP3
171. program would terminate in subroutine SHALF if the number of basis functions for the calculation was greater than 300 However the maximum number of basis functions allowed in a normal GAMESS calculation is 2047 To make GAMESOL program limitations the same as GAMESS program limitations which consequently makes GAMESOL transparent to normal GAMESS users the maximum number of basis functions allowed in subroutine SHALF is now 2047 This modification requires that the parameter MXBAS be set to 2047 throughout the file smx src In addition the array that stores the square root of the overlap matrix is dimensioned to 2047 2047 1 2 The parameter MXATM which defines the maximum number of atoms in GAMESOL was inconsistently defined in various subroutines used in GAMESOL The maximum number of atoms allowed in GAMESS is 500 so to make GAMESOL program limitations consistent with GAMESS program limitations all occurrences of MXATM in GAMESOL specific code has been set to 500 Mayer s bond order formula for unrestricted wave functions was incorrectly implemented in version previous to 3 0 1 GAMESOL now supports energy calculations using unrestricted wave functions but not analytical gradients GAMESOL Version 3 1 August 2002 Authors J D Xidos J Li T Zhu G D Hawkins J D Thompson Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version June 11 2000 CM2 CM3 L wdin RLPA SM5 42R a
172. put testA20b 91 The T20ASRP and T20ACSRP files used to run testA20a and the T20BSRP file used to run testA20b should be kept in the scratch directory Two test cases are performed using SM5 42 HF MIDI 6D ISCRF 1 e geometry optimization of 2 4 pentadione in acetonitrile solution testB 13 e transition state optimization for the Sy2 reaction of NH3 and CH3Cl in aqueous solution the Menschutkin reaction testB14 Two numerical Hessian calculations are carried out in testB14 one before the optimization to generate a good Hessian guess and one after the optimization to ensure that the stationary point is a first order saddle point Test case testB 16a performs a generalized Born electrostatic solvation energy and gradient evaluation using L wdin partial atomic charges and UHF 6 31G d f for peroxyl radical in aqueous solution Test case testB 16b performs a generalized Born calculation of the electrostatic contribution to the free energy of solvation by RHF MG3 Test case testB 16a tests the use of Cartesian f functions in the basis set and the use of unrestricted wave functions in generalized Born calculations Test case testB 16b tests the use of spherical harmonic f functions in the basis set Test cases testB 17a testB 17b testB 17c and testB17d all perform an SM5 42 MIDI 6D calculation of water solute in acetone solvent using ISCRF 2 e testB17a tests the ICREAD 1 option e testB17b tests the ICSAVE 1 option e testB17c
173. r a given solute solvent system using the SM5 42 or SM5 43 continuum solvation model It also uses GAMESSPLUS to calculate P which is defined by the free energy of self solvation of the solute the free energy of solvation of the solute in a pure solution of itself using SM5 42 or SM5 43 Instead of using SM5 42 or SMS 43 to predict the free energy of self solvation this utility program can also take as input a user defined value for the pure solute vapor pressure The solubility of a solute can be calculated with SM5 42 for the following restricted and unrestricted wave functions HF MIDI HF cc pVDZ HF MIDI 6D HF AM1 HF 6 31G d HF PM3 HF 6 31 G d B3LYP MIDI B3LYP 6 31G d and with SM5 43 with HF 6 31G d and B3LYP 6 31G d using either a restricted or unrestricted formalism This section and several subsections below of this manual provide a stand alone introduction to the solubility utility program for users who just want to calculate solubilities Such users should first install GAMESSPLUS according to the instructions given in the sections entitled Notes on Running GAMESSPLUS and Updating and Compiling GAMESSPLUS of this manual Users of this utility program should give the appropriate references described in the section entitled GAMESSPLUS Reference Then everything else they need to know is in the following few sections The SM5 42 and SM5 43 continuum solvation models SM5 42 and SM5 43 are universal Solvation Models that use
174. r sequentially that is on one CPU or in parallel on one N core processor in this case the number of CPUs will be equal to N that is 2 for a dual core processor 4 for a quad core one We have successfully tested the whole GAMESSPLUS test suite in parallel on 4 32 CPUs except subtest A because the original GAMESS cannot run semiempirical calculations in parallel and testE14 because it fails due to some run time error unrelated to GAMESSPLUS We have also successfully tested the original GAMESS test suite in parallel except exam05 exam23 exam25 exam27 exam32 exam39 and exam42 which cannot be run in parallel with the current version of GAMESS However we encourage the user to always make sure that the computational results are identical regardless of sequential or parallel execution 83 GAMESSPLUS Test Suite The GAMESSPLUS test suite is located in the directories EEQMTests and SMxTests of the GAMESSPLUS distribution There are two subdirectories Tests and Output in each of these directories Input files of test calculations and scripts to run the test suite and analyze the results are located in the Tests directory Output files corresponding to a successful installation of GAMESSPLUS are located in the Output directory In the following sections a discussion of the contents of the test suite how to run the test suite and how to verify that GAMESSPLUS is installed correctly by analyzing the test suite is presented Note Becau
175. rallel jobs For serial jobs ddikick x spawns two GAMESS processes although only one process actually does anything The running of GAMESSPLUS v2010 2 has been tested in both parallel and serial modes using the full GAMESSPLUS and GAMESS test suites Representative Performance Data on Running GAMESSPLUS in Parallel The performance data below were obtained on an SGI Altix cluster using up to 16 Intel Itanium2 Madison class processors We performed an SM8 RHEF 3 21G single point energy and analytical gradient calculation for the heme A molecule C49Hs606N4Fe in its singlet spin state The system contains 116 atoms 452 electrons and 672 atomic orbitals The option DIRSCF TRUE in SCF was used The timing is given in minutes 1 proc 2 procs 4 procs 8 procs 16 procs setup Hiickel guess 0 9 0 7 0 7 0 7 0 6 57 RHF iterations 299 5 179 8 120 7 90 2 71 3 properties 0 3 0 3 0 3 0 3 0 3 1 2 electron gradients 80 4 78 9 77 6 78 8 80 0 total CPU time 381 1 259 7 199 3 170 0 152 2 total wall time 381 9 261 8 206 2 173 4 241 8 The user should take the liberty to decide if there is a need to run GAMESSPLUS in parallel for a particular task on a particular platform Although the SCF calculation is quite scalable there is no speed up for gradients with the current version of GAMESSPLUS The scalability also depends on the quality of a network interconnecting different nodes or CPUs within the same node In general we recommend to run GAMESSPLUS eithe
176. rations of organic carbon for soil and aqueous solution respectively Typically a standard state of 1 yg of solute g of organic carbon is used for C and 1 mol L is used for Cy The GAMESSPLUS soil sorption utility program calculates Ko according to Koc Poil ace AGEs 2 where poi is the density of soil in g mL AG is the standard state free energy associated with transferring a solute from the gas phase to aqueous solution and AG Soi 18 the standard state free energy associated with transferring a solute from the gas phase to soil In the GAMESSPLUS Soil sorption utility program calculated AG values are for a standard state of 1 mol L in both the gas and aqueous phase and AG pi Values are for a standard state of 1 mol L in both the gas phase and in soil Using the above relationship between Koc gt Psoil gt AGG and AG the GAMESSPLUS Soil sorption utility program can soil calculate Koc several ways First given a value for pi gt the GAMESSPLUS Soil sorption utility program can calculate AG and AGsoil gt and then use these calculated values in eq 2 to determine Koc For calculating AG gt the SM5 42 o aqueous continuum solvation model is used For A vil the SM5 42 universal continuum solvation model is used along with a set of solvent descriptors that have been empirically optimized for modeling bulk soil for a description of the SM5 42 model for soil as well as a discussion of
177. rc to the gamess source directory where x is the number of the given version of GAMESSPLUS for instance x v2010 2 3 Go to the gamess tools directory and copy file actvte code to file actvte f then modify file actvte f based on the instructions given there Compile actvte f to generate an executable file called actvte x 77 actvte f o actvte x 4 Go to the gamess ddi directory and edit the compddi script In this file modify the set TARGET line to list the appropriate machine type Compile the distributed data interface portion of the GAMESS code by typing compddi When this compilation is completed move the file named ddikick x to the gamess directory 5 In the gamess directory edit the three compile script files in the gamess directory compall comp and ked In all three files modify the set TARGET line to list the appropriate machine type and the chdir line to list the directory name where GAMESS resides Add the lines comp smx comp nddosrp comp ghodum comp eeqm after the comp zmatrix line in compall Finally add smx o nddosrp o ghodum o eeqm o to the list of object files in file ked near the end of the file On some platforms the FORTRAN optimization level may have to be lowered for some of the source files This can be determined by running the full test suite and checking for any discrepancies between results obtained and those provided in the distribution see the next section
178. rds 267 270 Because the gradients of the effective Born radii and the gradients of the solvent accessible surface areas are each only used locally in one specific subroutine they do not need to be written to disk so they are no longer stored on the DICTNRY file GAMESSPLUS Version 4 0 September 2003 Authors J D Xidos J Li T Zhu G D Hawkins J D Thompson Y Y Chuang P L Fast D A Liotard D Rinaldi C J Cramer and D G Truhlar GAMESS version January 14 2003 R3 A utility program for calculating solubilities with the GAMESSPLUS program has been added This utility program computes the solubility for a solute in a given solvent by computing the standard state free energy of solvation of the solute in the solvent using the SM5 42R solvation model and by computing the pure solute vapor pressure of the solute which is defined by the free energy of self solvation of the solute with SM5 42R In addition the user can specify a value for the vapor pressure instead of calculating it with SM5 42R The solubility utility program is a PERL script called solubility pl It has been added to the GAMESSPLUS distribution package No modifications have been made to the GAMESSPLUS source code GAMESSPLUS Version 4 1 January 2004 Authors J Pu J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast D A Liotard D Rinaldi J Gao C J Cramer and D G Truhlar GAMESS version July 3
179. re read from namelist MM IUCMM 0 The MM coordinates are in bohr IUCMM 1 The MM coordinates are in angstrom default RCUT Defines a cutoff distance r ut in angstrom for the QM MM electrostatic interaction Default 100 0 58 ITAPER Specifies whether the TINKER tapering function is used for the QM MM electrostatic interactions or not See Section The TINKER tapering function for long range electrostatic interactions ITAPER 0 The TINKER tapering function is not used default ITAPER 1 The TINKER tapering function is used CTAPER Defines the fraction of r with respect to r See Section The TINKER tapering function for long range electrostatic interactions 7a CTAPER X w The default ut value is 0 65 which is the same value as that in TINKER Namelist MM Namelist MM defines the charges and coordinates of the MM atoms Namelist MM is effective only if IRDMM 1 in namelist EEQM The format of namelist MM is as follows MM N MM MM MM MM atom Q x yi Zi MM MM MM MM atom Q x y Z gt MM MM MM MM atom Q X y3 Q MM MM MM MM atom m NMM X yMM J ym Z yM END N is the number of MM atoms atom a is the name of A th MM atom A 1 2 N A ov is the partial charge in unit of e of A th MM atom x y and z are the Cartesian coordinates of the A th MM atom whose units are determined by IUCMM in namelist EEQM Example of namelist MM SMM 4 N 0 59717
180. rence when the number is negative of bonds will be constrained Example IFZBND 1 1 2 0 1 3 1 2 and 1 3 bonds will be constrained IFZBND 1 1 2 1 3 The sum of 1 2 and 1 3 bonds will be constrained IFZBND 1 1 2 1 3 The difference between 1 2 and 1 3 bonds will be constrained IFZBND 1 1 2 3 4 5 6 7 8 The R R R5 Rg value will be constrained where R is the distance between a and b Integer array that specifies bond angles to be constrained Currently GAMESSPLUS cannot constrain sums or differences of bond angles 0 is required between two constraints when user wants to constrain two or more bond angles Example IFZANG 1 2 1 3 2 1 3 angle will be constrained IFZANG 1 2 1 3 0 5 4 6 2 1 3 and 5 4 6 angles will be constrained Integer array that specifies torsional angles to be constrained Currently GAMESSPLUS cannot constrain the sums or differences of torsional angles 0 is required between two constraints when user wants to constrain two or more torsional angles 64 Example IFZTOR 1 1 2 3 4 1 2 3 4 torsional angle will be constrained IFZTOR 1 1 2 3 4 0 5 6 7 8 1 2 3 4 and 5 6 7 8 torsional angles will be constrained INITCN Specifies whether internal coordinates to be constrained will be initialized or not INITCN 0 Initialization will be not performed Internal coordinates calculated by Cartesian coordinates in namelist DATA are regarded as constraint values default INITCN 1 Ini
181. riginal version of the QM MM program included in eeqmmm src in the Code subdirectory and for permission to include it in modfified form in GAMESSPLUS 75 A Note on GAMESS Versions In updating GAMESS it is important to understand GAMESS versions GAMESS versions are primarily determined by the date printed in the box at the top of the output However that is not a unique version indicator A complete unique specification of a version of GAMESS therefore requires specifying not only the date in the box at the top of the output but also whether or not any routines have dates later than this and if so which routines and what dates The present version of GAMESSPLUS is designed for use with the April 11 2008 R1 version of GAMESS For users with this version of GAMESS updating and compiling can be accomplished using the Standard Method for Updating and Compiling GAMESSPLUS For users with other versions of GAMESS updating and compiling can usually be accomplished using the Makepatch Method for Updating and Compiling GAMESSPLUS 76 Standard Method for Updating and Compiling GAMESSPLUS Instructions for modifying the April 11 2008 R1 version of GAMESS to produce a GAMESSPLUS code that can carry out CM2 CM3 CM4 SM5 42 SM5 43 SM6 SM8 NDDO SRP CM2 SRP and EEQM calculations are summarized as follows for all other versions of GAMESS use the Makepatch Method for Updating and Compiling GAMESSPLUS 1 Obtain the Ap
182. ril 11 2008 R1 version of GAMESS from Iowa State University and obtain GAMESSPLUS from the University of Minnesota 2 Place both files in the same directory e g jsmith gamessplus and gunzip and untar them The files will untar into the directories jsmith gamessplus gamess and jsmith gamessplus gmsplus x where x is the number of the given version of GAMESSPLUS for instance x v2010 2 Move into the gmsplus directory and execute modgms modgms This script might prompt you for a few bits of system information and then it will compile GAMESSPLUS If GAMESSPLUS compiled successfully the script should end with something like if x v2010 2 done with all compilations GAMESS will be linked into the executable image gamessplus v2010 2 x chdir object xlf o gamessplus v2010 2 x q64 W1 m WI1 bloadmap Iked map gamess o unport o messages but no errors from linker End of GAMESPLUSS INSTALL The executable gamessplus v2010 2 x should now be in the jsmith gamessplus gamess directory Makepatch Method for Updating and Compiling GAMESSPLUS Instructions for modifying versions of GAMESS other than the April 11 2008 R1 version to produce a GAMESSPLUS code that can carry out CM2 CM3 CM4 SM5 42 SM5 43 SM6 SM8 NDDO SRP CM2 SRP and EEQM calculations are summarized as follows 1 Obtain GAMESS from Iowa State University and obtain GAMESSPLUS from the University o
183. rix F without the Gp correction will be used This inconsistency makes the use of Eq 3 fail to evaluate the energy weighted density matrix correctly This bug only exists for UHF and ROHF cases For RHF cases GAMESSPLUS v4 0 evaluates the W matrix using Eq 2 for one set of orbitals only In GAMESSPLUS v4 1 we fix this bug When SCRF is on GAMESSPLUS v4 uses Eq 2 instead of Eq 3 to compute W matrix Modifications have been made in module grd src Modified implementation for gradients of the Mayer bond order based on UHF wave functions In GAMESSPLUS v4 0 the Mayer bond order has been correctly implemented as follows both of which are correct for both RHF and UHF wave functions B gt gt P s PS P s P s 5 AcA wcB 2 gt Y S P S PES P S J 6 AcA wcB where P and PA are the alpha and beta spin density matrices respectively P P PP is the spinless density matrix and PS P PP is the spin density matrix Note that for RHF wave functions the spin density matrix P is zero by definition However the gradients of the bond order are not implemented consistently with Eq 5 in GAMESSPLUS v4 0 Therefore the GB CM2 UHF or SM5 42 UHF gradients based on CM2 charges were incorrect In the GAMESSPLUS V4 1 the form of the 0B OR part of OGp OR the energy gradient changes due to the implementation of Eq 5 for the Mayer bond order Necessary modifications have bee
184. s described as VM E K b ba gt Ko 0 0a bonds angles K E ee 33 gt 1 c08 np 7 E dihedrals non bonded r r r pairs where b 0 and r are bond length bond angle dihedral angle and distance between non bonded atoms respectively The other quatities in Eq 33 are parameters Note that the user can use any force field that has the form described in Eq 33 e g TIP3P and OPLS as the MM potential energy function In this manual we call the force field described by Eq 33 the AMBER force field for simplicity There are many versions of the AMBER force field The AmberTools manual recommends the ff03 force field ref DWO3 and ff99SB force field refs WCOO and HA06 for proteins The default water model in the AmberTools program is TIP3P ref JC83 For non protein molecules one can use the general AMBER force field GAFF ref WW04 GAMESSPLUS reads AMBER parameter topology and coordinate inputs generated by the AmberTools program In the current version of GAMESSPLUS the link atom method is used when the QM MM boundary cuts a covalent bond The link atoms QL usually hydrogen or fluorine are always located on the Q1 M1 bonds where Q1 and M1 denote the QM and MM boundary atoms respectively 26 The position of the link atoms can be determined in two possible ways The first way is as proposed by Morokuma and co workers ref DK99 In this type of link atom placement the ratio of the Q1 QL b
185. se the use of the GHO AIHF functionality of GAMESSPLUS requires usage of CHARMM a separate test suite is provided by the CGPLUS package for testing GHO AIHF through the CHARMM GAMESSPLUS combination package http comp chem umn edu cgplus see the CGPLUS v1 0 Users Manual Description of Test Suite for EEQM There are totally 150 test calculations in the EEQM test suite with IRDMM 0 located in EEQMTests Three molecules were selected to test the EEQM calculations with a site site representation of the QM MM electrostatic interactions In all cases calculations were performed using the MPWX density functional where X is the percentage of Hartree Fock exchange and the 6 31G d basis set The three molecules are water test1 a t methyl chloride test2 a t and the transition state of the Sn2 reaction of methyl chloride with chloride anion test3 a 1 Calculation type test cases EEQM energy calculation testx a t where x 1 3 and a 0 25 428 606 or 999 and t 1 gas phase or 6 embedded EEQM gradient calculation testx a t where x 1 3 and a 0 25 428 606 or 999 and t 2 gas phase or 7 embedded EEQM Hessian calculation testx a t where x 1 3 and a 0 25 428 606 or 999 and t 3 gas phase or 8 embedded EEQM CRK calculation with respect to testx a t where x 1 3 and a 0 25 428 606 coordinates or 999 and t 4 gas phase or 9 embedded EEQM CRK calculation with respect to testx a t where x
186. stretching parameters for interactions within the QM region are ignored Current version of GAMESSPLUS can read only an old format of the parameter topology input which has no comment lines If one has a new format of the parameter topology input whose first line starts VERSION comment the input file must be converted to the old format by new2oldparm program which is one of programs included in AmberTools Usage of new2oldparm program is as follows new2oldparm lt input file name gt output file name The current version of GAMESSPLUS does not support QM MM calculations with water cap periodic boundary conditions including the Ewald method for long range electrostatic interactions the generalized Born implicit solvent model or the polarizable model Therefore if such options are specified in AMBTOP they are ignored Example of namelist AMBTOP SAMBTOP 5817 17 3334 2555 5439 3474 10449 8021 0 0 28185 628 2555 3474 8021 51 105 51 35 1 0 0 0 0 0 0 0 0 24 1 0 N H1 H2 H3 CA HA CB HB2 HB3 CG HG2 HG3 SD CE HE1 HE2 HE3 C O N H CA HA CB HB CG2 HG21HG22HG23CG1 HG12HG13CD1 HD11HD12HD13C O N H CA HA CB HB2 HB3 CG OD1 ND2 HD21HD22C O N H CA HA CB HB1 HB2 HB3 C O N H CA HA CB HB CG2 HG21HG22HG23CG1 HG12HG13CD1 HD11HD12HD13C OoOO OoOO OoOO OoOO OoOO OO Note There are many versions of the AMBER force field The AmberTools Manual recommends the ff03 force field ref D W03 and ff99S
187. stricted and unrestricted HF DFT and hybrid DFT methods are available but only the 6 31 G d or the 6 31 G d p basis set should be used Note that CM3 for the BLYP and B3LYP methods and CM3 1 for the HF MIDI method in the above table uses a different mapping function for compounds that contain N and O than CM3 for MPWX and HF does For more information see Parameterization of Charge Model 3 For AM1 PM3 BLYP and B3LYP by Thompson J D Cramer C J Truhlar D G J Comput Chem 2003 24 1291 and Accurate Partial Atomic Charges for High Energy Molecules with the MIDI Basis Set by Kelly C P Cramer C J Truhlar D G Theor Chem Acc 2005 113 133 HFE Defines the fraction of Hartree Fock HF exchange to be used when ICMD 315 to ICMD 319 or ICMD 416 to ICMD 419 For example if a calculation of CM3 charges by the mPW1PW91 MIDI method were to be carried out then ICMD would be set to 315 and HFE 45 would be set to 0 25 The HFE keyword only defines the fraction of HF exchange for a given calculation in the charge and solvation model portion of GAMESSPLUS not in the entire electronic structure part of the GAMESSPLUS program thus when using the MPWX functional the HFE keyword must also be specified in the DFT namelist This keyword is only active when ICMD 315 319 or ICMD 415 419 or ICMD 500 510 and in these instances HFE must be explicitly specified by the user in both the DFT and t
188. such that all of the SM6 atomic surface tension parameters are tested In all cases calculations were performed using MPWX where X is the percentage of Hartree Fock exchange The eight molecules are nitroethyne testC1 a 1b methyl disulfide testC2 a 1b hydrogen peroxide 92 testC3 a 1b water testC4 a 1b hydrazine testC5 a 1b acetamide testC6 a 1b hydrogen cyanide testC7 a 1b and hydrogen thionophosphate testC8 a 1b In the names above a is the value of X in MPWX between 0 and 999 and b is the value of ICMD between 416 and 419 Two important issues that arise when optimizing geometries is the level of accuracy that should be used for both the SCF convergence tolerance this is also an issue for single point calculations and the gradient convergence tolerance For both single point calculations and geometry optimizations in GAMESSPLUS by default the SCF convergence is set to 10 5 a u the gradient convergence tolerance is set to 10 4 a u bohr for the maximum value of the gradient and to 1 3 the value of tolerance for the maximum value of the gradient for the root mean square of the gradient Our tests show that in many cases the default options in GAMESSPLUS are suitable for performing geometry optimizations in solution However the user should be aware that in some cases the default options may lead to convergence problems In these cases depending on the non default options specified the calculated results should be
189. t of coefficients read in CCMSRP and DCMSRP RLPA is not used ICMD 999 Use the set of coefficients read in CCMSRP DCMSRP and ZCMSRP RLPA is used HFE Defines the fraction of Hartree Fock HF exchange to be used when ICMD 315 to ICMD 319 or ICMD 416 to ICMD 419 This option is the same as HFE in namelists GMSOL and CM2 ZCMSRP Determines the parameters used in the RLPA charge calculation when ICMD 999 The RLPA charge is given by Q RLPA Q LPA Z gt exp a Ras gt ZY exp a7 Ras b a b a where Z is a empirical parameter Y is the L wdin population that is associated with the diffuse basis functions on atom a is the diffuse orbital exponent on atom a and R is the distance between atom a and b ZCMSRP i determines Z for atomic number i For example if one uses Z 0 11 for a Cl atom one sets ZCMSRP 17 0 11 The default is that all the ZCMSRP are zero DCMSRP CCMSRP Determines the parameters used in the CM2 CM3 CM4 charge calculation when ICMD 998 and ICMD 999 These are the same as C and D in namelist CM2SRP The CM2 CM3 CM4 charge is given by Q z o3 a Bi Di C Ba b a 57 where Q is the partial atomic charge from either LPA or RLPA B is the Mayer bond order between atom a and b and D and C are empirical parameters DCMSRP i and CCMSRP 7 determine D and C for atomic pair i The list of the atomic pairs is the same as that in namelist CM2SRP For example if one uses D
190. te new patch files for other versions of GAMESS bassto src patch inputa src patch rhfuhf src patch dft src patch inputb src patch statpt src patch dftxca src patch intl src patch symorb src patch gamess src patch int2a src patch 74 grd1 src patch mpcegrd src patch grd2a src patch mpcint src patch grd2b src patch mpcmol src patch grd2c src patch mthlib src patch QMMM Tests contains three subdirectories called AmberTools Tests and Output The subdirectory AmberTools contains some files used to make AMBER parameter topology and coordinate files See the section entitled Short Tutorial for Making AMBER Paramter Topology and Coordinate Files The subdirectory Tests contains the QM MM test suite input inp files and the subdirectory Output contains the corresponding output samples log SMxTests contains two subdirectories called Tests and Output The subdirectory Tests contains the SM lt x solvation test suite input inp files and the subdirectory Output contains the corresponding output samples log The directory SMxTests Tests also contains a script compare pl used to verify correct installation of GAMESSPLUS scripts used to run the SMx test suite run bat rungmsplus and run pbs the T20ASRP and T20ACSRP files used to run testA20a the TZOBSRP file used to run testA20b and the CDS_ Param file used to run testB13c and testB 13d Acknowledgment The authors are grateful to Prof Shigehiko Hayashi for providing the o
191. tehascbdecipadenivcacsedee cones saataassizacnudacbucaselsvodanetsiatanshieds en tondueaueosentee 5 Extended ADSI AEE soisessa iisisti itsin a s i i a a i 12 L wdin Population Analysis and Redistributed L wdin Population Analysis scsscccssssssssessseees 12 Charge Models Based on Class IV Charges CM2 CM3 CM4 and CM4M sssccssssscssssessssssseccssscees 12 SM5 42 SM5 43 SM6 SM8 SM8AD and SM8T Solvation Models ccssssccssssssscessssccessssccsessssceeees 13 Incorporating temperature dependence into the SMx models SMB ce eeeseseceseeseeeceseeeeceeeecsaeceeeaecaseeesaeceeeeaeeess 15 A comment on using gas phase geometries to calculate solvation free energies eee ceeecsecseeeeeeeeesecseeeeereeeeaeeees 15 Why use SM5 42 SM5 43 SM6 SM8 or SM8AD ooo ceeeceesececeecssseeeereecseceeeeesaeceeeeesaeeeceeeessaeceeeecsaeceeeeeaeeeesees 16 Analytical gradients and geometry optimization in liquid phase solutions c ccesceesceeseeseeeesceseeeseceaeensecnnecneenaeens 17 Notation for Solvation Models cssscscsscssssssscssscscssccssssesssscssesessscsscssssssesscsssseneesecsssesseneessssssseesecsscseeees 18 Solvent Parameters scscssssssssscsssscscscsssscessssscsscsscsssssscsscsssseseesecsssenecsscssesscssesssessesssseneesesssesssesessocsseses 18 NDDO and CM2 Specific Reaction Parameters SRP Models ssssccssssccssscsssssssssccssssssssccsssssssssscssnees 18 Solubility CalCu
192. tests the ISREAD 1 option e testB17d tests the ISREAD 2 option Test case testB21 tests the calculation of the bond order using an unrestricted wave function In particular testB21 performs an SM5 42 UHF 6 31G d energy calculation of C H in water Test case testB22 calculates the electrostatic contribution to the aqueous free energy of solvation of water using RLPA charges The corresponding free energy gradient is also calculated in testB22 TestB23 tests the new CM3 mapping scheme for compounds that contain N and O In particular testB23 is a calculation of the electrostatic contribution to the aqueous free energy of solvation of pyramidal formamide using the generalized Born method and CM3 charges calculated by B3LYP 6 31G d The free energy gradient which is calculated analytically is also calculated in testB23 Test cases testB24 and testB25 test the MPW1K functional Test cases testB26 and testB27 both test the B3LYP functional In testB26 the version II VWN correlation functional is used B3LYP3 the version of B3LYP implemented in Gaussian testB27 uses the version V VWN correlation functional B3LYP5 the version of B3LYP implemented in GAMESS Test case testB28 tests the MPW1IPW91 functional also called MPW25 and test case testB29 tests the MPWX functional and the use of the HFE keyword in the DFT namelist Subset C Subset C contains 96 input files for testing the SM6 solvation model Eight molecules have been selected
193. th the new HFE keyword Note that X is a percentage and HFE is a fraction Thus X 100 HFE This keyword is entered in the DFT namelist for gas phase calculations and also in the CM2 or GMSOL namelist for CM3 CM4 SM5 43 or SM6 calculations A test job test 29 has been added to the test suite that tests the MPWX and HFE keywords The keywords IRADII ISTS and SOLVRD were added The IRADII keyword specifies the set of atomic number dependent radii that are used to build the molecular cavity The ISTS keyword determines whether SM5 or SM6 type functionals are used The SOLVRD keyword specifies the solvent radius that is used for the SASA The default value has been set to 0 40 A Coulomb radii for SM6 were added In previous versions of the code a default value of 0 0 A was assigned to elements for which an intrinsic Coulomb radii had not been optimized In version 4 7 the default atomic radius for elements for which an intrinsic Coulomb radius has not been optimized is Bondi s value for the van der Waals radius when available and 2 0 A for all other atoms The atomic radii used in the SCRF calculation intrinsic Coulomb radii are now printed out by default 106 The CM3 parameter sets for MPWX MIDI ICMD 315 MPWX MIDI 6D ICMD 316 MPWX 6 31G d ICMD 317 MPWX 6 31 G d CMD 318 and MPWX 6 31 G d p CMD 319 were added and tested The CM3 and CM3 1 parameter sets for HF MIDI were added and tested CMD 301 and
194. tial atomic charges Class II charges can be calculated For calculations using the 6 31 G d and 6 31 G d p basis sets gas phase and liquid phase redistributed L6wdin population analysis RLPA partial atomic charges can be calculated for all restricted and unrestricted HF DFT and hybrid DFT methods available in GAMESS e Gas phase and liquid phase CM2 class IV charges can be determined for the following combinations of electronic structure theory and basis set using either a restricted or an unrestricted formalism AMI PM3 HF MIDI B3LYP MIDI HF MIDI 6D BPW91 6 31G d HF 6 31G d HF 6 31 G d BPW91 MIDI HF cc pVDZ BPW91 MIDI 6D BPW91 DZVP e Gas phase and liquid phase CM3 class IV charges can be determined for the following combinations of electronic structure theory and basis set using either a restricted or an unrestricted formalism AM1 PM3 HF MIDI 6D HF 6 31G d MPWX MIDI MPWX MIDI 6D MPWX 6 31G d MPWX 6 31 G d MPWX 6 31 G d p BLYP 6 31G d B3LYP MIDI 6D B3LYP 6 31G d B3LYP 6 31 G d MPWYX is a method that uses the mPW exchange functional of Adamo and Barone Adamo C Barone V J Chem Phys 1998 108 664 the PW91 correlation functional Perdew J P Electronic Structure of Solids 91 Zieesche P Eshrig H Eds Akademie Berlin 1991 and a percentage of HF exchange X Note that MPWX includes the following special cases MPWO0 mPWPW91 MPW6 MPWIS MPW25 mPWI1PW91 MPW42 8 MPWIK MPW60 6
195. tialization will be performed Internal coordinates will be modified according to FRZBND FRZANG and FRZTOR xk Parameters below are effective only if INITCN 1 FRZBND Real number array that specifies the constraint values in A corresponding to IFZBND For example when IFZBND 1 1 2 0 3 4 5 6 and FRZBND 1 1 5 0 0 the 1 2 bond length and the difference between 3 4 and 5 6 bonds will be fixed at 1 5 and 0 0 A respectively FRZANG Real number array that specifies the constraint values in degree corresponding to IFZBND For example when IFZANG 1 1 2 3 0 4 5 6 and FRZBND 1 90 0 120 0 the 1 2 3 and 4 5 6 angles will be fixed at 90 0 and 120 0 degrees respectively FRZTOR Real number array that specifies the constraint values in degree corresponding to IFZTOR For example when IFZTOR 1 1 2 3 4 0 5 6 7 8 and FRZBND 1 90 0 120 0 the 1 2 3 4 and 5 6 7 8 torsional angles will be fixed at 90 0 and 120 0 degrees respectively Special Notes on Basis Sets MIDI basis set The MIDI basis also called MIDIX is similar to the 3 21G basis set with the following key differences e With the exception of Si Br and I the MIDI s and p orbitals have different exponents whereas the 3 21G valence s orbitals are constrained to have the same exponents as the p orbitals e The bang indicates that there is a d shell on every atom except for carbon and hydrogen atoms and a p shell on Li There are no p polarization functi
196. ts as arguments a number of possible keywords that are entered after the namelist name in a free format style that can span over multiple lines Each keyword accepts either character or numeric values A namelist is terminated with a SEND Only input between a namelist name and its corresponding END will be read from a GAMESS input file Below is a brief summary the namelists that are used by the GAMESSPLUS Soil sorption utility program SOILDENSITY Soil density in g mL By default Koc values are calculated using a soil density of 0 11 g mL in eq 2 which is a representative value of soil density measured for a variety of soil types from Eastern North America Federer C A Turcotte D E Smith C T Can J For Res 1993 23 1026 It is recommended that this value be used when calculating K oc values SCF Determines the level of electronic structure theory that will be used to calculate transfer free energies AMI Use AM1 This is the default method ABINITIO Use HF MIDI DELTAGW User supplied value for aG If the DELTAGW namelist is left blank or is not present agg will be calculated using the SM5 42 continuum solvation model and this value will be used in eq 2 to calculate Koc Alternatively a user supplied value for AG can be entered and this value will be used in eq 2 to calculate Koc Note that when entering a value for AG in the DELTAGW namelist it must be in units of kcal mol Also note that the GAMESSPLUS Soi
197. ture ETGAS Accepts an inputted value for the gas phase energy in hartrees The ETGAS keyword must be defined for IGAS 2 or 3 the calculation will stop otherwise Note that if IGAS 0 or 1 an inputted value of ETGAS will be ignored Theoretically the gas phase SCF energy used in the evaluation of the free energy of solvation is that of the gas phase equilibrium geometry Inputting an energy for a non equilibrium structure will lead to theoretically questionable results ICREAD Controls the reading of atomic charges from an external file ICREAD 0 Do not read in atomic charges default ICREAD 1 Read in atomic charges from file CM2CHG case sensitive ICSAVE Controls the saving of the evaluated CM2 CM3 or CM4 charges to an external file ICSAVE 0 Do not save the CM2 or CM3 charges default ICSAVE 1 Save the evaluated CM2 or CM3 charges in file CM2CHG case sensitive ISREAD Controls the reading of surface tension parameters from an external file ISREAD 0 Use the surface tension coefficients provided in the code default ISREAD 1 Read and use the surface tension coefficients provided in a file called CDS_Param case sensitive and use the surface tension coefficients provided in the code for any surface tension coefficient that is not provided by the user in the file CDS_Param ISREAD 2 Read and use the surface tension coefficients provided in a file called CDS_Param and set all other surface tensi
198. tz S T Elbert M S Gordon J H Jensen S Koseki N Matsunaga K A Nguyen S J Su T L Windus M Dupuis and J A Montgomery J Comput Chem 14 1347 1993 Elsevier style M Higashi A V Marenich R M Olson A C Chamberlin J Pu C P Kelly J D Thompson J D Xidos J Li T Zhu G D Hawkins Y Y Chuang P L Fast B J Lynch D A Liotard D Rinaldi J Gao C J Cramer D G Truhlar GAMESSPLUS version 2010 2 University of Minnesota Minneapolis 2010 based on the General Atomic and Molecular Electronic Structure System GAMESS as described in M W Schmidt K K Baldridge J A Boatz S T Elbert M S Gordon J H Jensen S Koseki N Matsunaga K A Nguyen S J Su T L Windus M Dupuis J A Montgomery J Comput Chem 14 1993 1347 ACS style Higashi M Marenich A V Olson R M Chamberlin A C Pu J Kelly C P Thompson J D Xidos J D Li J Zhu T Hawkins G D Chuang Y Y Fast P L Lynch B J Liotard D A Rinaldi D Gao J Cramer C J Truhlar D G GAMESSPLUS version 2010 2 University of Minnesota Minneapolis 2010 based on the General Atomic and Molecular Electronic Structure System GAMESS as described in Schmidt M W Baldridge K K Boatz J A Elbert S T Gordon M S Jensen J H Koseki S Matsunaga N Nguyen K A Su S J Windus T L Dupuis M Montgomery J A J Comput Chem 1993 14 1347 T
199. ue of ICDS with the wave function for which the corresponding CDS parameter set has been optimized If ICDS is not given then 800 is the default ICDS Method which corresponds to set of coefficients 0 All coefficients equal zero calculate bulk electrostatics only 1 SMS 42 HF MIDI 2 SMS 42 HF MIDI 6D 3 SM5 42 HEF 6 31G d 4 SM5 42 BPW91 MIDI ICMD 5 SMS 42 BPW91 MIDI 6D 6 SMS 42 B3LYP MIDI 7 SMS5 42 BPW91 6 31G d 8 SMS 42 HF 6 31 G d 9 SM5 42 HF cc pVDZ 10 SMS 42 BPW91 DZVP 11 SMS 42 AM1 12 SMS 42 PM3 303 SM5 43 HF 6 31G d 313 SMS5 43 B3LYP 6 31G d 315 SMS 43 MPWX MIDI 316 SMS 43 MPWX MIDI 6D 317 SMS5 43 MPWX 6 3 1G d 318 SMS 43 MPWX 6 31 G d 319 SM5 43 MPWX 6 31 G d p 416 SM6 DFT MIDI 6D 417 SM6 DFT 6 31G d 418 SM6 DFT 6 31 G d 419 SM6 DFT 6 31 G d p 800 SM8 801 SM8AD 43 Selects the set of CM2 CM3 or CM4 CM4M coefficients used for the evaluation of the CM2 or CM3 or CM4 CM4M charges Always required ICMD Wave function which corresponds to set of coefficients 0 All coefficients equal zero calculate Lowdin charges 1 CM2 HF MIDI default for ICMD 2 CM2 HF MIDI 6D 3 CM2 HF 6 31G d 4 CM2 BPW91 MIDI 5 CM2 BPW91 MIDI 6D 6 CM2 B3LYP MIDI 7 CM2 BPW91 6 31G d 8 CM2 HF 6 3 1 G d 9 CM2 HF cc pVDZ 10 CM2 HE DZVP 11 CM2 AM1 12 CM2 PM3 300
200. ulation which allows the solvent induced change in the solute electronic wave function to be optimized variationally The Gcps term is not a self consistent term it has no effect on the solute electronic wave function GCDS is given by Gc D gt Ak Ox 3 where Ax is the solvent accessible surface area of atom k this depends on the solute s 3 D geometry and is calculated by the Analytical Surface Area ASA algorithm as described in D A Liotard G D Hawkins G C Lynch C J Cramer and D G Truhlar J Comput Chem 1995 16 422 440 and as included in recent versions of AMSOL GAMESSPLUS HONDOPLUS OMNISOL MN GSM ZINDO MN and DGSOL and ox is the atomic surface tension of atom k The atomic surface tension ox is itself a function of the solute s 3 D geometry and a small set of solvent descriptors discussed in detail below The linear parameters in the functional forms for the atomic surface tensions are called the surface tension coefficients The functional forms themselves are the same in all SMx x 5 42 and 5 43 models Usage Input for the GAMESSPLUS solubility utility program is similar to the input for a typical GAMESSPLUS calculation with some additional modifications discussed below In order to use the utility program some knowledge of how to run a GAMESS GAMESSPLUS calculation is required A brief description of GAMESS GAMESSPLUS input that is important to run this utility program is given below For more detail
201. ure on the bulk electrostatic contributions to the free energy of solvation is accounted for using a temperature dependent dielectric constant T which was computed using the following equation T 249 21 79T 00072T7 4 where T is the temperature of the aqueous solvent This is a empirically derived equation found in the CRC Handbook of Chemistry and Physics 76 edition ed Lide D R 1995 CRC Press New York The variation of the free energy of solvation due bulk electrostatic contributions is quite small The majority of the temperature dependence of aqueous free energies of solvation must by accounted for using AGcps In SM8T the AGgps term mimics the thermodynamic equation for the temperature dependence of free energies of solvation where the thermodynamic properties the heat capacity and the entropy of solvation have been replaced parameterized atomic surface tensions Gcps T T 298 9 Agog a r 298 T In ag oe 5 k k where of and of are atomic surface tensions with identical functional forms to those of ox but the parameters are different Caution should be used in assigning any physical meaning to the atomic surface tensions shown above While the sum VAR appears to be the solute s entropy of solvation k and the sum SA ar appears to be the solute s heat capacity it must be pointed out that some of the k temperature dependence of the free energy of solvation has been accounted for in the electrostatic term Ad
202. using gas phase geometries as well as carry out geometry optimization in the liquid phase using analytical gradients The EEQM energy calculations with a site site representation of the QM MM electrostatic interaction enable one to calculate the electronic energy in the presence of an external electrostatic potential such as the electrostatic potential from a solvent or a molecular mechanics region In these calculations the external electrostatic potential distribution is described as the collection of the values of the external electrostatic potential at the locations of the QM nuclei The first and second derivatives of the EEQM energy with respect to coordinates and external electrostatic potentials can be calculated GAMESSPLUS can carry out QM MM geometry optimization with a site site representation of the QM MM electrostatic interaction The QM MM geometry optimization routine in GAMESSPLUS was originally developed by Hayashi and Ohmine ref HOOO and modified by Higashi and Truhlar refs HTO08 and HT09 The AMBER force field is used for the MM subsystem For the QM MM electrostatic interaction around the QM MM boundary advanced algorithms such as the balanced redistributed charge algorithm are available GAMESSPLUS can also perform constrained geometry optimization in Cartesian coordinates by a projection operator method The current version of GAMESSPLUS can constrain bond lengths the sums or differences of bond lengths bon
203. ution and can be found in the directory gmsplus_soil Below are the input files and output files for these three calculations Input testl inp SDATA Koc calculation for benzene using calculated values for deltagw and deltagsoil Cl C 6 0 000000 1 208141 697522 6 0 000000 1 208141 697522 6 0 000000 000000 1 395043 C 6 0 000000 1 208141 697522 Cc 6 0 000000 1 208141 lt 697 522 6 0 000000 000000 1 395043 H 1 0 000000 2 160477 1 247356 H 1 0 000000 2 160477 1 247356 H 1 0 000000 000000 2 494707 H 1 0 000000 2 160477 1 247356 H 1 0 000000 2 160477 1 247356 H 1 0 000000 000000 2 494707 SEND SSCF AM1 SEND SSOILDENSITY 0 11 SEND test2 inp SDATA Koc calculation for benzene using an experimental value for deltagw Cl C 6 0 000000 1 208141 697522 C 6 0 000000 1 208141 697522 6 0 000000 000000 1 395043 6 0 000000 1 208141 697522 C 6 0 000000 1 208141 697522 C 6 0 000000 000000 1 395043 H 1 0 000000 2 160477 1 247356 H 1 0 000000 2 160477 1 247356 H 1 0 000000 000000 2 494707 H 1 0 000000 2 160477 1247356 H 1 0 000000 2 160477 12247356 H 1 0 000000 000000 2 494707 SEND SSCF AM1 SEND SSOILDENSITY 0 11 SEND SDELTAGW 0 87 SEND test3 inp SSCF ABINITIO SEND SSOILDENSITY 0 11 SEND SDATA Koc calculation for benzene using calculated values for deltagw and deltagsoil El C 6
204. ven set of radii By default the van der Waals radii of Bondi are used when defined in cases where the atomic radius is not given in Bondi s paper Bondi A J Phys Chem 1964 68 441 a radius of 2 0 A is used The SASA is the area generated by rolling the spherical solvent molecule on the van der Waals surface of the molecule The SASA is calculated with the Analytic Surface Area ASA algorithm see Liotard D A Hawkins G D Lynch G C Cramer C J Truhlar D G J Comput Chem 1995 16 422 By default the solvent radius is set to 0 40 A see Thompson J D Cramer C J Truhlar D G J Phys Chem A 2004 108 6532 for a justification of this value for the solvent radius but the user can specify a different value for the solvent radius including zero which yields the van der Waal s surface area with the keyword SolvRd A solvent radius of 0 0 A is recommended for predicting solvation free energies with SM5 42 while the default value of 0 40 A is recommended for predicting solvation free energies with SM5 43 SM6 SM8 and SM8AD See the section entitled GAMESSPLUS Keywords for more details e Liquid phase calculations based on gas phase geometries can be performed with SM5 42 for the following restricted and unrestricted Hartree Fock DFT and adiabatic connection method wave functions i e hybrid DFT wave functions that employ spherical harmonic or Cartesian d functions HF MIDI B3LYP MIDI HF MIDI 6D BPW91
205. yield accurate gas phase geometries Third for other solutes such as transition states solutes with low barrier torsions multiple low energy conformations weakly bound complexes and in cases where one or more solvent molecules are treated explicitly more expensive levels of theory might be needed to yield accurate geometries Finally solvation energies obtained using gas phase geometries can be added conveniently to gas phase energies for separable equilibrium solvation dynamics calculations In some cases geometry optimization in the presence of solvent is important In these cases one can also apply the SM5 42 SM5 43 SM6 SM8 or SM8AD models at a solute geometry R that is not an approximation to an equilibrium gas phase geometry This type of calculation corresponds to the fixed R solvation energy which is still given by AGg R of equation 1 Evaluation of this quantity for geometries that do not correspond to an equilibrium structure is useful for dynamics calculations because the potential of mean force is given by W R V R AGs R 6 where VI is the gas phase potential energy surface which is itself given by the sum of the gas phase electronic energy and the gas phase nuclear repulsion energy If one applies the SM5 42 SM5 43 SM6 or SM8 models to a geometry optimized in solution and subtracts the gas phase energy at a geometry optimized for the gas phase one obtains the true solvation energy for the given method Furt

Download Pdf Manuals

Related Search

Related Contents

  Massive Suspension light 37760/55/10  Gear Head PB5000BLK  結腸直腸がん患者の FOLFOX6, FOLFIRI 療法における携帯型  Philips Disney Table lamp 71705/33/26    T625 TDR - Francis MERCK sur le NET  Report - Sealclub.net  Yamaha PSR-7 Musical Instrument User Manual  Sony ACC-FH50A Operating Instructions  

Copyright © All rights reserved. Failed to retrieve file