This action might not be possible to undo. Are you sure you want to continue?

Welcome to Scribd! Start your free trial and access books, documents and more.Find out more

From: http://www.gaussian.com/iops.htm

Nov. 15 2005

Index

Overlay 1 IOp(1/5) IOp(1/6) IOp(1/7) IOp(1/8) IOp(1/9) IOp(1/10) IOp(1/11) IOp(1/12) IOp(1/13) IOp(1/14) IOp(1/15) IOp(1/16) IOp(1/17) IOp(1/18) IOp(1/19) IOp(1/20) IOp(1/21) IOp(1/22) IOp(1/23) IOp(1/24) IOp(1/25) IOp(1/26) IOp(1/27) IOp(1/28) IOp(1/29) IOp(1/30) IOp(1/32) IOp(1/33) IOp(1/34) IOp(1/35) 1 1 1 1 1 2 3 3 4 4 4 4 5 5 5 5 6 6 6 7 7 7 7 7 8 8 9 9 9 9 9

IOp(1/36) IOp(1/37) IOp(1/38) IOp(1/39) IOp(1/40) IOp(1/41) IOp(1/42) IOp(1/43) IOp(1/44) IOp(1/45) IOp(1/46) IOp(1/47) IOp(1/48) IOp(1/49) IOp(1/52) IOp(1/53) IOp(1/54) IOp(1/55) IOp(1/56) IOp(1/57) IOp(1/58) IOp(1/59) IOp(1/60) IOp(1/61) IOp(1/62) IOp(1/63) IOp(1/64) IOp(1/65) IOp(1/66) IOp(1/67) IOp(1/70) IOp(1/71) IOp(1/72)

10 10 10 10 11 11 11 11 12 12 12 13 13 13 13 13 14 14 14 14 15 15 15 15 15 16 16 16 17 17 17 17 17

i

95. 96.IOp(1/73) IOp(1/74) IOp(1/75) IOp(1/76) IOp(1/77) IOp(1/78) IOp(1/79) IOp(1/80) IOp(1/81) IOp(1/82) IOp(1/83) IOp(1/84) IOp(1/85) IOp(1/86) IOp(1/87) IOp(1/88) IOp(1/89) IOp(1/90) IOp(1/91) IOp(1/92) IOp(1/94. 98) IOp(1/101. 97. 104) IOp(1/105) IOp(1/106) IOp(1/107) IOp(1/108) IOp(1/109) IOp(1/110) IOp(1/111) IOp(1/112) IOp(1/113) IOp(1/114) Overlay 2 ii 18 18 18 19 19 19 19 19 20 20 20 20 20 20 21 21 21 21 22 22 22 22 22 23 23 23 23 24 24 24 24 24 25 IOp(2/9) IOp(2/10) IOp(2/11) IOp(2/12) IOp(2/13) IOp(2/14) IOp(2/15) IOp(2/16) IOp(2/17) IOp(2/18) IOp(2/19) IOp(2/20) IOp(2/29) IOp(2/30) IOp(2/40) IOp(2/41) Overlay 3 IOp(3/5) IOp(3/6) IOp(3/7) IOp(3/8) IOp(3/9) IOp(3/10) IOp(3/11) IOp(3/12) IOp(3/13) IOp(3/14) IOp(3/15) IOp(3/16) IOp(3/17) IOp(3/18) IOp(3/19) IOp(3/20) 25 25 25 25 26 26 26 27 27 27 27 27 28 28 28 28 29 29 30 31 32 32 32 34 34 34 34 35 35 35 36 36 36 . 102. 103.

IOp(3/21) IOp(3/22) IOp(3/23) IOp(3/24) IOp(3/25) IOp(3/26) IOp(3/27) IOp(3/28) IOp(3/29) IOp(3/30) IOp(3/31) IOp(3/32) IOp(3/33) IOp(3/34) IOp(3/35) IOp(3/36) IOp(3/37) IOp(3/38) IOp(3/39) IOp(3/40) IOp(3/41) IOp(3/42) IOp(3/43) IOp(3/44) IOp(3/45) IOp(3/46) IOp(3/47) IOp(3/48) IOp(3/49) IOp(3/51) IOp(3/52) IOp(3/53) IOp(3/54) 36 37 37 37 37 37 38 38 38 38 38 38 39 39 39 39 40 40 40 40 40 41 41 42 42 42 42 43 44 44 44 44 45 IOp(3/55) IOp(3/56) IOp(3/57) IOp(3/58) IOp(3/59) IOp(3/60) IOp(3/61) IOp(3/62) IOp(3/63) IOp(3/64) IOp(3/65) IOp(3/66) IOp(3/67) IOp(3/70) IOp(3/71) IOp(3/72) IOp(3/73) IOp(3/74) IOp(3/75) IOp(3/76) IOp(3/77) IOp(3/78) IOp(3/79) IOp(3/80) IOp(3/81) IOp(3/82) IOp(3/83) IOp(3/84) IOp(3/85) IOp(3/86) IOp(3/87) IOp(3/88) IOp(3/89) 45 45 45 45 45 46 46 46 46 46 46 47 47 47 47 48 48 48 49 49 50 50 50 50 51 51 51 51 51 52 52 52 52 iii .

IOp(3/90) IOp(3/91) IOp(3/92) IOp(3/93) IOp(3/94) IOp(3/95) IOp(3/96) IOp(3/97) IOp(3/98) IOp(3/99) IOp(3/100) IOp(3/101) IOp(3/102) IOp(3/103) IOp(3/104) IOp(3/105) IOp(3/106) Overlay 4 IOp(4/5) IOp(4/6) IOp(4/7) IOp(4/8) IOp(4/9) IOp(4/10) IOp(4/11) IOp(4/13) IOp(4/14) IOp(4/15) IOp(4/16) IOp(4/17) IOp(4/18) IOp(4/19) IOp(4/20) iv 53 53 53 53 54 54 54 54 54 54 55 55 55 55 55 55 56 57 57 57 58 58 58 59 59 60 60 60 60 60 61 61 61 IOp(4/21) IOp(4/22) IOp(4/23) IOp(4/24) IOp(4/25) IOp(4/26) IOp(4/28) IOp(4/29) IOp(4/31) IOp(4/33) IOp(4/34) IOp(4/35) IOp(4/36) IOp(4/37) IOp(4/38) IOp(4/43) IOp(4/44) IOp(4/45) IOp(4/46) IOp(4/47) IOp(4/48) IOp(60-62) IOp(4/63) IOp(4/64) IOp(4/65) IOp(4/66) IOp(4/67) IOp(4/68) IOp(4/69) IOp(4/71) IOp(4/72) IOp(4/80) IOp(4/81) 61 62 62 62 62 63 63 63 63 63 63 64 64 64 64 64 64 65 65 65 65 65 66 66 66 66 66 67 67 67 67 67 68 .

IOp(4/82) IOp(4/110) Overlay 5 IOp(5/5) IOp(5/6) IOp(5/7) IOp(5/8) IOp(5/9) IOp(5/10) IOp(5/11) IOp(5/12) IOp(5/13) IOp(5/14) IOp(5/15) IOp(5/16) IOp(5/17) IOp(5/18) IOp(5/19) IOp(5/20) IOp(5/21) IOp(5/22) IOp(5/23) IOp(5/24) IOp(5/25) IOp(5/26) IOp(5/27) IOp(5/28) IOp(5/29) IOp(5/30) IOp(5/31) IOp(5/32) IOp(5/33) IOp(5/34) 68 68 69 69 70 70 71 71 72 72 72 72 73 74 74 75 76 76 77 77 77 78 78 78 78 79 79 79 79 80 80 80 80 IOp(5/35) IOp(5/36) IOp(5/37) IOp(5/38) IOp(5/39) IOp(5/40) IOp(5/41) IOp(5/42) IOp(5/43) IOp(5/45) IOp(5/47) IOp(5/48) IOp(5/49) IOp(5/50) IOp(5/51) IOp(5/52) IOp(5/53) IOp(5/55) IOp(5/56) IOp(5/57) IOp(5/55-58) IOp(5/59) IOp(5/60-62) IOp(5/63) IOp(5/64) IOp(5/65) IOp(5/70) IOp(5/71) IOp(5/73) IOp(5/74) IOp(5/75-78) IOp(5/79) IOp(5/80) 81 81 81 81 81 82 82 83 83 83 83 83 84 84 84 84 85 85 85 85 85 86 86 86 86 87 87 87 87 87 88 88 88 v .

IOp(5/81) IOp(5/82) IOp(5/83) IOp(5/84) IOp(5/85) IOp(5/86) IOp(5/87) IOp(5/88) IOp(5/89) IOp(5/90) IOp(5/91) IOp(5/92) IOp(5/93) IOp(5/94) IOp(5/95) IOp(5/96) IOp(5/97) IOp(5/98) IOp(5/99) IOp(5/100) IOp(5/101) IOp(5/102) Overlay 6 IOp(6/15) IOp(6/16) IOp(6/17) IOp(6/18) IOp(6/19) IOp(6/20) IOp(6/21) IOp(6/22) IOp(6/23) IOp(6/24) vi 88 88 89 89 89 89 90 90 90 90 90 90 91 92 92 92 93 93 93 93 94 94 95 95 96 96 96 96 96 97 97 97 97 IOp(6/25) IOp(6/26) IOp(6/27) IOp(6/28) IOp(6/29) IOp(6/30) IOp(6/31) IOp(6/32) IOp(6/35) IOp(6/36) IOp(6/37) IOp(6/38) IOp(6/39) IOp(6/40) IOp(6/41) IOp(6/42) IOp(6/43) IOp(6/44) IOp(6/45) IOp(6/46) IOp(6/47) IOp(6/48) IOp(6/49) IOp(6/51) IOp(6/52) IOp(6/53) IOp(6/54) IOp(6/55) IOp(6/56) IOp(6/57) IOp(6/58) IOp(6/59) IOp(6/60-62) 98 98 98 98 98 99 99 99 99 100 100 100 101 101 101 102 102 102 102 102 103 103 103 103 103 104 104 104 104 104 104 105 105 .

IOp(6/63) IOp(6/64) IOp(6/65) IOp(6/70) IOp(6/71) IOp(6/72) IOp(6/73) IOp(6/74) IOp(6/75) IOp(6/76) IOp(6/77) IOp(6/78) IOp(6/79) IOp(6/80) IOp(6/81) IOp(6/82) Overlay 7 IOp(7/6) IOp(7/7) IOp(7/8) IOp(7/9) IOp(7/10) IOp(7/11) IOp(7/12) IOp(7/13) IOp(7/14) IOp(7/15) IOp(7/16) IOp(7/18) IOp(7/25) IOp(7/28) IOp(7/29) IOp(7/30) 105 105 105 105 106 106 106 106 106 106 107 107 107 107 107 107 109 109 109 109 109 110 110 110 110 111 111 111 111 112 112 112 112 IOp(7/31) IOp(7/32) IOp(7/40) IOp(7/41) IOp(7/42) IOp(7/43) IOp(7/44) IOp(7/45) IOp(7/52) IOp(7/53) IOp(7/60-62) IOp(7/63) IOp(7/64) IOp(7/65) IOp(7/70) IOp(7/71) IOp(7/72) IOp(7/74) IOp(7/75) IOp(7/76) IOp(7/77) IOp(7/87) Overlay 8 IOp(8/5) IOp(8/6) IOp(8/7) IOp(8/8) IOp(8/9) IOp(8/10) IOp(8/11) IOp(8/12) IOp(8/13) IOp(8/14) 112 113 113 113 113 113 114 114 114 114 114 115 115 115 115 115 115 116 116 116 116 116 117 117 117 117 117 117 118 119 119 119 119 vii .

IOp(8/15) IOp(8/16) IOp(8/17) IOp(8/18) IOp(8/19) IOp(8/20) IOp(8/22) IOp(8/27) IOp(8/28) IOp(8/29) IOp(8/30) IOp(8/31) IOp(8/32) IOp(8/35) IOp(8/36) IOp(8/38) IOp(8/39) IOp(8/40) IOp(8/41) IOp(8/42) IOp(8/43) IOp(8/44) IOp(8/45) IOp(8/46) IOp(8/47) Overlay 9 IOp(9/5) IOp(9/6) IOp(9/7) IOp(9/8) IOp(9/9) IOp(9/10) IOp(9/11) viii 119 120 120 120 120 120 121 121 121 121 122 122 122 122 122 123 123 123 123 123 124 124 124 124 125 126 126 126 126 127 127 127 128 IOp(9/12) IOp(9/13) IOp(9/14) IOp(9/15) IOp(9/16) IOp(9/17) IOp(9/18) IOp(9/19) IOp(9/20) IOp(9/21) IOp(9/22) IOp(9/23) IOp(9/25) IOp(9/26) IOp(9/27) IOp(9/28) IOp(9/30) IOp(9/31) IOp(9/36) IOp(9/37) IOp(9/38) IOp(9/40) IOp(9/41) IOp(9/42) IOp(9/43) IOp(9/44) IOp(9/45) IOp(9/46) IOp(9/47) IOp(9/48) IOp(9/49) IOp(9/60-62) IOp(9/70) 128 128 128 129 129 129 129 130 130 130 130 130 131 131 132 132 132 132 132 132 133 133 133 134 134 134 135 135 135 135 135 136 136 .

IOp(9/71) IOp(9/72) IOp(9/73) IOp(9/74) IOp(9/75) IOp(9/81) IOp(9/82) IOp(9/83) IOp(9/84) IOp(9/85) IOp(9/86) Overlay 10 IOp(10/5) IOp(10/6) IOp(10/7) IOp(10/8) IOp(10/9) IOp(10/10) IOp(10/11) IOp(10/13) IOp(10/14) IOp(10/15) IOp(10/16) IOp(10/17) IOp(10/18) IOp(10/19) IOp(10/20) IOp(10/21) IOp(10/22) IOp(10/28) IOp(10/29) IOp(10/30) IOp(10/31) 136 136 136 136 136 137 137 137 137 137 138 139 139 139 140 140 140 140 140 140 141 141 141 141 142 142 142 142 143 143 143 143 143 IOp(10/32) IOp(10/45) IOp(10/46) IOp(10/47) IOp(10/48) IOp(10/60-62) IOp(10/63) IOp(10/72) IOp(10/73) IOp(10/74) IOp(10/75) IOp(10/76) IOp(10/77) IOp(10/78) IOp(10/79) Overlay 11 IOp(11/5) IOp(11/6) IOp(11/7) IOp(11/8) IOp(11/9) IOp(11/10) IOp(11/11) IOp(11/12) IOp(11/13) IOp(11/14) IOp(11/15) IOp(11/16) IOp(11/17) IOp(11/18) IOp(11/19) IOp(11/20) IOp(11/21) 144 144 144 144 145 145 145 145 145 146 146 146 146 146 147 148 148 148 148 148 148 149 149 149 150 150 150 150 151 151 151 151 151 ix .

IOp(11/22) IOp(11/23) IOp(11/24) IOp(11/26) IOp(11/27) IOp(11/28) IOp(11/29) IOp(11/30) IOp(11/31) IOp(11/32) IOp(11/33) IOp(11/39) IOp(11/40) IOp(11/41) IOp(11/42) IOp(11/43) IOp(11/45) IOp(11/46) IOp(11/53) IOp(11/60-62) IOp(11/63) IOp(11/70) IOp(11/71) IOp(11/75) Overlay 9999 IOp(9999/5) IOp(9999/6) IOp(9999/7) IOp(9999/8) IOp(9999/9) IOp(9999/10) IOp(9999/11) IOp(9999/12) x 152 152 152 152 152 152 153 153 153 153 154 154 154 154 154 155 155 155 155 155 155 156 156 156 157 157 157 157 158 158 158 158 159 IOp(9999/13) IOp(9999/14) IOp(9999/15) IOp(9999/16) IOp(9999/17) IOp(9999/18) IOp(9999/33) 159 159 159 159 159 159 160 .

L103.LT. L113. SCAN THE LST PATH. L112) = Min(20. 10**-7 in L117. L109) = Min(40. L109. L109. RMS EST. CONVX=4*CONVF -1 0 N ConvF = 1/600 HARTREE/BOHR OR RADIAN CONVF = 0. CONFV. 0 NSTEP = Max(20. L105. 10**-N. IOp(1/6) L102. L112: MAXIMUM STEP SIZE ALLOWED DURING OPT. L107.NVar+20) (L113. Default value for single-points: 10**-5 in L116. L117: Convergence on electric field/charges -1 0 N Default value for optimizations: 10**-7. L114) N NSTEP = N IOp(1/7) L103.NVAR+10) (L102.LT. L113. IOp(1/8) L103. DISPLACEMENT . L105.Overlay 1 IOp(1/5) L103 MODE OF OPTIMIZATION 0 1 N FIND LOCAL MINIMUM FIND A SADDLE POINT FIND A STATIONARY POINT ON THE ENERGY SURFACE WITH N NEGATIVE EIGENVALUES OF THE 2ND DERIVATIVE MATRIX L107: MODE OF SEARCH 0 1 LOCATE THE MAXIMUM IN THE LST PATH. L109. L114: MAXIMUM NUMBER OF STEPS (OR NUMBER OF STEPS FOR AN LST SCAN). L114: CONVERGENCE ON THE FIRST DERIVATIVE AND ESTIMATED DISPLACEMENT FOR THE OPTIMIZATION RMS FIRST DERIVATIVE . L112.NVAR+10) (L103. L105. L112.0003 HARTREE/BOHR OR RADIAN CONVF = N*10**-6 L116. 1 .

Do III and Allow Surface To "Relax" in Solution if no spheres Whether to evaluate densities using orbitals or density matrix. 0 1 YES. 0 1 2 00 10 20 Whether to update trust radius (DXMaxT. = 0. No.2 Bohr or Radian (L105). IV . Use only the Center of NuclearCharge Use Interlocking Spheres Order of Adam's-Bashforth-Moulton (ABM) predictor-corrector method to use in solving diff. default Yes). Default is to use density.0001 fs. 1000 2000 3000 4000 N0000 10000 20000 Apprx. N DXMAXT = 0. L117: Whether to delete points which are too close together: 0 2 No . max is 9. Which type of basin to use to partition the density isosurface. NO.1 IOp(1/9) L103: Use of Trust radius. Default is Yes for minima. Whether to scale or search the sphere when reducing the step size to the trust radius (Default Scale. Search. Number of small steps per ABM step to be used in starting ABM and when "slow down" is needed Which approximation to make. Default is 4 GradVne GradRho Don't Use Basins. 0 1 2 3 4 N0 N00 N000 surface. L121: Time step.3 Bohr or Radian (L103. in ABM. = 0. L114). II .Do-Self Polarization.01 * N L117: General control.Don't Do Self-Polarization or "Compensation" Apprx. Read or CalcFC). Default is 4. Apprx. Default is III for Tomasi (interlocking spheres) and IV for general for the grad RHO or Vne trajectories. Yes. Apprx.).0 DXMAXT = 0. eqns. I . III . N*0. scale for transition states. search for minima. Use MOs. no for TS. L107: WHETHER TO MAINTAIN SYMMETRY ALONG THE SEARCH PATH.Do Self-Polarization and Compensation. Use density. Default is 5.3 Bohr or Radian (L113. = 0. But No Compensation. Estm or UnitFC). default 0.1 BOHR OR RADIAN (L103.

IOp(1/11) L103: TEST OF CURVATURE. TEST.J=1. DON'T TEST. Read I.1 -N 0 N Yes. L114: Input of initial Hessian: All values must be in atomic units (Hartree.05 Angstroms) Yes. Estimate force constants using valence force field. L105. BOMB THE JOB IF THE SECOND SECOND DERIVATIVE MATRIX HAS THE WRONG NUMBER OF NEGATIVE EIGENVALUES. Read cartesian forces and force constants from the checkpoint file are convert to internal Read cartesian forces followed by cartesian force constants (both in format 6F12.0**-N L121: Whether to read in initial velocities: 0 1 2 3 Default (same as 1) Generate random initial velocity Read in initial cartesian velocity (Bohr/sec) Read in initial MW cartesian velocity (sqrt(amu)*Bohr/sec) IOp(1/10) L103.NVAR) (8F10. using a default criteria (0.8) from input stream.F20. Use semiempirical force constants. coordinates. Second derivative matrix calculated analytically. L112. Read ((FC(I. followed by a blank line. Read from checkpoint file in internal coordinates.0) (L103 only).0D-6 (N=20) 2.6) (L103 only). 0 1 2 3 4 5 6 7 8 9 10 Use defaults (not valid for L109). Approx 1.I). (5I3.J. only recognized by 103).I=1. (not valid for L109).FC(I.001) Step size for ABM method in Trudge for isodensity method. and radians). using a (10**-N Angstroms) criteria. Bohr. Use unit matrix (default for L105. End with a blank card. L113.000 + N*(0. How close to get to the isosurface in search.J). 3 .010 1.J). L109. Use unit matrix throughout. L117: Scaling Factor for Determining Overlaps of VDW atoms -1 0 N Turn off scaling Default is 1. 0 1 2 DEFAULT (TEST for z-matrix or cartesian TS but not for LST/QST or for minimum).

L115: Type of Hessian Update: 0 1 2 3 4 5 6 7 8 9 Default (9 for L103 minimization.0 N 0. use initial Hessian). Powell for L113 and L114). number of bad steps to allow before attempting a linear minimization (i. otherwise BFGS). 7 for L103 TS.linear only starts with the N+1st.1/N IOp(1/12) L103: OPTIMIZATION CONTROL PARAMETERS 0 1 USE DEFAULT VALUES READ IN NEW VALUES FOR ALL PARAMETERS (SEE INITBS) IOp(1/13) L103.NDtrP) 0 NN MM00 No multi-time stepping Iterate density constraints NN times per step Do gradient once every MM steps IOp(1/14) L103: Max.05 (N=2) 0. D2Corr (BFGS) D2Corr (Bofill Powell+MS for transition states). FMAXT = 0. D2Corr (Old. D2Corr (New if energy rises. 1 for minima). only in L103 and L115).L114. no quadratic step). D2Corr and L115. safeguarding positive definateness (not inL103 or L115) D2Corr (New.1 * N 4 .L109: ABORT IF DERIVATIVES TOO LARGE -1 or 0 N No force test at all. only in L103 and L115).e. Powell (not in L103).L113. BFGS (not in L103) BFGS. D2Corr (No update. L121: Multi-time step parameter (NDtrC. Allow N -. 0 N Default (0 for TS.. IOp(1/15) L103.

Do not define H-bonds Define H-bonds with no related coordinates (default) Define H-bonds and related coordinates Reduce the number of redundant internals Define all redundant internals Old definition of redundant internals. will occur. Microiterations for pure MM. Skip MM atoms in internal coordinate definitions and do microiterations the new way. in L120. Read the AddRedundant input section for each structure. Do full optimization in pruned distance matrix coords. Do full optimization in redundant internal coords with large molecular tools.IOp(1/16) L103. IF THE LIMIT IS EXCEEDED. Skip MM atoms in internal coordinate definitions and do microiterations the old way.L113. / N IOp(1/18) L103: Coordinate system.L114: MINIMUM ALLOWABLE MAGNITUDE OF THE EIGENVALUES OF THE SECOND DERIVATIVE MATRIX. THE SIZE OF THE EIGENVALUE IS REDUCED TO THE MAXIMUM. in L103.L114: MAXIMUM ALLOWABLE MAGNITUDE OF THE EIGENVALUESOF THE SECOND DERIVATIVE MATRIX.0 HARTREE / BOHR**2 OR RADIAN**2 EIGMAX = 0. SIMMILAR TO IOp(16) 0 N EIGMIN = 0. IOp(1/19) L103: SEARCH SELECTION 5 . Do full optimization in redundant internal coord. Default (2000000). AND PROCESSING CONTINUES.L113. No optimization Do optimization in cartesian coordinates. 0 N EIGMAX = 25. 0 1 10 20 30 40 50 100 1000 2000 3000 10000 20000 100000 0000000 1000000 2000000 3000000 4000000 Proceed normally Second derivatives will be computed as directed on the variable definition cards. Do optimization in Z-matrix coordinates.1 * N IOp(1/17) L103.0001 EIGMIN = 1. done in L402. Include MM atoms in internal coordinate definitions (no microiterations).

Newton-Raphson only. THESE INCLUDE FMAXT. Read in a re-ordering vector from the input. L109. RFO if not.L113. IOp(1/22) L107: Whether to reorder coordinates for maximum coincidence. L106.L114: Search Selection: 0 1 P-RFO OR RFO STEP ONLY (DEFAULT) P-RFO OR RFO STEP FOR "WRONG" HESSIAN OTHERWISE NEWTON-RAPHSON IOp(1/20) L101. STEEPEST DESCENT AND LINEAR ONLY WHEN ESSENTIAL. 0 1 2 Yes. GDIIS and linear GDIIS only. Quadratic if curvature is correct. EIGMAX AND EIGMIN. No linear search. 0 1 NORMAL MODE. Linear as usual. Quadratic if curvature is correct. L110: INPUT UNITS 0 1 2 3 ANGSTROMS DEGREES BOHRS BOHRS DEGREES RADIANS ANGSTROMS RADIANS IOp(1/21) L103. RFO and linear. Assume reactant order equals product order. L115: KIND OF SEARCH: 0 1 6 BOTH DIRECTIONS AND GENERATE SEARCH VECTOR FORWARD DIRECTION AND GENERATE S. Newton-Raphson and linear. L113. RFO without linear. VECTOR . L108. LINEAR AND STEEPEST DESCENT. First-order simultaneous optimization. DXMAXT.0 2 3 4 5 6 7 8 9 10 11 13 Default (same as 6).L114: EXPERT SWITCH. RFO if not. EXPERT MODE: CERTAIN CUTOFFS USED TO CONTROL THE OPTIMIZATION WILL BE RELAXED.

2 3 4 5 6 7 BACKWARD DIRECTION AND GENERATE S.6 FORWARD DIRECTION AND READ S.d-5.d-7. the field in /Gen/ must be cleared each time.d-7). -1 0 1 2 3 Read in values Default (same as 1). gradient 1. Normal accuracy for HF (energy and gradient both 1. IOp(1/26) Accuracy of function being optimized: -NNMM Energy 10**-(NN). VECTOR 8F10.d-4) Fine grid accuracy for DFT (Energy 1. VECTOR FORWARD DIRECTION AND READ S. VECTOR 8F10. Energy + Forces. No.e.6 BACKWARD DIRECTION AND READ S. IOp(1/25) Wether SCRF is used with numerical polarizability: 0 1 No. VECTOR 8F10.6 IOp(1/23) L112: Derivative availability.d-6) IOp(1/27) = IJKL (i. Standard grid accuracy for DFT (Energy 1. VECTOR BOTH DIRECTIONS AND GENERATE S. VECTOR 8F10. Yes. round angles within 0. 0 1 2 Default (Yes). Gradient 10**-(MM). gradient 1. Energy + Forces + Force constants IOp(1/24) Whether to round tetrahedral angles. Yes. 1000*I+100*J+10*K+L) 7 .001 degree.6 BOTH DIRECTIONS AND READ S. 0 1 2 Energy only.

either initial guess of the minimizing structure or transition structure Input 2 structures. GET CARTESIAN COORDINATES ONLY FROM THE CHECKPOINT FILE. R and P are used to guide the QST optimization of the TS. model A. Abort job if model builder generates a z-matrix with too many variables. BY Z-MATRIX BY DIRECT COORDINATE INPUT (must set IOp(29) in L202). charge and multiplicity.01*I. L= 3 TS. The first one is reactant the second one is the product. Default. N.05) Input 3 structures. same as -1. but read new values for some variables from the input . QSTRad = 0.1*(10-J)*P. K = 1-9 J = 1 J = 2 J = 3 J = 4 I = 0-9 Interpolation of initial guess of TS between R and P (TS=0. Default (same as 100).Transition state searching using QST and redundant internal coordinates L= 0. default QSTrad = 0. R and P are used to guide the QST optimization of the IOp(1/28) L103: Number of translations and rotations to remove during redundant coordinate transformations: -2 -1 0 N 0. the first one is the reactant. 7 10 000 100 200 1000 2000 8 Get all input (title. The values of the TS coord are estimated by interpolating the sturcture of R and P. Set all optimization flags to optimize. The union without QST.1*J*R + 0. Print details of the model building process. Normal (6 or 5 for linear molecules). structure) from the checkpoint file. the second one is the product. By model builder. By model builder. model B. The third one is the initial guess of the transition structure. default J=5) LST constraint in internals QST constraint in internals LST constraint in distance matrix space QST constraint in distance matrix space Control parameters for climbing phase of QST (e. Do not abort job if model builder generates a z-matrix with too many variables. IOp(1/29) L101: SPECIFICATION OF NUCLEAR CENTERS 0 1 2 3 4 5 6 stream.g. of the two redundant coordinates are taken as the redundant coords for the TS.1 L= 2 Input one structure. Read optimization flags in format 50L1 after the z-matrix. GET Z-MATRIX AND VARIABLES FROM THE CHECKPOINT FILE. Get Z-matrix from the checkpoint file.

hack to make IRCs work with Cartesian input). PUNCH. Do not retain symbolic constants. same as 10000. Rebuild the coordinate system. Generate a symbolic z-matrix using all Cartesians if none is present on the checkpoint file (a Same as one. IOp(1/33) L101: L102 L103 L106 L109 L110 L113 L114 0 1 OFF ON DEBUG PRINT IOp(1/34) L101 L102 L103: DEBUG + DUMP PRINT 0 1 OFF ON IOp(1/35) RESTART (L102-L112). (2+3) Purge all flags but keep the coordinate definition. Mark Z-matrix constants as frozen variables rather than wired-in constants. 9 . IOp(1/30) L103: ARE THE READ-WRITE FILES TO BE UPDATED? THIS OPTION IS SET FOR THE LAST CALL TO 103 IN FREQUENCY CALCULATIONS IN ORDER TO PRESERVE THE VALUES OF THE VARIABLES FOR ARCHIVING. It also suppresses error termination on large gradients.3000 4000 5000 00000 10000 20000 100000 200000 Purge flags except the frozen variables. 0 1 DON'T PUNCH. but retain the redudant internal coordinate definitions. Default. 0 1 YES NO IOp(1/32) TITLE CARD PUNCH CONTROL.

0001*N (angstroms in L106. GET GEOMETRY. 0 N -1 -N>1 Use internal default (0. SUPPRESS CHECKPOINTING. L103. In L106: differentiate Nth derivatives once. au in L111).001 Angstroms in L106. L109. 1 for others). L108. L111). IOp(1/37) D2E CLEANUP (obsolete) 0 1 NO CLEANUP. 0. Use step-size of 0. THIS IS THE LAST POINT AT WHICH ANALYTIC SECOND DERIVATIVES WILL BE DONE. IOp(1/39) Step size control for numerical differentiation. In L110 and L111: differentiate energy N times. WAVEFUNCTION. In L106: differentiate wrt electric field. FIRST POINT OF A RESTART. 10 . L111. 0. and L112 but not L102. 0 1 N>1 . Path step size in L115. In L103: Initial entry of guided optimization using N levels. free-format. L109.005 A in L109. L110. and L105). L107. Initial entry. Read stepsize (up to 2 for L106.0001*N atomic units everywhere.001 Use step-size of 0.01 Angstrom in L110. 0 1 NORMAL CHECKPOINT OF OPTIMIZATION. 0. In L106: differentiate wrt nuclear coordinates. ET. L110. IOp(1/38) Entry control option (currently only by L106.0 1 NORMAL OPTIMIZATION. L110. electric field au in L111). DELETE THE D2E FILE AND THE BUCKETS AND TRUNCATE THE READ/WRITE FILES. N0 000 100 200 Continuation of run. (L106. L109. IOp(1/36) CHECKPOINT. FROM THE CHECKPOINT FILE. In L106: differentiate wrt field and nuclear.

L116: Whether to read initial E-field: 0 1 2 Start with 0. centers will be on atoms Read-in centers and radii on cards Force Merz-Kollman radii (Default) Force CHELP (Francl) recommended radii. Default is 6. Read from checkpoint file. L117: Cutoff to be used in evaluating densities. 0 N 1. IOp(1/41) Step number of optimization from which to take geometry. L115: Number of points along the reaction path in each direction.Rad).0.0D-10 1. Initial 11 by a blank line. -1 0 N Pick up analytic second derivatives every time. Just update.Rad). . terminated Read in replacment radii for selected atoms as pairs (I. Recalculation the Hessian every N steps. execpt for CalcAll. -1 for the initial geometry IOp(1/42) L103. Read in replacement radii for selected atom types as pairs (IAn. 0 1 2 3 Dipole Quadrupole Octapole Hexadecapole L117: How to define Radii 0 1 2 10 20 30 100 200 Default is 11 Use internally stored Radii. Force CHELPG (Breneman) recommended radii.Rad) or (Symbol. terminated by a blank line.0D-N IOp(1/43) L116: Extent of Reaction Field. L114: Hessian recalculation. The default. Read from input stream.IOp(1/40) L113.

M = NM/10 IOp(1/44) IRC Type 0 1 2 3 L117: Default (same as 3). 0 1 Do not read isotopes.M au = NM/10 L121: Seed for random number generator (ISeed) -1 0 N Use system time initialize iseed (Note each run will give different results) Use default seed value (ISeed = 398465) Set random number seed to N IOp(1/45) Read isotopes in L115. Cartesian. Read Isotopes. This parameter should be chosen with the parameter Cont in mind 0 NM 10. 12 .used in Trudge only to eliminate. from the outset. 0 NM 0. points which clearly lie in another basin. Mass-weighted.1 N. Maximum distance between a nucleus and its portion of the isosurface . IOp(1/46) Order of multipoles in numerical SCRF: 0 1 2 3 Dipole Quadrupole Octapole Hexadecapole.radius of spheres to be placed around attractors to "capture" the gradient trajectories.0 au N. Internal. The final radius is then automatically optimized separately for each atom.

IOp(1/47) Number of redundant internal coordinates to allow for. 3 Restore point M from data on the rwf. No electrostatics included in the model systems Do full square for testing. Default: 1 3/2/1 dont/QST3/QST2 optimize TS structure (for QST input). 0 N Default: 50000 N. Default: 1 2/1 unimolecular/bimolecular reaction. IOp(1/53) L120: Action of each invocation of L120: 0 1 2 there. IOp(1/49) Options to IRC path relaxation (IJKL) L K J I 2/1 dont/do optimize reactant structure. Default: 1 2/1 dont/do optimize product structure. Use atomic charge type N-1 during microiterations. Default: unimolecular IOp(1/52) L101 and L120: Type of ONIOM calculation: 0/1 2 3 00 10 20 100 N000 One layer. normal calculation Two layers Three layers Default (20) Include electrostatics in model systems using MM charges. 1 20 Do IRCMax Include zero-point energy. IOp(1/48) IRCMax control. 13 Do nothing Set up point MM on rwf from initial data Set up point MM on rwf from initial data and restore point MM on chk file if ONIOM data is present . The default is MK charges.

use gradients to find the next geometry. gradient. and hessian Restore point MM from RWF but do not create a new model system. NN = MaxLev**2 + 1 (currently 17) to restore real system. hessian) of point NN of the ONIOM Next point to do is MM. energy. IOp(1/55) L103: Options for GDIIS: ICos*1000+IChkC*100+IMix*10+Method form. Calc Level High | | | Mid | | | Low S M L 1--3--6 system size 2--5--8 4--7--9* IOp(1/54) Whether to recover initial energy during IRCMax from chk file: 0 1 No. Amber. Use displacements to find the next geometry. Save necessary information (some rwf's. IOp(1/56) Set of atom type names to parse: 0 1 2 3 Accept any. grid.4 5 6 7 NN0 MM000 Integrate energy Integrate energy and gradient Integrate energy. 0 1 Default. Amber allowing any symbol. gradients. for use with parameters in input stream. IOp(1/57) Whether to produce connectivity: 14 . L115: IRC optimization. Yes. Dreiding/UFF.

Connectivity input is in terms of z-matrix entries. Useful for treating the full system as having electrons only on the QM atoms. IOp(1/60) Interpret extra integer and fp values in z-matrix as scan information. NN < 50. Yes. otherwise 3). Read modifications. Yes. IOp(1/59) Update of coordinates in L103 0 1 2 Default (1 for large opt. read from rwf file. NN NN fragments. generate connectivity. 2 for regular) New versions. read from checkpoint file.0 1 2 3 4 5 10 100 Default (4 if reading geom from chk file and connectivity is there. 0 1 2 Default (No). read from input stream Yes. IOp(1/58) IRCMax control in L115. IOp(1/61) How ONIOM should leave the rwf at the end of each geomtry: 0 1 2 Default (1). Old version. Normal: leave the rwf set up for the low-level calculation on the real system. IOp(1/62) Counterpoise control. Yes. 15 . MOMM: leave the rwf set up for the real system. No. Yes. but with NBasis and NBsUse for the high-level calc on the model system. including dummy atoms. No.

MM2 (NYI). Use the first when there are equivalent matches. Read modifications to parameter set. abort when different parameters match to the same degree. Lowest 2 digits then have no meaning. Use soft and hard-wired. then the default is to apply soft parameters with higher priority. 0 1 10 100 1000 10000 Do all (default) Non-bonded Stretching Bending Torsion Out-of-plane 16 . UFF. With soft parameters.IOp(1/63) Step in counterpoise calculation: MNN NN = 0 1-NFrag M = order of derivatives (1=Energy. Supermolecule Fragments with ghost atoms NFrag+1 . 2=Gradient. MMFF (NYI).2*NFrag -. Use only soft. AMBER. soft has priority. Quartic fitting field (NYI). MM3 (NYI). Use soft and hard-wired. corresponding to the 'classes' in FncInf.lone fragments IOp(1/64) Molecular mechanics force field selection: 0 1 2 3 4 5 6 7 000 100 200 300 0000 1000 00000 10000 20000 None. Use only hard-wired. hard-wired has priority. Do not read modifications to parameter set. IOp(1/65) Control of which terms are included in MM. If IOp(67)=3. Use the last when there are equivalent matches. Dreiding.

Do for all atoms regardless of typing. 0 1 2 3 Default: 2 if reading geom from chk file. Do for all atoms regardless of initial charge. reading from input if requested by IOp(64). over-written the values in AtChMM. Default (10) Do for atoms which were not explicitly typed. IOp(1/72) L103: Algorithm choice for microiterations. Pick up non-standard parameters from chk file. 0 1 2 00 10 20 000 100 200 Default (2.IOp(1/66) Whether to generate QEQ charges. 1==> 221) Do QEq. Copy from chk file. Default (100) Do for atoms which have charge specified or defaulted to 0. IOp(1/70) L118 Type of sampling (Nact) 0 1 2 3 4 Defalt (same as 3) Orthant sampling Microcanonical normal mode sampling Fixed normal mode energy Local mode sampling ( now only Nact = 0 or 3 OK ) IOp(1/71) Whether to print out input files for each structure along an IRC: 0 1 No. else 1. Don't do QEq. IOp(1/67) Source of MM parameters. 17 . Generate here. Yes. or to use the values already there.

L121: Lagrangian constrain method for ADMP (ICType) Half*Gamma*Tr[(P*P-P)**2] + Lambda*[Tr(P)-Ne] + Eta*Tr(P*P-P) 0 0) 1 2 3 4 5-7 8-11 Use Lambda and Eta only. (Gamma=0) Use Lambda, Eta, Gamma. Gamma = .2 Use Lambda, Eta, Gamma. Gamma = 1. Constraints for scalar Mass case: Use exact constraint Sum(ij)[Vij*(P**2-P)ij] Iterative Scheme same as 4. Different initial guesses. 7 is default for scalar mass case. Mass-weighting constraints. Documentation maybe found in DVelV1. 10 is default. Default Same as 7 if no Mass-Weighting (IOp(76) < 0) Same as 10 if Mass-Weighting (IOp(76) >

Constraints for tensorial Mass:

IOp(1/73)

L103: NInit for microiterations. L121: Initial Kinetic energy of the Nuclei (EStrtC) 0 N>0 N<0 Default (.1 Hartree) N*micro-Hartree 0.0 Hartree

IOp(1/74)

Charge scaling for charge embedding in ONIOM. IJKLMN 6th through 1st nearest neighbors of current layer scaled by I*0.2, J*0.2, etc. 0 ==> 5 (no scaling); all layers are scaled by at least as much as ones farther out. The default is 500. M L0 K00 Factor for charges one bond away from link atom Factor for charges two bonds away from link atom Factor for charges three bonds away from link atom IJ etc. The actual factors used are: 0: 1.0

1: 0.0 2: 0.2 3: 0.4 4: 0.6 5: 0.8 6-9: 1.0

IOp(1/75)

ADMP control flag (ICntrl) 0 1 2 3 00 10 20 Standard ADMP Read converged density at every step Fix the nuclear coordinates Test time reversability (MaxStp must be even) Default (20). Read stopping parameters from input. Do not read stopping parameters.

18

IOp(1/76)

+/- XXXXZYYYY = Ficticous electron mass (EMass) YYYY IOp(76)>0 IOp(76)<0 Z XXXX BoxMas=0 Default (1000) YYYY*.0001 AMU MW core functions more than valence functions. YYYY*.0001 AMU. Use uniform scaling for all basis functions (Note YYYY > 9999 makes no sense) Mass-weighting option. If IOp(76)<0, Z is meaningless. If PBC: Mass of Box Coordinates (BoxMas) = XXXX*.0001 AMU Box coordinates not propagated (default).

IOp(1/77)

Initial Kinetic energy of the density matrix (EStrtP) (For UHF, Alpha and Beta each get half this energy) and Option Number to compute initial kinetic energy. Format of Input: XXYYYY (six digits) IWType = XX N = YYYY (For UHF, Alpha and Beta each get half this energy) 0 N>0 Default (0.0 Hartree) N*micro-Hartree IWType is used to figure out how the initial velocity is is computed (in gnvelp).

If XXYYYY < 0 : Initial velocity = 0.0 Hartee (i.e., currently same as N=0 above)

IOp(1/78)

Sparse in L121 -N 0 1 Sparse here with cutoff 10**(-N), full elsewhere Use full matrices or spase based on standard settings. Use sparse fixed form

IOp(1/79)

IRCMax convergence in L115 Stopping criteria in L118 and L121.

IOp(1/80)

L106: 0/1/2 Cartesian/Normal mode/Internal coordinate differentiation. 2 is NYI. L118: .eq.1 to surpress the 5th order correction after surface hop has been made in Trajectory Surface Hopping calculations. Needs also IOp(10/80=1) Nuclear Kinetic Energy Thermostat Option. (Currently only Velocity scaling is implemented) 19

0 11XXXXX 1000000

No Thermostat. Velocity scaling, but only for the first XXXXX simulation steps. (This options is useful, Velocity scaling, all the way through the simulation.

if thermostating in only required during equilibration.

IOp(1/81)

Nuclear KE thermostat in ADMP -- temperate is checked and scaled every IOp(81) steps.

IOp(1/82)

Temperature for nuclear KE thermostat in L121.

IOp(1/83)

Whether to read in frequencies for electric and magnetic perturbations. 0 1 2 Default (No). Yes. No.

IOp(1/84)

Differentiation of frequency-dependent properties. 0 N No. Mask for which properties on file 721 will be differentiated.

IOp(1/85)

Band gap calculation in PBC ADMP: 0 1 2 Default (No). Diagonalizae Fock matrix to get band gap, evolution, etc. No.

IOp(1/86)

Printing for NMR for ONIOM. 0 1 20 Default (1). Print tensors and eigenvalues.

for backwards compatibility. as required by IOp(80).2 Print eigenvectors as well. 21 . IOp(1/87) ONIOM integration of density. IOp(1/90) To read in the velocity in cartesian coordinates Nuclear Kinetic Energy Thermostat Option. Read isotopes from rwf. 4 if geometry read from chk) Use most abundant isotopes.M. Integrate densities specified by following digits: Density to use from gridpoint 1 Density to use from gridpoint 2 etc. Integrate current densities. in Kelvin. 0 1 2 K0 L00 M000 Do not integrate. Read isotopes from input. IOp(1/89) Maximum allowed deviation from average nuclear KE during ADMP.L. The temperature and pressure are read first. Read isotopes from chk. Average energy (in microhartree) to be maintained during Simulation.etc: 0: SCF 1: MP first order 2: MP2 3: MP3 4: MP4 5: CI one-particle 6: CI 7: QCI/CC 8: Correct to second order IOp(1/88) Whether to read in atomic masses (isotopes): 0 1 2 3 4 Default (1 if geometry read from input. K.

104) Phase control in L115 and L118: N1. Coupled macro step. direct /w MM Hessian incore. Quadratic micro-iterations. Quadratic micro-iterations. 96. 97. fully direct. Quadratic micro-iterations. Coupled macro step. direct /w raw MM Hessian incore. direct /w full Hessian incore. fully direct. Also in microhartree. Default(15). N2. full diagonalization. 103. Do not use dynamic convergence criteria for the micro-iterations. 102. IOp(1/94. direct /w prepared Hessian incore. Coupled macro step. full diagonalization. 95. N4 IOp(1/105) Reaction direction 00 22 Default (Same as 10) . IOp(1/92) Maximum allowed deviation from average nuclear KE specified in IOp(81).IOp(1/91) Thermostat Option. Regular micro-iterations. 98) IOp(94): IOp(95): IOp(96): IOp(97): IOp(98): <0 0 1 2 3 4 5 10 20 30 40 50 Davidson control for quadratic micro-iterations (see MMOpt2) RFO/Davidson control for quadratic micro-iterations (see MMOpt2) Davidson control for coupled QM/MM macro step (see MMOpt2) RFO/Davidson control for coupled QM/MM macro step (see MMOpt2) Control of quadratic micro-iterations and coupled QM/MM quadratic macro step. N3. Coupled macro step. Regular non-coupled macro step. IOp(1/101. Quadratic micro-iterations.

0001 IOp(1/108) Gradient magnitude for DVV stopping criteria (Crit1) 0 N Default (N=15) N*0.0001 IOp(1/107) Error tolerance for DVV time step correction (Error) 0 N Default Error=0.0001 IOp(1/109) Force-Velocity angle for DVV stopping criteria (Crit2) 0 N Default (90 Degrees) Use N Degrees 23 .04 (N=400) v0 is set to N*0.003 (N=30) Error=N*0.10 20 0 1 2 00 10 20 000 100 200 300 0000 1000 2000 Forward direction Reverse direction Damped-Velocity Verlet (DVV) options for Dynamic Reaction Path Following Default (Same as 2) Use DVV Do not use DVV Default (Same as 10) Follow the rxn path in the forward direction Follow the rxn path in the reverse direction Default (Same as 200) Time step correction not used Time step correction used but not to recalculate current DVV step Time step correction used and current DVV step recalculated Default (Same as 1000) Use DVV stopping criteria Do NOT use DVV stopping criteria IOp(1/106) Damping constant for DVV Dynamic Rxn Path following (v0) 0 N Default v0=0.

IOp(1/113) Pressure for thermochemistry. N/1000 degrees. unless read in). Effectively disables the scaling Default (50) Scale up or down to maximum change in a variable of N/1000 IOp(1/112) Temperature for thermochemistry.IOp(1/110) Scaling of rigid fragment steps during microiterations. 0 1 2 -n Do not scale Scale with 1/NRA Scale with 1/Sqrt(NRA) Scale with 1/n (NRA = number of atoms in fragment) IOp(1/111) Step-size to use with steepest descent when L103 is having trouble: -N -1 0 N Scale up to RMS step of N/1000 if DXRMS is less. unless read in). IOp(1/114) Scale factor for harmonic frequencies for use in thermochemistry and harmonic vibration-rotation analysis. 0 N Default (standard temperature. 0 N Default (1 atomosphere. N/1000000. 0 N Default (1 unless specified by IOp in overlay 7 or read in). 24 . N/1000 atmospheres.

Overlay 2 IOp(2/9) Printing of distance and angle matrices. Print dihedral angles. using the z-matrix for connectivity info. Print dihedral angles.471 WILL BE SET TO ACOS(-1/3). 0 1 2 00 10 20 30 000 100 200 300 0000 1000 2000 Default: same as 2. Do not print dihedral angles. Do not print the angle matrix. Default: same as 20. Use cutoffs instead of the z-matrix for determining which angles to print. DO NOT TEST FOR SUCH ANGLES. 0 1 2 Default (print if 50 atoms or less) Print Don't print IOp(2/12) CROWDING ABORT CONTROL 25 . IOp(2/11) PRINTING OF Z-MATRIX AND RESULTANT COORDINATES. using z-matrix connectivity if possible. ANGLES WITHIN 0. Print the angle matrix. Default: print only for small cases Do not print the cartesian coordinates in the input orientation Do print the cartesian coordinates in the input orientation IOp(2/10) TETRAHEDRAL ANGLE FIXING 0 1 2 Default (don't test). using a distance cutoff for connectivity info.001 DEGREE OF 109. Print distance matrix. Default: same as 100. Do not print the distance matrix.

26 . this is not a full optimization). Note that symm is still called.0 1 2 3 Default (same as 1). 2 3 4 5 Bring the molecule to a symmetry orientation. symmetrize the resulting coordinates. Abort the run for zero atomic distances only Abort the run if any atoms are within 0. for PBC.. ATOMS WILL NOT TAKE THE ATOMIC NUMBERS. This is not correct for the job if they are dependant. Perform the test. and confirm that the new structure has framework group. Do not perform the test. Default (same as 2). SO THEY ARE NOT PUNCHED. YES. 0 1 2 3 10 linear independance check. IOp(2/13) PUNCH COORDINATES. test the variables for linear independance and abort the Compute nuclear forces as well as second derivatives for the test. 0 1 2 NO YES. THE ATOMIC NUMBERS AND COORDINATES ARE PUNCHED IN FORMAT (I2. the same symmetry. IN FORMAT SUITABLE FOR COORD INPUT TO Gaussian. then confirm symmetry Recover the previous symmetry operations from the rwf. Call Symm once with loose cutoffs. the molecule is not oriented. same as 0 for molecules but turns on assignment of space group ops. Leave symmetry in whatever state it is presently in.5 A. Unconditionally turn symmetry off. If internal coordinates are in use. -1 0 1 Turns on symmetry.3E20. 100 Abort the job if the number of z-matrix variables is not exactly the number of degrees of freedom (i. and will determine the However. Don't even call symm.12) IOp(2/14) Internal coordinate linear independance. NOTE.e. but then disable further use of symmetry. IN 'ATOMS' FORMAT (3E20. IOp(2/15) SYMMETRY CONTROL. with tight cutoffs. but is useful for debugging the derivative transformation routines. but do not abort the job. Do not abort the run regardless of 0 distances.12).

IOp(2/17) Tolerance for distance comparisons in symmetry determination. 0 N>0 N<0 Default (determined in the symmetry package. use same tolerance for orientation. IOp(2/19) Largest allowed point group. IOp(2/18) Tolerance for non-distance comparisons in symmetry determination. IOp(2/20) Number (1-3 for X-Z) of axis to help specify which subgroup of the type specified in IOp(19) to use. Keep going and leave symmetry on. 10**N. Keep going and leave symmetry on.6 00 10 20 100 Same as 5. but get symmetry info from the chk. 0 1 2 3 Abort the job.d-8). Default (10) Do re-orientation for PBC. Keep going. currently 1. IOp(2/16) action taken if the point group changes during an optimization. 10**N. 0 N>0 N<0 Default (determined in the symmetry package. 10**-N. as Hollerith string. Suppress re-orientation for PBC. use same tolerance for orientation.d-7). 27 . 10**-N. using the new symmetry. currently 1. Turn on symmetry operations for PBC. using the old symmetry.

format (50I2) IOp(2/40) Save (initial) structure and possible constraints in rwf 698: 0 1 2 3 Default (No).IOp(2/29) Update of coordinates from current Z-matrix. IOp(2/30) Read in vector of atom types (for debugging). Pick up structure from rwf698 on the chk file. Yes. Default (no constraint unless reading constraint from chk file). N/1000 Hartree/Bohr**2 28 . 0 1 2 3 Default (1) No. Read in structure from input stream. Yes. IOp(2/41) Force constants for Harmonic constraints. 0 1 No Yes. -2 -1 0 N Read in force constants for each cartesian coordinates No constraints. Yes. but remove z-matrix.

MacLean-Chandler (12s. 6 7 8 9 10 11 12 13 14 15 16 LANL ECP basis sets.Overlay 3 IOp(3/5) TYPE OF BASIS SET. EPR-II basis sets. rest LANL1MB. Stevens/Basch/Krauss/Jasien/Cundari ECP basis sets for H-Lu.6}Z) and augmented if IOp(7)=10.2p) on H . (VARIOUS IMPLEMENTATIONS MAY OMIT SECOND ROW ATOMS. EPR-III basis sets.5-31G.Ar CBS basis #4 -.p) on P.6-31+g(d.Be 2SDF. whether intended for use in expanding AOs (IOp(5)) or in expanding the density (IOp(82)).p) on H.Ar CBS basis #2 -.6-31+G(d. type selected by IOp(6) (=0-4 for V{D. S. UGBS basis set.6-31G. He 6-311+G(2df) on Li . Literature citations in SDDPot. Dunning/Caltech basis sets.Si 6-31+G(df. Coreless: Li. G3MP2large basis set. GENERAL--SEE ROUTINE GenBas FOR INPUT INSTRUCTIONS.52111) for second row. Literature citations in CEPPot. use daggers if any polarization CBS basis #3 -.p) on H . Cl CBS basis #5 -. 0 1 2 3 4 MINIMAL STO-2G TO STO-6G EXTENDED 4-31G. Ahlrichs TZV basis sets.6-31G (VALENCE FUNCTIONS ONLY) MINIMAL STO-NG EXTENDED LP-N1G (VALENCE BASIS FOR CORELESS HARTREE-FOCK PSEUDOPOTENTIALS) EXTENDED 6-311G (UMP2 FROZEN CORE OPTIMIZED) BASIS for first row. Type selected by IOp(6) for H-Ar. CBS basis #1 -.T.Ne 6-311+g(3d2f) on Na .Large APNO basis set CBS basis #6 -. IOp(3/6)=5 for MTsmall basis set. 17 18 19 20 21 22 23 24 25 26 Stuttgart/Dresden ECP basis sets.9p)-->(631111.Q. 5 SPLIT VALENCE N-21G (OR NN-21G) BASIS FOR FIRST OR SECOND ROW ATOMS. Ahlrichs SV basis sets. B-Ne 2MWB. IOp(6) specifies type. 29 . Type selected by IOp(6).6-311++G(2df.Core correlation basis set Dunning cc basis sets.) SEE IOp(6) FOR DETERMINATION OF THE NUMBER OF GAUSSIANS IN THE INNER SHELL. The same numbers are used for all basis sets. MIDI! basis sets. USE IOp(8) TO SELECT 5D/6D. G3large basis set.5. IOp(6) selects options.Ne 6-311++g(3d2f) on Na .

Smedley. Used for counterpoise calculations. LANL2 ECP (where available.27 28 DGauss basis sets. Dunning valence double-zeta.STO-NG-VALENCE. Chem. LANL1 ECP. 1662 (1981) and J. Chem. Phys.LP-N1G. otherwise LANL1). for density basis (generated here from 1) Same as 2. NOTE IF IOp(5)=3 AND IOp(6)=8 . When IOp(5)=9 (CEP basis) this option selects the type (H-Ar only): 0 1 30 CEP-4G. they are read along with the basis set information. THIS OPTION IS USED TO CONTROL WHERE THE BASIS IS TAKEN FROM: 0 1 2 3 4 1x READ GENERAL BASIS FROM THE INPUT STREAM. selected by IOp(6) Auto-generated. LP-31G FOR LI. If non-standard ECPs are in use.NA. 85.AL LP-41G FOR OTHER ROW1 AND TWO ATOMS. Same as 1. . See Rappe. MBS. SHC ECP on second row). for density basis (generated here from 2) Read from the alternate file and remove functions/ECPs for inactive atoms. READ THE GENERAL BASIS FROM THE RW-FILES AND MERGE WITH THE COORDINATES IN BLANK COMMON TO PRODUCE THE CURRENT BASIS. DZ. J. otherwise LANL1). When IOp(5)=6 (LANL basis and potentials) this selects the type: 0 1 2 3 0 1 2 LANL1 ECP.N-31G. where one wants to modify the basis differently during different steps. Phys. IOp(3/6) NUMBER OF GAUSSIAN FUNCTIONS N STO-NG. DEFAULT OPTIONS IOp(6)=0 IF IOp(5)=0: N=3 STO-3G IF IOp(5)=1: N=4 4-31G IF IOp(5)=2: N=3 STO-3G (VALENCE) IF IOp(5)=3: N=3 IF IOp(5)=5: N=3 WHEN IOp(5)=7 (GENERAL BASES). useful only for density basis sets. WAG basis (Dunning VDZ on first row. 3546 (1981). LANL2 ECP (where available.BE. Read the general basis from the checkpoint file. DZ. MBS. When IOp(5)=8 (Dunning bases) this option selects the type: Dunning full double-zeta. N-21G. and Goddard. This option is useful when doing general basis geometry optimizations or properties using a wavefunction on the checkpoint file. 85. CEP-31G.B.MG.

A DIFFUSE S FUNCTION ON HYDROGENS. Three sets of p-functions and one set of d-functions. Functions from extended Huckel theory. D-FUNCTIONS ON HEAVY ATOMS (2ND ROW ONLY FOR 3-21G). CBS-Q d(f). 0 1 2 3 4 5 6 7 8 9 10 20 100 200 300 400 500 600 700 1000 NONE.p polarization basis Tight d for VnZ+1 (W1 theory) A SET OF DIFFUSE SP FUNCTIONS ON HEAVY ATOMS.2 CEP-121G.p) -. Three sets of d functions and one set of f-functions. TWO SETS OF P-FUNCTIONS AND ONE SET OF D-FUNCTIONS ON HYDROGENS. 2 D-funcs. Three sets of p-functions. 2d. When IOp(5)=17 (Stuttgart/Dresden ECP bases) this option selects the type according to: 6 7 0 1 2 3 SDD SDD for Z > 18. TWO SETS OF D-FUNCTIONS AND ONE SET OF F-FUNCTIONS ON HEAVY ATOMS. ONE SET OF P-FUNCTIONS AND ONE SET OF D-FUNCTIONS ON HYDROGENS. When IOp(5)=7 (DGauss basis sets): 1 2 3 4 5 DGDZVP DZVP2 DGTZVP DGA1 (fitting basis) DGA2 (fitting basis) IOp(3/7) DIFFUSE AND POLARIZATION FUNCTIONS. Augment non-hydrogens only (cc basis sets only). along with LP-31G potential. Three sets of d functions. P-FUNCTIONS ON HYDROGENS. When IOp(5)=26 (Coreless basis) this selects the choice of basis (the same ECPs are used regardless): Default (3) Primitives which match the ECPs.2d on 2nd and later atoms. ONE SET OF D-FUNCTIONS AND ONE SET OF F-FUNCTIONS ON HEAVY ATOMS (indicates an extra tight 2df with ccp basis sets. VSTO-4G basis for 1st row.d. 31 . 2 SETS OF P-FUNCTIONS ON HYDROGENS. on heavy atoms (scaled up/down by a factor of 2 from the standard single D value). 1d on 1st row atoms. D95 and no ECP otherwise. Three sets of d functions and two sets of f-functions.d.

6D/7F 5D/7F 5D 5D 5D 5D 5D/7F IOp(3/9) Where 308 should store dipole velocity integrals. Do not split S=P AO shells. Uncontract the density basis Uncontract both basis sets. 32 . the values for H-Ar can be determined by Slater's rules: H=1. 0 SELECTION DETERMINED BY THE BASIS N-31G N-311G N-21G* STO-NG* LP-N1G* LP-N1G** GENERAL BASIS 1 2 10 20 FORCE 5D.2. Na-Ar=(0.7. D.95)/3 None. FORCE 7F. Li-Ne=0. Add ghost atoms to /B/ so that every shell is on a separate center. 0 -1 N Usual place (572). He=1. For VSTO-nG. FORCE 6D. F. shells. Split S=P AO basis shells into separate S and P shells. FORCE 10F.325*(IA-1). Read in general basis data in addition to setting up a standard basis.65*I-4. Massage the data in Common /B/ and Common /Mol/. Split S=P=D= AO shells into S=P. Store in RWF N.IOp(3/8) SELECTION OF PURE/CARTESIAN FUNCTIONS. IOp(3/10) Modification of internally stored bases (default 12000): 0 1 10 100 1000 2000 10000 20000 100000 200000 300000 DEFAULTS STO-NG STANDARD SCALE-FACTORS. Do not split AO S=P=D Uncontract the AO basis. Write over the dipole length integrals (518).

00 2SP 1. 38.53 14.40 2SP 3SP 0.79 6.43 17.99 0.15 FIRST ROW ATOMS ATOM B C N O F 1S 1.68 4.00 1.56 13.24 1.36 4.75 1.01 1. N-31G (ALSO N-31G* AND N-31G**) STANDARD SCALE-FACTORS HYDROGEN H 1S 1.50 1.83 5.00 1.65 9.01 LP-N1G SCALE=1.70 1.75 1.26 6.00 1.88 3.67 7.00 1.74 1.67 6.59 12.00 1.0 FOR LI-AR (INNER AND OUTER) STANDARD POLARIZATION EXPONENTS FOR N-31G* AND N-31G** BASES 33 .00 0.70 1.PHYS.12 1.72 1.ATOM H HE LI BE B C N O F NE NA MG AL SI P S CL A 1S 1.CHEM.64 10.66 8. 2686 (1963) OUTER SHELL HAS BEEN SELECTED ON THE BASIS INNER SHELLS ARE BEST ATOM VALUES OF NUMEROUS OPTIMIZATION STUDIES ON VARIED SMALL MOLECULES.00 3SP 0.00 2SP 1.61 11.55 2.47 16.80 1.00 2SP* 1.98 1.15 1.31 5.00 1.00 1.68 5.20 1S* 1.98 1.00 3SP* 1.90 4.03 1.99 1.95 2.50 15.48 3.02 1.04 0.33 J.05 2.10 2.69 3.69 2.25 2.90 2.00 1.98 0.98 0.00 SECOND ROW ATOMS ATOM P S CL 1S 1.

IOp(3/12) Flag for semi-empirical runs. SCRF calculation -.39 IOp(3/11) CONTROL OF TWO-ELECTRON INTEGRAL STORAGE FORMAT. RAFFENETTI '3' INTEGRAL FORMAT.1 0. translation vectors and d functions properly: 1 2 3 MNDO/AM1. starting with electric field.8 STANDARD POLARIZATION EXPONENTS FOR STO-NG* BASIS. SUITABLE FOR USE WITH OPEN SHELL RHF SCF AND THE POST-SCF USE ILSW TO DECIDE BETWEEN RAFFENETTI 1 AND 2. PROCEDURES. The magnitude is specified by IOp(3/15). IOp(3/13) Nuclear center whose Fermi contact terms are to be added to the core hamiltonian. but not yet accepted by them.ATOM H LI BE B C-NE VALUE 1. 0 1 2 3 9 REGULAR INTEGRAL FORMAT IS USED.2 0.09 0. ZINDO/S. Read coefficients of field. CAN ONLY BE USED WITH THE CLOSED SHELL SCF. ZINDO/1. RAFFENETTI '2' INTEGRAL FORMAT. 34 . IOp(3/14) Addition of electrostatic integrals to core hamiltonian. blank terminated. CNDO/2.multiply moments by fudge factor for charged species. to account for sparkles.6 0. INDO/2. MG AL-CL VALUE 0. up through 34 elements (hexadecapoles) in free format. ATOM NA. SUITABLE FOR USE WITH THE OPEN SHELL (UHF) SCF.4 0. RAFFENETTI '1' INTEGRAL FORMAT IS USED. 0 -1x -6 No.

xzz. Read components of moments off rwf 521 on chk file. quadrupole.zzzy. IOp(3/15) Magnitude of electric field. octopole.z components of electric field. yzz.yyxz. LANL1 potentials.yz electric field gradient. IOp(3/16) Pseudopotential option 0 1 2 Default.yyyy.zz.xxxy. (These correspond to dipole. and hexadecapole perturbations). N N * 0. LANL2 potentials.zzz. xxxz. followed by xx. and the electric field is stored in Gen(2-4). Read components of moments off rwf 521. and xxxx. read 12 cards with x.yyy.zzzx. Stevens/Basch/Krauss CEP potentials.xxyz.xxyy.0001.SDD for Z > 18. No. The nuclear repulsion energy is also modified appropriately. Dresden/Stuttgart potentials .xxzz. xxx. ECPs if defined with the basis set.SDF 35 . 1-34 Just component number n in the above order with magnitude given by IOp(3/15). unused READ IN FROM CARDS (SEE PINPUT FOR DETAILS) Dresden/Stuttgart potentials .yyz. based on IOp(3/5).xz.SDD combination Dresden/Stuttgart potentials .xxz. IOp(3/17) SPECIFICATION OF PSEUDOPOTENTIALS -1 0 1 2 3 4 5 6-7 8 9 10 11 Read potential in old format.yyyz. Yes.y. xxy.-5 -4 -3 -2 -1 Read components of electric field only from /Gen/ on checkpoint file.yy. USE INTERNALLY STORED 'CORELESS HARTREE-FOCK' Goddard/Smedley SECE/SHC potentials. no ECP otherwise. D95V. Yes.yyzz. zzxy field third derivatives in format (3D20.xy.xyy. read if general basis.10). Default.xyz field second derivatives.zzzz. yyyx. Read components of electric field only from /Gen/.

MHF (second set) Dresden/Stuttgart potentials .CHF). Alternative potentials for coreless basis.MHF (first set) Dresden/Stuttgart potentials .MWB (second set) Dresden/Stuttgart potentials . WITH A SUBSTITUTION POTENTIAL OF TYPE N WHEREVER SUCH A SUBSTITUTION POTENTIAL EXISTS. 0 N Default (Machine dependant. IOp(3/21) Size of buffers for integral derivative file. N integer words. PRINT DON'T PRINT IOp(3/19) SPECIFICATION OF SUBSTITUTION POTENTIAL TYPE 0 N DONT USE ANY SUBSTITUTION POTENTIALS REPLACE THE STANDARD POTENTIAL OF THIS RUN (EG. No longer used.MWB (third set) Pseudopotentials for all coreless basis. 55296 words on Cray). 36 .SHF Dresden/Stuttgart potentials . IOp(3/20) Size of buffers for integral file. 16384 integer words on VAX.MDF Dresden/Stuttgart potentials . IOp(3/18) PRINTING OF PSEUDOPOTENTIALS 0 1 2 PRINT ONLY WHEN INPUT IS FROM CARDS or if GFPrint was specified.12 13 14 15 16 17 18 19 20 Dresden/Stuttgart potentials . N integer words. 0 N Default (3200 integer words).MWB (first set) Dresden/Stuttgart potentials .

Print as GenBas input. IOp(3/23) Disable use of certain basis functions. on the RWF. Sleazy. IOp(3/25) NUMBER OF LAST TWO-ELECTRON INTEGRAL LINK. LINK NUMBER. IOp(3/26) ACCURACY OPTION. TEST. TEST. INTEGRALS ARE COMPUTED TO 10**-10 ACCURACY. Use looser cutoffs in L314. IOp(3/24) Printing of gaussian function table. Used only in L312. We are re-using integrals produced earlier in the current calculation. USE OLD very inaccurate CUTOFFS IN LINK 311. Print old-fashioned table. -2 -1 0 1 >0 Use integrals from a previous job read /IBF/ from the checkpoint file. terminated by a blank line. use the /IBF/ already WE ARE NOT USING TWO-ELECTRON INTEGRALS. Print shell coordinates. 0 1 2 10 20 DEFAULT. Read in a list of basis function numbers in Format (10I5). 0 1 NO PRE-CUTOFF.IOp(3/22) CONTROL OF THE PRE-CUTOFF IN THE TWO-ELECTRON D-INTEGRAL PROGRAM. and set their dialgonal core Hamiltonian elements to +100. 0 1 Use all basis functions. 0 1 10 100 1000 Default (don't print). DO ALL INTEGRALS AS WELL AS POSSIBLE in L314. Print in more readable format. 37 .0. DO ALL INTEGRALS AS WELL AS POSSIBLE in L311. Direct SCF. PRE-CUTOFFS DESIGNED FOR THE 6-31G* BASIS. STO-3G.

DISCARD INTEGRALS WITH MAGNITUDE LESS THAN 10**-N. No. SYMMETRY HAS BEEN TURNED OFF ELSEWHERE. IOp(3/31) USE OF SYMMETRY IN COMPUTING GRADIENT (Obsolete). IOp(3/30) CONTROL OF TWO-ELECTRON INTEGRAL SYMMETRY.L311 does nothing. Yes. use IsAlg. SP integrals in link 311. 0 N DISCARD INTEGRALS WITH MAGNITUDE LESS THAN 10**-10. 10**-N. IF THEY DON'T. IOp(3/28) Special SP code control. AND SET2E WILL ALSO TURN IT OFF HERE.IOp(3/27) HANDLING OF SMALL TWO-ELECTRON INTEGRALS. IOp(3/29) Accuracy in L302: 0 N Default (10**-12). d and higher elsewhere. HOWEVER. THE SET2E WILL INTERROGATE ILSW TO SEE IF THE SYMMETRY RW-FILES EXIST. IOp(3/32) Whether to check the eigenvalues of the overlap matrix: 0 1 2 38 Default (4). 0 1 TWO-ELECTRON INTEGRAL SYMMETRY IS TURNED OFF. TWO-ELECTRON INTEGRAL SYMMETRY IS TURNED ON. NOTE. . 0 1 2 Default. All integrals with d's -.

Quadrupole. COMMON/B/ IS DUMPED AT THE BEGINNING OF EACH INTEGRAL LINK. PRINT TWO-ELECTRON INTEGRALS IN STANDARD FORMAT. Hexadecapole. Yes. and reduce expansion space if linear dependence is found (NYI). Dipole.3 4 5 Yes. CONTROL WORDS PRINTED (AS USUAL). Compute absolute overlap over both contracted and over primitive functions. Yes. 0 1 2 3 NO DUMP. using SVD to reduce expansion space. IOp(3/33) INTEGRAL PACKAGE PRINTING. IOp(3/34) DUMP OPTION. Octopole. Default (same as 20). PRINT ONE-ELECTRON INTEGRALS. COMBINATION OF 1 AND 3. PRINT TWO-ELECTRON INTEGRALS IN DEBUG FORMAT. 0 1 3 4 5 6 NO INTEGRALS ARE PRINTED. IOp(3/35) LAST INTEGRAL DERIVATIVE LINK (No longer used in overlay 3). IOp(3/36) Maximum order of multipoles to compute in L303: -1 0 1 2 3 4 00 10 20 30 None Default (dipole). Compute absolute overlap over contracted functions. ADDITIONALLY. 0 N WHATEVER LINK STARTS WRITING THE INTEGRAL DERIVATIVE FILE SHOULD ALSO CLOSE IT. ADDITIONALLY. COMBINATION OF 1 AND 4. THE INTEGRALS ARE PRINTED (STANDARD FORMAT). 39 . Do not compute absolute overlaps. and use Schmidt orthogonalization to reduce expansion space. IS THE NUMBER OF THE LAST TWO-ELECTRON INTEGRAL DERIVATIVE PROGRAM.

0 1 2 Default (No). neglect four center integrals. same as 1. Default in 308. Leave alone. same as 1. IOp(3/39) Initialization of force and force constant rwfs. PRISM. PRISM. No. Do only overlap and not other 1e integrals. neglect three center two-electron integrals as well. neglect 1e integrals with diatomic differential overlap. Rys. 0 40 No NDDO . neglect 2e integrals with diatomic differential overlap.IOp(3/37) Whether to sort integrals in L320. neglect three center one-electron integrals. Explicit spdf code. IOp(3/40) Neglect of integrals: 0 1 2 3 10 20 30 keep all integrals. IOp(3/41) Various semi-empirical methods. 0 1 Initialize. Yes. IOp(3/38) Algorithm for 1e integrals: 0 1 2 00 10 20 Default in 302.

12). Original INDO/2 Beta and HDiag Parameters.J=1. Load all the integrals into memory.I). followed by ((HDiag(J.J=1. IOp(3/42) How to form NDDO core hamiltonian in L317: 0 1 2 Default (same as 1).I=1.3). Geometric mean in NDDO. Do ZINDO/1.I). Read parameters for atomic numbers 1-18 in the order Scale (D20. Default (unit overlap matrix). Default parameters (same as 5). or read-in parameters). Use Slater's rules scale factors. Same as 1. Do CNDO/2. followed by ((Beta(J.18) Read parameters from rwf. 1400000 Regular SCF with separate K.12). 1500000 J as usual but NDDO for K. 1200000 First-order XC. Read parameters from chk. 1000000 Do Harris functional. Use the unit matrix for the overlap.I=1. Read the integrals sequentially. geometric mean for NDDO/1 Arithmetic mean in NDDO.3). 41 .18) (Format 3D20. 1300000 Second-order XC (NYI).1 00 10 20 000 100 200 300 400 500 0000 1000 00000 10000 20000 100000 200000 300000 400000 NDDO Default use of NDDO beta parameters (arithmetic mean for indo parameters. Do ZINDO/S. GNDDO/1 parametrization. for testing. Do INDO/2. Use STO-3G scale factors. 1100000 Do Harris functional scaling atomic densities for current charge and multiplicity. Use the real overlap matrix. IOp(3/43) Solvent charge distribution to add to Hamiltonian: 0 1 2 3 4 None Read charges and DBFs from input stream in input orientation Read from RWF. Read from Chk.

Force units of Bohr for coordinates. only 1e. 0 1 2 use S**-1/2. Use expanded matrix logic for PBC exact exchange.0) integrals. 0 1 2 3 4 5 6 keep all integrals. Order of multipoles in SCRF for L303. IOp(3/46) Whether to abort the job if badbas detects an error: 0 1 2 Default (yes). just orthogonalize functions on the same center. IOp(3/44) integral rejection using L318. . Do not transform 2e integrals. Use unit matrix (for debugging). neglect four center and 3 center (ab|ac) integrals. If negative. neglect four center and three center (0. Force use of only the OS path for all calculations.T and S unchanged. the perturbation is computed separately and stored in the third and fourth matrices in the core Hamiltonian rwf.no (ab|xx) and no NDDO on 2e and V ints only -. neglect four center transformed integrals. IOp(3/45) Transformation matrix in L318. -1 Bit flags: 0 1 42 If bit 0 is set (use AllowP array) then read in a list of allowed paths. yes IOp(3/47) Flags for use in PRISM and CalDFT throughout the program.5 10 20 Read charges and DBFs from input stream in standard orientation Force units of Angstroms for coordinates. NDDO approximation -. no.

Do allocation for parallel 2e integrals but run sequentially. Forbid use of FoFCou. Turn off MP-based cutoffs in FF part of NFx. Force square loops. Reverse normal choice of diagonal/canonical sampling in Prism and PrmRaf. Make all shells large in XC quadrature. Do not symmetry reduce grid points on unique atoms. Turn on angular offsets in XC grid generation. 43 . currently only affects CalDFT. Reverse normal choice of Scat20 vs. Use Mura radial grid instead of Euler-2 grid. The default is diagonal only on vector machines. Make all atoms large in XC quadrature. Do allocation for parallel XC but run sequentially. Force use of FoFCou. Forbid use of gather/scatter digestion even for small numbers of density matrices. Skip consistency checks for XC quadrature Do not do extra work to use cutoffs better. No longer used. IOp(3/48) Options for FMM: RRLLNNTTWW RR: LL: NN: TT: WW: Range (default 2) LMax (default from tolerance) Number of levels (default 8) Tolerance (default 18) IWS (default 2). even if not doing FMM. Force all near field in FMM. Turn off vFMM. Default is to use replicated Turn off Schwartz during FMM/NFx calculations. Turn on use of precomputed XC weights. Do nuclear contribution in FoFCou even for non-PBC Do not use special Coulomb algorithm in FoFCou. Force single matrix code in CPKS. Trace input and output using Linda/subprocess. matrices only on Fujitsu and NEC. Reserved for more control of scatter/gather. currently only in PrismC. replicated Fock matrices.2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Reverse choice of whether to precompute distance matrix during numerical quadrature. Turn off use of Sqrt(P) in density-based cutoffs.

Turn off parallelism in FMM (does not use parallel logic). so L302 does not do ECP ints. Do not save as many multipole expansions as possible in memory. IOp(3/51) Parameters for NF exchange and box length (MMMMNN): 00 NN 0000xx MMMMxx no NFx NFx range NN (R+n with n=NN-1). No Cutoffs 10**-N. UseUAB. IOp(3/52) Turn off normal evaluation of ECP integrals. 44 . Set up for parallel FMM but run loops sequentially. Split primitives for better boxification. Do not default to FMM. Old routines will be used. IOp(3/53) Accuracy in ECP integral evaluation: 0 -1 N Default. Uncontract all shell pairs. if 128 set. Convert to sparse storage under FoFCou for testing.0 Bohr). Force FMM on. Box length MMMM/1000 Bohr. Set by PsmSet to indicate whether the NAtoms test for defaulting FMM was passed. Default UseUAB/Use 256. Default box length.IOp(3/49) More options for FMM: 1 2 4 8 16 32 64 128 256 512 1024 2048 4096 8192 Indicates whether FMM can be used by FoFCou. based on geometry (but minimum for molecules 3. ECP integrals are evaulated in L302. Turn on FMM print. 0 1 Default: if needed. Apply symmetry to derivative distributions (NYI).

IOp(3/57) No. 45 . crude accuracy (5x10**-5) Yes. No Yes. cutoff 5 x 10 ** (N+100) Yes. intermediate accuracy (5x10**-7) Yes.IOp(3/54) Type of core density to use with ECPs: -1 0 1 2 None Default (None) Non-relativistic Relativistic IOp(3/55) Use of sparse storage: N<-100 -3 -2 -1 0 N Yes. cutoff 10**(-N) IOp(3/56) Cutoff for intermediate matrices during sparse operations: 0 N 100 times smaller than storage cutoff. IOp(3/59) Threshold for throwing avay eigenvectors of S: 0 N Default (10**-4) 10**-N. default accuracy (10**-10). 10**(-N). IOp(3/58) Cholesky control options. of core electrons for Stuttgart/Dresden ECP's.

set to N.IOp(3/60) Control of orthogonalization and simplification of ccp basis sets. IOp(3/65) Number of k-points: -1 46 Just Gamma point. 0 1 2 Default (1). IOp(3/63) Debug option to test point charge FMM. -2 -1 0 N Set to zero Set to -1. . Lambda are basis functions on different atoms.Lambda) atom--atom cutoff criterion (angstroms). IOp(3/64) Set value for ILSW derivative flag. 0 1 No. (Defaults to 15 angstroms). IOp(3/62) Maximum allowed error in S over orthogonalized basis functions: 0 N Default (10**-9). Nu are basis functions on the same atom. Orthogonalize and remove primitives with 0 or small coefficients. 10**-(N).Nu) cutoff). Leave alone. F(Mu. Mu. IOp(3/61) Sparse Semiempirical Hamiltonian Cutoffs in L302: XX XX00 F(Mu. Yes. Orthogonalize and remove primitives with 0 coefficients. (Defaults to no F(Mu. Only active if IOp(3/39)=0.Nu) atom--atom cutoff criterion (angstroms) Mu.

Flag for macroiterations (IPCM). D-PCM. Generate COSMOTHERMO output. IOp(3/66) Over-ride setting of NThInc in lineary dependence cutoff: -1 0 N 0 Don't change. SCI-PCM. Yes. Read setting from checkpoint and modify them by reading from the input stream. IOp(3/71) IDeriv level flag (for SCRF setup). Onsager. with standard values. Old logic for NRecip=N. IOp(3/67) Electric-field dependent functions: 0 1 2 3 Default (on if already present in basis read from rwf or chk. Read setting from the input stream. with read-in values. 0 1 2 3 4 5 0100 1000 2000 2100 2200 2300 3000 4000 10000 Default (1) Use defaults. C-PCM. 47 . Read from rwf. IVC-PCM Cramer/Truhlar solvation model. Yes. Read setting from checkpoint. No.N -N About N points. Set to N. otherwise off). IOp(3/70) SCRF flag. IEF-PCM.

Becke "Half and Half": 0. BA3PBE. PBE1PBE mPW3PBE. Becke3 with Perdew 91 correlation. Default. PBE3PBE. B98. Becke 3 with Perdew 86 correlation. Vosko-Wilk-Nusair method 5 correlation.5 LSD Do only coulomb part. mPW1LYP. LG1LYP. HCTH407.IOp(3/72) Solvent type flag (for SCRF setup). Lee-Yang-Parr correlation. Becke3 using VWN/LYP for correlation. mPW1PBE. B1B95. B1LYP. skip exchange-correlation. IOp(3/74) Type of exchange and correlation potentials: -24 -23 -22 -21 -20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 00 00 01 02 03 04 05 06 48 O3LYP. same as 100. No correlation. Becke "Half and Half" with LYP/VWN correlation. VWN 80 (LSD) correlation VWN 80 (LSD) + Perdew 86 correlation . B97-1. HCTH93.5 HF + 0. B97-2. HCTH147. BA1PBE. Perdew 81 + Perdew 86 correlation. mPW91PW91. IOp(3/73) ONIOM system flag (for SCRF setup). Perdew 81 correlation.

B97-1 coefficients.0 and 1.0 for DFT and HF. same as B1B95. X-alpha exchange (alpha= 0. 49 . Hartree-Fock-Slater exchange (Alpha = 2/3). 205 is Local Spin Density. III radial points. -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 O3LYP coefficients.7) Becke 1988 exchange. PW91 exchange Gill 96 exchange PW86 exchange mPW exchange PBE exchange BA exchange VSXC exchange So 100 is Hartree-Fock. Coefficients of 0.72 c=0. HCTH coefficints. Becke 3 coefficients: aLSD + (1-a)HF + b(dBx) + VWN + c(LYP-VWN). Note that Becke actually used Perdew correlation rather than LYP. 200 is Hartree-Fock-Slater. JJJ angular points. IOp(3/75) Number of radial and angular points in numerical integration for DFT: 0 IIIJJJ Use a special grid designed for efficiency (default).5 HF + 0.5 Xc + Corr Coefficients of 0 and 0 (no exchange). respectively. with a=0. Becke "Half and Half" 0. and 402 is BLYP.81.8 b=0. LG exchange.07 08 09 10 11 18 19 20 100 200 300 400 500 600 700 800 900 1000 1100 1200 OS1 correlation PW91 PBE VSXC Bc96 VWN5+P86 LYP+VWN5 for scaling KCIS Hartree-Fock exchange. B97-2 coefficients. mPW91PW91 coefficients. IOp(3/76) Mixing of HF and DFT. B98 coefficients.

IOp(3/77) Mixing of local and non-local exchange: -1 0 MMMMMNNNNN MMMMM/10000 non-local plus NNNNN/10000 local.0 MMMMMNNNNN Default: pure HF. Sign is applied to the local term. Use Becke weights with default cutoff of 30 au. IOp(3/80) Range for microbatching in DFT. 50 . 1 to set up for CAS-MP2 or 2 to do spin-orbit calculation. Use Becke weights with cutoff N Bohr. 0 for both. In L510. Default (coefficients of 1 and zero as determined by IOp(3/42) IOp(3/78) Mixing of local and non-local correlation: -1 0 MMMMMNNNNN MMMMM/10000 non-local plus NNNNN/10000 local. 0 -1 -2 -3 -M<-3 N Default (SS weights) Use SS weights. Use Savin weights. 1000000000 turns of microbatching logic. Use SS weights with XCal = M/1000. Default (coefficients of 1 and zero as determined by IOp(3/42) IOp(3/79) Range cutoff in Becke weights. DFT or mixed in accord with IOp(3/76) Mixture of MMMM/10000 DFT exchange and NNNNN/10000 HF exchange. Negative to turn off screening of basis functions and grid points. Sign is applied to the local term. 0 for both.

N Discard functions with L>=N. where MaxTyp is the highest AO angular IOp(3/84) Equivalent of IOp(3/7) for density basis. IOp(3/85) Pure vs. 0 1 2 Default (pure for read-in basis). -1 0 N None. Use only AO primitives squared in fititing basis. See comments for IOp(3/5) and IOp(3/6) 28=Generate automatically from AO basis. for debugging. For auto-generated basis sets: -1 0 momentum. Same numbering of basis sets as for AO basis. Cartesian. IOp(3/83) Equivalent of IOp(3/6) for density basis. Do not split shells Split F and higher shells away from S=P=D. 51 . Keep all generated functions.IOp(3/81) Frequency-dependence (if any) for XC functional. including 7=General basis. Also static limit. For auto-generated basis sets: 0 1 2 10 20 Default (22) Use all products of AOs. 0 1 2 3 Default (same as 1). Keep all functions with angular momentum up to MaxTyp+1. Default (-1). Cartesian functions in density basis. but returning zero for imaginary response contributions. Gross-Kohn form. Pure. IOp(3/82) Fitting density basis set for Coulomb in DFT. None (static limit).

IOp(3/86) Discard basis functions based on angular momentum: 0 N No. Solve iteratively with symmetric block-diagonal preconditioning. in IOp(3/10).. 1 otherwise). Do not split S=P density shells. forming Z = modified A^-1.. D. solving iterative with stored A. Do not use density fits. Discard density basis functions with angular momentum >= N.. Use fits. IOp(3/87) Discard density basis functions based on angular momentum: 0 N No. no formation of A. IOp(3/89) Set up for density fitting. This is also done if requested Split S=P density basis shells into separate S and P shells.. Use fits. Iterative. Do not split density S=P=D. shells. Massage the data in Common /B/ and Common /Mol/. Solve iteratively with no preconditioning Solve iteratively with diagonal preconditioning. with A formed to generate preconditioning. . IOp(3/88) Modification of internally stored density basis. Discard basis functions with angular momentum >= N. solving iterative with direct products.. density shells into S=P. Form inverse matrix once. Form A' over neutral distributions via direct products. Read in general basis data in addition to setting up a standard basis. 0 1 10 100 1000 2000 10000 20000 None. 0 1 2 3 4 5 6 7 1xx 2xx 3xx 4xx 52 Default (102 if a fitting set has been included and pure DFT is being used. F.. Use fits. Add ghost atoms to /B/ so that every shell is on a separate center. . Form A' over neutral distributions via multiplies by A. Split S=P=D=.

Solve non-iterative using precomputed A'^-1. AOs have J-normalization. Put all functions into a single block in forming the preconditioning matrix. Douglass-Kroll-Hess 0th order. DBFs have J-normalization. default NN=06. for debugging. 53 . Douglass-Kroll-Hess 2nd order. AOs have normalization as for AOs. DBFs have normalization as for AOs. 10**(-NN) on generalized inverse. AO coefficients are for raw primitives. Same as 2 but exponents are 100x smaller. Include nuclear charge distributions in DBF set. al Very tight single s Gaussians. IOp(3/92) Whether read-in basis sets are in terms of normalized primitives? 0 1 2 3 10 20 30 Default (12). IOp(3/93) Nuclear charge distribution: 0 1 2 3 4 10x Default (1. IOp(3/90) Thresholds for density fitting MMNN 10**(-MM) on iterative solution.5xx 6xx 1xxx Solve iteratively with non-symmetric block-diagonal preconditioning. default MM=09. for debugging. DBF coefficients are for raw primitives. unless scalar relativistic) Point nuclei Single s Gaussians using formula of Quiney et. RESC. IOp(3/91) Scalar relativistic core Hamiltonian: 0 1 2 3 4 Default (1) Non-relativistic.

At least N cells total. Multiply usual range by M. IOp(3/96) Number of PBC cells for DFT: 0 N As many as look significant. IOp(3/95) Minimum number of PBC cells. No more than N matrices.Mxxx Use method M to handle nuclear charges during density fitting. At least N. At least N. IOp(3/94) Range of PBC cells in Bohr. . based on number of cells having overlap with cell 0. Based on range estimate (IOp(3/94)). -N 0 N At least N cells in each direction. IOp(3/99) Whether to set up precomputed quadrature grid in L302: 0 54 Default (2 if doing DFT. -1 otherwise). N Bohr. IOp(3/98) Maximum number of density matrices in PBC. 0 N Default. IOp(3/97) Number of PBC cells for exact exchange: 0 N As many as look significant. 0 N -M default (100).

Default is 6 (using 5 previous solutions plus the current right-hand side to generate the initial guess). IOp(3/102) Number of density fittings solutions to save from previous SCF iterations. storing only grid parameter Yes. Negative to use projected equations rather than least-squares. IOp(3/104) Maximum number of iterations during iterative density fitting. 0 1 Default (re-use if present). IOp(3/103) Maximum number of vectors allowed in expansion space during iterative density fitting. Reuse. 55 .NDBF+100). and point coordinates. IOp(3/105) Re-use of PBC cell data. storing grid parameters.1000). IOp(3/101) Maximum range of cells -N N No more than N in each direction No limit. Yes. No more than N total. storing grid parameters and weights. N. weights. IOp(3/100) Minimum Number of PBC cells for PBC-MP2 0 N Same as for HF exchange.-1 1 2 3 No Yes. Default is Max((1000. Default is Max(NDBF/2.

This number is scaled up appropriately if symmetry is in use. 0 N Default (60) N atoms for the C1 case. 56 .2 3 Do not reuse. Read from chk file. IOp(3/106) Override default number of atoms threshold for turning on FMM (for debugging). to compensate for the loss of some symmetry with FMM.

3. Don't check MOs for othornormality.2. Default MO checking (yes). on the checkpoint file. Guess from model Hamiltonian. 0 1 2 00 10 20 000 100 200 FORCE PROJECTED GUESS. in which case Read guess from the checkpoint file. IGuess values of 10-14 are generated internally and are the sparse versions of 0 and 5-8. Check MOs for othornormality.1 corresponding to 1.4 above. Default orthogonalization (perform) Schmidt orthogonalize guess orbitals. Re-use Fock matrices instead of orbitals. SUPPRESS PROJECTION. Read generalized density specified by IOp(38) from the chk file & generate natural orbitals from Huckel is used. EVEN WHEN BASES ARE IDENTICAL. Re-use orbitals not Fock matrices. Renormalize and orthogonalize the coefficients which are currently on the read-write files. FORCE PROJECTED GUESS. chosen via IOp(11). Huckel guess (only valid for NDDO-type methods). EVEN WHEN BASES ARE IDENTICAL. Read intermediate SCF results which are on the checkpoint file. 57 . 9 it. otherwise 2).2. Renormalize and orthogonalize intermediate SCF results which are on the RWF. Note that variable IGuess here has 4.5 * V. Projected ZDO guess. This uses the Harris functional unless atoms heavier than Xe are present.Overlay 4 IOp(4/5) Type of guess: 0 1 2 3 4 5 6 7 8 it. Read generalized density specified by IOp(38) from the rwf file & generate natural orbitals from Default. Suppress orthogonalization. Default (1 for PBC without alter. 100 1000 00000 10000 20000 Convert Guess=Check to Guess=Restart or to generating guess depending on what if anything is Use the simultaneous optimization recipe: S**-0. IOp(4/6) FORCED PROJECTION WHEN GUESS IS READ FROM CARDS (401).3.

IOp(4/7)

SCF CONSTRAINTS (401,402,403). -1 0 1 2 3 4 5 6 Ignore ILSW and determine on the fly. USE ILSW TO DETERMINE. REAL RHF. REAL UHF. COMPLEX RHF. COMPLEX UHF. COMPLEX, BUT USE ILSW TO DECIDE WHETHER RHF/UHF. REAL ROHF.

IOp(4/8)

ALTERATION OF CONFIGURATION (401). 0 1 2 10 100 DO NOT ALTER CONFIGURATION. READ IN PAIRS OF INTEGERS in free format INDICATING WHICH PAIRS OF MO'S ARE TO BE INTERCHANGED. PAIRS ARE READ UNTIL A BLANK CARD IS ENCOUNTERED. Read in a permutation of the orbitals. READ ALTERATION INFORMATION FROM THE READ-WRITE FILE. Use alpha orbitals for guess for both alpha and beta.

NOTE IF THE CONFIGURATION IS ALTERED ON AN OPEN SHELL SYSTEM, TWO SETS OF DATA AS DESCRIBED ABOVE WILL BE EXPECTED, FIRST FOR ALPHA, SECOND FOR BETA.

IOp(4/9)

SCF symmetry control (401). 0 1 2 3 4 5 10 20 100 200 1000 58 Default, same as 104 Read groups of irreducable representations to combine in the SCF. These are read before any orbitals and before alteration commands. Use no symmetry in the SCF. Pick up the symmetry mixing information from the Alteration read-write file. use the full abelian point group, as represented by the symmetry adapted basis functions produced by link 301. Initial guess orbital symmetries are assigned. (Use symmetry in SCF if possible, but do not assign initial guess abelian symmetries). Localize all occupied orbitals together and all virtual orbitals together Localize the orbitals within the selected or defaulted symmetry. Assign orbital symmetries for printing in full symmetry. Do not assign orbital symmetries in full symmetry. Force the guess orbitals to have the Abelian symmetry. This option can cause the symmetry

adapted basis function common blocks to be modified.

IOp(4/10)

Orbitals to mix to form complex guess (401). 0 1 Mix the HOMO with the LUMO. Read from cards (2I3) pairs of integers indicating which pairs of orbitals are to be mixed. Reading

is terminated by a blank card. NOTE THE SAME CONSIDERATIONS FOR OPEN SHELL SYSTEMS WHICH APPLIED IN IOp(8) APPLY HERE, ALSO.

IOp(4/11)

Type of Guess (401): For iterative ZDO Guess: -1 0 1 2 3 4 5 6 Force old path using old Huckel. Best available (6,4 in order of preference). Old Huckel. CNDO. INDO. New Huckel. Iterative extended Huckel. Harris, converted to IGuess=3 and IZDO=3 here.

For unprojected single diagonalization guess: 0 1 2 3 000 100 200 300 400 500 1000 n0000 MMMM00000 Use functional MMMM Default (same as 1). Use bare core matrix. Dress core Hamiltonian with QEq-based density. Use Harris Functional. Default, same as 2. Use SG1 and 10^-6 accuracy in Harris guess Use FineGrid and 10^-8 in Harris functional. Use UltraFine and 10^-8 in Harris functional. Use user's IRadAn and 10^-8 in Harris functional. Use (199,974) and 10^-12 in Harris functional. Save energy in Gen(43) for Harris functional. Force IDoV=n in HarFok.

59

IOp(4/13)

MIXING OF ORBITALS (401). 0 1 NO MIXING. LUMO = LUMO + HOMO (ALPHA) AND LUMO = LUMO - HOMO (BETA). NOTE THAT THIS WILL USUALLY DESTROY BOTH SPACIAL AND ALPHA/BETA SYMMETRY. THE MIXING IS DONE AFTER ANY ALTERATIONS.

IOp(4/14)

Reading of specific orbitals (401). 0 1 No. Yes. For alpha orbitals, read one card with the format for the orbitals, followed by zero or

more sets of IVec (I5) -- vector to replace. If IVec is -1, all NBasis vectors follow. (Vector(I),I=1,NBasis) -- vector in the specified format. Input is terminated by IVec=0. For beta orbitals, the same format as for alpha is used. Note that if alter is also specified, the replacements are read before the corresponding alterations (thus the order is alpha orbitals, alpha alterations, beta orbitals, beta alterations).

IOp(4/15)

Spin-state for initial guess (401). 0 N UHF). Use multiplicity in /Mol/. Use multiplicity N. This is useful for generating guesses for open-shell singlets or unusual

spin states involving orthogonal orbitals by treating them as high-spin in the guess (which only does

IOp(4/16)

Whether to translate basis functions of read in guess (401). 0 1 2 3 Default (same as 2). Use the basis functions as is. Translate to the current atomic coordinates. Translate to the current atomic coordinates, and determine an overall rotation to provide to the read- in orbitals.

IOp(4/17)

Number of open-shell orbitals (not electrons) in 402. 0 N 60 #open electrons. N. Number of electrons in the CAS space.

Use pseudo-diagonalization. MNDO.default full CAS. IOp(4/20) Type of model (402): (This is also tested in 401 to see whether atomic number greater than 102 are special flags). 3/4 point on unless Pulay or Camp-King. No Camp-King. AM1. Use slater determinants. 0 1 2 10 20 100 200 1000 2000 Default (no Pulay. L405: Trucation level for excitations -. use pseudodiagonalization). Just list configurations. Use determinant basis with Sz=b/2. Pulay. 61 . No Pulay (DIIS). IOp(4/21) SCF type (402). 0 1 2 3 4 5 Default (AM1). no Camp-King. MINDO/3. Camp-King. IOp(4/19) L402: Spin change in CI (default based on multiplicity). L405: Number of orbitals in the CAS space.IOp(4/18) Number of orbitals in CI in 402. CNDO. Flags for MCSCF (L405): 1 10 100 1000 Read options from input stream. No 3/4. Default is number of open shells. INDO. No pseudo-diagonalization. 3/4.

0 1 2 12 100 1 No. Restart 2nd derivatives. using specified orbitals. Yes. 0 N Default. 403). IOp(4/22) Derivatives? (402). Update. IOp(4/25) Wavefunction (402). 2nd derivatives. Don't update. /Mol/.10000 100000 Write unformatted file (NDATA) of symbolic matrix elements. RHF/UHF from IOp(5). and ILSW on the rwf (402). (For straight semiempirical calculations). (For Opt=MNDOFC).MNDO. Single determinant. General CI. Closed-shell 1/3 CI (only for MINDO3. 0 1 2 3 4 5 -N 62 Default (Same as 1).AM1). Do 1st derivatives analytically if possible. (For Opt=MNDOFC).MNDO. with N microstates read in. . ROHF (NYI). NRow in L405. NDiag in L405. eigenvalues. N. Update second force array instead of first. General CI.10 binary switches in L405.AM1). Biradical 1/2 CI (only for MINDO3. IOp(4/24) Whether to update orbitals. Write formatted file of symbolic matrix elements. Update. More flags for MCSCF: IFlag2 IOp(4/23) Number of iterations (402. 0 1 2 3 10 Default (don't update). but don't convert from Lowdin orbitals.

No Yes.IOp(4/26) Whether to mix orbitals in generated guess density: 0 -3 au. IOp(4/34) DUMP OPTION. PRINT EVERYTHING. IOp(4/33) PRINTING OF GUESS. default 10**-7). Yes. IOp(4/31) Root to solve for in CI (402) (Default is 1). mix valence orbitals and an equal number of virtuals. TURN ON ALL POSSIBLE PRINTING. mix valence occupieds with 0. Equal occupations of the lowest N virtuals and high N occupieds. IOp(4/29) NC in L405. 0 1 2 NO PRINTING. mix all equally. -2 -1 N Yes.15 IOp(4/28) SCF Convergence (10**-N. PRINT THE MO COEFFICIENTS. 63 . 0 1 NO DUMP.05 au (according to ZDO) of the HOMO & virtuals within 0.

A flag is automatically sent to L510 to tell it to compute the remaining atels directly. IOp(4/43) IDiEij = Switch for direct matel calculation. since there can be many thousands in a large CAS. Also L510 must use Lanczos. IOp(4/38) Generalized density to use for natural orbitals: N Density number N. option generates data for use in MC-SCF generation of zeroth order H note: for b=0 ie no unpaired spins forces use of Clifford Algebra Spinors 64 . This will be the only way to print configs in a direct matel calc. IOp(4/44) 1 Prepare input for Mp2 implies IOp(21)=10 Slater Det. Overlap assumed to be unity. IOp(4/37) Selection of old MNDO parameters in L402: 0 1 2 Defaults. 0 1 No. reformat ZIndo integrals and wfn into rwf. Old Si parameters. but all configurations are printed. 0 1 For normal route. Copy on disk is used. 2 As option 1. Configs printed as normal. Eij's calculated here and stored on disk. The configurations will not be listed unless see below. IOp(4/36) ZIndo reformating. with all matels calculated here and stored on disk. Old S parameters.IOp(4/35) Overlap matrix. 0 1 2 Default (copy on disk is used). This type of computation can only be done in a CAS comp. For direct route. Yes.

-n There are n pairs: 2*n orbitals and electrons of the specified CAS are to be considered to be GVB type orbitals when generating configs / matels. This occupies as much space as a full CAS in this link. normal CAS calculation. IRanWt. L510 will execute normally. This is a normal GVBCAS calculation. implicitly. Use Lewis dot structure guess density. IOp(4/48) Whether to do (sparse) Conjugate Gradient methods in 402: 0 1 2 No.instead of simple determinants IOp(4/45) Ipairs = number of GVB pairs in GVBCAS. This is the GVBCAS test mode. 0 n Default. Default (-1 if sparse is turned on) Yes. There are n pairs: 2*n extra orbitals and electrons will be added into the active space later. and sets up L510 to compute extra matels etc. and IRanGd. but is smaller subsequently. Yes. IOp(60-62) Over-ride standard values of IRadAn. -1 0 N No. L405 performs a CAS on the inner space. IOp(4/46) CI basis in CASSCF: 1 2 3 10 Hartree-Waller functions for singlets Hartree-Waller functions for triplets Slater Determinants Write SME on disk IOp(4/47) Convert to sparse storage after generating guess. Use diagonal guess density. 65 . No pairs. use the Lewis dot structure to generate a sparse guess directly. Use threshold 10**-N. Yes.

Yes. Turn on inversions/improper torsions Turn on torsions. 0 N Default (no cutoff). initial threshold 5 x 10 ** (N+100) IOp(4/66) Dielectric constant to be used in MM calculations. 2 otherwise. 10**-N. initial threshold 5x10-5. Turn on non-bonded terms. r**-2 Coulomb. 10**-N. initial threshold 10**(-N) Yes. 1==> 221) Do QEq. 0 1 66 Default (211 if UFF.IOp(4/63) Flags for which terms to include in MM energy: 0 1 2 10 100 1000 10000 100000 Default (111111) Turn on all terms. 0 N Eps = 1. Turn on all terms. Turn on bond stretches. IOp(4/65) Tighten the zero thresholds as the SCF calculation proceeds. Eps = N / 1000. 0 1 N N N<-100 Default: Yes. No variable thresholds. IOp(4/67) Whether to use QEq to assign MM charges. IOp(4/64) Cutoff for MM non-bonded term.0. Turn on angle bending. r**-1 Coulomb. .

Harris guess. Default (20) Do for atoms which were not explicitly typed. Default (10 for PBC. 0 IJKLMNOP List of up to 8 irrep numbers to include. Do not do the extra guess. Do the extra guess and store as the initial Fock matrix. 20 otherwise). Default (200) Do for atoms which have charge specified or defaulted to 0. Save the Harris guess as an initial Fock matrix. All IOp(4/80) The maximum conjugate gradient step size (MMNN) 67 . IOp(4/68) Convergence criterion for microiterations in L402: 0 N Default. Do for all atoms regardless of initial charge. Just generate orbitals from the Harris guess. IOp(4/72) Irreps to keep in MCSCF CI-wavefunction. IOp(4/69) Whether to do a new additional guess in addition to reading orbitals from the rwf: 0 1 2 3 00 10 20 Default: yes if no Guess=Alter. 10**(-N). Do for all atoms regardless of typing. and not a small geometry step. Do the extra guess regardless.2 00 10 20 000 100 200 Don't do QEq. IOp(4/71) Write out AM1 integrals in 402 0/1 No/Yes.

Maximum Number of purification cycles per CG iteration. (defaults to 4 CG cycles). Nu are basis functions on the same atom. (MM=00 defaults to 20 cycles. IOp(4/110) Scaling of rigid fragment steps during microiterations. Lambda are basis functions on different atoms. No preconditioning.Lambda) atom--atom cutoff criterion (angstroms) Mu. (defaults to 15 angstroms). Don't use CG DIIS Use CG DIIS. 68 .NN IOp(4/81) Sparse SCF Parameters MM NN00 PP0000 Maximum number of SCF DIIS cycles. (defaults to no F(Mu. MM=01 turns DIIS off) F(Mu. (defaults to 3 cycles).0000 MMNN No maximum step size Step size of MM.Nu) cutoff). IOp(4/82) Conjugate-Gradient Parameters MM NN00 00000 10000 000000 100000 0000000 1000000 Maximum Number of CG cycles per SCF iteration.Nu) atom--atom cutoff criterion (angstroms) Mu. F(Mu. Polak-Ribiere CG minimization Fletcher-Reeves CG minimization Use diagonal preconditioning in Conjugate-Gradient.

Compute 2e integrals.(How to Obtain the Integrals) 0 1 2 3 4 5 6 NNNNNx Incore or Direct(FoFDir) according to available Memory. HGP df (no sp done here at all). L510: Direct MCSCF control (L510). Compute 2e integrals and store in-core. Sleazy (10**-6) Cutoffs. Read the integrals off disk OR Incore (Acording to Memory) Compute 2e integrals(Using FoFDir). HGP d. Compute SCF Fock matrices. Force Rys only. Rys df (for debugging). Form full Fock matrix every time. HGP spd. Use option NNNNN in control of 2e integral calculation. BraKet only. rest not done here. Cutoffs for high accuracy. Force FoFDir. Do not compute operator matrices. NNNNN=ICntrl with values as below. Compute 2e integrals (Use TrnDir + FoFDir = 4 Can. The default for integrals or second derivatives. Form delta-F each iteration -. Force HGP only. Rys f. 0 1 2 3 4 NNNNNx 0000000 1000000 2000000 Default (same as 1).Overlay 5 IOp(5/5) Direct SCF control (L502.: ICntrl = Algorithm control: 0 1 2 3 4 5 6 7 8 10 20 30 40 100 200 Default for MCSCF is (1522). Default -. HGP sp. Compute 2e integrals and forbid in-core.only in L502.incremental Fock matrix formation only for direct SCF. forbidding incore Force conventional Something obsolete use option NNNNN in control of 2e integral calculation. Rys f (no sp done here at all). No cutoffs. Cutoffs for 10**-10 accuracy. 69 . L508). BraKet up to L=8. Calcs). Read the integrals off disk. The default for first derivatives.

300 400 500 600 700

Compute CIS operators Compute WTilda terms. GVB: DA==>FJ,FK. Compute regular integrals and load into R03, in canonical form if IOpcl=0 and square form if Compute raffenetti integrals: IOpCl=0 IOpCl=1 IOpCl=2 IOpCl=3 IOpCl=4 1000 2000 3000 4000 Load R1. Load R1 and R2. Load R1 and R03. Load R1, R2, and R03. Load R2 and R03. Do not compute forces. Compute forces. Make derivative Fock matrices Make derivative Fock matrices and form contributions to polarizability derivatives (ie 6 sets of forces will be returned in FXYZ, and 3 extra sets of densities must be supplied in PA,B). 5000 10000 Compute forces using including CIS 2PDM terms. Compute second derivatives.

IOpcl=1. NMatS is used as the dimension of R0 if IOpcl=1.

0000000 1000000 2000000

Default -- incremental Fock matrix formation only for direct SCF. Form full Fock matrix every time. Form delta-F each iteration -- only in L502.

IOp(5/6)

Convergence (RMS density except in L506 (SQCDF), L508(rms rotation gradient), and L510 (Energy)). 0 N 10**-8, except 10**-7 for PBC. 10**-N.

L510: CONVERGENCE CRITERIUM (ACC) FOR THE ENERGY IN THE MCSCF 0 N Acc = 10**(-8) Acc = 10**(-N)

IOp(5/7)

Maximum number of iterations. 0 128, except 512 in L503 and L508.

L510: MAXIMUM NUMBER OF ITERATIONS TO BE DONE (MaxIt) 0 N -1 MaxIt=64 (Default Value) MaxIt=N It does only a CI calculation. Options other than the standard SCF ones:

70

IOp(5/8)

SELECTION OF THE PROCEDURE OF DIRECT MINIMIZATION (L503). 0 1 2 4 5 STEEPEST DESCENT WITH SEARCH PARAMETERS DEFAULT STEEPEST DESCENT WITH SEARCH PARAMETERS READ (SEE BELOW) CLASSICAL SCF (ROOTHAAN'S METHOD OF REPEATED DIAGONALIZATION CONJUGATE GRADIENTS WITH SEARCH PARAMETERS DEFAULT CONJUGATE GRADIENTS WITH SEARCH PARAMETERS READ: MAX. NUMBER OF SEARCH POINTS (I1) MIN. NUMBER OF SEARCH POINTS (I1) INITIAL STEPSIZE, TAU (G18.5) SCALING FACTOR FOR SUBSEQ. TAU (G20.5) Q (G20.5) Search method (L508). 0 1 2 3 00 10 20 000 100 200 Default (123). Steepest descent. Scaled steepest descent. Quadratic convergence (after rotation gradient is sufficiently small). Default linear search (full search). Do a full linear search to locate a minimum. Do a linear search only if the energy goes up after the initial step. Default handling of wrong curvature (switch direction). Reverse direction if curvature in NR step direction is wrong. Take pure NR steps, even if curvature is wrong.

Flags for L510: 1 10 100 1000 10000 1000 IRdF2, read damping coefficients. IFrzCI, freeze CI coefficients after 1st iteration. Read unformatted symbolic matrix elements from NDATA instead of rwf. Read in damping factors from cards. Use Levy damping. Read Fock matrix restriction matrix.

IOp(5/9)

SWITCH TO CLASSICAL SCF AFTER DENSITY MATRIX HAS ACHIEVED A CERTAIN CONVERGENCY (L503 only). 0 1 2 NO YES, CRITERION DEFAULT 10(**-3) YES, CRITERION READ IN (FORMAT G16.10)

Number of pair iterations (L504, L506). -1 None; coefficients are frozen at initial values (L504: causes coefficients to be read in in order 71

11 12 22). 0 L510: 5.

IOp(5/10)

IVShft Level shifting: -N -2 -1 0 N Dynamic level shifting to achieve a gap of -0.001*N Dynamic level shift to a default goal (same as -200) No level shifting. Default: -200 for diagonalization calculations, -1 for sparse diagonalization replacements, and if energy DIIS is turned on. Shift by 0.001*N

IOp(5/11)

3 and 4 Point Density extrapolation control (L501,L502, L503 has only 4 point, L505). 0 1 BOTH 3-POINT & 4-POINT EXTRAPOLATION PERFORMED WHEN APPLICABLE. THREE-POINT EXTRAPOLATION IS INHIBITED, BUT THE PROGRAM WILL STILL PERFORM FOUR-POINT EXTRAPOLATION WHEN POSSIBLE. (IE. DISABLED).

IOp(5/12)

Whether to allocate only two N**2 arrays for RHF: 0 1 2 Default (No). Yes. No.

Number of GVB pairs (L506). If non zero, the number of orbitals in each pair is read in format (30I2). Each pair consists of the highest available occupied from the guess (after high spin orbs are accounted for) and the lowest available virtuals. If <0, pair coefficients are read; otherwise standard initial values are used.

IOp(5/13)

Action on convergence failure (L502): 0 1 2 72 Default (2). Continue the run even on non-convergence. The ILSW flag for convergence failure is set. Terminate on non-convergence.

ADMP. State average density matrices. Just recompute band structure from stored real-space Fock matrix. (Obsolete. 200 1000 2000 10000 ADMP. Terminate after computing the 2e terms at the first iteration. Turn the current RHF run into a uhf run at the end of this link. Skip core-valence Fock matrix elements. first cycle: use initial AO densities. Fit the converged density even if fitting is not in use during the SCF. REORDERING OF THE ORBITALS (MAINTAINING CONTINUITY OF THE WAVEFUNCTION ALONG THE SEARCH PATH. use IOp(17)). CALCULATION IS BYPASSED.L510: MCSCF flags: 2 10 100 XX000 Generate MOs using UHF natural orbitals. Skip valence-virtual Fock matrix elements. later cycles: transform the density from L121 before calculating the energy and Fock L510: Flags for MCSCF: 1 10 100 1000 10000 100000 Skip valence-valence Fock matrix elements. 73 .(Calculates directly Nact*(Nact+1)/2 Fock matrices by contracting the AO integrals with the Density matrices. Skip core-valence Fock matrix elements. 0 1 2 ON BESSEL CRITERION ON STRONGER INDIVIDUAL-OVERLAP CRITERION OFF None.turned on automatically in FoFDirINFC Number of Frozen Core Orbitals IRdNT. L503). 0 1 2 CALCULATION IS PERFORMED (PROVIDED OF COURSE THAT ENOUGH SPACE EXISTS IN THE RW-FILES). CONTROL OF ANNIHILATION OF SPIN CONTAMINANTS (L502). CALCULATION IS PERFORMED. Use Generalized energy-weighted density routines regardless. Also redoes the fit at the end even if using fits during SCF. Use full diagonalization method rather than Lanczos. IRdNLp. Do not use GEW routines even for CP.-For testing purposes. (More input from cards See below) IOp(5/14) Special functions in L502: 0 1 10 20 100 matrices. CONTINGENT ON SPACE. AND THE SYSTEM RW-FILES FOR THE APPROPRIATE DENSITY MATRICES ARE UPDATED (USEFUL IF ONE WANTS A POPULATION ANALYSIS). Number of rows in an initial transformation of MO.

Pseudo-diagonalization whenever possible. 0 1 NO YES Selection of OCBSE vectors (L506). Use Abelian symmetry in diagonalization. Do not use Abelian symmetry in diagonalization. keep occupation of each irrep the same as the initial guess. DiagD. By energy least change. PDM. 0 1 DONE TAU IS KEPT FIXED IOp(5/16) Diagonalization method (L502): 0 -N 1 2 3 4 5 6 7 8 1xx 2xx Default (1 for full matrices. Lanczos starting vector in L510: -1 74 Read in eigenvector. 2 for L501 and L506). No. KyDiag. SNRDMS. CEM. 4 for sparse). Force formation of the Fock matrix using full storage. Yes. keep overall wavefunction the same as the initial guess. Yes.(ILzVec=-1) See below . CGDMS. but doing the minimal amount of orbital switching to accomplish this.IOp(5/15) Apply Abelian symmetry constraints on orbitals. Sign Matrix Method. Default (use Abelian symmetry in diagonalization). 0 1 2 By eigenvalue. 0 1 2 3 00 10 20 Default (1 for L502. By orbital least change. CONTROLS THE AUTOADJUSTMENT OF TAU (L503). INHIBIT PERFORMANCE OF MINIMIZATION OF ALTERNATE WAVEFUNCTION PROVIDED BY SECOND ORDER PROCEDURES (L503). Force formation of the Fock matrix using sparse storage. Pseudo-diagonalization with real diagonalization every Nth cycle.

D-5 FUZZY AND DEGEN READ IN (2D20. Warning!! should be used only for large jobs where Hessian does not fit in memory.O. 0 1 2 5 10 100 200 300 400 500 Orthogonalize C. in preparation for CPMCSCF. VECTORS 0 1 FUZZY=1. Use of symmetry (in L502 and L508) and linear equation convergence (in L508): 0 1 2 3 10 20 30 40 100 200 0000 1000 2000 Default (1032 for 502. 1012 for 508). symmetric.14) Selection of virtual orbitals (L506). then Schmidt. Same as 30 in 502 but 20 in 508. 2nd order iteration at end.D-10. Choose LinEq convergence based on orbital gradient.0 N C(1) = 1. Just schmidt. Don't orthogonalize. Prepare for CPMCSCF(FREQ): Direct method with no storage of Hessian.LE. 2nd order iteration using RFO type step + level shift and prepare for non-direct CPMCSCF 75 . Don't use natural orbitals each iteration. If 2E symmetry is on. Force the density matrix to have full symmetry at the first iteration. Bad for 1st order method.V by separate Lowdin.LE.0 IOp(5/17) CONDITION OFF-DIAGONAL TERMS OF THE Fock MATRIX (L503).O. 2nd order iteration using RFO type step + level shift. be symmetric. 0 1 Virtuals obtained by diagonalization of hamiltonians.FUZZY. Virtuals obtained by Schmidt orthogonalization to occupieds. If 2E symmetry is on. Use full 2nd order convergence. symmetrize Fock matrices and require proper density matrix symmetry. Tighten convergence by an extra factor of 10. replicate integrals so that density matrices and wavefunctions need not If 2E symmetry is on. SET TO ZERO IF Abs(F(I. then Schmidt. DEGEN) M. Force the density matrix to have full symmetry at every iteration. Lowdin orthogonalize C+O and V. symmetrize them to ensure that they are exactly Do not symmetrize the density matrices. choose between replicating integrals and symmetrizing the Fock matrix based on whether the current density matrix is symmetric. Always use tight convergence.J)). DELETE COUPLING TERMS BETWEEN ALMOST DEGENERATE (DELTA E . DEGEN=2.0 C(N) = 1. L510: MCSCF flags. Default (1000) If the density matrices pass the symmetry test.

Forbid in-core: force re-reading of integrals even if they fit in 2 buffers if conventional. Read CI vector and use it every iteration. FOCK-MATRIX ELEMENT Abs(F(I.S.J)). Force Square in-memory storage.(IRdCIV) Use full diagonalization method rather than Lanczos. 30000000 Do SA and prepare for Gradient of Energy difference. Turn off damping. 2 if possible. Force allocation for 1 or 2 buffer case conventional case (VV.10000 100000 1000000 Attempt to control root flipping in CI.O. otherwise 1. do not convert to in-core if direct and enough memory for in-core is available. 40000000 Do SA and prepare for SA Second Derivative Computation (terms involving 2nd order orbital rotation derivatives not included) IOp(5/18) L502: Mixing when doing damping: -3 -2 -1 0 N MO damping at all iterations. 10**-N. (100-N)/100 old density.14) Damping (L506) Maximum rotation gradient for Newton-Raphson in L508 (above this value. IOp(5/19) Over-ride integral storage control (L501. L508): -1 0 1 2 3 4 1x 2x 3x 76 Choose the best given amount of memory available. Force Lower-triangular in-memory storage.SPAN ARE CONSIDERED TO BE ZERO 0 1 STHRS=1.d-2).GE. Default (-1 unless reoptimizing during Stable=Opt).8) 20000000 Do SA and prepare for SA-CPMCSCF. Force computation of raff 1 and 2 integrals even for RHF. Dynamic selection of density damping based on band gap and DIIS error. Save generated integrals on disk (file 610).D-7 STHRS AND SPAN READ IN (2D20.D-4.STHRS ELEMENTS Abs(F(I.LE. L506. Do not save integrals (same as 0x). L502.(the weights 8F10. L503: CUTOFF CRITERIA IN SYMMETRY DETERMINATION OF M. 10000000 Use State Average density matrices. SYMMETRY IS DETERMINED IF LARGEST OFF-DIAGONAL M. . scaled steepest descent is used): 0 N Default (1.ne. N/100 new density.J)).IBuf2E). SPAN=5.O.

PRINT F(1).001) Damp if error > 10**-N For MO damping: 0 N Default. no more than 1/3 virtual component for any occupied at each iteration. 0 1 ABORT RUN VIA LNK1E. just quit when extrapolated convergence is reached. IOp(5/21) DIIS error for density damping. 0 1 NO YES IOp(5/20) Final non-DIIS iteration (L501. IRdNLp. ACTION IF OTEST DETECTS PROBLEMS (L503). Maximum N/1000 virtual component. do a final unextrapolated diagonalization after convergence is reached. L504). CONTINUE RUN. L502. L502. No. MCSCF flags: 2 10 Generate MOs using UHF natural orbitals. (READ ONE CARD WITH START.END 2I2) (L503). 0 1 2 Default (no). Extrapolation control in L506.T. 77 . Yes. maximum virtual mixing for MO damping: For density damping: 0 N Default (Damp if error > 0. 0 Default (1042) for calculations using diagonalization (2) for calculations using sparse diagonalization replacements. N Rotations are turned on when SQCDF is below 10**(-N). IOp(5/22) Use of DIIS extrapolation (L501. L504). Orbital rotation control (L506).

Apply rotations sequentially. used (DIIS error/10^-N) for Use print level M in DIIS. Yes. Read from chk. commutator. Yes. deciding on the fly between the two forms. Freeze all closed. otherwise. Yes. mixture of energy and Use energy DIIS when commutator gives huge coefficients. IOp(5/24) Orbital freezing (L506). Switch from energy to commutator when error is 10^(-N) in method 3. 0 1 Default (exponentiate rotation angles).1 2 3 4 5 10 20 30 40 1xx Nxxx Mxxxx No. 0 1 Optimize all orbitals. L509). IOp(5/25) Rotation application (L506). with "FON" version of Fermi broadening. Regular DIIS Energy-based mixing Energy DIIS when DIIS error has increased significantly or is above threshold Energy DIIS when DIIS error has increased significantly. Yes. with "pFON" version of Fermi broadening. 3 78 3rd root of CAS(2. 0 1 2 Read from input stream. so that pair and hamiltonian information can be reused (L506. Optimize only 2nd and higher naturals. IOp(5/23) Flag for later points of an optimization. weight of energy DIIS in method 4. Read from rwf. IOp(5/26) Type of calculation (L504). Orbital mixing control in L506. high spin and first natural orbitals. with Fermi broadening as well.2) .

L505.2 1 0 -1 -2 Excited singlet as 2nd root of CAS(2. If zero. Force closed shell. Default: let FoFDir decide. 79 . Force UHF. 0 1 2 Default (Yes. GVB as CAS(2. IOp(5/27) Whether to do closed-shell calculation in L502. IOp(5/28) L510: Root of CI to use in MCSCF (IState) 0 -1 N Defaults to Istate=1 Read IState from cards (see below) IState = N IOp(5/29) Use of rafinetti integrals during direct SCF. Closed/Open control for L511: 0 1 2 Default.2) GVB(1/2) Orthogonal open-shell singlet. error if Multip>1. Yes. IOp(5/30) Whether to symmetrize final orbitals using abelian symmetry operations (L502. All integrals are done as regular integrals. ROHF Triplet (a debugging option). It will never use Raffenetti for SCF. If -1. -1 0 1 N All integrals done as Raffenetti. No Yes (used for RHF direct SCF). the unpaired orbitals are assumed to be high spin. not needed in L506). an open-shell singlet is assumed. 0 1 Default (Yes). if mulitplicity 1). closed for Multip=1.2). Integrals with degree of contraction greater than or equal to N are done are regular integrals. Number of hamiltonians to read in (L506).

use loose integral cutoffs. doing an inexpensive pass 0 and then full accuracy in pass 1. Yes. N. Use approximation I. then one iteration with next grid up. convergence on either energy or density and always do incremental No. L510): 0 1 2 3 4 5 6 00 N00 I000 00000 10000 20000 Default (No). SAME AS IOp(33)=2. Do iterations with sleazy XC grid. BUT AT THE END OF EACH ITERATION. Force use of general DBF logic. ONLY SUMMARY RESULTS ARE PRINTED (WITH POSSIBLE CONTROL FROM THE 'NO. IOp(5/31) How many vectors to form at a time during microiterations in L508 (NYI) and L509: 0 N Default (3 in L509). SAME AS IOp(33)=3. O. SAME AS IOp(33)=1. No longer used.PRINT' OPTION). IOp(5/33) PRINT 0 1 2 3 4 IOp(33) PRINT OPTION. Yes. Force use of 1c instead of general DBF logic. Use general DBF logic only if the DBF rwf is present. Thresholds similar to DGauss for convergence and integrals. IOp(5/32) Sleazy SCF (L502. REGULAR SYSTEM DEFAULTS APPLY HERE. Foc formation. 80 . THE EIGENVALUES AND THE M. The default is CoarseGrid for iterations and SG1 for final energy. BUT ADDITIONALLY THE DENSITY MATRIX IS PRINTED. BUT ALL MATRIX TRANSACTIONS ARE PRINTED (BEWARE!!! MUCH OUTPUT EVEN ON SMALL MOLECULES. Decide between 1 and 4 based on details of the calculation. COEFFICIENTS ARE PRINTED AT THE END OF THE SCF.) IOp(5/34) DUMP OPTION. 0=normal 1=Linear approximation to Xc. No longer used.2 No.

0 1 2 Default (checkpoint only if direct). -1 0 N Do not do incremental Fock formation. then 2 digits more Converge to 10**-5 with integrals good to 10**-6 first. Checkpoint. No. Yes. Yes. Don't checkpoint. 0 1 2 Default (No).IOp(5/35) Whether basis is orthonormal (L501. allows different default actions for PBC. do integrals at same accuracy as convergence until final iteration. IOp(5/39) New On-Fly symbolic matrix element generator. IOp(5/36) Whether to checkpoint after every SCF cycle. REQUIRE 'NOFULLDIAG' Remember: the first digit indicating the type of function to be used. VarAcc forbidded because of guess=read. then full convergence. Default (every 20 for direct). L502). accurately. must be set. No. IOp(5/37) Frequency at which to do full Fock formation instead of incremental (L502). decide based on details of problems. do integrals 3 digits more accurately than current convergence. 1 2 3 Hartree-Waller functions for singlets Hartree-Waller functions for triplets Slater Determinants 81 . Every Nth cycle. VarAcc allowed. IOp(5/38) Whether to vary integral cutoffs during direct SCF: 0 1 2 3 4 5 6 Default (same as 1). Yes.

WARNING !!! ratio (zz+k-1)/k must be equal to n. Update surface on pass 2 iterations only.. Converge only on RMS density change. number specified in nroot=n. Maximum allowed change is 10**N larger than RMS. convergence only on maximum). Maximum allowed changed is same as RMS (i. Default=1 (no reduction) ll0000000 Davidson iteration at which to scale back the number of vectors WARNING !!! For overflow reasons.50) Maximum dimension of iterative subspace.e. Multipoles of order N. IOp(5/41) Whether to converge on maximum density change as well or instead of RMS: 0 N -1 -2 N0 in L510. store field in Gen(2-4) Default (same as 10000). Update surface every iteration in pass 1 only. but re-use 1e matrix instead of surface terms. Default is DON'T EXCLUDE any Use cutoff = 10**(-yy) on the product Integr*DenMat. Update surface and restart DIIS when within 10**-2 of convergence. Same as 3. value must be 0<=ll<=20 Default=0 82 22. Update surface every iteration. Converge on energy to 10**(N)*RMS-density-accuracy Also control of iterative diagonalization . Default is DON'T EXCLUDE any integral Use sum of the first IState roots of a Reduced Hamiltonian as guess for Lanczos Use IState-th root of a Reduced Hamiltonian as guess for Lanczos Save first IState roots on disk for Davidson (this option will automatically call Davidson instead Print S**2 Print S**2 and its orbital components of Lanczos) IOp(5/40) Use of reaction field. only used now for Onsager and control of details of SCIPCM -N 0 N 00000 10000 20000 30000 40000 50000 Multipoles of order N. Option xx is used also by Lanczos if IOp(39)=10000n or 20000n Maximum dimension of reduced Hamiltonian used as guess Default=Min(NSec. Default=60 Number of vectors provided in input BEWARE !!! Davidson will execute zz updating per iteration. Default=IVEC Reduction factor between number of guess vectors provided and number of vectors wanted at the end (1<=k<=9). L510: xx yy00 zz0000 k000000 Davidson options.xx0 integral yy000 100000 200000 300000 1000000 2000000 Use cutoff = 10**(-xx) on integral value to exclude contributions. increment field in Gen(2-4) No.

7: the displacement in geometry in internals in Angs. If negative reads the vector from rwf.IOp(5/42) Number of orbitals to localize in L510 1 n Localize all active orbitals Localize first n (strongly occupied!) orbitals IOp(5/43) L509: Whether 5th order terms are treated explicitly 0 1 2 Default: set to 1 All 5th order terms are treated implicitly (Debug option) 5th order GG and WG terms are explicitly computed in L715 L510: DFT corrections to MCSCF on last iteration 0 1 2 No Yes. IOp(5/47) In L510. 1 to set up for CAS-MP2 or 2 to do spin-orbit calculation. 10 Include the CSF contribution to the orbs for the DerCpl.8 f10. If positive reads vector from input 4f20. 1 2 Prepare data for Mp2 (l906 obsolete) Compute transition spin density and SO coupling IOp(5/48) Options to be passed to CalDFT: 83 . Uses MC-SCF density to compute B88 + LYP energy (These are hard-wired since they were the only choices that gave sensible results) Replaces diagonal elements of MC-SCF CI with B88 + LYP energy IOp(5/45) Numerical Derivative Coupling calculation(for testing) 0 1 No Yes (Needs NonStd root and two cards in input stream): i3 the other vector which coupled with iVec.

also none. Diagonalization Conjugate gradient. 0 1 Yes canonicalize. Speeds up convergence in CP-MCSCF Do not Canonicalize (turn this on to maintain compatability with previous versions of code.1 Float(1/i) . or 22 if sparse is set in ILSW). Linear storage (only in Fock formation if diagonalization). Square storage (only in Fock formation if CG). N words. L510: Option for using reorthogonalization procedure in Lanczos 0 1 No Yes IOp(5/49) Use of sparse storage and Conjugate Gradient optimization instead of N**2 memory and diagonalization. L510: Option for using lanczos in CPMCSCF calculations 0 1 2 No Yes Use lanczos except for the last iteration IOp(5/50) L510: Option for setting the maximum number of lanczos iterations in Direct CPMCSCF IOp(5/51) L510: Canonicalize MC-SCF orbitals by diagonalization of Core and Virt Fock operators.N Control flag for CalDFT is N.) IOp(5/52) Amount of memory to allocate to stashing integrals. -1 0 N None Default. L510: configuration cutoff for mp2 0 i 84 . 0 1 2 10 20 Default (11.

N of each.5-0.5). L510: See below IOp(5/55-58) L510: Switching on a State Averaged calculation graduately. In that case.1 unless doing stable=opt.0) to (0. (Or reversed. Default is 0.) When IOp(55).001 .lt. L510: see below IOp(5/56) A0 for Onsager SCRF. Energy difference smaller than IOp(56)*0.5). the number of steps for switching on/off is -IOp(55). The state averaging will be switched off (graduately) when the degenerate region is left. N>0 N<0 Solvent type N. L510: See below IOp(5/57) First iteration at which to level shift and do FON. Usefull for optimizations or trajectory calculations where only a part of the surface is (nearly) degenerate. when the SA calculation graduately is switch off.001 .IOp(5/53) PCM input and solvent type. N N/1000 Bohr. IOp(55) Number of steps over which the SA coefficients are brought from (0.050 Threshhold for switching off. 0 Default .0-1. Default is 0. in link 1003 a normal frequency calculation is performed when the optimization is in 85 . Energy difference larger than IOp(57)*0.5-0. then start after instability searches. the calculation is started with coefficients (0. which is usefull when a trajectory is started at a degenerate region. default parameters Dielectric constant |N|/1000 IOp(5/55) How many HOMOs and LUMOs to solve for after CG: 0 N None.075 When set. IOp(56) IOp(57) IOp(58) Threshhold for switching on.0.

4/0.9 0. and IRanGd.6 0. Yes.8 0.8 0. regardless of weather that continues to decrease after the switch point.a region of (0. for a hypothetical example with nroot=2: cycle 4 5 6 7 8 old SA 0.0-1.0001 (checks Offdiagonal element). 1: 2: Normal computation when (0. IOp(5/63) Whether to do FMM.7 0.1/0.3/0.5/0.1 Egap <limit <limit >limit >limit >limit Note: trajectory option set in overlay 10: IOp(10/99)=n -.2/0. instead of a State Averaged second derivatives calculation.3/0. only used with IOp(97)=11 IOp(5/60-62) Override standard values of IRadAn.2/0. 0 1 Use global default.8 0.7 0. 86 . this can cause jobs sensitive to SA to fail to converge The new method checks the gap each cycle and will decide to increase or decrease the SA based on that result.0).Threshold for a determining an adiabatic hop threshold=n*0. Turn off FMM here regardless. use N. IRanWt.5 crash new SA 0.0-1.2/0. IOp(5/64) Over-ride default value of FMFlags 0 N No.0) (default) Allways SA second derivatives THESE OPTIONS MUST BE SET IDENTICAL FOR OVERLAY 10! (only IOp(10/55) needs to be set in overlay 10) IOp(5/59) IOp(59) 0 1 (Default) No action Turn on new more flexible SA options: The old method switches SA on when you reach certain energy gap.0/0.

Occupy lowest NE at each k point regardless of the energies Improved LT with quadratic corrections Original LT method No concern for corrections Smearing Marzari method I Smearing Marzari method II 87 . 6000K for PBC) N degrees IOp(5/71) Number of steps to apply simulated annealing (L502): 0 N Default -. Use Cholesky basis for CP orthonormal transform. 0 N No. use N. or temperature for broadening (PBC and IOp(74)=[1-4]xx). -2 -1 0 N None Start at a high temperature (limited only by DIIS error). IOp(5/74) Type of k-point integration: 0 1 2 3 9 10 20 90 100 200 Default (911). temperature broadening for metals.IOp(5/65) Over-ride NFx parameter. IOp(5/70) Maximum Initial temperature for FON (non-PBC). Use LT method (interpolation) Occupy entire points (used together with broadening) Full points for insulators. Default (3000K = 10 milliHartrees for non-PBC.10 steps FON / 20 steps PFON N steps IOp(5/73) Options for ADMP: 0 1 2 Default (2) Use Lowdin basis for CP orthonormal transform. Yes.

No preconditioning. beta electrons. Convergence criterion 0 N 88 Defaults to 10**(-7) 10**(-N) . (defaults to 4 CG cycles).NN IOp(5/81) Conjugate-Gradient Parameters MM NN00 00000 10000 000000 100000 0000000 1000000 Maximum Number of CG cycles per SCF iteration. IOp(5/79) Range around Fermi level where temperature distribution will be applied if broadening is turned on for PBC. Don't use CG DIIS Use CG DIIS Polak-Ribiere CG minimization Fletcher-Reeves CG minimization Use diagonal preconditioning in Conjugate-Gradient. 0 Default. a value will be chosen in ZInLT1. (defaults to 3 cycles).300 400 500 900 First order Hermite-Gaussian of Paxton and Methfessel Gaussian smearing Classical Fermi-Dirac broadening No broadening (this will be Gaussian broadening with small T) IOp(5/75-78) Number of alpha electrons. IOp(5/82) C.G. and beta orbitals for fractional occupation. IOp(5/80) The maximum conjugate gradient step size -1 0 MMNN No maximum step size Default maximum (. alpha orbitals. Maximum Number of purification cycles per CG iteration.8) Step size of MM.

Level shift only if the HOMO-LUMO gap is zero or insignificant (>-0. Turn off 3 and 4 point extrapolation if DIIS is on. then 4). Limit is N cycles. Turn off level shift after energy rises. Level shift as much as necessary for HOMO>LUMO.IOp(5/83) Maximum SCF DIIS vectors 0 N Default (20). Reduce DIIS space whenever energy is above the lowest energy. Reduce DIIS space when energy rises from previous cycle. No. up to twice the goal No longer used. 89 . Yes. IOp(5/85) Over-riding of maximum cycles for XQC. Level shift to a maximum of 2*Goal. Reduce DIIS space when energy goes above the lowest energy. Use SCF DIIS with N vectors IOp(5/84) Restart in L509. 0 1 No. Default (1 except during Stable=Opt. Keep level shift after energy rises. Level shift only if the HOMO-LUMO gap is zero. Level shift to a maximum of the Goal. Turn on dynamic level shift from the beginning Turn on dynamic level shift only after FON is over.1) Level shift only if the HOMO-LUMO gap is zero or insignificant (>-0.1). -1 0 N Default for first step (128). IOp(5/86) Strategy options 000000 0 1 2 3 4 10 20 100 200 1000 2000 3000 4000 5000 6000 N0000 100000 Default (101100). Just continue as usual if energy goes up. Restart using Fock matrix in L502.

this and ZFormV should be moved to L302. variable=kPrint 0 1 2 3 Print only summary information Print the a(t) vector and probability for each csf Print almost everything for debugging Print everything for debugging warning! this is a lot of stuff and you will only be able to do a few cycles IOp(5/89) Linearly dependent basis control for PBC. . Yes. Retain 3 and 4 point extrap. 1 2 Yes. separate Coulomb and NFx exchange for PBC). do preliminary AM1 calculation and compare with guess from previous step in geometry optimization. 10**-N. IOp(5/92) Whether to use FoFDir or FoFCou for exact exchange: 0 90 Default: normal processing based on FMM for non-PBC. IOp(5/88) L510: controls the amount of printing. The energy is only checked after FON has been turned IOp(5/87) Accuracy criterion in Fock matrix formation: 0 N Default. do preliminary AM1 calculation. set in FoFDir/FoFCou/CalDSu based on accuracy part of IOp(5).200000 off. if DIIS is on. Typically 10^-10 for molecules and 10^-12 for periodic systems. IOp(5/90) Whether to generate sparse guess here. IOp(5/91) Control option for chebyshev sparse control.

FoFCou for Coulomb. L510: has different meaning depending on if you are using IOp(97)=22 or IOp(97=23) If IOp(97)=22. N iterations. separate FoFCou/NFx for exchange. where: Zyy yy Z xxx Threshold for a hop down determined by probability of being on the upper state (=x*0. wait time is xxx*0.01) Determines a variable Ulimt which lies between 01 and 99 Determines if this probability is halved after the first hop default Zyy=25. value=xxx. threshold for a hop up (after a hop down) = 0. threshold =0. variable=IWait.1 2 FoFCou for Coulomb.75 in ES (or 0.25 in GS) and if Zyy=125.125 in GS How long to wait before checking for a hop after going through a hop. Value = xxxzyy. L510: flag hopping controls starting and stoping options (x=0 or x=1) variable=iBack for first hop being up: xxxx0 xxxx1 variable=iStNow xxx0x xxx1x variable=iSpace xx0xx xx1xx variable=iEnd x0xxx xNxxx 0xxxx Nxxxx Use full space CI basis Use reduced space in projection of alpha number of cycles to carry on before stopping the timedep code after exiting the hopping region Default stop 6 iterations later Stop time dep on cycle N after exit (if IOp(92) is negative then stop immediately) No effect Stop time dep on cycle N allowable values 1-9 Use Energy gap criteria to start timedep Start timedep imediately Hopping down (forward) Hopping up (backward) variable=iFcTD to stop the time dependent code on cyle Z IOp(5/93) Number of initial iterations for which damping is allowed: 0 N Default (10).1 fempto seconds If IOp(97)=23.01) or the lower state (=(1-x)*0. which determines what type of gradient calculation to do: xxN Choose the basis variable=mBasis 0 1 2 3 xNx (Default) same as 1 Use a(t) basis Use mcscf basis orthogonal to a(1) and a(2) Use currnet mcscf vectors to check the code must be used with mHTest=2 91 Do with diagonalisationof Ecc variable=mTDGrd: . separate FoFDir for exchange.

on the real system PCM/ONIOM calcn. and for pass 0 grid in L502.05 switch timdep off only used with IOp(97)=22 and IOp(97)=23 IOp(5/95) Option for using Davidson in RFO calculations 0 1 Yes No use Lanczos (not recommended) IOp(5/96) Over-ride IRadAn for CPHF-like step in L509. Ras1. Ras3 orbitals are empty. RAS control in L510. where ww 92 Number of Ras1 orbitals . In the reference space. We also need to define the maximum number of holes in the Ras1 space (ie the number of electrons that can be excited out of the Ras1 subspace.0 1 2 N00 0 1 2 information (Default) same as 1 Only check Ecc if MCSCF energies are almost degenerate Force check on Ecc by diagonalising it (Default) no testing Testing construction of ecc Testing construction of the ecc or exx portion of hessian using the mcscf vector 3 Calculate but do not use TD gradient Testing options. variable=mHTest IOp(5/94) PCM/ONIOM calculation 0 1 2 Standard PCM calculation PCM/ONIOM calcn. The CAS active space is subdivided into three RAS active subspaces.04 switch timdep on. on the model system L510: Threshold for turning propagation method on and off yyy xxx The first three digits determine the energy gap for turning off The last three digits determine the energy gap for turning on Threshold =xxx*0. default yyy=5 ie deltaE>0. Use grid N. 0 N Use default. and the maximum number of electrons in the Ras3 space: zzyyxxww. Ras2 and Ras3.01 (checks energy gap). Ras1 orbitals are doubly occupied. default xxx=4 ie deltaE<0.

No. No. 0 Default: 32. Vector following or Root following hop alone (Needs option 80) Make a hop based on the secular equation (adiabatic hop) this option also includes a hop decision Debugging Option: Propagation of the wavefunction is switched on.2. if available). Yes. 22 23 Propagation of the wavefunction. IOp(5/99) Grid for numerical k-integration in FT-LT method. Needs option 80 and option 99. Yes. 0 1 2 Default (No). No.xx yy zz Maximum number of holes in Ras1 Number of Ras3 orbitals Maximum number of electrons in Ras3 IOp(5/97) Whether to update precomputed grid data with timing information. but the hop is determined by based on the vector following method (diabatic hop) (Needs option 99 and 80) the diabatic or adiabatic criteria (whichever determines a hop first). hop is based on the probability of being in a specific state Propagation of the wavefunction but no hopping. 93 . Yes. Hopping threshold during trajectories with L510. 0 1 2 10 11 21 Default (Yes. for detailed control of the propagation conditions see options 92-96. the molecule continues in a mixed state (Needs (Needs also option 80) (detailed control is determined by options 92-96) option 71 and 80) IOp(5/98) Whether to save eigenvalues and orbitals at all k-points.12.8 for 1.3d IOp(5/100) Tight convergence during CGDMS: 0 1 2 Default (No).

IOp(5/101) SDif test on numerical accuracy of PBC diagonalization. L510 Notes These options must be set in multiple links: L1003 iop(97) iop(55-59) iop(80) yes yes yes yes yes no no no yes L510 L118 These options must be set for the following links: IOp l118 l510 l1003 55-58 no yes yes 80 yes no yes 97 no yes yes 98 no no yes 99 no no yes 94 . 0 -1 N>0 Default (10) No test Abort of SDif is larger than N. IOp(5/102) Maximum number of configurations for CAS-MP2: 0 N Default (1000). N.

number of increments and vector.X1.Y-field. READ ADDITIONAL CENTERS. Read in centers at which to evaluate the potential from the rwf. Read grid. 0 1 2 4 NO ADDITIONAL CENTERS. If KTape is 0.Z-coord. Evenly spaced rectangular grid. ONE CARD PER CENTER WITH THE X. An arbitrary list of points. field.Y2. The potential (NEFG=3). potential and field (NEFG=2). containing: X-coord. Only one card is needed: N.LTape. do not default cube. N1 records will be written to unit KTape.12). Y AND Z COORDINATES IN ANGSTROMS (FREE FORMAT). with N2 values in each record.Y-coord.Z1 N2.XO. and field gradient (NEFG=1) will be computed and written along with the coordinates to unit KTape in format (4F20. -.12). 8 16 32 128 Do potential-derived charges.KTape The coordinates of N points in Angstroms will be read unit LTape in format (3F20. Read in a set of cards specifying a grid of points at which the electric potential will be computed. READ IN COORDINATES AS FOR 1. IF MORE THAN ONE OF THESE IS REQUESTED. Three cards are required: KTape. Thus if NEFG=3 for each point there will be 4 cards written per point.output unit and coordinates of one corner of grid. Constrain the dipole in fitting charges.number of increments and vector. POTENTIAL. or potential. it defaults to 51. B. STARTING AT EACH POINT.X2.XY-EFG.NEFG.Z2 -.YO.Overlay 6 IOp(6/15) SPECIFICATION OF ADDITIONAL CENTERS.ZZ-EFG.XX-EFG YY-EFG.Y1. EVALUATE THE PROPERTIES ONLY AT EACH ATOMIC CENTER.XZ-EFG YZ-EFG Note that either form of grid should be specified with respect to the standard orientation of the molecule. THE LISTS ARE IN SEPARATE INPUT SECTIONS IN THE ORDER LISTED BELOW. -.Z-field.Potential X-field.ZO N1. 95 . Two forms of specifications are allowed: A. LOCATED THE NEAREST STATIONARY POINT IN THE ELECTRIC.

0 1 2 -N COMPUTE ALL CONTRIBUTIONS TO SELECTED PROPERTIES. IOp(6/20) How to do electrostatic-potential derived charges: 0 -1 1 2 3 00 10 20 30 100 96 Default (1).Rad).IOp(6/16) Cutoffs in L602. Force Merz-Kollman radii. IOp(6/18) Whether to update dipole rwf 0/1 yes/no. CHELPG point selection. Exact is already in z-matrix orientation. one per line. Read a list of points at which to fit. COMPUTE ONLY THE NUCLEAR CONTRIBUTION. 0 1 2 Default. This is like IOp(9) in L9999. Default radii are those defined with the selected method. 0 1 Use full accuracy in calculations at specific points. COMPUTE ONLY THE CONTRIBUTION OF SHELL N.Rad) or (Symbol. Force CHELP (Francl) recommended radii. Exact is still in standard orientation. use as-is. Merz-Kollman point selection CHELP point selection. COMPUTE ONLY THE ELECTRONIC CONTRIBUTION. but use sleazy cutoffs in mapping a grid Do all points to full accuracy. so rotate. IOp(6/19) Whether to rotate exact polarizability before comparing with approximate (which will be calculated in the standard orientation). of points. Force CHELPG (Breneman) recommended radii. Read in replacement radii for selected atom types as pairs (IAn. terminated . IOp(6/17) DEBUGGING CONTROL (L602). same as 1.

chk file. Use only active atoms in the fit. terminated by a blank line. 97 . L602. Fit united atoms (heavy atoms only) rather than all atoms. Density values. Density values. Use the density matrix for the current method. Density for the excited state given by IOp(29). gradients and divergence. Transition density between the states given by IOp(29) and IOp(30).Rad). IOp(6/22) Selection of density matrix (currently only in L601. or the HF density if the one for the current method Use the density matrix for method N (see Link 1 for the numbering scheme).0 Read density matrices from . All available transition densities. IOp(6/24) Frozen core: -N 0 1 2 Freeze N orbitals. Use all available density matrices. Default (Yes).chk file.ge. Density values and gradients. is not available. IOp(6/23) Density values to evaluate over grid in L604: 0 1 2 3 Default (same as 3). IOp(6/21) Operation of L603: 0 1 2 Default (same as 2). Yes. L604): -1x +1x -5 -4 -3 -2 -1 N. Optimize density basis set. Read density matrices from .by a blank line. 200 1000 10000 Read in replacment radii for selected atoms as pairs (I. Read in density basis functions and compute populations. No.

Spin. Do Mulliken populations. IOp(6/27) Choice of population analysis: 0 1 2 10 20 100 Default (12). Yes: mark as well. IOp(6/28) Mark SCF density as current density. Don't do Mulliken populations. Yes.requires 2e integrals. Beta. No. O(N**4)). quantum mechanically (no self terms . 98 . 0 1 No: save SCF density. IOp(6/29) Excited state to use if requested by IOp(22). IOp(6/26) Which density to use in L602 and L604: 0 1 2 3 4 Default (same as 1). Alpha. classically (including self terms . Do Minimal population analysis. Total. but do not mark. Do bonding Mulliken Populations.requires 2e integrals. Yes.IOp(6/25) Whether to compute coulomb self-energy in L601: 0 1 2 O(N**5)). and only available for HF. Don't do bonding Mulliken Populations.

s. default 10). This precludes calculations of atomic property derivatives with respect to nuclear displacements.01*IDcInt are interpreted as lone pair MOs rather than bond MOs (1. 1: Do not print the atomic occupancies. Name this file 'rpac. Determine zero-flux surfaces (IDoZrF). default 9). IOp(6/32) CONTROL PARAMETERS FOR COVBON in L609 (NOT TO BE CHANGED UNDER MOST CIRCUMSTANCES): 10000*MItLoc+1000*ITlLoc+100*IDcInt+IPrLoc. 99 . Use spin density. where MItLoc ITlLoc IDcInt IPrLoc MItLoc*NOrb*(NOrb-1)/2 is the maximum number of iterations in localization of (spin)orbitals (1. Localized (spin)orbitals with atomic occupancies less than 0. Transition denstiy between state IOp(29) and state N. Store only alpha NOs.11' IOp(6/35) WHAT TO DO: 0 1 2 4 10 100 Determine attractors. ring points.. Compute energies of electrostatic interactions between AIMs (IDoPot). Compute atomic overlap matrices (IDoAOM).9. 0: Print the atomic occupancies of localized (spin)orbitals (default)..9. Yes.. Store only beta NOs. 10. using alpha and beta separately for UHF. L606: naming of RPAC interface file. Compute charges of AIMs (IDoAtC). and cage points.. Compute kinetic energies and multipole moments of AIMs (IDoPrp).99.. default 6). 0 1 Make this a scratch file.**(-ITlLoc) is the convergence criterion for (spin)orbital localization (1.. IOp(6/31) Whether to determine natural orbitals from densities: 0 1 2 3 4 5 No. L605. attractor interaction lines. Yes. using total density.IOp(6/30) 2nd excited state for transition density: 0 N Transition denstiy between state IOp(29) and g.

Note that analytical force constants have to be available! Compute localized orbitals and bond orders (IDoLoc)..**(-IPrNAt) in integrations over atomic basins (099. where INoZer IPrNDe IPrNAt 0: Ignore (spin)orbitals with zero occupancies (default). If necessary.9. Neglect primitive contributions below 10.. Note that IDoSRe should be set to 1 in order to obtain correct results! Also note that analytical polarizabilities have to be available but force constants have to be absent! 2000 10000 20000 100000 200000 400000 1000000 2000000 3000000 Compute derivatives of atomic properties with respect to nuclear displacements as well (IDoNuD).200 400 1000 Compute other atomic matrix elements (IDoAMa). where MxBpIt SBpMax NGrd LookUp Maximum number of iterations in trial path determination (1. Maximum value of the control sum (1. 1: Do not ignore (spin)orbitals with Neglect primitive contributions below 10. Compute atomic orbitals in molecule (IDoAOs). Number of grid points in critical point search (1. default 10). default 20).. augment valence electron densities with nonrelativistic core contributions Abort if pseudopotentials have been used (IHwAug=3). 99.**(-IPrNDe) in evaluations of electron density and its derivatives (0 default 7). Length of Fourier expansion for the trial path (1. and IEpsIn=100. This option sets IPrNDe=5. IPrNAt=5. Reduce accuracy so atomic charges can be computed more rapidly (IQuick). (IHwAug=1).999.. Use numerical instead of analtyic integration.99. AND CAGPNT in L609 (NOT TO BE CHANGED UNDER MOST CIRCUMSTANCES): 1000000*MxBpIt+100000*SBpMax+1000*NGrd+LookUp. default 100). IOp(6/38) CONTROL PARAMETERS FOR ZRFLUX AND OIGAPI in L609 (NOT TO BE CHANGED UNDER MOST CIRCUMSTANCES): 100 . zero occupancies. IOp(6/36) CONTROL PARAMETERS FOR NEGLECT OF ORBITALS AND PRIMITIVES in L609: 10000*INoZer+100*IPrNDe+IPrNAt. Include zero-flux surface relaxation terms in all atomic matrix elements (IDoSRe) Compute derivatives of atomic properties with respect to electric field (IDoSeP). default 2).99. which is a default anyway (IHwAug=0).. No other properties can be calculated. If necessary. IOp(6/37) CONTROL PARAMETERS FOR ATINLI.. RNGPNT. Use numerical instead of analtyic integration and use reduced cutoffs. default 7).. augment valence electron densities with relativistic core contributions..

IOp(6/41) Number of layers in esp charge fit... 10. This link must have been invoked with IOp(40) = 0 or 1 prior to invoking it with IOp(40)=2. Read input data to control NBO analysis.9. IOp(6/40) Control of link 607.**(-5-IToler) is the tolerance for the intersection search (1.. N layers.. 0 N Default (4).. default 2). 0. 0 1 2 Default NBO analysis .. Delete selected elements of NBO Fock matrix and form a new density. default 2).....99. 10..9. 10*INTrig is the number of sine and cosine functions in the trial function for surface sheets 10. default (1. (1. IRScal is the scaling factor in the nonlinear transformation used in the intersection search 10.9. 3 Read the deletion energy produced by a previous run with IOp(40)=2 and print it.9.9. 10*INInGr is the initial number of grid points in theta and phi in the adaptive integration subroutine (1. default 6).*IRtFSe is the safety factor used in the intersection search (1. default 2). default 2)... 101 . default 5).. default 2).. IOp(6/39) More CONTROL PARAMETERS FOR ZRFLUX AND OIGAPI in L609 (NOT TO BE CHANGED UNDER MOST CIRCUMSTANCES): 1000000*IToler+100000*INInGr+10000*INInCh+1000*IEpsSf+10*IEpsIn+INTrig IToler INInGr INInCh 2).9.*IPraIn is the cut-off for zero-flux surfaces (1. default 6).**(-IGueDi) is the initial displacement from the critical point in the Runge-Kutta integrations (1.. default 2).9.9. default 5). 5+INInCh is the initial number of sampling points in the intersection search (1. must be >=4...9.100000*INStRK+10000*IHowFa+1000*IGueDi+100*IPraIn+10*IRScal+IRtFSe INStRK IHowFa IGueDi IPraIn IRScal IRtFSe 10*INStRK is the number of steps in the Runge-Kutta integrations along gradient paths (1. default 2)..don't read input..9.9.. IEpsSf IEpsIn INTrig IEpsSf is the safety factor used for patches with surface faults in the adaptive integration subroutine (1.0001*IEpsIn is the target for integration error (1.. IHowFa is the maximum distance in the Runge-Kutta integrations along gradient paths (1.. whose energy can then be computed by one of the SCF links..

50 is recommended. 0 -1 N Default . points per unit area.10**-3 Read from input.for tight accuracy. 102 .4/Sqrt(#layers)) 0.IOp(6/42) Density of points per unit area in esp fit. same as 2. N*10**-4. 0 N Default (1). IOp(6/44) Type of calculation in L604: 0 1 2 3 10 Default. IOp(6/46) Threshold for molecular volume integration. 0 N Default (0.01*N. Compute the molar volume Evaluate the density over a cube of points Evaluate MO's over a cube of points Skip header information in cube file. IOp(6/45) Number of points per bohr**3 for Monte-Carlo calaulation of molar volume -1 0 N Read from input Default (20) N points . IOp(6/43) Increment between layers in MK charge fit.

N*0.01 . Fine grid. 3 points/Bohr. IOp(6/48) Use of cutoffs 0 N Default (10**-6 accuracy for cubes. on non-SCF densities. IOp(6/52) Number of radial integration points in L609: 0 N Default (100). up to N IPs and EAs. 6 points/Bohr. Medium grid. 12 points/Bohr. -N>4 Grid using 1000 / N points/Bohr.for debugging. 1 digit better than desired acuracy for volumes). Yes. IOp(6/51) Whether to apply Extended Koopman's Theorem (EKT): 0 N -1 -2 Default (No). all possible IPs and EAs. Yes. 103 . on non-SCF densities. Coarse grid. N. 10**-N IOp(6/49) Approximate number of points per side in cube in l602/l604: 0 N -1 -2 -3 -4 Default (80) N points Read from cards. No.IOp(6/47) Scale factor to apply to van der Waals radii for the box size during volume integration: 0 N Default.

Yes. IOp(6/58) Grid to use in generating tables of density and potential. also read in name for input file with a different grid and compare. 104 .IOp(6/53) Distribution of radial points in L609: 0 N Default (cubic) Polynomial of order N. N. Must be an unpruned grid. 0 Default (99001). Yes. 0 1 No. read in name for output file. IOp(6/57) Whether to generate data over a grid using the total SCF density: 0 1 2 3 No. IOp(6/55) Number of inner angular points in numerical integration in L609: -1 0 N 0 (no inner sphere) 302 N point Lebedev grid (see AngQad). 0 N Default (100000). IOp(6/56) Whether to read in density matrix from input stream in L608. Yes. IOp(6/54) Maximum number of domains. Output in the form of data statements.

Do correct energies and 0th order approximation Do correct energies and 0th-1st order approximations Do correct energies and 0th-2nd order approximations IOp(6/60-62) Over-ride standard values of IRadAn. and IRanGd. Yes. -1 0 N Zero. Do test as usual. 105 .IOp(6/59) Approximations to Exc -1 0 1 2 3 Test superposition of atomic densities using L608: Do correct energies. of all tensor components. of isotropic value. Suppress test. IOp(6/63) Suppress number of electrons test in XC quadrature in L608 (for debugging with small grids): 0 1 2 Default (do test). Default (1 pmm). of diagonal tensor elements and isotropic value. IOp(6/65) Threshold for printing of NCS contributions. Yes. N/1000 ppm IOp(6/70) Control of L610. IOp(6/64) Natural Chemical Shielding Analysis: 0 1 2 3 No. IRanWt. Yes.

0 1 2 Default (use if present and choose is selected in NBO input). Yes. Don't use. Save NBOs in place of regular MOs. regardless of other terms. NMR data is available.D-6). Yes.IOp(6/71) XC functional in L610. if open-shell. Use. IOp(6/76) Threshold for linear dependence in L607. 10**(-N). Save NLMO occupieds and NBO virtuals. IOp(6/72) Whether to read isotopes for hyperfine interractions and do hyperfine terms in L602: 0 1 2 3 4 Default (1). 0 N 106 Default (1. Yes. Suppress re-orthogonalization. and other terms are being computed No. IOp(6/74) Whether to use Gaussian connectivity in choosing Lewis structure for NBO. IOp(6/75) model for CM2 charges. . reading isotopes IOp(6/73) Whether to save orbitals from NBO: 0 1 2 3 10 Default (No). Save NLMOs in place of regular MOs.

IOp(6/82) Tensors for hyperfine spectra. 0 1 2 Default (No). IOp(6/78) Use MOs instead of density in AtmTab. No. No. Use MOs.d-4 N * 10^-5. IOp(6/80) Whether to calculate Lowdin charges and Mayer bond orders. Yes. 0 1 2 Default (No). for all orbitals.IOp(6/77) Restraint in charge fitting in L602: 0 -1 N None. No. 2. Use density. 107 . IOp(6/79) Whether to calculate Hirshfeld charges. for the top 5 occupieds and lowest 5 virtuals. 0 1 2 Default (2). if doing other orbital results). Yes. IOp(6/81) Print kinetic energy of orbitals? 0 1 2 3 Default (yes. Yes. Yes.

108 . 2 Default. compute if there are 100 or fewer atoms Compute QEq tensors and for open-shell systems compute isotropic and anisotropic splitting Do not compute tensors.0 1 tensors.

Default scale factor is #N (1=HF. only harmonic thermochemistry. Increment the 1e derivative matrices with ecp terms. 2=CBS4=0. IOp(7/7) USE OF INTERNAL COORDINATES. If M=3. 109 . 3=CBSQ=0.91671.12. Print higher precision normal modes. Yes. Do not form 1e derivative matrices. Do not the ecp contribution to the SCF forces. If M=2. Read hindered rotor parameters from input. Print normal mode displacements in redunant internals. with most common isotopes. with read-in isotopes. 1/1.Overlay 7 IOp(7/6) operation of link 705 (NYI). but neglect first derivatives in conversion of second derivatives to internal coordinates. No.91844) If M=1. do hindered rotor analysis. 0 1 2 YES NO Yes. Print both HP modes and internal displacements. 0 1 2 10 20 Default (12). Yes. 0/1 yes/no. Form the ecp contribution to the SCF forces. IOp(7/9) Whether to rotate derivatives back to the z-matrix orientation. IOp(7/8) Harmonic frequency calculation: 0 1 2 3 10 20 30 Nxx Mxxx No.

FoFDir: Rys spdf. Generate 2PDM from CIS square 1PDM (for debugging) . FoFDir Rys df. Use both HF and external 2PDM. Berny SP. IOp(7/12) Selection of density matrix. do only second derivatives. 0 N Usual SCF density. PRISM df. Rys df. FoFDir: PRISM spdf. Rys f. Berny SP. Old vector Rys SPDF. Scalar Rys SPDF. Use generalized density number N for both the one-electron integral derivatives and the corresponding 2PDM terms. FoFDir HGP df. FoFDir HGP d Rys f. Scalar Rys DF. FoFDir: HGP sp. FoFDir Rys df (same as 7). Berny SP. FoFDir: HGP spdf. Berny SP.IOp(7/10) First/second derivative control. FoFDir: Berny SP. 0 1 2 do only first derivatives. Use external 2PDM. Berny SP. old vector Rys DF. FoFDir: HGP spd. do both. IOp(7/13) Contraction with two-particle density matrices: 0 1 2 3 4 110 Default (same as 1). Berny SP. Use HF 2PDM. IOp(7/11) control of integral derivative algorithm: 0 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Default use IsAlg to decide.

while 6-7 imply use of the generalized density derivatives in L701. The types of derivatives are given by IOp(15). 111 . Magnetic Field Ith order. Normal (6 or 5 for linear molecules).5 6 7 10 Generate 2PDM from CIS square 1PDM and use HF/Z 2PDM as well. IOp(7/15) The nature of the perturbation(s). Default. Contract with external 2PDM derivatives. DO AS ACCURATELY AS POSSIBLE in L703. Generate simulated density derivatives. Use sleazier cutoffs in L703. The types of derivatives are Leave the external 2PDM on the disk instead of deleting it. Nuclear Kth order. N. IOp(7/14) State for CIS gradients. 0 IJK 1000 Default (1st order nuclear and electric field). given by IOp(15) 0-5 imply use of the generalized density in L701. Defaults to 1. and 11 are valid in overlay 7. DO AS ACCURATELY AS POSSIBLE in L708. Only 1. IOp(7/18) Derivative accuracy option: 0 1 2 10 20 100 200 Compute to 10**(-8) accuracy. Electric field Jth order. IOp(7/16) Number of translations and rotations to remove during redundant coordinate transformations: -2 -1 0 N 0. DO AS ACCURATELY AS POSSIBLE in L702. 10. same as -1. Use sleazier cutoffs in L708. Form derivative 2PDM from CIS and HF derivative density matrices. USE THE ORIGINAL 'BERNY' VALUES in L702.

Read derivatives from checkpoint file (in Z-matrix orientation). same as 2. 0 1 2 6 00 Normal. PSCF freq=numer). Normal + Generate estimated initial force constants. Third. Use FXYZ in conversion of forces constants to internal coordinates (PSCF opt with HF 2nd deriv). 112 .IOp(7/25) Type of derivatives available. (HF freq. 0 1 2 10 First. Second. IOp(7/28) SKIP OPTION TO DEFER INTEGRAL EVALUATION TO L703. Compute HF forces from D2E file and increment both FX and FXYZ (non-O11 PSCF grad and HF freq). Normal NUCLEAR REPULSION ONLY (USEFUL FOR TESTING). DON'T USE. Default method for initial force constants IOp(7/30) USE OF SYMMETRY IN OVERLAY 7: 0 1 USE (SUBJECT TO AVAILABILITY). IOp(7/31) Handling of forces contributions. DO ALL GRADIENT INTEGRALS IN L703 IOp(7/29) MODE OF USE OF L716. Use FX in conversion of force constants to internal coordinates. 0 1 00 10 Just use the forces in IRWFX. 0 2 COMPUTE AS NORMAL.

16. Apply NDDO approximation.12 format in compressed form. and derivatives in cartesians and redundant internals. IOp(7/42) 1PDM: 0 N Use SCF total density. Neglect 2e integrals with diatomic differential overlap. Punch energy. IOp(7/43) 2nd order simultaneous optimization flag. Forces are punched in 3D20. . (other 2nd derivative options must also be set appropriately) 113 . coordinates.true. first and second derivatives in both cartesian and internal coordinates. Neglect four center integrals. Punch nuclear coordinate derivatives. Neglect 1e integrals with diatomic differential overlap. Neglect three center two-electron integrals as well. 0 1 Evaluate integrals correctly. coordinates. Punch energy.IOp(7/32) PUNCH OPTION. Neglect three center one-electron integrals. IOp(7/41) NDDO flag. one card per atom. Force constants and third derivatives are punched in 4E20.false. 0 1 .8. Use generalized density N. Do punch only if second derivatives are available. Punch energy. 0 1 2 3 4 5 1x None. IOp(7/40) Neglect of integrals (only option 1 works in Overlay 7): 0 1 2 3 10 20 30 Keep all integrals.12 format. forces and lower triangular force constants in format 6F12. Punch energy in format D24. and derivatives in redundant internals only in compressed form. Do only overlap and not other 1e integrals.

No. Update forces for a uniform electric field. 0 1 Yes. IOp(7/53) Convert forces over shells to field-dependent dipole and forces over atoms (for debugging): 0 1 10 No.d-6. IOp(7/45) Controlling the projection of the reaction path. and IRanGd. Update forces for the self-consistent reaction field (SCRF) method Update forces for a uniform electric field. Project using the Newton-Raphson step.2 . IOp(7/52) Whether ECP integrals should be done in L701 as usual. (debugging option: compute fifth order WG and GG terms in L715) IOp(7/44) Handling of an applied electric field. 114 .true. Project using forces if the RMS force is larger than 1. with forces done the usual way for CIS or MP2 2nd derivatives. Project the reaction path and compute 3N-7 frequencies. The point is a stationary point. Compute optimimum lambdas. Yes. IOp(7/60-62) IOp(60-62) Over-ride standard values of IRadAn. -1 0 1 2 Do not add electric field terms to forces. IRanWt. 0 1 2 3 Do not project.

IOp(7/64) Type of simulated spectrum in output. 0 1 2 3 Default (1). No. Turn off FMM here regardless. -1 0 1 No. if ref structure is present and has non-zero force constants).IOp(7/63) Whether to do FMM. Yes. 0 1 Use global default. IOp(7/71) Do vibrational 2nd order perturbation: 0 1 No Yes. Currently lots of hacks to determine where we are in the process instead of different values of this option. Yes. Lines Lorenzians Both IOp(7/65) Harmonic constraints with respect to initial structure during geometry optimization. IOp(7/72) Read additional parameters for anharmonic computations 0 No 115 . Default (Yes. IOp(7/70) Do vibro-rotational analysis: 0 1 2 Default (No).

IOp(7/75) Threshold for printing redundant internal contributions to normal mode displacements: 0 N -1 Default (10%) 10**-N Zero (all printed). 10^-10 for molecules. Prohibit FoFCou. Turn off 1c logic for 1c DBF case. Copy fit density over real density and do P(Fit)*Jx*P(Fit). 116 . Yes. 10**(-N). The threshold is automatically lowered for each mode until 90% of the absolute displacements are included. 10^-12 for PBC. IOp(7/87) Accuracy in FoFDir/FoFCou/CalDSu: 0 N Default.1 Yes IOp(7/74) Non-equilibrium PCM gradients: 0 1 No. Omit subtraction and do P(Fit)*Jx*P. IOp(7/76) Over-ride use of FoFCou in L703: 0 1 2 Normal processing. IOp(7/77) Debuging options for DBFs: 0 1 2 3 4 Normal processing. Force FoFCou. Clear real density and do -1/2 P(Fit)*Jx*P(Fit).

0 1 2 Default (No).(IA/JB).(IJ/KL). IOp(8/8) Whether to delete MO integrals in L811. Do not test SCF convergence (mainly used for testing). Full transformation if this is consistent with MaxDisk. THE COMPLETE SET OF TRANSFORMED INTEGRALS.(IJ/KA). IOp(8/7) SCF convergence test. No. QCISD.Overlay 8 IOp(8/5) Whether to pseudocanonicalize ROHF orbitals. 0 1 2 3 4 5 6 7 BUCKETS FOR MP2: (IA/JB). BD: (IJ/AB).(IA/JB). Full transformation if this is consistent with MaxDisk. -1 0 Yes. BUCKETS FOR CID OR MP3: (IJ/AB). BUCKETS FOR SEMI-DIRECT MP4DQ. CISD. IOp(8/9) Debug control (L802): 117 .(IK/KL). 0 1 Test that SCF has convergd. CISD or MP4SDQ or MP4SDTQ. otherwise same as 3. BUCKETS FOR STABILITY: (IA/JB).(IJ/AB). BUT INCLUDES (IA/BC). Yes. No. IOp(8/6) Bucket selection. otherwise same as 4.

virt-rep1. A negative value for the end deletes the top virtuals. The largest noble gas core and main group d's are frozen.. Force sorting for output bucks. same as 4. . Use all MOs. a card is read in indicating the start and the end. The window is specified by IOp(37-38). Generate and test RInt3 array (L804). (Upper triangle of symmetry compressed integrals for IOp(6)=5 or 4 only!) Reorder MOs. If IOp(37) is 0. except that the outer sp electrons of 3rd row and later alkalai and alkalai earth elements are retained. The window is recovered from file 569 on the checkpoint file. eigenvalues and symmetry assignment vectors according to ther representations IOp(8/10) Window is selected as follows: -N 0 N 1 2 3 4 90 91 92 93 94 000 118 Use the top N occupieds and lowest N virtuals. Accumulate MP2 force constant terms in direct fashion Write the MO basis first derivative ERI's to disk Force fully in-Core algorithm (L804 only). Force semi-direct mode 4 if IOp(6)=3. Force semi-direct mode 1. L811): 0 1 2 3 10 20 30 100 200 1000 2000 3000 4000 00000 10000 20000 Operate normally. 1 <= N <= 89 selects frozen-core type N: The largest noble gas core is frozen. Force N orbitals per pass.. Default. Eigenvalues and symmetry assignment vectors will be put in correspondence with vectors. Force semi-direct mode 2.. Read a list of orbitals to freeze.0 -N Operate normally. Force transformed integrals in Core algorithm.occ-rep2. The next to the largest noble gas core is frozen. Default (10000) Do not symmetry compress transformed integrals. G2 frozen-core: the largest noble gas core and main group d orbitals are frozen. The window is recovered from rwf 569.. core virtuals are also frozen.) 30000 100000 Symmetry compress transformed integrals only if RHF. Force output bucket in Core antisymmetrization. Do symmetry compress transformed integrals (buckets) (This will cause windowed MOs. Force semi-direct mode 3 if IOp(6)=3. Direct Transformation Control (L804. Default (200). For basis sets with double-zeta cores. reordered in the order of representations like occ-rep1..virt-rep2.. Force semi-direct transformation.

Used to select output mode. Use overlap with atomic core orbitals from Core Ham to choose core orbitals. Read 'old-style' drt input cards. 0 1 2 Default. 119 . IOp(8/14) Control of drt input. e. Also. IOp(8/15) Control of DRT output. IOp(8/12) Calculation of frozen-Core contributions. same as 10.g. Kill the job if any mo coefficients are greater than 1000. IOp(8/11) MO coefficient. Calculate 2 J . and number of electrons test. This is done when IOpCl=0 or when IOpCl=1 and the calculation is rohf or gvb. kill a frozen-core job if there there is significant core-valence mixing in the canonical orbitals 00 10 20 Default.10x 20x 30x Use orbital energies to choose core orbitals. IOp(8/13) Control of output. 0 1 Take necessary input from Gaussian data structures.0 or the smallest difference between occupied and virtual orbital energies is less than 0. if there is only 1 electron or 1 virtual spin-orbital. Output transformed integrals for DRT-CI calculation.. 0 1 No. Use overlap with atomic core orbitals from Harris to choose core orbitals. 0 1 Output to Gaussian system buckets. orbital energy. Kill the job if there is no correlation energy.K over deleted orbitals and add to Core-Hamiltonian.001. same as 2 Just print a warning message. Suppress such a test (CPHF may still be done for such a case).

x = -Sab. Non-canonical.x = (Fij. 0 N Default let program decide. Do both.EjSij.x/2 Canonical virtuals. Nuclear Kth order.x = -Sij. Uab.x/2 IOp(8/19) The nature of the perturbation(s) in L811: 0 IJK Default (1st order nuclear and electric field). Excitation level = N. Canonical. Uij.x) / (Ei-Ej) Note that this blows up for degenerate orbitals and is intended primarily for debugging. IOp(8/18) Which type of derivative transformation to do in L811: 0 1 2 3 4 5 6 Default.x . (only if integral derivative file is being written) Excitation level for SDGUGA-CI. IOp(8/20) Which terms to include in L811: 0 120 Default (same as 11).x.x Uji. Non-canonical. Canonical occupieds. Magnetic Field Ith order. Uij. IOp(8/17) Specification of integral block size for GUGA CI programs.x = 0.x. Uij. same as 3. Electric field Jth order. Uij. except canonical in frozen-active blocks. 0 N Default excitation level = 2. Integral Block Size = N.x = -1/2 Sij.x = -Sij. Non-canonical.x = -1/2 Sij.0 1 2 Write DRT output to RW-files. . IOp(8/16) Maximum number of orbitals per pass in L811. Uij. Write DRT output to Fortran unit 'drttap'.

IOp(8/29) Maximum number of perturbations per batch in L811: (only applies if integral deriv file is written) -3 -2 -1 0 1 N Do not use batching logic. Do one batch. Do as many in a batch as can be overlapped with sorting space for half transformed integrals. independant of MAXDISK. IOp(8/28) Hack number of occupieds for full ci using links 921 or 922: -1 0 N Transform all orbitals (after freezing Core) as occupieds (i.6 are the currently implemented options) 121 Use as much as desired. but use multi-batch logic. -1 0 N Use as much as needed for maximum efficiency.e.1 10 MO derivative times integral term. Look there for more information. L805. -3 -2 code. This only applies if the integral derivatives are not stored. N words. Do a single atom at a time (minimum disk usage). MO times integral derivative term. subject to the limit imposed by MAXDISK (IOp(27)).. This will determine the number of times AO integrals and derivatives are evaluated unless overridden by IOp(31).. IOp(8/22) These options control the in-Core post-SCF link. Requested disk usage. Transform N orbitals (after frozen Core) as occupieds (i. IOp(8/27) Maximum amount of disk to use in L804 and L811: 0 N Unlimited.. Default (same as 1). Default (-1) N evaluations and hence N coarse tiled batches (1.e. set NOA=NOB=NROrb in No. set NOA=NOB=N for purposes of transformation).. N triplets. Use an amount which is similar to the maximum disk usage in other parts of the MP2 frequency . transformation).

i<=jab alpha-beta only for closed-shell). No. . Explicit control of the fine tile batch size (largely a debugging option. 0 1 NO YES IOp(8/32) Whether to do CI in the interacting space only. Set up /Orb/ for a full window but then blank the wavefunction coefficients in L804. IOp(8/31) PERFORM PRIMITIVE POST-SCF OPERATIONS (NOT CURRENTLY FUNCTIONAL). Yes. IOp(8/36) Whether to update force constants with the MP2 product of MP2 integral derivatives term (only applies if integral derivative file is not written). 0 N Let the program choose (make it as large as possible) maximum fine tile batch size. (Yes if Ix is not stored. IOp(8/35) Output format for closed-shell and debugging control: (only for when integral derivative file is written) 0 1 2 10 Default (consistent with integrals for open-shell. order g i. no otherwise). only for no-Ix routines). 0 1 2 Default (all spin-eigenfunctions). All. Default on whether to make Poo and Pvv for MP2. 0 1 2 00 122 Default (Yes). Interacting only.IOp(8/30) Type of window. Set up for full but zero Core MOs.=j a<=b) for closed-shell AA and AB consistent with integrals. up to 9. 0 1 -1 Default. Set up /Orb/ as indicated by IOp(10). Do extra debugging computations. Test window. Store only the unique AB integral derivatives (gO2V2/4.

0 1 2 3 00 10 20 Default. IOp(8/39) Localized orbital method adopted in SAC/SAC-CI. yes otherwise.10 20 Yes. IOp(8/40) Handling of ROHF window: 0 1 2 Default (2). Boys method. use unity as rotation. Force not to rotate MOs. Force to rotate MOs. 0 No 123 . Population method. transformation. No. Take reference MOs from disk if available. forming pseudo-canonical alpha and beta orbitals and doing UHF Treat as RHF. Boys + population method. Rotate MOs to compute the displaced overlap. IOp(8/38) Integers specifying the window to use. No for the 1st geometry of opt. Use ROMP2 approach. Yes. transforming only alpha orbitals. IOp(8/42) Whether to reorder MOs during potential surface exploration. but save MOs. IOp(8/41) Transformation of spin-orbitals (alpha only) within occupied and unoccupied orbital subspaces by minimum orbital-deformation (MOD) method. No. No localization. 0 1 2 3 Default. If C1. No.

MOs are canonical HF orbitals. IOp(8/44) K-point specification for MP2 band correction. Yes. N-th occupied band at the k-point for which the hoco is highest (by default) or at k-point N-th virtual band at the k-point for which the luco is lowest (by default) or at k-point specified specified by IOp(44) by IOp(44) IOp(8/46) Indicates special case of non-HF calculation. Input orbitals are not canonical HF and pseudocanonical orbitals must be generated here for the 124 . K-point at which to apply correction. IOp(8/45) Type of quasiparticle job: 0 -1 1 -N N Band gap. 0 N Use the k-point for which the hoco is highest and luco is lowest. (for SAC-CI single point calculation) Use orbital energies in ordering Don't use orbital energies in ordering Use second moments in ordering Don't use second moments in ordering Use dipole moments in ordering Don't use dipole moments in ordering IOp(8/43) Number of Laplace points to use. Default . Use N points and set up for band gap correction. N -N Use N points for MP2. Electron affinity. Ionization potential. 0 1 post-SCF.1 2 00 10 000 100 0000 1000 Yes.

Full list. Full list. 125 . but blank contributions from inactive atoms.IOp(8/47) Whether 804/811 should generate results compressed over active atoms: 0 1 2 3 Default (2). Active atoms. No difference from 2 for overlay 8.

CI WITH ALL DOUBLE SUBSTITUTIONS. ECID IN CID. AND EUMP4 IN MP4 CALCULATIONS YES. FULL FOURTH ORDER PERTURBATION THEORY IN THE SPACE OF SINGLE.D18. BD. CCSD. FOURTH ORDER PERTURBATION THEORY IN THE SPACE SINGLE. CID. ECISD IN CISD EUMP3 IN MP3. CCD. MP4(SDTQ). DAVIDSON RESTARTS WITH CURRENT EIGENVECTORS AS INITIAL GUESSES. FOURTH ORDER PERTURBATION THEORY IN THE SPACE DOUBLE AND QUADRUPLE SUBSTITUTIONS. WITH THE CORRELATION ENERGY. DOUBLE. WITH EUMP3. IOp(9/6) L913: CRITERIA FOR TERMINATION OF THE ITERATION 0 -1 N DEFAULT CONVERGENCE CRITERION AND MAXCYCLE READ IN MAXCYCLES AND CONVERGENCE CRITERION (I2. IOp(9/7) UPDATE THE ENERGY IN COMMON/GEN/ 0 1 126 YES. MP4(DQ). COUPLED CLUSTER THEORY WITH DOUBLE SUBSTITUTIONS. L914: MAXIMUM NUMBER OF EXPANSION VECTORS IN DAVIDSON SCHEME 0 N 200 VECTORS N VECTORS **** NOTE: WHEN EXPANSION VECTORS EXCEED THE MAXIMUM. MP3. CONFIGURATION INTERACTION WITH ALL SINGLE AND DOUBLE SUBSTITUTIONS. TRIPLE AND QUADRUPLE SUBSTITUTIONS. QCISD. DOUBLE AND QUADRUPLE SUBSTITUTIONS. MP4(SDQ).13) Max N cycles.Overlay 9 IOp(9/5) METHOD 0 1 2 3 4 5 6 7 8 9 CIDS. THIRD ORDER PERTURBATION THEORY. COUPLED CLUSTER THEORY WITH SINGLE AND DOUBLE SUBSTITUTIONS. .

0 Default: L913 single point: 10**-7 energy. IOp(9/10) Test flag in link 902 Whether to do "fake" frozen-core (i. 127 . DD4RQ (closed-shell). L914: State of interest: 0 N WE ARE NOT DOING GRADIENTS. Only active in L914. UMP41R. default 1 in 907 and 10 in 919. L913: Whether to use fast routines: 000 1 2 10 20 30 40 000 100 200 Default (no Slava. with a full transformation). Original code (DD1. 10**-8 energy. same as 1. FP OR CIS-MP2 WE ARE INTERESTED IN THE NTH EXCITED STATE IOp(9/9) Convergence criterion (on energy for L913.3. UMP41. L913 gradient: L914 gradient: N 10**-N. 10**-6 wfn. follow /Orb/. The defaults are 22 for RCI.2.2. 10**-5 wfn. 10**-6 wfn.2 7 YES. Use DD1.3. 11 for UCI. DD4UQ Use DD1R. 42 for RQCI. Slava routines. UMP42. Number of roots in 907 and 919. UMP42.4) for first iteration Use DD[1-3]R and UMP4xR (closed-shell) on 1st iteration Original code for 2nd and later iterations. UMp41U. Use DD[1-3]R and UMP4xR (closed-shell). UMP43. fast and R where possible). Term and method selection for debugging in 906. NO IOp(9/8) L902: Constraint on output wavefunction for stability calculations (see link 902). wavefunction for L914). Default. L914 single point:n: 10**-4 wfn. UMP43.e. and 31 for UQCI. WITH EMP4(SDQ) OR EMP4(DQ) IF SINGLES ARE NOT AVAILABLE. 0 No.. Original routines.

EMax. Include triples? IOp(9/12) Test flag in l902. Force direct computation of contributions. Yes. BD(T). but with two ranges on the same line Force N**3 algorithm in GFSCMA. Read EMin. Fourth-order triples (NYI). and pole strength warning level on one line. but save the amplitudes. Link 909 only. Yes. Test flag in l902. Normal production of intermediates (in-core if possible). Same as 2. for open-shell. Fourth and fifth order singles and triples . and store full /Orb/ back on disk. Read window of MOs to refine in the same format as 801. IOp(9/13) Symmetry constraint of output wavefunction from stable=opt: 0/1 yes/no. IOp(9/11) Flags for Green's function calculations: 0 1 2 00 10 100 1000 10000 Normal use of MO integrals. Force direct computation of contributions. IOp(9/14) Non-iterative corrections: ICNonI 0 1 2 3 128 No. Spin projection control in L913: 0 1 2 Default (1) Do basic projection. note number of frozen core and virtual and reset /Orb/ for full.1 2 3 For AO usage (NYI here). .QCISD(T). Force use of sort for intermediates.

IN THE FORMATION OF A NEW WAVE FUNCTION. L913. otherwise spilling algorithm in Tran4D. IOp(9/18) ITERATION SCHEME: DE= (IN A(S)=W(S)/(DE-DELTA(S)) I. Do a maximum of (-N-6) occupieds per pass. Do Lagrangian in L906. Try to minimize integral evaluations as for -3. requires 2OVN words of memory for E2. -2 -1 0 M Force a single integral evaluation (two for UMP2) using disk-based algorithm. IOp(9/16) L906: Control of (Semi-) Direct MP2: -N -6 -5 -4 -3 to disk. IOp(9/15) Type of derivative information generated: 0 1 2 None. default: in-core if possible. but also force use of the fully out-of-core Force a single integral evaluation as for -2. Default (same as -3) Use disk storage for partially transformed integrals handling M occupieds at once. 2N**3 words for derivatives). (IOp(5)). using the fully out of core allgorithm. included Z-amplitudes if necessary. Force the fully in-core algorithm.E. using fully direct methods if possible. suppress in-core storage. Do gradient in L913. L914: Control of in-core integrals for W(Tilda): -6 -3 0 1 Force in-core storage. FOR METHOD = 129 . IOp(9/17) Auto-adjustment of tau in L918. but also force use of the fully out-of-core algorithm Try to minimize integral evaluations. 0 USE DE DEPENDING ON THE METHOD USED. Use AO integral algorithm (L914 only).4 Same as 2. Force in-memory algorithm (fully direct MP2. Functional to use in L914. in Tran4D. Do AO derivatives and Lagrangian in L906. but do E4T as well.

IOp(9/21) Guess for eigenvector of y-matrix in link 902. 10**-N.5) DE IS NOT REALLY NEEDED SINCE THE WAVE FUNCTION FORMED NEVER GETS USED. QCISD using RLE. IOp(9/20) Whether to update the total energy with the MP2 energy in L901. 0 1 Yes. Use A as guess for Z. Reset RLE for Z iterations. IOp(9/22) Conversion factor in L919. Use RLE. No (used in HF second derivative calculations). 130 .3. Use scaled A as guess for Z. Do not extrapolate. 0 1 2 3 4 5 00 10 100 Default: CI using old extrapolation. Use BFGS.0. ALWAYS. FOR METHOD . IOp(9/23) Localization of orbitals in L919. ALWAYS. Use DIIS.1->. 1 2 W(0)/A0.->DE = 0. 1. -1 0 N Read in factor in format D20.DE = W(0)/A0. IOp(9/19) EXTRAPOLATION. 0.10.4.GT. Use old extrapolation for CI. NOTE THAT FOR PERTURBATION METHODS (METHOD=2. Default of 10**-8.

In-core method. AT EACH CYCLE YES. Localize occupieds. AT FIRST CYCLE AND AT END IOp(9/26) NORMALIZATION OF THE WAVEFUNCTION 0 1 NORMALIZED TO A(0)=1. AT ONE CYCLE GIVEN BY INPUT (I3) YES. Rettrup-Davidson RPA. Choose configurations by simple truncation. Localize virtuals. IOp(9/25) PRINT PAIR CONTRIBUTION AND WEIGHT TO CORRELATION ENERGY 0 1 2 3 4 NO YES. Correction to CIS in L914: 0 -2 -1 1 2 No CIS-DFT (in primitive energy code) CIS-MP2 (in primitive in-core program) CIS-MP2 (in MO Basis disk routine) CIS-DFT (in production code). AT THE END OF CI YES. Read in configurations. Triplet states.0 1 2 3 00 10 20 000 100 0000 1000 00000 10000 None. Out-of-core method. SUM(S) A(S)**2 = 1 (ALL S) NOTE: PERTURBATION THEORETICAL RESULTS ARE VALID WITH NORM=0 ONLY 131 . Localize both. Default (same as 10). The functional is given by IOp(17). Jorgensen-Linderberg Hermetian RPA. Maximum order of perturbation theory in L921 and L922. Singlet states.

Do not print coefficients. The maximum dimension is 25. 132 . 22 otherwise). IOp(9/36) Compute the T1 Diagnostic of T. IOp(9/28) PRINTING OF DOMINANT CONFIGURATIONS. force fully direct method by default. Lee IOp(9/37) Maximum dimension for the QCISD extrapolation.IOp(9/27) Maximum amount of disk to use in L906: -1 0 N No disk. IOp(9/31) Print vectors and matrices in 902 and 918 0/1 no/yes.J. Compute the CI one-particle density matrix. 0 -3 -2 -1 N Default (print coefficients 0. Print all coefficients every iteration.0001*N. Use as much disk as needed for a single pass. IOp(9/30) Calculation of the one-particle density matrices: 00 1 2 10 20 Default (21 for CI. Do not form the CI one-particle density matrix. Scan the 'A' vector and print all coefficients having coefficients greater than 0. N words. MP2 energy). Scan the 'A' vector and print all coefficients. Compute the density correct to second order (NOT the same as the density corresponding to the Do not compute the density correct to second order. Same for DIIS extrapolation.1 and above). For BFGS extrapolation.default size is ten.

the previous job indicates wavefunction not just expansion vectors has converged). CASSCF. OR NUMBER OF STATES TO PRINT OUT INFORMATION FOR WHEN USING DODIAG: 0 N -N DEFAULT TO 2 LOWEST N STATES READ IN PRINCIPLE COMPONENT OF N GUESSES (DAVIDSON). THRESHOLD FOR PRINTING EIGENVECTOR COMPONENTS in L914: 0 N ITHR = 1 ITHR = N WHERE THRESHOLD = GFLOAT(10)**(-ITHR) IOp(9/41) L914: NUMBER OF STATES TO SEEK WHEN USING DAVIDSON. For BFGS extrapolation. IOp(9/40) Reference wavefunction for MP2 in L906: 0 1 2 Default (HF). L914: Pick out guesses from restart file or othogonalize guesses to the states already on restart file (IOp49 must be set to 1 or 2 for this option to be valid) 0 N -1 Just take guess from restart file Make N additional orthogonal guesses to those present Read which N states to use (free format integers) *** WARNING: The states on the restart file MUST be orthogonal to the convergence requested (ie.IOp(9/38) Minimum dimension for the QCISD extrapolation. HF. 1 . the only dimension is IOp(39) L913: Type of convergence test 0 . The maximum dimension is eight.Default: energy and gradient. For DIIS extrapolation.Converge on energy and gradient 3 .Converge on gradient only Convergence on gradient is for extrapolated CI and QCISD procedures. the smallest dimension one and the default is three.Converge on energy only 2 . FORMAT I5 ON LAST CARD BEFORE EOF 133 .

Force Davidson not to half the number of states at iteration 2. DEFAULT IS: 3 (RHF REFERENCE STATE) (UHF REFERENCE STATE) MATRIX AA. BASIS -- IOp(9/43) How to handle subsequent Davidson Iterations in L914: 0 1 2 If this is not a restart. and transitions densities among all excited 134 .BB AB BA AB BA METHOD -|-> FORCE DAVIDSON IN A. AO basis. 222.BB AB BA AA. transition densities from ground. IOp(9/44) Density matrix control for filling RWF 633 in L914: 0 1 2 3 states Same as 2 Do densities of each excited state Do densities and transition densities from ground Do densities.IOp(9/42) METHOD AND MATRIX BLOCKS TO WORK ON in L914 (See below) -NNN 1 2 3 0 333 BITS 1 NYI NYI 2 20 200 3 30 300 Mapped directly to NNN below. MO Mapped to 3. Mapped to 2. a restart.O.BB 10 100 AA. If this is Force Davidson to half the number of states at iteration 2. or 20 as appropriate. In-core. 333.O. then half the number of states at the second iteration. or 30 as appropriate. then don't. BASIS --|-> FORCE DODIAG TO FIND ALL ROOTS --|-> FORCE DAVIDSON IN M.

0 1 Use Phycon to convert to eV's Use old conversion to eV's IOp(9/46) Control of Davidson Convergence in L914: <0 0 N Use Ortvec convergence only Converge on the number of roots .IOp(41) Converge on Ci Amplitudes for N lowest states IOp(9/47) Control of Davidson Iterations in L914: 0 1 2 Usual Don't do any iterations (guess=print) Stop after first iteration IOp(9/48) RESTRICTION ON TYPES OF ROOTS (DAVIDSON RHF ONLY) 0 1 2 3 GUESS ONLY SINGLETS Same as 0 GUESS BOTH SINGLETS AND TRIPLETS GUESS ONLY TRIPLETS NOTE: A SINGLET GUESS MAY RESULT IN A TRIPLET ROOT IN EXTREME CASES (SMALL NUMBER OF ROOTS SOUGHT) IOp(9/49) INITIAL GUESS VECTORS 0 1 2 3 4 MAKE A GUESS BASED ON DIAGONAL ELEMENTS USE GUESS VECTORS ALREADY ON RWF USE GUESS VECTORS ALREADY ON CHK GENERATE GUESSES FROM CIS DENSITIES on CHK GENERATE GUESSES FROM CIS DENSITIES on RWF 135 .IOp(9/45) Debug option for comparing previous results in L914.

Yes. 0 1 2 Default (No). IOp(9/74) Over-ride default choice of frequency dependence of the XC functional in L914: 0 N Use default value. no for stability). IOp(9/70) 1 to force TDHF in L914. use equilibrium. IOp(9/71) Whether to do an extra iteration after Davidson convergence. IOp(9/75) Whether to save amplitudes and integrals in L906: 136 . if doing excited states. IOp(9/73) Whether to do non-equilibrium solvation in L914: 0 1 2 Default (Yes. 0 1 No. Yes.IOp(9/60-62) Over-ride standard values of IRadAn. IOp(9/72) Whether to computed frequency-dependant polarizabilities. No. and IRanGd. Use form N (see IOp(88) in overlay 5). IRanWt. Yes. No.

0 -N N Default . IOp(9/84) Save CBS localized orbitals to RWF (this will overwrite the SCF orbitals. intended for visualization). 0 N Use the default. Save amplitudes and integrals. save them. don't save (default). Population Boys+Population. Use 10**(-N) IOp(9/83) Localization Method.5 (assuming CBS-4 calculations. i. Minimal population. IOp(9/85) Flags for SAC-CI 137 . Get the lowest energy value between CBS(N) and CBS(NVirt). NStart. -1 0 1 2 3 4 5 10 100 No localization. IOp(9/81) Minimum number of Pair Natural Orbitals (PNO) to start the extrapolations from.e. Calculate the extrapolated value at N only. Boys. Save amplitudes. 6-31+G(d'. Yes. Localize core even if not needed.p')). Do 2nd order. No localization. Default (4). 0 1 No. IOp(9/82) Convergence tolerance for CBS localization.0 1 2 Save only if doing second derivatives (SqS12 set).

Yes. No. 138 .IOp(9/86) Whether L906 should generate data compressed to active atoms during mp2 frequencies with ONIOM: 0 1 2 Default (2).

Use new Px/Wx code and don't save S1 but do save F1. D E(CID) / D R CALC. D E(MP4) /D R Default CPHF usage (Z-vector unless HF D2E) Full 3*NAtoms CPHF. D E(MP2) / D R CALC. No contributions to the force constants are done here). Set up for GIAO MP2 calculation. 0 1 2 00 10 20 30 100 1000 10000 20000 No. Do hyperpolarizabilities for second-harmonic generation. Use old Px/Wx digestion code. save as little data as possible. Don't process any derivative terms here.Overlay 10 IOp(10/5) CALCULATION OF FIRST DERIVATIVES OF POST-SCF ENERGIES. Do DFT 3rd derivatives. Compute dipole derivatives using only electric field CPHF and F(x) matrices.e. IOp(10/6) Calculation of the second derivatives of the SCF energy. Test Z-Vector using full CPHF. do D2 E(SCF) / D R(I) D R(J) Setup For MP2 2nd Derivatives (i.just set up here unless doing HF 2nd derivatives simultaneously. Only implemented for closed-shell and UHF. D E(MP3) / D R Calc. 0 1 2 3 4 5 6 7 8 9 00 10 20 30 000 100 200 NO CALC. Yes. D E(CISD) / D R Calc. 139 . Use new Px/Wx code but save both S1 and F1 over MOs. Default derivative processing . Default: use new Px/Wx digestion code if possible. Setup for external processing of W and Z. D E(CCSD/QCISD) / D R Calc. D E(CIS) / D R Calc. Z-Vector method. Available for RHF and UHF. D E(BD) / D R Calc. Partially coded but NYI for high-spin ROHF. Compute F1 and S1 derivative terms here. D E(CCD) / D R Calc.

never invert directly. IOp(10/10) Control of CPMCSCF during avoided crossing/conical intersection searches.D-11 for Z-Vector CPHF. Solve all equations together. Nuclear Lth order. 140 . 0 IJKL Default (1st order nuclear and electric field). The max element is tested against 10* this value. Yes. Nuclear magnetic moment Ith order. except 1. Magnetic Field Jth order.D-9. IOp(10/11) Largest matrix for direct inversion in LinEq2. IOp(10/9) Whether to compute Born-Oppenheimer corrections. Electric field Kth order. IOp(10/13) The nature of the perturbation(s). 0 -1 N Default invert directly if there is enough memory. the secondary solution. 0 N Default: 1. 1. Expand each variable in a separate expansion space. Always use DIIS.D-N. 0 1 No. 0 1 2 3 Default (same as 2).A) contributions.IOp(10/7) RMS CONVERGENCE ON C1(I. N. IOp(10/8) Selection of linear equation solution method. possibly reverting to the old (one variable at a time) method in Invert the A matrix directly.

141 . Do C1. but delete at end of link. but use of electric field expansion vectors for nuclear coordinate CPHF can cause errors of up to 1 cm**-1 with current tolerances. which is used only for one term in polarizability derivatives and for which the accuracy requrirements are less stringent. IOp(10/17) Frozen-core: 0 1 2 3 Default (use AO 2PDM for Lagrangian only if orbitals are frozen in /Orb/). Note that because of numerical instabilities in the simultaneous solution method. but leave the full version of /Orb/ on the disk. C2. otherwise). =2 Save vectors at end. S1. and S2 off the AO 2PDM. Delete vectors at end of each CPHF. Pass vectors from 1st to 2nd order CPHF. IOp(10/16) Convergence in secondary linear equations (only for simultaneous solution). Convergence is 10**(-N) for max and rms. IOp(10/15) What to do with expansion vectors from the linear equations. Update dipole. Don't update dipole Update polarizability. This may be acceptable in the electric field second order CPHF. for debugging frozen-core with integrals over the full window. 0 N Use standard machine tolerance (MDCutO) on maximum and rms. Ignore old vectors.eq. Convert /Orb/ to full. Don't update polarizability.IOp(10/14) Whether to update dipole and polarizability derivatives. Default (Use old vectors if available). reusing old expansion vectors for new B vectors can reduce accuracy. Force 2nd order cphf for polarizability derivatives. 0 1 2 10 20 100 Default (yes if IOp(5). Save as 2.0). 0 1 2 3 00 10 20 Default (=1 if IOp(8)=1 and electric field only and no derivatives are being computed. This option is normally used to pass 1st order electric field results to the second invokation of 1002 during frequency calculations. Use old vectors if available.

Note that the appropriate rwf (588) must be present in any case. Use option MN in control of 2e integral calculation. IOp(10/20) How to handle 2e integral contributions: 0 1 2 3 4 MNx Default (decide on the fly). Only a single set of products AX are computed. Read the 2e integral files.e. The A-matrix is neglected. No. even if MO ones are available. and hence the U-matrices are set equal to the B-matrices (i.e. independent of convergence criteria. The U-matrices are set to zero. Compute the 2e integrals when needed. Simultaneous solution is implied. Spq. and full MO Fock matrix derivatives in permanent rwfs. 142 . MO if possible.. 0 1 2 Default (yes). IOp(10/21) Whether to store Uai. Disables use of symmetry to reduce the size of the CPHF problem here. uncoupled Hartree-Fock is used). IOp(10/19) Whether overlap (S1) terms must be included. i. even if present. force AO or direct. 0 1 2 10 Default (No). Yes. Don't use integrals. Yes. 0 1 2 3 CPHF is done correctly. Force use of AO integrals. No.IOp(10/18) Whether to do correct or approximate CPHF. Save magnetic MO deriviatives.

-N 0 1 N All integrals done as Raffenetti if there are N or more matrices. Nth excited state. IOp(10/28) State for CPMCSCF: 0 N Default (ground state). Yes. all as regular if there are Default: let FoFDir decide. 143 . recover ints if available on rwf 610. No. less than N. Integrals with degree of contraction greater than or equal to N are done are regular integrals. Octopole. 0 1 2 3 4 Default. Force in-core storage. IOp(10/29) Use of rafinetti integrals during direct SCF.IOp(10/22) Which multipole (electric field) perturbations to include? Only used if J part of IOp(13) is non-zero. Quadrupole (electric field gradient. all 6 cartesian components. All integrals are done as regular integrals.do if possible in direct calculation. Dipole (uniform electric field). Force recomputation. Uniform electric field (dipole) only. IOp(10/31) Whether to use symmetry to reduce the number of CPHF equations: 0 1 2 Default (yes). Hexadecapole. IOp(10/30) In-core storage of 2e integrals: 0 1 2 Default .

IOp(10/46) Whether to calculate dipole and rotational strengths (VCD). -1 0 1 2 4 8 16 None. Use single gauge origin . No. and hence the molecule's other magnetic properties. 0 1 2 3 No (Default) Yes No Do only optical rotational strength. Whether to read D2E file in link 1003: 0 1 2 Default (No). Yes. Use IGAIM method . 144 . No. Use single gauge origin . IOp(10/45) Type of Gauge Transformations to perform to calculate the current distribution within the molecule. Default (16 if doing magnetic CPHF). Use GIAOs. Yes.gauge origin coincident with the nucleus of the integrated atomic regions.IOp(10/32) Whether to apply interchange in link 1004: 0 1 2 Default (No). 0 1 2 Default (No) Yes.the coordinates of which are read in (in Angstroms). Use CSGT method. No. IOp(10/47) Whether to do spin-spin coupling constants.the gauge used to calculate the angular momentum perturbed wavefunctions.

Compress. No. read in frequencies. unless both electric and magnetic properties are requested). use more aggressive cutoffs in Xc integration.IOp(10/48) Whether to operate only over perturbations involving active atoms. lower cutoffs suitable only for CPHF/CPKS. IRanWt. Update regardless. 0 1 2 10 20 Default: Use FMM if turned on globally. N. 145 . Yes. but blank contributions for inactive atoms. 0 N Default (1000). compress if overlay 11 did). IOp(10/63) Changing defaults. Use FMM if turned on globally. and IRanGd. Use local. Do not update. Don't compress. Use global cutoffs. calculation is performed. IOp(10/73) Maximum number of CPHF cycles. Yes. Update frequency-dependent property file if frequency-dep. IOp(10/72) Whether to do frequency-dependent properties: 0 1 2 3 4 00 10 20 Default (No. use more aggressive cutoffs in integrals and FMM unless doing NFx. Turn off FMM here regardless. Don't compress. with formalism for frequency-dependent XC response. IOp(10/60-62) Over-ride standard values of IRadAn. 0 1 2 3 Default (For nuclear. Yes.

MOD orbital derivatives. 0 N Use global value for this job step. Solve using DiagD. Canonical MO derivatives. 0 1 2 Default (No). 0 1 2 00 10 20 Default (1). 0 1 2 Default (1). IOp(10/77) Test CPHF results by checking the CPHF equations using the complete MO Fock and density derivatives. Solve with SimEqn. Print tensors and eigenvalues. IOp(10/76) Over-ride general choice of exchange-correlation frequency dependence.IOp(10/74) Whether to do non-equilibrium solvation. Yes. 146 . Print eigenvectors as well. Use type N (see IOp(88) in overlay 5). Yes. IOp(10/78) Whether to solve CPHF equations for MOD method. 0 1 2 Default: Only if frequency-dependent. Default (20). No. IOp(10/75) Print during NMR. No.

for testing restarts. 147 .IOp(10/79) Stop the link at selected points. MNN Stop at pass M (default 1). restart point NN.

Yes. PRODUCE A D2E FILE. IOp(11/7) IFTPDM: WHETHER OR NOT TO CONTRACT INTEGRAL DERIVATIVES WITH A 'READ-IN' TWO-PARTICAL DENSITY-MATRIX. but generate and write out the HF 2PDM here for debugging purposes. YES. IOp(11/9) IDOUT: FIRST-DERIVATIVE OUTPUT OPTION. IOp(11/6) IFHFFX: WHETHER OR NOT TO CONTRACT INTEGRAL DERIVATIVES WITH HARTREE-FOCK DENSITY MATRIX TERMS TO PRODUCE HARTREE-FOCK TWO-ELECTRON CONTRIBUTION TO THE FORCES.Overlay 11 IOp(11/5) IFWRT: DERIVATIVE INTEGRAL WRITE OPTION. Yes. 0 1 No. 0 1 DO NOT PRODUCE A D2E FILE. also contracted electric field density matrix derivatives to form the two-electron integral derivative contribution to the polarizability derivatives. YES. 0 1 2 NO. 0 1 2 NO. I0 148 WHETHER OR NOT TO USE THE CONTENTS OF IRWFX. . Yes. IOp(11/8) IFF1: WHETHER OR NOT TO COMPUTE F1 OVER AO'S. CONTAINS I2*100+I1*10+I0.

ADD IN CONTENTS OF FX2. Lagrangian. 149 . PROCESSING OF TPDM CONTRIBUTIONS. Yes. FoFDir: Prism spdf. Illegal here.0 1 I1 0 1 2 3 I2 0 1 NO. TAKE HF CONTRIBUTIONS FROM FX1 (A LA IFHFFX). MERELY SET THE ARRAY TO ZEROES. Form 1/2(F+H) term in link 1110. Illegal here. Illegal here. TAKE HF CONTRIBUTIONS FROM F1 (A LA IFF1). Illegal here. Illegal here. NONE. Illegal here. Illegal here. IOp(11/10) Whether to compute Fock matrices. YES. PROCESSING OF TWO-ELECTRON HARTREE-FOCK CONTRIBUTIONS. (forms the 1/2(F-H) term in link 1110). use IsAlg to decide. IOp(11/11) Control of integral derivative algorithm: 0 2 3 4 5 6 7 8 9 10 11 12 13 Default Illegal here. Illegal here. Use generalized density number N for both the one-electron integral derivatives and the corresponding 2PDM terms. IOp(11/12) Selection of 1PDM in L1102 and L1110: 0 N Usual SCF density. NONE. IF NOT THERE. Illegal here. Scalar Rys SPDF. and SCF energy in L1110: 0 1 No.

Nuclear Kth order. Force old algorithm.ne.1 calculate nuclear contribution.ne. 0 IJK Default (1st order nuclear and electric field).IOp(11/13) Flags for L1112: 0 1 2 00 10 20 000 100 200 300 0000 1000 2000 Default for Ix==>Sx (same as 1). same as 1. 0 1 10 Default: compute dipole derivative matrices only.ne.ne. Formation of Ux*I*T terms. IOp(11/14) The nature of the perturbation(s). Force O2V2 method. Use (2g+O)V2 memory algorithm even if O2V2 memory is available. Use L(x) and Ux*I. Electric field Jth order. Magnetic Field Ith order. i4 .0 load fxyz from rw-files if it exists.0 calculate one-electron contribution. i3 . Old gOV3 I/O algorithm. N**4 I/O algorithm.eq. Also compute dipole derivative integral contribution to the HF dipole derivatives. IOp(11/15) Controls output of derivatives to rw-files. Formation of Fx*T*T terms: default is to choose based on available memory.0 forces out-of-core algorithm IOp(11/16) Mode of operation of L1102. Also compute HF contribution to the dipole moment. Use Ix. i4*10000+i3*1000+i2*100+1i*10+i0 i0 . 150 . default. i2 .0 controls output of 'old' format. Force old N**5 I/O algorithm. Default Ix*T algorithm (1) Force new algorithm. i1 .

IOp(11/17)

Frozen-core in L1111: 0 1 2 3 10 Default (use AO 2PDM for Lagrangian only if orbitals are frozen in /Orb/). Do C1, C2, S1, and S2 off the AO 2PDM. Convert /Orb/ to full, for debugging frozen-core with integrals over the full window. Save as 2, but leave the full version of /Orb/ on the disk. Form the derivative integral contribution to the Lagrangian as well. This is stored on disk as RL(NBasis,NBasis,NAt3,IOpCl+1) in rwf 1001.

IOp(11/18)

Save AO 2PDM from L1111. 0 N No. Save the AO 2PDM on rwf N. It is (NTT,NTT) and includes factors (2-Delta(ij))(2-Delta(kl)). It doesn't include any normalization factor.

IOp(11/19)

Whether to delete MO integrals after 1112: 0 1 2 Default (Yes). Yes. No.

IOp(11/20)

How to handle 2e integral contributions in L1112: 0 1 2 3 Default (same as 1). Read the 2e integral files, MO if possible. Compute the 2e integrals when needed. This link must have been built with the non-dummied version of FoFDir and associated integral routines. Force use of AO integrals, even if MO ones are available. MNx Use option MN in control of 2e integral calculation.

IOp(11/21)

Size of buffers for integral derivative file. 0 N Default (Machine dependent; see DSet2E). N integer words.

151

IOp(11/22)

In-core option in 1112.

IOp(11/23)

Use of rafinetti integrals during direct term in L1112: -N 0 1 N All integrals done as Raffenetti if there are N or more matrices; all as regular if there are Default: let FoFDir decide. All integrals are done as regular integrals. Integrals with degree of contraction greater than or equal to N are done are regular integrals. less than N.

IOp(11/24)

Output of 1102: 00 1 10 Default (01). Contract with density matrix to form dipole derivative contributions. Store dipole derivative matrices on disk.

IOp(11/26)

PROGRAM ACCURACY OPTION. 0 1 DO INTEGRALS ECOMOMICALLY TO 10**(-10) ACCURACY. 'TEST' OPTION BYPASS CUTOFFS.

IOp(11/27)

INTEGRAL RETENTION PARAMETER. 0 N RETAIN INTEGRALS GE 10**(-10) IN THE D2E FILE (IF SELECTED) AND/OR 10**(-10) IN THE INTEGRAL HEAP IF IFF1=1 AND MODE=2. RETAIN INTEGRALS GE 10**(-N).

IOp(11/28)

Location or generation of MO 1 and 2 PDMs for L1111: -7 -6 -5 152 Compute QCISD 2PDM Compute CCD 2PDM Compute CIS 2PDM

-4 -3 -2 -1 0 N

Compute CISD 2PDM. Compute CID 2PDM. Compute MP2 2PDM. Compute HF DMs. Default (RWFs 626, 627, and 628). RWFS N (1PDM), N+1 (W), and N+2 (2PDM).

IOp(11/29)

What to do: 1 10 100 200 it. 1000 10000 20000 Suppress writing alpha, beta, and spin density rwfs. Form and sort the 2PDM derivatives rather than the 2PDM. Generate replicated 2PDM copies for testing. Transform 1PDM and Lagrangian from MO to AO. Transform 2PDM from MO to AO. Sort AO 2PDM into shell order. If back transformation has not been requested, the double-length AO 2PDM is expected in file 1001. The sorted 2PDM is left in file 602. Form the contribution of the 2PDM to the forces right here. Note that if the 2PDM is also to be left behind, it will be over 6d/10f and have the HGP d and f scale factors in

IOp(11/30)

What to compute using integrals or D2E file. 1 10 Energy. Gradient.

IOp(11/31)

Whether to use symmetry in Rys integral derivatives in L1110: 0 1 Yes. No.

IOp(11/32)

Whether to do 2PDM or just Lagrangian in L1111: 0 1 2 3 4 Compute Full Gradient Compute Full Gradient (Same as Default). Compute Density Only. Compute Density and W Only. Compute 2PDM only, no density or W. 153

Compute over active atoms only. . PRINT COMPUTED FIRST-DERIVATIVES. 0 1 2 154 Default: compressed. 0 1 Evaluate usual integrals. Force compressed form. Neglect four center integrals. Do only overlap and not other 1e integrals. Force expanded form. IOp(11/39) Compression of derivative matrices: 0 1 2 3 Default (2). Compute over the full list of atoms. Compute over the full list of atoms.IOp(11/33) IPRINT IOp(33) 0 1 2 NO PRINTING. but blank contributions for inactive atoms. Evaluate matrices in the NDDO approximation. PRINT OPTION. IOp(11/42) Compressed file formats. Neglect three center one-electron integrals. IOp(11/40) Neglect of integrals (only option 1 works in Overlay 10): 0 1 2 3 10 20 30 Keep all integrals. PRINT F1 MATRICES. Neglect 1e integrals with diatomic differential overlap. Neglect 2e integrals with diatomic differential overlap. Neglect three center two-electron integrals as well. IOp(11/41) NDDO flag.

155 . Save gradients to disk. needed for non-canonical methods. 0 1 N Default. but using the out of core algorithms. IOp(11/46) Whether to include orbital rotation gradient terms for SAC-CI. IOp(11/45) Force NAt3 instead of NAt3+3 storage of matrices (for debugging): 0 1 No. 0 1 Use global default. Yes. IOp(11/60-62) Over-ride standard values of IRadAn. and IRanGd. Yes. IOp(11/43) Batching in overlay 11. For Rys in L1110. IOp(11/63) Whether to do FMM. Do at least N passes. Do at least one pass. Convert 1PDM to canonical representation. Turn off FMM here regardless.3 Compressed Sx but separate H1 and F1. N is 0/1/2 for default/in-core/out-of-core. IRanWt. IOp(11/53) Convert forces over shells to field-dependent dipole and forces over atoms (for debugging): 0 1 No. 0 1 2 No. smallest possible number of passes.

No. Ignore fitting density and just process real density in L1110.IOp(11/70) Whether to allow cavity to move in PCM derivatives. Print eigenvectors as well. Print tensors and eigenvalues. Only works using 1C shell pairs for the density basis and only with cartesian functions. Copy fitting density over real density. 0 1 2 Default (No). 0 1 2 Default (1). IOp(11/71) Debugging option for DBF derivatives: 0 1 2 Normal processing. IOp(11/75) Print during NMR. Yes. 156 .

if a new version was not generated in this step). Restart a multi-step job. Save data on the chk file. GVB2P5 trans file to unit 14.E. SAVE THE PERMANENT INFORMATION (MOS. 1 2 3 4 5 THE RUN IS NOT AN OPTIMIZATION.) ON THE CHECKPOINT FILE.e. Yes. DELETE THE RESTART INFORMATION IF THE RUN IS FINISHING NORMALLY (I. PolyAtom output in working precision to Fortran unit 8. IF THE ERROR TERMINATION ILSW BIT IS NOT SET).. 0 1 No. SO BOTH THE PERMANENT AND RESTART FILES ARE IN THE CHECKPOINT FILE. Do not write anything to the checkpoint file. recovering data from the checkpoint file and figuring out which job step to run next and whether it needs restart if an optimization or numerical frequency.Overlay 9999 IOp(9999/5) CONTROLS HANDLING OF THE CHECKPOINT FILE: 0 THE RUN IS AN OPTIMIZATION OR FREQUENCY RUN. but don't remove extra data (i. WFN file output with GIAO magnetic orbital derivatives. No GVB2P5 trans file. Archive data from the checkpoint file. BASIS SET INFO ETC. 157 . Use natural orbitals in WFN file IOp(9999/7) Controls whether MOs are written to the polyatom integral tape in LANL style. WFN file output WFN file output with magnetic orbital derivatives. IOp(9999/6) Controls output of Fortran unformatted files for other programs: 0 1 00 10 100 200 300 1000 No PolyAtom output.

An input deck for HONDO. use defaults.. in the standard orientation. 158 . Read a list of atoms to use in the Pickett input. New format.3D20. A GAMESS input deck.e. No.12). IOp(9999/9) Controls archiving of dipole moment and other electic field derivatives. 0 1 2 Archive all as is. and isotopes during multi-step energy calculations: 0 1 2 Default (same as 1). Derivatives (forces and force constants) in format (2X. 0 1 2 4 8 16 32 64 128 256 512 1024 2048 Nothing. pressure.3D20. but rotates to z-matrix orientation first. Archive all. Both. in format suitable for guess=cards. The molecular orbitals. Old format. IOp(9999/10) Controls punching of assorted information (i. except for archiving from the chk file. This is independant of normal archiving to the main file. Use natural orbitals in WFN file. Don't archive. Atomic numbers and coordinates in format (I3.12). IOp(9999/11) Which type of database to update: 0 1 2 3 Default (3). orientation.IOp(9999/8) Reading temperature. The natural orbitals generated by link 601. Output hyperfine tensors as input to Pickett's program (sent to the output file). formatted output to unit 7). These are in the Z-matrix The archive entry. Title. Yes. A WFN file for PROAIMS.

Yes. and update rwfs. Yes. Yes. No. 159 . IOp(9999/18) How many virtual orbitals to include in the WFN file. Go back to Link 1. remove /ZMat/ and /ZSubst/ from the rwf and chk files. No. IOp(9999/14) Whether to attempt to express the final optimized structure in terms of the input z-matrix: 0 1 2 3 Yes if there are 20 or fewer atoms. IOp(9999/16) Treat the job as type (Info(7)) given by IOp(16). IOp(9999/17) Treat as MSJDon=IOp(17) step in a multi-step job. IOp(9999/13) Whether this is the end of the job step: 0 1 2 3 Default (Yes).IOp(9999/12) Flag for coordinate optimization: 0 1 No. IOp(9999/15) Act as though in multi-step job type IOp(15). Yes.

160 . IOp(9999/33) CONTROLS DEBUG PRINT: 0 1 NO DEBUG PRINT. Include all virtual orbitals. Include N virtual orbitals.0 -1 N Default (None). DEBUG PRINT.

- Optimal Unit Commitment Problem Solution Using Real Coded Particle Swarm Optimization Technique (1)
- ENHANCEMENT OF POWER SYSTEM SECURITY USING PSO-NR OPTIMIZATION TECHNIQUE
- Optimal Film
- Optimum Coordination of Overcurrent Relays in Distribution System
- Controlling the Movement of Crowds in Computer Graphics by Using
- Atomic Theory
- AberdeenGroup_StrategicSourcing0415
- Doc
- Quantum Physics Problems
- XANES introduction
- 1-s2.0-S1389128613000832-main
- DK2015_CH01
- Momento
- Opf With Enhanced GA
- rwer ass
- The Transportation Problem
- Test 3 Cheat Sheet
- Democritus
- Jonathan Tennyson and Brian T. Sutcliffe- Discretization to Avoid Singularities in Vibration-Rotation Hamiltonians
- Introduction to Linear Programming
- PSO
- p-frodov
- 1 Atomic Structure Periodic Table - C - Atomic Structure-i
- Periodic Trends Lesson Plan
- Om0010 Operations Management 3rd Sem Mba Spring 2015 Smu Solved Assignment
- Geo Calc
- Sentence corrections practice
- VSEPR Shortcut
- 供应链成熟模型.PDF
- 2thestructureoftheatomicstructure 120131040501 Phpapp02 (Repaired)

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd