摘要:
Paper More reliable partial atomic charges when using diffuse basis sets Jason D. Thompson, James D. Xidos, Timothy M. Sonbuchner, Christopher J. Cramer* and Donald G. Truhlar* Department of Chemistry and Supercomputer Institute, 207 Pleasant St. SE, University of Minnesota, Minneapolis, MN 55455-0431. E-mail: cramer@chem.umn.edu, truhlar@umn.edu; Fax: 612-626-2006, 612-626-9390; Tel: 612-624-0859, 612-624-7555 Received 2nd July 2002, Accepted 30th August 2002 Published on the Web 9th September 2002 We present a method that alleviates some of the sensitivity to the inclusion of diffuse basis functions when calculating partial atomic charges from a Lo�wdin population analysis. This new method locally redistributes that part of the Lo�wdin population that comes from diffuse basis functions so that the final charges closely resemble those calculated without diffuse functions.We call this method the redistributed Lo�wdin population analysis (RLPA). The method contains one parameter for each atomic number, and we optimized the parameter for the 6-311G(d) basis set. The method has been tested on compounds that contain H, Li, C, N, O, F, Si, P, S, Cl, and Br. For a test set of 398 compounds with experimental and high-level theoretical dipole moments, the dipole moments derived from the charges obtained by standard Lo�wdin population analysis have errors 35% larger than those obtained by the corresponding RLPA using the same basis set. In judging the quality of the RLPA with respect to the test set of dipole moments, we have also found that dipole moments derived from Mulliken population analysis have errors 120% larger than those derived from RLPA for the same basis set.The new method is particularly successful for the 207 systems containing only first row atoms (H, C, N, O, F) for which the errors in the dipole moments computed from the partial atomic charges obtained by standard Lo�wdin and Mulliken analysis are respectively 115 and 419% larger than those obtained by RLPA. Introduction Analysis of the electronic wave functions of molecules can provide useful insight into electronic structure and molecular interactions, including electrostatics and solvation, and quantitative analysis of electronic wave functions is very important for predictive molecular modeling.1 One particularly intuitive quantity (at least from the standpoint of a chemist) that can be obtained from the wave function is the partial charge on an atom.However, the partial atomic charge is not a quantum mechanical observable, so there is no unique method to extract it from a wave function. As a consequence, a large number of methods have been proposed to calculate quantum mechanical partial charges.1–47 For example, the electrostatic potential, which is an observable, can be directly calculated from the wave function, and one can calculate partial atomic charges that best reproduce this electrostatic potential in some region of space surrounding the molecule. Methods based on fitting the electrostatic potential are, however, often ill conditioned,29,31 especially for interior atoms in a molecule.Two other popular methods for determining partial charges from the wave function are those proposed by Mulliken2–4 and Lo�wdin.5–8 These two methods offer a prescription for partitioning the electron density into atomic contributions. Mulliken’s method accounts for the non-orthogonality of a set of basis functions by partitioning density arising from the overlap of basis functions centered on different atoms equally among the atoms. Lo� wdin’s method attempts to eliminate the arbitrariness of equal partitioning of overlap densities by transforming to a symmetrically orthogonalized basis before the population analysis is carried out. Both Mulliken’s and Lo� wdin’s prescriptions for assigning partial atomic charges are sensitive to basis set size and, in particular, they often predict unphysical charges when extended basis sets or basis sets that include diffuse functions are used.This is apparently a consequence of the fact that a large or diffuse basis set on a given atom can mathematically describe electron density on neighboring atoms. For example, with a large, diffuse basis set, one can obtain a reasonably accurate electronic wave function for methane even with all of the basis functions centered only on carbon,48 and either Mulliken or Lo�wdin analysis based on such a wave function would assign a partial charge of 24 to carbon and 11 to each of the hydrogen. There are many applications that may require the use of diffuse basis functions, while at the same time requiring reliable partial atomic charges.For example, it has been shown49–51 that density-functional theory (DFT) methods are in many instances more accurate for predicting barrier heights, energies of reactions, and conformational energies when diffuse functions are included. Theoretical predictions of reaction rates, reaction energies, or conformational energies that take account of solute–solvent interactions by self-consistent reaction- field (SCRF) models such as the generalized Born approximation, 52–55 which operationally requires partial atomic charges whose dependence on density-matrix elements can be expressed analytically, may therefore require a method to obtain realistic partial atomic charges from a calculation including diffuse functions in the basis set. It is our goal here to devise a practical scheme to repartition the diffuse part of the electron density in a (modified) Lo�wdin population analysis when the popular 6-311G(d)56–59 basis set is used, such that the resulting charges are similar to Lo�wdin charges calculated from the non-diffuse 6-31G(d)56–58 basis set.These two basis sets differ in that each non-hydrogenic atom has one set of diffuse s and p functions (with the same orbital exponent) in the former but not in the latter. We will also consider a charge-dependent property, the DOI: 10.1039/b206369g PhysChemComm, 2002, 5(18), 117–134 117 This journal is # The Royal Society of Chemistry 2002dipole moment, to further evaluate the quality of these new charges. Theory The Lo�wdin charge, qk (Lo�wdin), on an atom k is5–8 qk(L�wdin)~Zk{X i[k (S1=2PS1=2)ii (1) where Zk is the nuclear charge, P is the electronic density matrix, S1/2 is the square root of the overlap matrix, and the summation index i runs over all basis functions centered on atom k.When the basis set for a particular calculation has diffuse functions, the second term in eqn. (1) can be readily partitioned into a contribution from orthogonalized tight basis functions and a contribution from orthogonalized diffuse basis functions. To correct for the fact that these diffuse basis functions carry excess charge that perhaps should not be assigned to the atom k, we will redistribute this charge to the other atoms based on the diffuseness of the function in question and on the geometry of the molecule. At this point we specialize the formulas to the case where each atom has at most one set of diffuse functions, which have orbital exponents that depend only on the atomic number Z.Let Zk and ak be the atomic number and orbital exponent, respectively, for atom k. The redistributed Lo�wdin population analysis charge, qK(RLPA), on an atom k, which we denote as the RLPA charge, is then given by qk(RLPA)~qk(L�wdin)zQZkYk X atoms k0=k e{akR2 kk0 {X atoms k0=k QZk0Yk0 e{ak0R2 k0k where Yk is the Lo�wdin population that is associated with the diffuse basis functions on atom k, R2 kk0 is the square of the interatomic distance (in atomic units) between atoms k and k’, and QZ is a parameter to be optimized for atomic number Z. The contribution to the population from diffuse functions on atom k is defined by Yk~ X diffuse i[k (S1=2PS1=2)ii where the summation is over diffuse basis functions on atom k.Since hydrogen does not have a diffuse basis function in the 6-311G(d) basis set, the first sum in eqn. (2) is omiitted if k is a hydrogen atom. Parameter optimization To determine the optimal set of QZ parameters, we will use a test set o, O, F, Si, P, S, Cl, and Br. Each compound in the test set has an experimental or high-level theoretical dipole moment, and we will use these dipole moment data to further assess the quality of the RLPA charges. Most of the test set has been explicitly detailed elsewhere; this previously defined test set consisted of 204 compounds that contain H, C, N, O, F, Si, P, S, Cl, and Br41 plus five compounds containing iodine that we removed from the present test set.A major part of the present study is the extension of this test set by adding an additional 194 molecules. These additions are summarized in the following paragraphs, and the entire test set is listed in the appendix. The original test set41 contained three pyramidal amides, formamide, acetamide, and N-methylformamide, and we used an experimental dipole moment for each of these three amides. We have added the three corresponding planar conformers of formamide, acetamide, and N-methylformamide and the pyramidal and planar conformations of N,N-dimethylformamide, E-N-methylacetamide, Z-N-methylacetamide, N,N-dimethylacetamide, and benzamide. For each conformation of each amide, we use theoretical dipole moments, where the theoretical methods are explained below.We have also replaced the experimental dipole moments for the pyramidal conformations formamide, acetamide, and N-methylformamide with theoretical dipole moments. We added 15 new compounds containing at most H, C, and O, all of which have experimental dipole moments, where the experimental data sources are explained below. We added 16 compounds that contain at most H, C, and N. For each one of these compounds, we use an experimental dipole moment. In addition to the amides discussed above, we also added 11 compounds that contain H, C, N, and O, including phenylurea and ten aromatic nitrogen heterocycles. Of these new compounds, nine have experimental dipole moments. We expanded the halogen test set by adding ten fluorinecontaining compounds, 13 chlorine-containing compounds, four bromine-containing compounds, and 19 halogenated bifunctional compounds. All of these new halogen compounds have experimental dipole moments.The sulfur subset of the test set was expanded by adding seven thiols, three sulfides, and 15 other sulfur-containing compounds. Three of the new thiols, all of the new sulfides, and 11 of the other sulfur-containing compounds have experimental dipole moments. We added 15 new phosphorus compounds, all of which have theoretical dipole moments. The silicon set was expanded from six compounds to a total of 36 compounds. The original six compounds in the silicon test set only contained C and H in addition to Si. The new test set for silicon now also contains O, F, and Cl.The new data for Si include eight theoretical dipole moments. The new test set includes lithium-containing compounds, four of which have experimental dipole moments, and 12 of which have theoretical dipole moments. Lastly, we added seven inorganic compounds, namely, nitric oxide, nitrogen dioxide, nitrosyl hydride, Z-diazene, SH, HSSH, and PH. We use theoretical dipole moments for Z-diazene, SH, HSSH, and PH. To summarize, the test set consists of 390 polar compounds and 398 dipole moments. This includes 382 non-amides for which we have experimental dipole moments in 323 cases and theoretical dipole moments in 59 cases. We also have eight amides, and each amide has a theoretical dipole moment for a planar conformation and a pyramidal conformation, resulting in a total of 16 dipole moments for these eight compounds.Our data set of experimental dipole moments has been assembled from several different sources and compilations.60–68 For molecules where an experimental dipole moment is used, except for molecules containing Li and P, mPW1PW9169/ MIDI!70,71 geometries were used. The mPW1PW91 density functional uses a modified version69 of Perdew and Wang’s exchange functional72 and a fraction of Hartree–Fock73 (HF) exchange of 0.25. (We note that the error in GAUSSIAN98 74 for this hybrid functional has been corrected in our version of GAUSSIAN98 as explained elsewhere.51) For amides, HF/MIDI! geometries were used. Both the minimum energy structure, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used.For all other molecules, we used mPW1PW91/MG3S geometries. Here, MG3S denotes the ‘‘modified G3 semidiffuse’’ basis set,51 which is obtained from the MG375 basis set, by deleting diffuse functions on H. Note that the MG3 basis set is also called the G3LargeMP276 basis set. When experimental dipole moments are not available, we use mPW1PW91/MG3S density dipole moments. A density dipole moment is obtained by integrating the electron density (2) (3) 118 PhysChemComm, 2002, 5(18), 117–134and dipole moment operator over all space. The full set of test dipole moments used in this work is given in the appendix. To determine the set of optimal QZ parameter values, we minimized the function, e, which is given by e~X a X k (qa,k(L�wdin){qa,k(RLPA)) 2 (4) with respect to the QZ parameters.In eqn. (4), the summation index a goes over all the molecules in a test set, which is some subset of our entire test set, the summation index k goes over all the atoms in each molecule a, qa,k(Lo�wdin) is the charge on atom k in molecule a derived from a Lo�wdin population analysis of the wave function calculated by mPW1PW91/6-31G(d), and qa,k(RLPA) is the RLPA charge on atom a in molecule k, which is calculated according to eqn. (2) by mPW1PW91/6-311G(d). Once we have determined the set of optimized QZ parameters, we further evaluate the quality of the new RLPA charges by calculating dipole moments from them. In particular, we use our test set of 398 dipole moments as a standard, and use them to test dipole moments obtained from the RLPA method, from standard Lo�wdin population analysis, from Mulliken population analysis, and from the density (i.e., density dipole moments).Results and discussion First the QZ values for C, N, O, and F were optimized by using a 207-molecule subset of our test set, where the subset consists of those molecules with no atom heavier than F. (Note that the 6-311G(d) basis set does not have any diffuse functions on H, and therefore we do not need a parameter for H.) Table 1 shows the values of the optimized QZ parameters. We then froze these values of QZ and attempted to optimize QZ for Li, second-row atoms, and Br by using the remaining molecules in the test set. For these atoms, no significant improvement could be obtained with any positive value of QZ, and so we set QZ 5 0 for Li and atoms heavier than F.Therefore, for these atoms, the first summation may be omitted in eqn. (2). Figures 1–5 show the Lo�wdin charges calculated by mPW1PW91/6-31G(d) and mPW1PW91/6-311G(d) and the RLPA charges calculated by mPW1PW91/6-311G(d) for 1,2-ethanediol, methylamine, the pyramidal conformation of N-methylformamide, fluoromethane, and for 1-chloropropane. Figures 1–4 show that the standard Lo�wdin charges on C, N, O, and F atoms calculated with 6-311G(d) are more negative than the corresponding charges calculated with 6-31G(d). Furthermore, the charges on H atoms are more positive when Table 1 QZ parameter values Z Parameter value 6 (C) 0.123 7 (N) 0.166 8 (O) 0.212 9 (F) 0.332 Fig.1 Partial atomic charges in 1,2-ethanediol. Beside each atom is a column of three numbers. These numbers correspond to the partial atomic charge calculated from Lo�wdin/mPW1PW91/6-31G*, RLPA/mPW1PW91/6-311G*, and Lo�wdin/mPW1PW91/6-311G, respectively. Fig. 2 Partial atomic charges in methylamine. See the caption to Fig. 1 for details. Fig. 3 Partial atomic charges in pyramidal N-methylformamide. See the caption to Fig. 1 for details. Fig. 4 Partial atomic charges in fluoromethane. See the caption to Fig. 1 for details. Fig. 5 Partial atomic charges in 1-chloropropane. See the caption to Fig. 1 for details. PhysChemComm, 2002, 5(18), 117–134 119they are calculated with 6-311G(d) than when they are calculated with 6-31G(d). The redistribution of te population, however, yields charges that agree well with Lo�wdin charges calculated with 6-31G(d).This is also the case for 1-chloropropane, which is shown in Fig. 5. The charges in 1-chloropropane are typical of what we observed for Li, second row atoms, and for Br, namely that the Lo�wdin charge on Cl from 6-311G(d) is not significantly different from the 6-31G(d) charge. Thus, there was no reason to redistribute any of the diffuse population coming from Li, second row atoms, and Br. Table 2 shows the root-mean square (RMS) errors (in debyes) of the density, Mulliken, and Lo�wdin dipole moments calculated by mPW1PW91/6-31G(d) and mPW1PW91/ 6-311G(d) and the RLPA dipole moments calculated by mPW1PW91/6-311G(d) for various subsets of the entire test set.The first row in Table 2 lists the various RMS errors over a subset of compounds that contain at most H, C, N, and O, which is denoted [H, C, N, O]. The rows in Table 2 that have entries of the form [H, C, N, O, F] 1 X correspond to subsets of the test set that contain one or more of the elements H, C, N, O, or F and element X. For all subsets of the test set, the density dipole moments are more accurate than dipole moments derived from Mulliken or Lo�wdin population analysis of the wave function. Table 2 shows that, upon addition of diffuse basis functions, there is no systematic improvement in the density dipole moments, i.e., the RMS error decreases for some subsets of the test set, while it increases for other subsets. When the entire test set is considered, the RMS error of the density derived dipole moments increases slightly when diffuse basis functions are included.Turning to the Mulliken and Lo�wdin methods, Table 2 shows that, when the 6-31G(d) basis set is used, dipole moments derived from the Lo�wdin population analysis are typically more accurate than dipole moments derived from the Mulliken population analysis for most of the subsets listed by a factor of up to about two. When diffuse basis functions are added, the Lo�wdin dipole moments are still up to two times more accurate than the Mulliken dipole moments, with the exception of subsets that contain either Li, Si, S, or P. However, over the test set that does not include Li, the Lo�wdin dipole moments are more accurate than the Mulliken dipole moments by almost a factor of two.Comparing the Lo�wdin and RLPA derived dipole moments, the new RLPA derived dipole moments are approximately 1.1 to 2.4 times more accurate than Lo�wdin dipole moments calculated with the 6-311G(d) basis set. The most substantial improvements afforded by the new method presented here are observed for subsets of the test set that only contain first row atoms. In particular, the third row of Table 2 shows the results for this 207-molecule subset that contains no atoms heavier than F. However, there are still improvements in compounds that contain Li, Si, P, S, Cl, and Br, even though none of the diffuse populations on these atoms are redistributed. Conclusions We have presented a modification to Lo� wdin’s prescription for calculating partial atomic charges when the 6-311G(d) basis set is used.These new charges are less sensitive to the presence of diffuse basis functions than charges from the usual Lo�wdin population analysis. Additionally, dipole moments from these new charges are more accurate than dipole moments obtained from Lo�wdin charges calculated by mPW1PW91/6-311G(d) for a wide range of compounds. In assessing the quality of the new charges on a large test set, we have also confirmed our previous finding41 that, in general, dipole moments obtained from Lo� wdin’s population analysis are more accurate than dipole moments obtained from Mulliken’s population analysis. We are confident that this new method will find utility in a broad range of applications where both the use of extended basis sets and reliable charges are required.Availability of the RLPA method The RLPA method is available in the latest versions of GAMESOL,77 HONDO/S 78,79 and MN-GSM,80 which is an add-on to GAUSSIAN98.74 Acknowledgements The authors are grateful to Paul Winget for contributions to this research project. This work was supported in part by the National Science Foundation under grant no. CHE00-92019 and CHE-9876792 by the U. S. Army Research Office under the Multidisciplinary University Research Initiative (MURI) program, grant no. DAAD19-02-1-0176. Appendix Table 3 gives the dipole moment test set, and Table 4 gives all dipole moments used in this work. Table 2 Root-mean square (RMS) errors in dipole moments (in debyes) calculated by the mPW1PW91 hybrid density functional using electron density, Mulliken population analysis, Lo�wdin population analysis, and redistributed Lo�wdin population analysis Densityc Mulliken PA Lo�wdin PA RLPA Type of compounda No.b 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) [H, C, N, O] 163 0.21 0.30 0.90 2.88 0.54 1.01 0.51 [H, C, N, O] 1 F 44 0.25 0.19 1.10 3.29 0.43 1.83 0.76 First row subset 207 0.22 0.28 0.95 2.97 0.52 1.23 0.57 [H, C, N, O, F] 1 Si 26 0.18 0.15 0.66 0.60 0.57 1.82 1.21 [H, C, N, O, F] 1 P 23 0.19 0.17 1.04 1.14 0.90 1.68 1.32 [H, C, N, O, F] 1 S 42 0.29 0.35 0.81 0.98 0.85 1.25 1.07 [H, C, N, O, F] 1 Cl 49 0.24 0.24 0.78 2.06 0.45 1.08 0.77 [H, C, N, O, F] 1 Br 16 0.23 0.23 0.55 1.45 0.23 1.12 0.96 [H, C, N, O, F] 1 Si 1 Cl 10 0.35 0.29 0.71 0.42 0.32 1.02 0.91 [H, C, N, O, F] 1 P 1 S 7 0.34 0.14 0.78 2.30 0.94 2.26 1.58 [H, C, N, O, F] 1 P 1 Cl 2 0.36 0.26 1.11 0.59 0.86 1.53 1.26 Subtotal 382 0.23 0.27 0.88 2.39 0.59 1.31 0.84 Li compounds 16 0.45 0.11 2.66 3.09 4.55 3.66 3.64 Total 398 0.25 0.27 1.01 2.42 1.08 1.48 1.10 aSee text for a detailed explanation of this column.bNumber of data in the training set for this row. cA ‘‘density’’ dipole is one calculated in the usual way from the electron density |Y|2 as an expectation value of the dipole moment operator. 120 PhysChemComm, 2002, 5(18), 117–134Table 3 Molecules, accurate dipole moments (in debyes, calculated or experimental) and geometries used for the parameterization of the CM3 model. For molecules where an experimental dipole moment is used, except molecules containing Li and P, mPW1PW91/MIDI! geometries were used.For amides, HF/MIDI! geometries were used. Both the minimum, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used (mPW1PW91/MIDI! geometries tended to not be pyramidal, motivating the switch to HF for coverage purposes). For all other molecules, we used mPW1PW91/MG3S geometries. These molecules are phenylurea, uracil, Z-diazene, eight silicon compounds (SiH, SIH3, CH2SiH3, CH3SiH2OH, H2Si(OH)2, H2SiO, H3SiOH, and dichloromethylsilane), formaldoxime, methanesulfonamide, methanesulfonic acid, methyl methanesulfenate, methyl methanesulfinate, CH2SH, CH2CHSH, CH3CH2(SH)CH3, CH3 SSH, CH3S, CH3(CS)CH3, (CH3)3CS, and all molecules containing Li and P Molecule Dipole Reference Geometry Inorganic compounds ammonia 1.472 1 mPW1PW91/MIDI! water 1.855 1 mPW1PW91/MIDI! nitric oxide 0.159 1 mPW1PW91/MIDI! nitrogen dioxide 0.316 1 mPW1PW91/MIDI! nitrosyl hydride 1.620 1 mPW1PW91/MIDI! Z-diazene 2.940 2 mPW1PW91/MG3S SH 0.8052 2 mPW1PW91/MG3S HSSH 1.1645 2 mPW1PW91/MG3S PH 0.4717 2 mPW1PW91/MG3S phosphine 0.574 1 mPW1PW91/MIDI! Alcohols, phenol methanol 1.700 1 mPW1PW91/MIDI! ethanol 1.440 1 mPW1PW91/MIDI! 1,2-ethanediol 2.4107 2 mPW1PW91/MG3S propargyl alcohol 1.130 1 mPW1PW91/MIDI! 2-propanol 1.580 1 mPW1PW91/MIDI! 1-propanol (trans) 1.550 1 mPW1PW91/MIDI! 1-propanol (gauche) 1.580 1 mPW1PW91/MIDI! 1,2-propanediol (CH3 gauche) 2.320 4 mPW1PW91/MIDI! 1,2-propanediol (CH3 anti) 2.568 4 mPW1PW91/MIDI! 1-butanol 1.660 1 mPW1PW91/MIDI! cyclopropanol (gauche) 1.460 4 mPW1PW91/MIDI! cyclobutanol 1.620 4 mPW1PW91/MIDI! phenol 1.224 1 mPW1PW91/MIDI! Ethers vinyl methyl ether (cis) 0.965 1 mPW1PW91/MIDI! methyl propyl ether (trans–trans form) 1.107 1 mPW1PW91/MIDI! dimethyl ether (methoxymethane) 1.300 1 mPW1PW91/MIDI! tetrahydrofuran 1.750 1 mPW1PW91/MIDI! diethyl ether (e! anisole (methoxybenzene) 1.380 1 mPW1PW91/MIDI! tetrahydropyran 1.580 1 mPW1PW91/MIDI! 1,3-dioxane 2.060 1 mPW1PW91/MIDI! 3,4-dihydro-2,4-pyran 1.400 1 mPW1PW91/MIDI! oxetane 1.940 1 mPW1PW91/MIDI! 3-methyleneoxetane 1.630 4 mPW1PW91/MIDI! Aldehydes formaldehyde 2.332 1 mPW1PW91/MIDI! ethanal (acetaldehyde) 2.750 1 mPW1PW91/MIDI! propanal (cis) 2.520 6 mPW1PW91/MIDI! butanal 2.720 1 mPW1PW91/MIDI! E-2-butenal 3.670 1 mPW1PW91/MIDI! Ketones propanone (acetone) 2.880 1 mPW1PW91/MIDI! 2-butanone 2.779 1 mPW1PW91/MIDI! cyclopentanone 3.300 1 mPW1PW91/MIDI! methyl phenyl ketone (acetophenone) 3.020 1 mPW1PW91/MIDI! cyclobutane-1,2-dione 3.831 4 mPW1PW91/MIDI! cyclobutanone 2.890 1 mPW1PW91/MIDI! cyclopropanone 2.670 1 mPW1PW91/MIDI! cyclopentadienone 3.132 1 mPW1PW91/MIDI! 4-cyclopentene-1,3-dione 1.680 4 mPW1PW91/MIDI! 3-cyclopentenone 2.790 1 mPW1PW91/MIDI! cyclohexanone 2.870 4 mPW1PW91/MIDI! Carboxylic acids s-cis-formic acid 1.425 4 mPW1PW91/MIDI! s-trans-formic acid 3.787 1 mPW1PW91/MIDI! ethanoic acid (acetic acid) 1.700 1 mPW1PW91/MIDI! formylformic acid (trans) 1.860 4 mPW1PW91/MIDI! propanoic acid (cis) 1.460 1 mPW1PW91/MIDI! acrylic acid ((C–C)-s-cis) 1.460 4 mPW1PW91/MIDI! acrylic acid ((C–C)-s-trans) 2.020 1 mPW1PW91/MIDI! 2-methoxyethanoic acid (gauche) 4.720 4 mPW1PW91/MIDI! acetoacetic acid 2.300 4 mPW1PW91/MIDI! Esters methyl ethanoate (methyl acetate) 1.720 1 mPW1PW91/MIDI! methyl methanoate (methyl formate) 1.770 1 mPW1PW91/MIDI! ethyl methanoate (cis) 1.810 1 mPW1PW91/MIDI! ethyl methanoate (trans) 1.980 1 mPW1PW91/MIDI! PhysChemComm, 2002, 5(18), 117–134 121Table 3 Molecules, accurate dipole moments (in debyes, calculated or experimental) and geometries used for the parameterization of the CM3 model.For molecules where an experimental dipole moment is used, except molecules containing Li and P, mPW1PW91/MIDI! geometries were used. For amides, HF/MIDI! geometries were used.Both the minimum, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used (mPW1PW91/MIDI! geometries tended to not be pyramidal, motivating the switch to HF for coverage purposes). For all other molecules, we used mPW1PW91/MG3S geometries. These molecules are phenylurea, uracil, Z-diazene, eight silicon compounds (SiH, SIH3, CH2 SiH3, CH3 SiH2OH, H2Si(OH)2, H2SiO, H3SiOH, and dichloromethylsilane), formaldoxime, methanesulfonamide, methanesulfonic acid, methyl methanesulfenate, methyl methanesulfinate, CH2SH, CH2CHSH, CH3CH2(SH)CH3, CH3 SSH, CH3S, CH3(CS)CH3, (CH3)3CS, and all molecules containing Li and P (Continued) Molecule Dipole Reference Geometry ethyl ethanoate (ethyl acetate) 1.780 1 mPW1PW91/MIDI! pentyl formate 1.900 1 mPW1PW91/MIDI! Other C, H, O compounds diethyl carbonate 1.100 1 mPW1PW91/MIDI! maleic anhydride 3.940 7 mPW1PW91/MIDI! b-propiolactone 4.180 1 mPW1PW91/MIDI! c-butyrolactone 4.270 1 mPW1PW91/MIDI! 2-methoxyethanol (gauche) 2.360 1 mPW1PW91/MIDI! benzyl alcohol 1.710 1 mPW1PW91/MIDI! fulvene 0.424 1 mPW1PW91/MIDI! diketene 3.530 1 mPW1PW91/MIDI! 3-oxetanone 0.887 1 mPW1PW91/MIDI! 2(5H)-furanone 4.905 4 mPW1PW91/MIDI! ketene 1.422 1 mPW1PW91/MIDI! methylketene 1.790 6 mPW1PW91/MIDI! Aliphatic amines, aniline methylamine 1.310 1 mPW1PW91/MIDI! dimethylamine 1.010 1 mPW1PW91/MIDI! ethylamine 1.304 1 mPW1PW91/MIDI! aziridine 1.890 3 mPW1PW91/MIDI! trimethylamine 0.612 1 mPW1PW91/MIDI! 2-aminopropane 1.190 1 mPW1PW91/MIDI! cyclopropyl amine 1.190 1 mPW1PW91/MIDI! propylamine 1.170 1 mPW1PW91/MIDI! piperidine (NH equatorial) 0.820 1 mPW1PW91/MIDI! piperidine (NH axial) 1.190 1 mPW1PW91/MIDI! 1,2,5,6-tetrahydropyridine (NH axial) 1.007 4 mPW1PW91/MIDI! 1,2,5,6-tetrahydropyridine (NH equatorial) 0.990 4 mPW1PW91/MIDI! aniline (NH2 pyramidal) 1.130 1 mPW1PW91/MIDI! Aromatic nitrogen heterocycles pyridine (azabenzene) 2.215 1 mPW1PW91/MIDI! 2-methylpyridine (2-methylazabenzene) 1.850 1 mPW1PW91/MIDI! 4-methylpyridine (4-methylazabenzene) 2.700 1 mPW1PW91/MIDI! pyrrole 1.767 1 mPW1PW91/MIDI! indole 2.100 7 mPW1PW91/MIDI! quinoline 2.290 1 mPW1PW91/MIDI! isoquinoline 2.730 1 mPW1PW91/MIDI! pyrimidine 2.334 1 mPW1PW91/MIDI! 2-methylpyrimidine 1.676 1 mPW1PW91/MIDI! imidazole 3.800 1 mPW1PW91/MIDI! pyridazine 4.220 1 mPW1PW91/MIDI! Nitriles hydrogen cyanide 2.985 1 mPW1PW91/MIDI! ethanonitrile (acetonitrile) 3.925 1 mPW1PW91/MIDI! dicyanomethane 3.730 6 mPW1PW91/MIDI! propanonitrile (propionitrile) 4.050 1 mPW1PW91/MIDI! butanonitrile (anti) 3.910 1 mPW1PW91/MIDI! butanonitrile (gauche) 3.730 1 mPW1PW91/MIDI! cyclopropane carbonitrile 4.131 4 mPW1PW91/MIDI! isobuteronitrile 4.290 1 mPW1PW91/MIDI! cyclobutane carbonitrile 4.040 5 mPW1PW91/MIDI! t-butyl cyanide 3.950 1 mPW1PW91/MIDI! pentanenitrile 4.120 1 mPW1PW91/MIDI! benzonitrile (phenyl cyanide) 4.180 1 mPW1PW91/MIDI! Imines propyleneimine (cis) 1.770 1 mPW1PW91/MIDI! propyleneimine (trans) 1.570 1 mPW1PW91/MIDI! Z-acetaldimine 2.058 4 mPW1PW91/MIDI! E-acetaldimine 2.560 4 mPW1PW91/MIDI! Z-N-ethylidene methanamine 1.498 5 mPW1PW91/MIDI! N-methylformaldimine 1.530 3 mPW1PW91/MIDI! Other C, H, N compounds diazomethane 1.500 1 mPW1PW91/MIDI! methyl azide 2.170 1 mPW1PW91/MIDI! 2-butenenitrile (trans) 4.750 5 mPW1PW91/MIDI! Z-2-butenenitrile 4.080 3 mPW1PW91/MIDI! cyanoacetylene 3.732 1 mPW1PW91/MIDI! acrylonitrile 3.920 1 mPW1PW91/MIDI! methacrylonitrile 3.690 1 mPW1PW91/MIDI! 2-cyanopyridine 5.780 1 mPW1PW91/MIDI! 122 PhysChemComm, 2002, 5(18), 117–134Table 3 Molecules, accurate dipole moments (in debyes, calculated or experimental) and geometries used for the parameterization of the CM3 model. For molecules where an experimental dipole moment is used, except molecules containing Li and P, mPW1PW91/MIDI! geometries were used.For amides, HF/MIDI! geometries were used.Both the minimum, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used (mPW1PW91/MIDI! geometries tended to not be pyramidal, motivating the switch to HF for coverage purposes). For all other molecules, we used mPW1PW91/MG3S geometries. These molecules are phenylurea, uracil, Z-diazene, eight silicon compounds (SiH, SIH3, CH2 SiH3, CH3 SiH2OH, H2Si(OH)2, H2SiO, H3SiOH, and dichloromethylsilane), formaldoxime, methanesulfonamide, methanesulfonic acid, methyl methanesulfenate, methyl methanesulfinate, CH2SH, CH2CHSH, CH3CH2(SH)CH3, CH3 SSH, CH3S, CH3(CS)CH3, (CH3)3CS, and all molecules containing Li and P (Continued) Molecule Dipole Reference Geometry 3-cyanopyridine 3.660 1 mPW1PW91/MIDI! 4-cyanopyridine 1.960 1 mPW1PW91/MIDI! cyanimide 4.280 1 mPW1PW91/MIDI! cyano allene 4.280 4 mPW1PW91/MIDI! cyclopentadiene-1-carbonitrile 4.250 4 mPW1PW91/MIDI! methylaminonitrile 2.640 4 mPW1PW91/MIDI! Amides and phenylurea formamide (NH2 planar) 3.955 8 HF/MIDI! formamide (NH2 pyramidal) 3.683 8 HF/MIDI! ethanamide (NH2 planar) 3.908 8 HF/MIDI! ethanamide (NH2 pyramidal) 3.642 8 HF/MIDI! N-methylformamide (NH2 planar) 4.290 8 HF/MIDI! N-methylformamide (NH2 pyramidal) 4.141 8 HF/MIDI! N,N-dimethylformamide (NH2 planar) 4.175 8 HF/MIDI! N,N-dimethylformamide (NH2 pyramidal) 4.113 8 HF/MIDI! E-N-methylacetamide (NH2 pyramidal) 3.991 8 HF/MIDI! E-N-methylacetamide (NH2 planar) 4.147 8 HF/MIDI! Z-N-methylacetamide (NH2 pyramidal) 3.667 8 HF/MIDI! Z-N-methylacetamide (NH2 planar) 3.869 8 HF/MIDI! N,N-dimethylacetamide (NH2 planar) 3.957 8 HF/MIDI! N,N-dimethylacetamide (NH2 pyramidal) 3.773 8 HF/MIDI! benzamide (NH2 planar) 3.624 8 HF/MIDI! benzamide (NH2 pyramidal) 3.393 8 HF/MIDI! phenylurea 3.898 2 mPW1PW91/MG3S Nitrohydrocarbons nitromethane 3.460 1 mPW1PW91/MIDI! nitroethane 3.230 1 mPW1PW91/MIDI! 1-nitropropane 3.660 1 mPW1PW91/MIDI! 2-nitropropane 3.730 1 mPW1PW91/MIDI! nitrobenzene 4.220 1 mPW1PW91/MIDI! Bifunctional H, C, N, O compounds methyl nitrate 3.100 7 mPW1PW91/MIDI! morpholine 1.550 1 mPW1PW91/MIDI! fulminic acid 3.099 1 mPW1PW91/MIDI! 3-iminofuran 1.503 1 mPW1PW91/MIDI! formaldoxime 0.294 2 mPW1PW91/MG3S hydroxylamine 0.590 1 mPW1PW91/MIDI! isoxazole 2.950 1 mPW1PW91/MIDI! 3-imino-2,3-dihydroisoxazole 1.770 4 mPW1PW91/MIDI! dimethylnitramine 4.610 7 mPW1PW91/MIDI! dimethylnitrosamine 4.010 7 mPW1PW91/MIDI! acetyl cyanide 3.450 7 mPW1PW91/MIDI! uracil 4.497 2 mPW1PW91/MG3S Fluorine-containing compounds fluorocyclohexane (axial) 1.810 1 mPW1PW91/MIDI! fluorocyclohexane (equatorial) 2.110 1 mPW1PW91/MIDI! 1,2-difluoroethane (gauche) 2.670 1 mPW1PW91/MIDI! 1-fluoropropane (gauche) 1.900 1 mPW1PW91/MIDI! 1-fluoropropane (trans) 2.050 1 mPW1PW91/MIDI! fluoromethane 1.858 1 mPW1PW91/MIDI! 1,1-difluoroethane 2.270 1 mPW1PW91/MIDI! fluoroethane 1.937 1 mPW1PW91/MIDI! 2-fluoropropane 1.958 1 mPW1PW91/MIDI! difluoromethane 1.979 1 mPW1PW91/MIDI! trifluoromethane 1.652 1 mPW1PW91/MIDI! 1,1,1-trifluoroethane 2.347 1 mPW1PW91/MIDI! pentafluoroethane 1.540 7 mPW1PW91/MIDI! 1,1,1,2,2,3,3-heptafluoropropane 1.620 7 mPW1PW91/MIDI! 2-fluoro-2-methylpropane 1.960 6 mPW1PW91/MIDI! Z-1,2-difluoroethene 2.420 1 mPW1PW91/MIDI! 3-fluoropropene (eclipsed) 1.765 1 mPW1PW91/MIDI! 3-fluoropropene (gauche) 1.939 1 mPW1PW91/MIDI! 1,1-difluoro-1-propene 0.889 1 mPW1PW91/MIDI! fluoroethene 1.468 1 mPW1PW91/MIDI! 2-fluoropropene 1.610 1 mPW1PW91/MIDI! cis-1-fluoro-1-propene 1.460 1 mPW1PW91/MIDI! 1,1-difluoroethene 1.389 1 mPW1PW91/MIDI! difluoroallene 2.070 1 mPW1PW91/MIDI! trifluoroethene 1.320 1 mPW1PW91/MIDI! 3,3,3-trifluoropropene 2.450 1 mPW1PW91/MIDI! PhysChemComm, 2002, 5(18), 117–134 123Table 3 Molecules, accurate dipole moments (in debyes, calculated or experimental) and geometries used for the parameterization of the CM3 model.For molecules where an experimental dipole moment is used, except molecules containing Li and P, mPW1PW91/MIDI! geometries were used. For amides, HF/MIDI! geometries were used. Both the minimum, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used (mPW1PW91/MIDI! geometries tended to not be pyramidal, motivating the switch to HF for coverage purposes). For all other molecules, we used mPW1PW91/MG3S geometries. These molecules are phenylurea, uracil, Z-diazene, eight silicon compounds (SiH, SIH3, CH2 SiH3, CH3 SiH2OH, H2Si(OH)2, H2SiO, H3SiOH, and dichloromethylsilane), formaldoxime, methanesulfonamide, methanesulfonic acid, methyl methanesulfenate, methyl methanesulfinate, CH2SH, CH2CHSH, CH3CH2(SH)CH3, CH3 SSH, CH3S, CH3(CS)CH3, (CH3)3CS, and all molecules containing Li and P (Continued) Molecule Dipole Reference Geometry fluoroallene 1.970 4 mPW1PW91/MIDI! fluorobenzene 1.600 1 mPW1PW91/MIDI! (trifluoromethyl)benzene 2.860 1 mPW1PW91/MIDI! o-fluorotoluene 1.370 1 mPW1PW91/MIDI! m-fluorotoluene 1.860 1 mPW1PW91/MIDI! p-fluorotoluene 2.000 1 mPW1PW91/MIDI! 1,3-difluorobenzene 1.510 1 mPW1PW91/MIDI! 1,2,3,4–tetrafluorobenzene 2.420 1 mPW1PW91/MIDI! 1,2-difluorobenzene 2.460 1 mPW1PW91/MIDI! fluoroacetylene 0.721 1 mPW1PW91/MIDI! acetyl fluoride 2.960 1 mPW1PW91/MIDI! 3,3,3-trifluoropropyne 2.317 1 mPW1PW91/MIDI! tetrafluoropropyne 1.710 4 mPW1PW91/MIDI! Chlorine-containing compounds chloromethane (methyl chloride) 1.893 1 mPW1PW91/MIDI! dichloromethane 1.600 1 mPW1PW91/MIDI! trichloromethane (chloroform) 1.040 1 mPW1PW91/MIDI! chloroethane 2.050 1 mPW1PW91/MIDI! 1,1-dichloroethane 2.060 1 mPW1PW91/MIDI! 1,1,1-trichloroethane 1.755 1 mPW1PW91/MIDI! pentachloroethane 0.920 1 mPW1PW91/MIDI! cyclopropylchloride 1.780 7 mPW1PW91/MIDI! 1-chloropropane (gauche) 2.020 1 mPW1PW91/MIDI! 1-chloropropane (trans) 1.950 1 mPW1PW91/MIDI! 1,3-dichloropropane 2.080 1 mPW1PW91/MIDI! 2-chloropropane (isopropyl chloride) 2.170 1 mPW1PW91/MIDI! 1-chloro-2-methylpropane 2.000 1 mPW1PW91/MIDI! 2-chlorobutane 2.040 1 mPW1PW91/MIDI! 1-chlorobutane 2.050 1 mPW1PW91/MIDI! 2-chloro-2-methylpropane 2.130 1 mPW1PW91/MIDI! 1-chloropentane 2.160 1 mPW1PW91/MIDI! chlorocyclohexane (axial) 1.910 1 mPW1PW91/MIDI! chlorocyclohexane (equatorial) 2.440 1 mPW1PW91/MIDI! chloroethene (vinyl chloride) 1.450 1 mPW1PW91/MIDI! 1,1-dichloroethene 1.340 1 mPW1PW91/MIDI! Z-1,2-dichloroethene 1.900 1 mPW1PW91/MIDI! Z-1-chloropropene 1.670 1 mPW1PW91/MIDI! E-1-chloropropene 1.970 1 mPW1PW91/MIDI! 2-chloropropene 1.647 1 mPW1PW91/MIDI! 3-chloropropene (allyl chloride) 1.940 1 mPW1PW91/MIDI! chlorobenzene 1.690 1 mPW1PW91/MIDI! 1,2-dichlorobenzene 2.500 1 mPW1PW91/MIDI! 1,3-dichlorobenzene 1.720 1 mPW1PW91/MIDI! o-chlorotoluene 1.560 1 mPW1PW91/MIDI! p-chlorotoluene 2.210 1 mPW1PW91/MIDI! chloroacetylene 0.444 1 mPW1PW91/MIDI! 3-chloropropyne 1.680 3 mPW1PW91/MIDI! Bromine-containing compounds bromomethane 1.820 1 mPW1PW91/MIDI! dibromomethane 1.430 1 mPW1PW91/MIDI! tribromomethane (bromoform) 0.990 1 mPW1PW91/MIDI! bromoethane 2.040 1 mPW1PW91/MIDI! 1-bromopropane 2.180 1 mPW1PW91/MIDI! 2-bromopropane 2.210 1 mPW1PW91/MIDI! 1-bromobutane 2.080 1 mPW1PW91/MIDI! 2-bromobutane 2.230 1 mPW1PW91/MIDI! 1-bromopentane 2.200 1 mPW1PW91/MIDI! 1-bromoheptane 2.160 1 mPW1PW91/MIDI! bromoethene 1.420 1 mPW1PW91/MIDI! bromobenzene 1.700 1 mPW1PW91/MIDI! bromoacetylene 0.230 1 mPW1PW91/MIDI! 3-bromopropyne 1.540 6 mPW1PW91/MIDI! Halogenated bifunctional compounds Bromotrifluoromethane 0.650 1 mPW1PW91/MIDI! Chlorofluoromethane 1.820 1 mPW1PW91/MIDI! Chlorodifluoromethane 1.420 1 mPW1PW91/MIDI! Fluorotrichloromethane 0.460 1 mPW1PW91/MIDI! Chloropentaflouroethane 0.520 1 mPW1PW91/MIDI! 1-chloro-1-fluoroethane 2.068 1 mPW1PW91/MIDI! 1,1-dichloro-2-fluoropropene 2.430 1 mPW1PW91/MIDI! 124 PhysChemComm, 2002, 5(18), 117–134Table 3 Molecules, accurate dipole moments (in debyes, calculated or experimental) and geometries used for the parameterization of the CM3 model.For molecules where an experimental dipole moment is used, except molecules containing Li and P, mPW1PW91/MIDI! geometries were used. For amides, HF/MIDI! geometries were used. Both the minimum, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used (mPW1PW91/MIDI! geometries tended to not be pyramidal, motivating the switch to HF for coverage purposes). For all other molecules, we used mPW1PW91/MG3S geometries. These molecules are phenylurea, uracil, Z-diazene, eight silicon compounds (SiH, SIH3, CH2 SiH3, CH3 SiH2OH, H2Si(OH)2, H2SiO, H3SiOH, and dichloromethylsilane), formaldoxime, methanesulfonamide, methanesulfonic acid, methyl methanesulfenate, methyl methanesulfinate, CH2SH, CH2CHSH, CH3CH2(SH)CH3, CH3 SSH, CH3S, CH3(CS)CH3, (CH3)3CS, and all molecules containing Li and P (Continued) Molecule Dipole Reference Geometry chlorotrifluoromethane 0.500 1 mPW1PW91/MIDI! dibromodifluoromethane 0.660 1 mPW1PW91/MIDI! dichlorofluoromethane 1.290 1 mPW1PW91/MIDI! 1-chloro-1,1-difluoroethane 2.140 7 mPW1PW91/MIDI! acetyl chloride 2.720 1 mPW1PW91/MIDI! carbonyl chloride 1.170 1 mPW1PW91/MIDI! carbonyl fluoride 0.950 1 mPW1PW91/MIDI! p-chlorophenol 2.110 1 mPW1PW91/MIDI! bis(trifluoromethyl)ether 0.540 6 mPW1PW91/MIDI! cyanogen fluoride 2.120 1 mPW1PW91/MIDI! trifluroacetonitrile 1.262 1 mPW1PW91/MIDI! cyanogen chloride 2.833 1 mPW1PW91/MIDI! 1-chloro-2-nitrobenzene 4.640 1 mPW1PW91/MIDI! 1-chloro-3-nitrobenzene 3.730 1 mPW1PW91/MIDI! 1-chloro-4-nitrobenzene 2.830 1 mPW1PW91/MIDI! 1-fluoro-4-nitrobenzene 2.870 1 mPW1PW91/MIDI! Thiols methanethiol 1.520 1 mPW1PW91/MIDI! ethanethiol (gauche) 1.610 1 mPW1PW91/MIDI! ethanethiol (trans) 1.580 1 mPW1PW91/MIDI! 1-propanethiol (trans) 1.600 1 mPW1PW91/MIDI! 2-methyl-2-propanethiol 1.660 1 mPW1PW91/MIDI! CH2SH 0.911 2 mPW1PW91/MG3S CH2CHSH 0.8258 2 mPW1PW91/MG3S CH3CH2(SH)CH3 1.7389 2 mPW1PW91/MG3S Sulfides, disulfides ethyl methyl sulfide (gauche) 1.593 1 mPW1PW91/MIDI! ethyl methyl sulfide (trans) 1.560 1 mPW1PW91/MIDI! dimethyl sulfide 1.554 1 mPW1PW91/MIDI! dimethyl disulfide 2.092 2 mPW1PW91/MG3S diethyl sulfide 1.540 1 mPW1PW91/MIDI! hydrogen sulfide 0.978 1 mPW1PW91/MIDI! Other sulfur-containing compounds CH3SSH 1.7985 2 mPW1PW91/MG3S CH3S 1.7531 2 mPW1PW91/MG3S CH3(CS)CH3 2.966 2 mPW1PW91/MG3S (CH3)3CS 1.9405 2 mPW1PW91/MG3S 1,3-dithiane 2.240 2 mPW1PW91/MG3S thietane 1.850 1 mPW1PW91/MIDI! thiacyclohexane 1.781 1 mPW1PW91/MIDI! thiophene 0.550 1 mPW1PW91/MIDI! 2-methylthiophene 0.674 1 mPW1PW91/MIDI! 2,5-dihydrothiophene 1.750 1 mPW1PW91/MIDI! 3-methylthiophene 0.914 1 mPW1PW91/MIDI! carbon monosulfide 1.958 1 mPW1PW91/MIDI! methyl-isothiocyanate 3.453 1 mPW1PW91/MIDI! carbon oxysulfide 0.715 1 mPW1PW91/MIDI! thioformaldehyde 1.649 1 mPW1PW91/MIDI! propanethial S-oxide 3.350 9 mPW1PW91/MIDI! thioacetaldehyde 2.330 4 mPW1PW91/MIDI! methylthiocyanate 3.340 7 mPW1PW91/MIDI! thiazole 1.650 7 mPW1PW91/MIDI! methanesulfonic acid 3.8396 2 mPW1PW91/MG3S methyl methanesulfenate 1.9979 2 mPW1PW91/MG3S methyl methanesulfinate 2.7726 2 mPW1PW91/MG3S methanesulfonamide 3.575 2 mPW1PW91/MG3S dicyanogen sulfide 3.020 6 mPW1PW91/MIDI! dimethyl sulfone 4.470 7 mPW1PW91/MIDI! dimethyl sulfoxide 3.960 1 mPW1PW91/MIDI! Phosphorus methylphosphine 1.100 7 mPW1PW91/MG3S dimethylmethylphosphonate 2.272 2 mPW1PW91/MG3S OPH2OH 2.704 2 mPW1PW91/MG3S trimethylphosphine 1.326 2 mPW1PW91/MG3S dimethylphosphine 1.230 6 mPW1PW91/MG3S CH3PH 1.3830 2 mPW1PW91/MG3S CH2PH2 0.5630 2 mPW1PW91/MG3S CH3CH(PH2)CH3 1.3443 2 mPW1PW91/MG3S CH3CH(PH)CH3 1.5127 2 mPW1PW91/MG3S (CH3)3PH2 1.3229 2 mPW1PW91/MG3S PhysChemComm, 2002, 5(18), 117–134 125Table 3 Molecules, accurate dipole moments (in debyes, calculated or experimental) and geometries used for the parameterization of the CM3 model.For molecules where an experimental dipole moment is used, except molecules containing Li and P, mPW1PW91/MIDI! geometries were used. For amides, HF/MIDI! geometries were used. Both the minimum, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used (mPW1PW91/MIDI! geometries tended to not be pyramidal, motivating the switch to HF for coverage purposes). For all other molecules, we used mPW1PW91/MG3S geometries. These molecules are phenylurea, uracil, Z-diazene, eight silicon compounds (SiH, SIH3, CH2 SiH3, CH3 SiH2OH, H2Si(OH)2, H2SiO, H3SiOH, and dichloromethylsilane), formaldoxime, methanesulfonamide, methanesulfonic acid, methyl methanesulfenate, methyl methanesulfinate, CH2SH, CH2CHSH, CH3CH2(SH)CH3, CH3 SSH, CH3S, CH3(CS)CH3, (CH3)3CS, and all molecules containing Li and P (Continued) Molecule Dipole Reference Geometry Multifunctional phosphorous compounds phosphorus nitride 2.747 1 mPW1PW91/MG3S phosphorus oxychloride 2.540 1 mPW1PW91/MG3S hypophosphorous acid 2.708 2 mPW1PW91/MG3S methylphosphonic acid 1.542 2 mPW1PW91/MG3S OPH3 3.710 2 mPW1PW91/MG3S OPH 2.345 2 mPW1PW91/MG3S (CH3)3PO 4.4539 2 mPW1PW91/MG3S PO 2.0266 2 mPW1PW91/MG3S PH2OH 0.707 2 mPW1PW91/MG3S phosphorus trichloride 0.560 1 mPW1PW91/MG3S phosphorus trifluoride 1.030 1 mPW1PW91/MG3S phosphoryl fluoride 1.869 1 mPW1PW91/MG3S FCP 0.279 4 mPW1PW91/MG3S Compounds containing S and P thiophosphorylfluoride 0.640 1 mPW1PW91/MG3S CH3P(O)(OCH3)(SCH3) 1.584 2 mPW1PW91/MG3S CH3P(O)(SCH3)2 1.544 2 mPW1PW91/MG3S OP(OCH3)(SCH3)2 2.8821 2 mPW1PW91/MG3S (CH3)3PS 4.9394 2 mPW1PW91/MG3S PS 0.6133 2 mPW1PW91/MG3S SP(CH3)(OCH3)2 2.5767 2 mPW1PW91/MG3S C, H, Si SiH 0.1404 2 mPW1PW91/MG3S SiH3 0.0075 2 mPW1PW91/MG3S CH2SiH3 0.6685 2 mPW1PW91/MG3S methylsilane 0.735 1 mPW1PW91/MIDI! dimethyl silane 0.713 4 mPW1PW91/MIDI! ethylsilane 0.810 6 mPW1PW91/MIDI! trimethylsilane 0.525 6 mPW1PW91/MIDI! vinylsilane 0.657 1 mPW1PW91/MIDI! phenylsilane 0.845 1 mPW1PW91/MIDI! C, H, O, Si silicon monoxide 3.098 1 mPW1PW91/MIDI! disiloxane 0.240 1 mPW1PW91/MIDI! methylsilyl ether 1.150 1 mPW1PW91/MIDI! ((CH3)3Si)2O 0.780 10 mPW1PW91/MIDI! CH3OSi(CH3)3 1.180 10 mPW1PW91/MIDI! CH3SiH2OH 1.677 2 mPW1PW91/MG3S H2Si(OH)2 0.952 2 mPW1PW91/MG3S H2SiO 3.804 2 mPW1PW91/MG3S H3SiOH 1.315 2 mPW1PW91/MG3S C, H, Si, halogen fluorosilane 1.268 10 mPW1PW91/MIDI! difluorosilane 1.540 10 mPW1PW91/MIDI! trifluorosilane 1.26 10 mPW1PW91/MIDI! fluoromethylsilane 1.710 10 mPW1PW91/MIDI! difluoromethylsilane 2.110 4 mPW1PW91/MIDI! trifluoromethylsilane 2.339 1 mPW1PW91/MIDI! SiH3SiH2F 1.230 3 mPW1PW91/MIDI! 1,1,1-trifluorodisilane 2.030 1 mPW1PW91/MIDI! chlorosilane 1.310 1 mPW1PW91/MIDI! dichlorosilane 1.170 1 mPW1PW91/MIDI! trichlorosilane 0.860 1 mPW1PW91/MIDI! CH3SiH2Cl 1.940 10 mPW1PW91/MIDI! trichloromethylsilane 1.910 1 mPW1PW91/MIDI! CH2ClSiH3 1.626 10 mPW1PW91/MIDI! CH3CH2SiHCl2 2.040 10 mPW1PW91/MIDI! Si(CH3)3Cl 2.080 10 mPW1PW91/MIDI! dichloromethylsilane 2.043 2 mPW1PW91/MG3S Cl3FSi 0.490 10 mPW1PW91/MIDI! Lithium compounds LiCl 7.129 1 mPW1PW91/MG3S LiF 6.327 1 mPW1PW91/MG3S LiNH2 4.853 2 mPW1PW91/MG3S LiO (2II) 6.840 1 mPW1PW91/MG3S LiOH 4.4975 1 mPW1PW91/MG3S LiS 7.017 2 mPW1PW91/MG3S LiSH 6.956 2 mPW1PW91/MG3S LiCH3 5.6778 2 mPW1PW91/MG3S 126 PhysChemComm, 2002, 5(18), 117–134Table 3 Molecules, accurate dipole moments (in debyes, calculated or experimental) and geometries used for the parameterization of the CM3 model.For molecules where an experimental dipole moment is used, except molecules containing Li and P, mPW1PW91/MIDI! geometries were used. For amides, HF/MIDI! geometries were used. Both the minimum, with a pyramidal NH2 group, and the transition state for inversion of the NH2 group through a planar structure were used (mPW1PW91/MIDI! geometries tended to not be pyramidal, motivating the switch to HF for coverage purposes).For all other molecules, we used mPW1PW91/MG3S geometries. These molecules are phenylurea, uracil, Z-diazene, eight silicon compounds (SiH, SIH3, CH2 SiH3, CH3 SiH2OH, H2Si(OH)2, H2SiO, H3SiOH, and dichloromethylsilane), formaldoxime, methanesulfonamide, methanesulfonic acid, methyl methanesulfenate, methyl methanesulfinate, CH2SH, CH2CHSH, CH3CH2(SH)CH3, CH3 SSH, CH3S, CH3(CS)CH3, (CH3)3CS, and all molecules containing Li and P (Continued) Molecule Dipole Reference Geometry LiOCH3 5.2496 2 mPW1PW91/MG3S vinyllithium 5.941 2 mPW1PW91/MG3S LiC3H5 3.767 2 mPW1PW91/MG3S LiC2H 6.061 2 mPW1PW91/MG3S LiC2H2Cl 4.300 2 mPW1PW91/MG3S LiC2H2F 6.910 2 mPW1PW91/MG3S LiC2H6N 5.679 2 mPW1PW91/MG3S LiCH3O 3.676 2 mPW1PW91/MG3S Nonpolar compounds ethene 0 mPW1PW91/MG3S benzene 0 mPW1PW91/MG3S Ref.1: CRC Handbook of Chemistry and Physics, ed. D. R. Lide, CRC Press, Boca Raton, FL, 81st edn., 2000. Ref. 2: mPW1PW91/MG3S calculation. Ref. 3: B. Stark, in Molecular Constants from Microwave Spectroscopy, Landolt-Bo�rnstein, New Series, Group II, ed. K.-H. Hellwege and A. M. Hellwege, Springer-Verlag, Berlin, 1967, vol. 4, p. 136. Ref. 4: B. Stark, in Molecular Constants from Microwave Spectroscopy, Landolt-Bo�rnstein, New Series, Group II, ed. K.-H. Hellwege and A. M. Hellwege, Springer-Verlag, Berlin, 1982, vol. 14a, p. 261. Ref.5: J. Demaison, W. Hu� nter, B. Stark, I. Buck, R. Tischer, M. Winnewisser, in Molecular Constants, Landolt-Bo�rnstein, New Series, Group II, ed. K.-H. Hellwege and A. M. Hellwege, Springer-Verlag, Berlin, 1974, vol. 6, p. 261. Ref. 6: R. D. Nelson, D. R. Lide, A. A. Maryott, Natl. Stand. Ref. Data Ser., United States National Bureau of Standards NSRDS-NBS 10, Washington, DC, 1967. Ref. 7: A. L. McClellan, Tables of Experimental Dipole Moments, W. H. Freeman, San Francisco, CA, 1963. Ref. 8: mPW1PW91/MG3S//HF/MIDI! calculation. Ref. 9: J. Z. Gillies, E. Cotter, C. W. Gilles, H. E. Warner, E. Block, J. Phys. Chem. A, 1999, 103, 4948. Ref. 10: R. J. Abraham, G. H. Grant, J. Comput. Chem. 1988, 9, 244. Table 4 Dipole moments (in debyes) of the test set used in this work Density Mulliken PA Lo�wdin PA RLPA Molecule Accurate 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) Inorganic compounds ammonia 1.47 2.08 2.12 1.97 2.39 1.66 2.02 1.66 water 1.85 2.19 2.35 2.39 2.79 2.10 2.50 2.02 nitric oxide 0.16 0.08 0.09 0.64 0.75 0.09 0.52 0.42 nitrogen dioxide 0.32 0.33 0.36 1.11 0.08 0.51 0.59 0.52 nitrosyl hydride 1.62 1.62 1.76 2.20 1.56 1.49 2.12 1.64 Z-diazene 2.94 3.02 3.27 2.74 2.91 2.10 2.65 2.06 SH 0.81 1.11 1.12 0.75 0.74 0.84 1.03 1.03 HSSH 1.16 1.58 1.59 1.20 1.15 1.14 1.44 1.44 PH 0.47 0.74 0.73 0.02 0.02 0.32 0.52 0.52 phosphine 0.57 0.97 0.90 0.44 0.41 0.61 0.95 0.95 Alcohols and phenol methanol 1.70 1.79 2.00 2.50 2.55 1.92 2.84 1.95 ethanol 1.44 1.64 1.85 2.33 2.50 1.76 2.56 1.68 1,2-ethanediol 2.41 2.40 2.63 3.40 3.19 2.53 3.64 2.52 propargyl alcohol 1.13 1.59 1.71 2.10 6.07 1.58 2.31 1.84 2-propanol 1.58 1.69 1.90 2.37 2.26 1.89 3.13 2.04 1-propanol (trans) 1.55 1.57 1.74 2.29 2.14 1.76 2.78 1.83 1-propanol (gauche) 1.58 1.60 1.75 2.23 2.41 1.60 1,2-propanediol (CH3 gauche) 2.32 2.65 2.84 3.69 3.99 2.81 3.72 2.76 1,2-propanediol (CH3 anti) 2.57 2.87 3.12 4.04 3.46 3.17 4.66 3.42 1-butanol 1.66 1.60 1.79 2.31 3.23 1.74 2.57 1.71 cyclopropanol (gauche) 1.46 1.60 1.76 2.32 2.37 1.67 2.43 1.70 cyclobutanol 1.62 1.55 1.72 2.22 2.56 1.59 2.33 1.60 phenol 1.22 1.39 1.45 1.98 2.40 1.30 1.94 1.46 Ethers vinyl methyl ether 0.97 1.01 1.10 1.72 1.75 1.24 2.52 1.38 methyl propyl ether 1.11 1.19 1.31 2.33 2.18 1.60 2.95 1.70 dimethyl ether 1.30 1.39 1.54 2.49 2.22 1.76 3.11 1.85 tetrahydrofuran 1.75 1.70 1.88 2.68 2.54 1.96 3.21 2.01 diethyl ether 1.15 1.34 1.50 2.33 1.67 1.68 3.10 1.68 anisole 1.38 1.36 1.42 2.25 5.39 1.58 2.61 1.52 tetrahydropyran 1.58 1.47 1.66 2.49 2.65 1.77 3.07 1.85 1,3-dioxane 2.06 2.11 2.36 3.72 3.12 2.64 4.53 2.76 3,4-dihydro-2,4-pyran 1.40 1.32 1.52 2.19 2.40 1.59 2.87 1.64 oxetane 1.94 2.05 2.24 3.09 3.00 2.36 3.58 2.37 3-methylene-oxetane 1.63 1.72 1.86 2.44 1.68 1.95 3.10 2.36 PhysChemComm, 2002, 5(18), 117–134 127Table 4 Dipole moments (in debyes) of the test set used in this work (Continued) Density Mulliken PA Lo�wdin PA RLPA Molecule Accurate 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) Aldehydes formaldehyde 2.33 2.28 2.55 2.77 2.67 1.92 3.15 2.34 ethanal 2.75 2.70 2.98 3.24 2.92 2.43 3.78 2.77 propanal 2.52 2.57 2.83 3.10 3.55 2.30 3.53 2.50 butanal 2.72 2.88 3.18 3.43 2.70 2.61 4.04 2.94 E-2-butenal 3.67 3.96 4.32 4.53 3.27 3.76 5.18 4.13 Ketones propanone 2.88 2.87 3.18 3.47 3.25 2.68 4.29 2.97 2-butanone 2.78 2.74 3.01 3.35 3.43 2.55 3.96 2.70 cyclopentanone 3.30 2.94 3.29 3.52 4.21 2.59 4.15 2.84 methyl phenyl ketone 3.02 3.03 3.27 3.47 5.49 2.68 4.12 3.14 cyclobutane-1,2-dione 3.83 3.73 4.07 4.80 5.64 3.21 5.35 3.84 cyclobutanone 2.89 2.81 3.11 3.40 4.28 2.42 3.96 2.79 cyclopropanone 2.67 2.81 3.09 3.46 5.44 2.29 3.82 2.84 Table 4 Dipole moments (in debyes) of the test set used in this work (Continued) Density Mulliken PA Lo�wdin PA RLPA Molecule Accurate 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) Nitriles hydrogen cyanide 2.99 2.95 3.09 3.51 3.33 1.57 2.34 1.90 ethanonitrile 3.93 3.86 4.07 4.67 3.86 2.80 3.78 2.91 dicyanomethane 3.73 3.77 3.94 4.85 4.64 2.64 3.43 2.58 propanonitrile 4.05 3.96 4.17 4.80 4.68 2.93 4.01 2.93 butanonitrile 3.91 3.93 4.15 4.71 4.98 2.88 3.95 2.77 butanonitrile 3.73 4.10 4.30 5.00 5.46 3.10 4.36 3.21 cyclopropane carbonitrile 4.13 4.25 4.46 5.09 6.89 3.17 4.21 3.16 isobuteronitrile 4.29 3.99 4.20 4.85 5.35 3.01 4.31 2.91 cyclobutane carbonitrile 4.04 4.22 4.44 5.09 7.60 3.20 4.33 3.14 t-butyl cyanide 3.95 4.00 4.21 4.89 5.80 3.07 4.61 2.83 pentanenitrile 4.12 4.20 4.42 5.11 6.71 3.16 4.38 3.28 benzonitrile 4.18 4.59 4.75 6.08 13.76 3.45 4.42 3.60 Imines propyleneimine (cis) 1.77 1.96 2.13 2.22 0.92 1.76 2.72 1.80 propyleneimine (trans) 1.57 1.77 1.92 2.07 2.27 1.46 2.04 1.36 Z-acetaldimine 2.06 2.53 2.79 2.51 1.22 1.94 2.81 1.99 E-acetaldimine 2.56 2.04 2.26 2.13 2.04 1.44 2.13 1.48 Z-N-ethylidene-methanamine 1.50 1.48 1.65 1.63 2.25 0.96 1.81 1.08 N-methylformaldimine 1.53 1.54 1.72 1.62 0.99 1.12 2.10 1.21 Other C, H, N compoundSTable 4 Dipole moments (in debyes) of the test set used in this work (Continued) Density Mulliken PA Lo�wdin PA RLPA Molecule Accurate 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) 1,2-difluoroethane (gauche) 2.67 2.48 2.93 4.40 4.64 3.35 5.27 3.23 1-fluoropropane (gauche) 1.90 1.63 1.93 2.74 3.02 2.17 3.48 2.10 1-fluoropropane (trans) 2.05 1.77 2.08 2.92 2.94 2.38 3.91 2.51 fluoromethane Table 4 Dipole moments (in debyes) of the test set used in this work (Continued) Density Mulliken PA Lo�wdin PA RLPA Molecule Accurate 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) Phosphorus compounds methylphosphine 1.10 1.42 1.38 0.24 0.37 0.24 0.78 0.52 dimethyl methylphosphonate Table 4 Dipole moments (in debyes) of the test set used in this work (Continued) Density Mulliken PA Lo�wdin PA RLPA Molecule Accurate 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) dimethyl sulfone 4.47 4.62 4.89 4.70 5.51 4.92 6.59 5.56 dimethyl sulfoxide 3.96 3.98 4.42 4.68 5.11 5.03 6.31 5.50 C, H, Cl compounds chloromethane 1.89 2.11 2.17 1.83 1.30 1.44 1.60 1.63 dichloromethane 1.60 1.89 1.91 1.66 1.18 1.27 1.28 1.31 trichloromethane 1.04 1.30 1.27 1.17 1.11 0.86 0.78 0.80 chloroethane 2.05 2.28 2.33 1.96 1.33 1.63 1.67 1.69 1,1-dichloroethane 2.06 2.29 2.31 2.08 1.37 1.66 1.65 1.58 1,1,1-trichloroethane 1.76 2.02 2.03 2.02 1.41 1.53 1.54 1.30 pentachloroethane 0.92 1.08 1.03 1.04 1.42 0.73 0.54 0.51 cyclopropyl chloride 1.78 2.06 2.09 1.70 1.57 1.31 1.16 1.13 1-chloropropane (gauche) 2.02 2.23 2.29 1.88 1.49 1.58 1.48 1.55 1-chloropropane References 1 J.Cioslowski, in Encyclopedia of Computational Chemistry, ed. P. v. R. Schleyer, N. L. Allinger, T. Clark, J. Gasteiger, P. A. Kollman, H. F. Schaefer III, Wiley, New York, 1998, vol. 5, p. 892. 2 R. S. Mulliken, J. Chem. Phys., 1935, 3, 564. 3 R. S. Mulliken, J. Chem. Phys., 1955, 23, 1833. 4 R. S. Mulliken, J. Chem. Phys., 1962, 36, 3428. 5 P.-O. Lo�wdin, J.Chem. Phys., 1950, 18, 365. 6 A. Golebiewski and E. Rzescowska, Acta Phys. Pol., 1974, 45, 563. 7 J. Baker, Theor. Chim. Acta, 1985, 68, 221. 8 T. Kar, A. B. Sannigrahi and D. C. Mukherjee, THEOCHEM, 1987, 153, 93. 9 L. C. Cusachs and P. Politzer, Chem. Phys. Lett., 1968, 1, 529. 10 C. A. Coulson, L. B. Redei and D. Stocker, Proc. R. Soc. London, Ser. A, 1962, 270, 357. 11 E. W. Stout, Jr. and P. Politzer, Theor. Chim. Acta, 1968, 12, 379. 12 P. Politzer and R. R. Harris, J. Am. Chem. Soc., 1970, 92, 6451. 13 M. E. Schwartz, C. A. Coulson and L. C. Allen, J. Am. Chem. Soc., 1970, 92, 447. 14 P. Politzer and A. Politzer, J. Am. Chem. Soc., 1973, 95, 5450. 15 W. B. Person and J. H. Newton, J. Chem. Phys., 1974, 61, 1040. 16 W. T. King and G. B. Mast, J. Phys. Chem., 1976, 80, 2521.17 F. L. Hirshfeld, Theor. Chim. Acta, 1977, 44, 129. 18 P. A. Kollman, J. Am. Chem. Soc., 1977, 99, 4875. 19 F. A. Momany, J. Phys. Chem., 1978, 86, 592. 20 S. R. Cox and D. E. Williams, J. Comput. Chem., 1981, 2, 304. 21 U. C. Singh and P. A. Kollman, J. Comput. Chem., 1984, 5, 129. 22 A. E. Reed, R. B. Weinstock and F. Weinhold, J. Chem. Phys., 1985, 83, 735. 23 L. E. Chirlian and M. M. Francl, J. Comput. Chem., 1987, 8, 894. 24 A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899. 25 P. H. Smit, J. L. Derissen and P. B. Van Duijneveldt, Mol. Phys., 1988, 37, 521. 26 B. H. Besler, K. M. Merz and P. A. Kollman, J. Comput. Chem., 1990, 11, 431. 27 C. M. Breneman and K. B. Wiberg, J. Comput. Chem., 1990, 11, 361. 28 K. M. Merz, J.Comput. Chem., 1992, 13, 749. 29 C. I. Bayly, P. Cieplak, W. Cornell and P. A. Kollman, J. Phys. Chem., 1993, 97, 10269. 30 R. Montagnani and J. Tomasi, THEOCHEM, 1993, 279, 131. 31 M. M. Francl, C. Carey, L. E. Chirlian and D. M. Gange, J. Comput. Chem., 1996, 17, 367. 32 B. Beck, T. Clark and R. C. Glen, J. Comput. Chem., 1997, 18, 744. 33 K. B. Wiberg and J. Wendoloski John, J. Am. Chem. Soc., 1978, 100, 723. 34 E. Reed Alan, R. B. Weinstock and F. Weinhold, J. Chem. Phys., 1985, 83, 735. 35 R. F. W. Bader, A. Larouche, C. Gatti, M. T. Carroll, P. J. MacDougall and K. B. Wiberg, J. Chem. Phys., 1987, 87, 1142. 36 J. Cioslowski, P. J. Hay and J. P. Ritchie, J. Phys. Chem., 1990, 94, 148. 37 B. H. Besler, K. M. Merz, Jr. and P. A. Kollman, J. Comput. Chem., 1990, 11, 431.38 J. Mullay, J. Comput. Chem., 1991, 12, 369. 39 J. D. Westbrook, R. M. Levy and K. Krogh-Jespersen, J. Comput. Chem., 1992, 13, 979. 40 K. B. Wiberg and P. R. Rablen, J. Comput. Chem., 1993, 14, 1504. 41 J. W. Storer, D. J. Giesen, C. J. Cramer and D. G. Truhlar, J. Comput.-Aided Mol. Des., 1995, 9, 87. 42 J. Li, T. Zhu, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. A, 1998, 102, 1820. 43 A. E. De Oliveira, P. H. Guadagnini, R. L. A. Haiduke and R. E. Bruns, J. Phys. Chem. A, 1999, 103, 4918. 44 B. Rousseau, A. Peeters and C. Van Alsenoy, THEOCHEM, 2001, 538, 235. 45 P. L. A. Popelier, Theor. Chem. Acc., 2001, 105, 393. 46 P. Bultinck, W. Langenaeker, P. Lahorte, F. De Proft, P. Geerlings, C. Van Alsenoy alenaere, J. Phys. Chem. A, 2002, 106, 7895.47 F. De Proft, C. Van Alsenoy, A. Peeters, W. Langenaeker and P. Geerlings, J. Comput. Chem., 2002, 23, 1198. 48 J. R. Hoyland, J. Chem. Phys., 1967, 47, 3556. 49 B. J. Lynch, P. L. Fast, M. Harris and D. G. Truhlar, J. Phys. Chem. A, 2000, 104, 4811. 50 B. J. Lynch and D. G. Truhlar, J. Phys. Chem. A, 2001, 105, 2936. 51 B. J. Lynch, Y. Zhao and D. G. Truhlar, manuscript in preparation. 52 O. Tapia, in Quantum Theory of Chemical Reactions, ed. R. Daudel, A. Pullman, L. Salem, A. Viellard, Reidel, Dordecht, 1980, vol. 2, p. 56. 53 S. C. Tucker and D. G. Truhlar, Chem. Phys. Lett., 1989, 157, 164. 54 W. C. Still, A. Tempczyk, R. C. Hawley and T. Hendrickson, J. Am. Chem. Soc., 1990, 112, 6127. 55 C. J. Cramer and D. G. Truhlar, J. Am. Chem. Soc., 1991, 113, 8305.56 R. Ditchfield, W. J. Hehre and J. A. Pople, J. Chem. Phys., 1971, 54, 724. 57 W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys., 1972, 56, 2257. 58 P. C. Hariharan and J. A. Pople, Chem. Phys. Lett., 1972, 66, 217. 59 T. Clark, J. Chandrasekhar, G. W. Spitznagel and P. v. R. Schleyer, J. Comput. Chem., 1983, 4, 294. 60 CRC Handbook of Chemistry and Physics, ed. D. R. Lide, CRC Press, Boca Raton, FL, 81st edn., 2000. 61 A. L. McClellan, Tables of Experimental Dipole Moments, W. H. Freeman, San Francisco, CA, 1963. 62 R. D. Nelson, D. R. Lide and A. A. Maryott, Natl. Stand. Ref. Data Ser., United States National Bureau of Standards NSRDSNBS 10, Washington, DC, 1967. 63 B. Stark, in Molecular Constants from Microwave Spectroscopy, Landolt-Bo�rnstein, New Series, Group II, ed.K.-H. Hellwege and A. M. Hellwege, Springer-Verlag, Berlin, 1967, vol. 4, p. 136. Table 4 Dipole moments (in debyes) of the test set used in this work (Continued) Density Mulliken PA Lo�wdin PA RLPA Molecule Accurate 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-31G(d) 6-311G(d) 6-311G(d) Lithium compounds LiCl 7.13 7.09 7.22 4.11 2.88 1.40 2.61 2.61 LiF 6.33 5.79 6.41 4.22 5.08 2.48 4.41 3.81 LiNH2 4.85 4.04 4.66 1.97 1.34 0.29 1.28 1.33 LiO 6.84 6.17 6.69 4.68 4.27 2.39 4.16 3.78 LiOH 4.50 3.80 4.40 2.16 2.05 0.05 1.66 1.65 LiS 7.02 6.97 7.17 3.42 3.09 0.96 2.14 2.14 LiSH 6.96 6.92 7.12 3.59 3.22 1.30 2.34 2.34 LiCH3 5.68 5.15 5.54 2.71 2.14 0.65 1.73 1.72 LiOCH3 5.25 4.79 5.14 2.48 3.14 0.47 1.01 1.72 vinyl lithium 5.94 5.53 5.86 3.24 2.32 1.03 2.45 2.51 LiC3H5 3.77 3.38 3.76 1.43 1.29 1.26 0.25 0.07 LiC2H 6.06 5.63 6.04 3.98 8.78 1.37 3.09 3.12 LiC2H2Cl 4.30 4.23 4.35 2.65 7.31 1.47 0.19 0.08 LiC2H2F 6.91 6.57 6.98 5.44 3.11 2.75 5.07 4.54 LiC2H6N 5.68 5.22 5.61 2.63 2.96 0.47 0.94 1.64 LiCH3O 3.68 3.38 3.53 0.63 1.65 2.32 0.90 0.76 aNH axial.bNH equatorial. cpl refers to the planar conformation. dpyr refers to the pyramidal conformation. PhysChemComm, 2002, 5(18), 117–134 13364 B. Stark, in Molecular Constants from Microwave Spectroscopy, Landolt-Bo�rnstein, New Series, Group II, ed. K.-H. Hellwege and A. M. Hellwege, Springer-Verlag, Berlin, 1982, vol. 14a, p. 261. 65 J. Demaison, W. Hu� nter, B. Stark, I. Buck, R. Tischer and M. Winnewisser, in Molecular Constants, Landolt-Bo�rnstein, New Series, Group II, ed.K.-H. Hellwege and A. M. Hellwege, Springer-Verlag, Berlin, 1974, vol. 6, p. 261. 66 R. J. Abraham and G. H. Grant, J. Comput. Chem., 1988, 9, 244. 67 P. R. Gerber, J. Comput.-Aided Mol. Des., 1988, 12, 37. 68 J. Z. Gillies, E. Cotter, C. W. Gilles, H. E. Warner and E. Block, J. Phys. Chem. A, 1999, 103, 4948. 69 C. Adamo and V. Barone, J. Chem. Phys., 1998, 108, 664. 70 R. E. Easton, D. J. Giesen, A. Welch, C. J. Cramer and D. G. Truhlar, Theor. Chim. Acta, 1996, 93, 281. 71 J. Li, C. J. Cramer and D. G. Truhlar, Theor. Chem. Acc., 1998, 99, 192. 72 J. P. Perdew, J. A. Chevary, S. H. Vosko, K. A. Jackson, M. R. Pederson, D. J. Singh and C. Fiolhais, Phys. Rev. B, 1992, 46, 6671. 73 C. C. J. Roothaan, Rev. Mod. Phys., 1951, 23, 69. 74 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle and J. A. Pople, Gaussian98 (rev. A.11), Gaussian Inc., Pittsburgh, PA, 2002. 75 P. L. Fast, M. L. Sanchez and D. G. Truhlar, Chem. Phys. Lett., 1999, 306, 407. 76 L. A. Curtiss, P. C. Redfern, K. Raghavachari, V. Rassolov and J. A. Pople, J. Chem. Phys., 1999, 110, 4703. 77 J. Li, T. Zhu, G. D. Hawkins, J. D. Thompson, Y.-Y. Chuang, P. L. Fast, J. D. Xidos, D. A. Liotard, D. Rinaldi, C. J. Cramer and D. G. Truhlar, GAMESOL–ver. 3.1, University of Minnesota, Minneapolis, MN, 2001 (based on the General Atomic and Molecular Electronic Structure System, GAMESS, described in the following reference); M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. J. Su, T. L. Windus, M. Dupuis and J. A. Montgomery, J. Comput. Chem., 1993, 14, 1347. 78 M. Dupuis, A. Marquez and E. R. Davidson, HONDO 99.6, 1999 (based on HONDO 95.3); M. Dupuis, A. Marquez and E. R. Davidson, HONDO 95.3, Quantum Chemistry Program Exchange (QCPE), Indiana University, Bloomington, IN, 47405. 79 H. Nakamura, J. D. Xidos, J. D. Thompson, J. Li, G. D. Hawkins, T. Zhu, B. J. Lynch, Y. Volobuev, D. Rinaldi, D. A. Liotard, C. J. Cramer and D. G. Truhlar, HONDO/s–ver. 4.1, University of Minnesota, Minneapolis, MN, 2002 (based on HONDO–ver. 99.6). 80 J. D. Xidos, J. Li, J. D. Thompson, G. D. Hawkins, P. D. Winget, T. Zhu, D. Rinaldi, D. A. Liotard, C. J. Cramer, D. G. Truhlar and M. J. Frisch, MN-GSM, ver. 2.1, University of Minnesota, Minneapolis, MN 55455-0431, 2002. 134 PhysChemComm, 2002, 5(18), 117
ISSN:1460-2733
DOI:10.1039/b206369g
出版商:RSC
年代:2002
数据来源: RSC