An Ab Initio Study of Complexes With N-H-N Hydr.gen Bonds by Barry C. Husowitz Submitted in Partial Fulfillment of the Requirements For the Degree of Master of Science in the Chemistry Program YOUNGSTOWN STATE UNIVERSITY August, 2002 An Ab Initio Study of Complexes With N-H-N Hydrogen Bonds Barry C. Husowitz I hereby release this thesis to the public. I understand this thesis will be made available from the OhioLINK ETD Center and the Maag Library Circulation Desk for public access. I also authorize the University or other individuals to make copies of this thesis as needed for scholarly research. 0..2. Signature: Approvals: - ~c.~ :Bal1)lC usowitz, stUdel1t Dr. Howard Mettee, Committee Member Dr. Timothy Wagner, Committee Member ~/ollo;).. I i Date Date Date ~c.~~c.~ 11 ABSTRACT Ab initio MP2/6-31+G(d,p) calculations have been perlonned on a series of hydrogen bonded complexes stabilized by N-H-N hydrogen bonds. These complexes have 2,5- and 3,4-disubstituted pyrroles as proton donors (with substituents H, F, and Be+ 1 ), and nitrogen bases including HCN, LiCN, NaCN, SCN-, OCN-, NH 3 , and N(CH 3 )3 as proton acceptors. Correlations have been established among the structures, binding energies, proton-stretching frequencies, and intensities of the proton-stretching bands of these complexes. The great majority of complexes are stabilized by traditional N-H...N hydrogen bonds, with proton-shared and ion-pair hydrogen bonds occurring only in charged complexes. iii ACKNO~DGEMENTS I would like to thank my advisor, Dr. Janet E. Del Bene, for her guidance on this project. I am proud to have studied under her. I am grateful for the knowledge she has imparted to me. This knowledge will benefit me for years to come. I am also grateful for the time she has dedicated, and for her patience and understanding. Secondly, I would like to thank the National Science Foundation (NSF) for financial support of this research, and the Ohio Supercomputer Center. Finally, I would like to thank my family and friends for believing in me, and for their constant encouragement throughout my life. This has made it possible for me to make it this far. I would especially like to thank my brother for his encouraging words and constant support. IV TABLE OF CONTENTS PAGE ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... iii TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI LIST OF TABLES vii I. INTRODUCTION. 1 Aims of Study II. METHODS... . 7 Basis sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Minimal Basis Set Double Zeta and Triple Zeta Basis Sets Double-split and Triple-split Valence Basis sets Augmented Basis Wavefunction Models . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Single Determinant Wavefunction Hartree-Fock Method Full Configuration Interactions (Full CI) Configuration Interaction Singles and Doubles (CISD) Coupled Clusters (CC) M¢ller-Plesset Perturbation Theory (MPn) Level of Theory for Studies of Hydrogen-Bonded Complexes 18 Geometry Optimization 18 Monomers Complexes Calculation of Harmonic Vibrational Spectra. . . . . . . . . . . . . . 20 'Reaction Energies 21 Electronic Binding Energies of Hydrogen-Bonded Complexes Binding Enthalpies of Hydrogen-Bonded Complexes Electronic Proton Affinities III. RESULTS AND DISCUSSION . . . . . . . . . . . . . . . ..... 22 Pyrrole and Disubstituted Pyrroles Nitrogen Bases: HCN, LiCN, NaCN, SCN-, OCN-, NH 3 and N(CH 3 )3 Complexes of Pyrrole with HCN and its Derivatives Complexes of 3,4-difluoropyrrole with HCN and its Derivatives Complexes of 2,5-difluoropyrrole with HCN and its Derivatives Complexes of 3,4-diberylliumpyrrole+ 2 with HCN and its Derivatives Complexes of Pyrrole and Disubstituted Pyrroles with NH 3 and N(CH 3 )3 IV. CONCLUSIONS....................... ..... 43 v V. REFERENCES................. 45 VI. APPENDICES Z-matrices for optimized monomers reported in Tables 1 and 2 Z-matrices for optimized complexes reported in Tables 5 and 6 Binding enthalpies(~Ho)for complexes in Tables 5 and 6 VI LIST OF FIGURES FIGURE 1. PAGE Computed harmonic vibrational spectra of ClH:4-R-pyridine complexes 4 2. Na-N b and Na-H distances versus computed MP2/6-31+G(d,p) binding energies of complexes with pyrrole as the proton donor to HCN and its derivatives . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. 31 3. Harmonic vibrational spectra of complexes of pyrrole with HCN and its derivatives 33 4. Na-N b distances and proton-stretching frequencies versus Na-H distances in complexes of pyrrole and disubstituted pyrroles with NH 3 and N(CH 3 h . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 42 vii LIST OF TABLES TABLE 1. PAGE MP2/6-31+G(d,p) electronic energies, zero point vibrational energies, Na-H distances, harmonic proton-stretching frequencies and band intensities for pyrrole and disubstituted pyrroles. . . . . . . . . . " 23 2. MP2/6-31+G(d,p) electronic energies, zero-point vibrational energies, and electronic proton affinities for the nitrogen bases . . . . . . . .. 24 3. MP2/6-31+G(d,p) electronic energies, zero-point vibrational energies, and frequency data for complexes of pyrrole and substituted pyrroles with HCN and its derivatives . . . . . . . . . . . . . . . . . . . " 25 4. MP2/6-31+G(d,p) electronic energies, zero-point vibrational energies, and frequency data for complexes of pyrrole and disubstituted pyrroles with NH 3 and N(CH 3 )3 . . . . . . . . . . . . . . . . . . . . . . 27 5. MP2/6-31+G(d,p) Na-N b and Na-H distances, electronic binding energies, harmonic proton-stretching frequencies and band intensities for complexes of pyrrole and substituted pyrroles with HCN and its derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 6. MP2/6-31+G(d,p) Na-N b and Na-H distances, electronic binding energies, harmonic proton-stretching frequencies and band intensities for complexes of pyrrole and substituted pyrroles with NH 3 and N(CH 3 h 39 1 I. INTRODUCTION: Hydrogen bonding is an important intermolecular interaction, which is ubiquitous in chemical and biochemical systems. Hydrogen bonding influences the properties of water in its various phases and of molecules in aqueous solution. The high boiling points of many solvents are a consequence of hydrogen bonding, and hydrogen bond formation influences products formed in reactions.! In biological systems such as DNA, the formation of nucleic acid base pairs involves hydrogen bonding. 2 Hydrogen bonds determine the structures of proteins 3 , and are a factor in enzymatic activity.4 It has been postulated that low barrier hydrogen bonds can provide stabilization during enzyme catalysis, which results in rate enhancement of the reaction. 4 In the most general case the hydrogen bond can be represented as A-H---B where A-H is the proton donor and B is the proton acceptor atom. In recent studies, three types of hydrogen bonds have been characterized: traditional, ion-pair, and proton shared. s - 7 In a traditional hydrogen bond between two neutral molecules, the A-H covalent bond of the proton donor remains intact in thecomplex~Ifproton transfer from A to B occurs, an ion-pair hydrogen bond is formed. In this type of hydrogen bond a covalent bond is now formed between Wand B, and what was previously the proton acceptor is now the proton donor. Intermediate between traditional and ion-pair hydrogen bonds is the proton-shared hydrogen bond. In this type of hydrogen bond the A-B distance is shorter than in traditional and ion-pair hydrogen bonds, but the A-H and B-H bond lengths are longer than the covalent A-H and B-W bonds in traditional and ion-pair hydrogen bonds, respectively. complex~complex~ 2 The first ab initio theoretical studies of hydrogen-bonded complexes were conducted in the late 1960's and the early 1970'S.8-12 In these studies a single determinant Hartree-Fock wavefunction was used with a small basis set. Geometry optimizations where carried out by freezing the monomer geometries and optimizing the intermolecular distance and intermolecular angles. The intermolecular coordinates were varied cyclicly and independently until convergence criteria were met. It was not until the 1980's that computationally efficient algorithms were developed for obtaining first and second derivatives of the energy with respect to the nuclear coordinates. The derivatives where first evaluated numerically, but later analytically.13-20 As a result, automated full geometry optimizations can now be easily carried out. The availability of analytical derivatives also permits routine calculation of harmonic infra-red (IR) spectra. An important experimental method for studying hydrogen-bonded complexes is IR spectroscopy. The IR spectra of hydrogen-bonded complexes are characterized by a shift to lower frequency of the A-H proton-stretching band compared to that of the corresponding monomer, and by a dramatic increase in the intensity of this band. 7 ,21-23 Ab initio calculations of vibrational spectra can help experimentalists gain insight into the properties of hydrogen-bonded complexes. Many ab initio calculations on various hydrogen-bonded complexes and their IR spectra have been performed. 7 ,22,24,25 Previous studies of hydrogen-bonded complexes that are most closely related to those investigated in this project were carried out by Del Bene, Person, and Szczepaniak on complexes between 4-substituted pyridines and hydrogen halides. 5 ,6 These complexes are represented as X-H·· R 3 where R represents the substituents Be+, CN, F, CI, H, CH 3 , NH 2 , Li, Na, S- and 0-, and X is a halogen atom. The complexes were optimized at the MP2/6-31+G(d,p) level of theory. In this study computed structures and vibrational spectroscopic properties were related to acid and base strength and hydrogen bond type. Figure 1 shows the computed spectra for the CIH:4-R-pyridine complexes. The spectra are arranged in order of increasing proton affinity of the substituted pyridine. The complexes of CIH:4-R-pyridines with R=CN, F, CI, H, CH 3 and NH 2 are stabilized by traditional hydrogen bonds, and a single strong CI-H stretching band appears in the computed harmonic spectra, with frequencies ranging from 2524 cm- I to 2156 em-I. For these complexes, as the proton affinity of the 4-substituted pyridine increases, the CI-N distance decreases, the CI-H distance increases, the binding energy ([I.E) increases, and the frequency (v) ofthe CI-H stretching band decreases. The complexes of CIH:4-Li-pyridine and CIH:4-Na-pyridine are stabilized by proton-shared hydrogen bonds. These complexes have very short Cl-N distances. Several strong bands appear in the computed spectra of these two complexes, with frequencies ranging from 1700 to 600 em-I. The multiple bands are due to coupling of the CI-H stretching mode to ring vibrational modes. As the proton affinity of the base increases further, proton transfer occurs in the complexes CIH:4-S-- pyridine and CIH:4-0--pyridine. The CI-N distance and the proton stretching frequency increase relative to complexes with proton-shared hydrogen bonds, x= eN x ::: F X ::: Cl x ::: H X:: CH3 X ::: II :x =Na X::: s- x::: 0- I 3400 2524 1 1 2434 1 2324 1 2271 1 2636 1 2826 , 2560 OJ • J700 , 850 4 Cl-H....N Cl...H...N } Cl..•.Fi-N o WAVENUMBER (em-i) Figure 1: Computed harmonic vibrational spectra of ClH:4-R-pyridine complexes, from reference 6. 5 and the N-H distance continues to decrease. The single strong band in the IR spectra corresponds to an N-H+ stretch, shifted downfield relative to the corresponding pyridinium ion. In related studies, Del Bene and Jordan examined hydrogen-bonded complexes between hydrogen halides as proton donors and NH 3 and N(CH 3 )3 as proton acceptors. 26 ,27 Examples of traditional, proton-shared and ion-pair hydrogen bonds were found. Systematic studies such as those reported for complexes with hydrogen halides have not been carried out on complexes with N-H-N hydrogen bonds. Therefore, in this project hydrogen-bonded complexes of pyrrole and substituted pyrroles with various nitrogen bases will be investigated. The proton-donating ability and proton-accepting ability of the hydrogen-bonded pair will be systematically varied by chemical substitution in an attempt to span the three hydrogen bond types. Ab initio calculations on isolated pyrrole have been carried out, and optimized structures and harmonic vibrational spectra were reported. 28 ,29 The only study of hydrogen-bonded complexes with pyrrole was reported by Jiang and Tsai. 3o These authors obtained structures and harmonic vibrational frequencies for pyrrole:HF complexes using second-order M(25ller-Plesset perturbation theory and density functional theory with various basis sets. Aims of Study: The current ab initio study focuses on N-H-N hydrogen bonds. Therefore in this work complexes with pyrroles and disubstituted pyrroles as proton donors to a series of nitrogen bases will be investigated. The complexes can be represented as 6 y z N-H"'B a z y where Y may be H, F, or Be+ and Z is H, giving 2,5-disubstituted pyrroles; or Z is H, F, or Be+ and Y is H, giving 3,4-disubstituted pyrroles. The nitrogen bases Bare HCN and its derivatives, LiCN, NaCN, SCN-, and OCN-, as well as NH 3 and N(CH 3 h. The specific aims of this work are 1. to systematically vary the proton-donating and proton-accepting abilities of the hydrogen-bonded pair in an attempt to span the three hydrogen bond types; 2. to determine the optimized MP2/6-31+G(d,p) structures of the monomers and complexes; 3. to calculate the harmonic vibrational spectra of monomers and complexes at the same level of theory; 4. to examine the effect of hydrogen bonding on the frequency and intensity of the proton-stretching band; 5. to correlate structural and spectroscopic properties with binding energies and hydrogen bond type; 7 6. to determine whether proton-shared or ion-pair hydrogen bonds can be formed in neutral complexes with N-H-N hydrogen bonds. II. METHODS: In order to calculate the electronic energy and electron distribution of a molecule, the wavefunction '¥ must be known. Both the energy E, and the wavefunction '¥, can be obtained by solving the nonrelativistic time-independent Schr6dinger equation H'¥ =E '¥ where H is the Hamiltonian operator. The Hamiltonian is defined as H=T+V where T is the kinetic energy operator and V the potential energy operator. The (1) (2) Hamiltonian operator in atomic units for n electrons in a field of m fixed nuclei is written as 31 - 33 1 11. 2 H= --1:: 'V. 2 i 1 11. 11. 1 +1::1:: i<:j f ij 11. In Z. In m Z Z 1:: 1:: _n+ 1:: 1:: _A_B i A f iA A <: B R AB (3) where the first term is the kinetic energy operator, and the remaining terms are the terms of the potential energy operator including the repulsion between each pair of electrons, the nuclear-electron attraction, and the nuclear-nuclear repulsion, respectively. The 8 Schr6dinger equation can be solved exactly only for the hydrogen atom or any one electron system. However, the results of this exact solution, namely the hydrogen atomic orbitals, are useful as a starting point for calculating the electronic distribution of other atoms and molecules which have more than one electron. 33 ,34 In this study, ab initio calculations will be performed to determine 'P and E. Ab initio means, "from first principles", that is, using only the fundamental constants and the atomic numbers of the nuclei, with no other adjustable parameters or data from experiment. However, within the ab initio framework it is necessary to choose a basis set to describe the atomic orbitals, and a wavefunction model. These will now be discussed in general, and the wavefunction model and basis set that will be used in this study will be identified. Basis sets: A basis set is a set of mathematical functions that describe orbitals on atoms. Basis sets are of three general types: minimal, split-valence, and augmented. A minimal basis set contains one basis function to describe each orbital of an atom in the valence shell and below. The advantage of using a minimal basis set is that the number of coefficients to be determined variationally is small, and this reduces the computational problem. However, the disadvantages are that the number of basis functions is not proportional to the number of electrons, anisotropic molecular environments are not correctly represented, and polarization effects are usually not well-described. 3 ! An improvement over a minimal basis set is a double-zeta (DZ) or triple-zeta (TZ) basis set. In a DZ basis set, each orbital in the valence shell and below is 9 represented by two basis functions, while for a TZ basis set each orbital is replaced by three basis functions. These basis sets are now twice and three times larger than a minimal basis set, respectively, and this increases the computational task. A valence double-split basis set is a DZ basis for orbitals in the valence space, but only a single basis function for each orbital below the valence shell. Similarly, a valence triple-split basis set is a TZ basis set for the valence shell, but has only one basis function per inner shell orbital. The most popular double-split valence basis set used as a starting point to construct larger basis sets is the 6-31G basis set,31 which contains one set of inner shell functions for each atomic orbital below the valence shell, and two sets of valence shell atomic functions. For example, for C, the 6-31G basis set has one function to describe the inner shell Is orbital, and two sets of sand p functions to describe the valence shell. The notation 6-31G means 6 gaussian functions are used to describe each inner shell function, 3 gaussians are used to describe each valence shell orbital in the first set, and 1 gaussian is used for each function in the second set. However, these basis functions are still severely limited. 22 ,31 In a molecular environment atomic orbitals are distorted and polarized due to the formation of bonds. To account for this nonuniform displacement of charge away from the nuclear center, basis functions are added with angular momentum quantum number~ one greater than the maximum value found in the valence shell. For carbon this means adding a set of d orbitals(~=2),and for hydrogen a set of p orbitals(~=1).The newly added functions are called polarization functions, and are required to describe anisotropic environments. Adding polarization functions to the 6-31G basis set gives 6-31G(d,p), 10 where one set of six Cartesian d functions are added to each nonhydrogen atom, and one set of p functions to each H atom. Further basis set improvements are made by adding diffuse functions. These are important for describing negatively charged species, lone pairs ofelectrons, and 11: electrons 31 . The diffuse basis functions are of sand p type for nonhydrogen atoms, and s for hydrogen. They have exponents that are considerably smaller than the other valence basis functions and provide a better description of electron density far removed from the nuclear centers. The notation for basis sets augmented with diffuse functions is a "+" symbol for diffuse functions on heavy atoms, and "++" for diffuse functions on hydrogen as well. Thus, the 6-31+G(d,p) basis set is an augmented valence double-split basis set with polarization functions on all atoms and diffuse functions on nonhydrogen atoms. Wavefunction models: Before discussing the wavefunction model in detail it is important to note that 1 1 electrons have spin ( + 2or - 2)' and that electron spin must be incorporated into 'P. The spin angular momentum is represented by a vector s that has the components of sx, Sy and Sz. Since these components follow the commutation relations of general angular momentum, only S2, the magnitude of the vector s and one other component (sx, Sy or sz) can be determined simultaneously. The z-axis component is normally chosen as this component. Thus, a one-electron wavefunction must describe both space and spin coordinates. If 1Vj (i) describes the spatial coordinates of electron i, the spin coordinate is described by31,33 1 Sz a(i) = +"211 a(i) Sz~(i)= -~11~(i) 11 (4) (5) h where 11- 2n and Sz is the operator of the spin angular momentum component along the z- axis. The many-electron wavefunction 'P must also be an eigenfunction ofthe many- electron spin operators S2 and Sz, whose eigenrelations may be written as 33 S2 'P = S(S+1)'P where M s is the sum of the individual spin eigenvalues (m s ). (6) (7) (8) S, the total resultant spin angular momentum, may have any positive half-integral value (0, Yz, 1,3/2,....), and M s is the component of S along the z-axis. For a given resultant spin S there are 2S+1 possible M s values. The quantity (2S+1) is the multiplicity. Thus for total S=O, the multiplicity is one (singlet, nondegenerate state). For S=1I2, (2S+1) is 2 (doublet, doubly-degenerate state), for S=l, 2S+1 is 3, (triplet, triply-degenerate state), and so forth. The total wavefunction 'P must describe both the space and spin coordinates of all the electrons, and must satisfy two requirements. The first requirement is that 'P must be ~(i)~~(i)~(i)~~(i) 12 antisymmetric. Antisymrnetry means that if two electrons are interchanged, then the wavefunction must change sign. The second requirement is that the Pauli exclusion principle must be obeyed. This means that no two electrons can be assigned to identical spinorbitals. These two conditions are most easily satisfied if the wavefunction is represented as a determinant, specifically the Slater determinant, which for an n-electron system with n even and no orbital degeneracies, is - - - l.hO) l.hO) *2(1) *20), ···,··, ljIn/2(l) *n/2(1) * 1(2) *1(2) *2(2) *2(2) ljIn/2(2) *n/2(2) "PO ,2, ... ,n) =(n !/112 - - - ljI 1(n) * 1(n) * zCn) * 2(n) ... .. .. .... ljIn/2 (n) *n/2(n) where '4Ji (a) represents the spinorbital (9) (10) and '4Ji (a) represents the spinorbital (11) In the Slater determinant the first row assigns electron one to all the possible spinorbitals, while the second row assigns electron two to all possible spinorbitals, and so forth. Each column assigns all electrons to a given spinorbital. Thus, the Slater determinant satisfies the Pauli-Exclusion Principle, since if two rows or two columns are identical then the 13 determinant is zero. It also maintains antisymmetry, since if rows i and j are interchanged the sign of the determinant changes. In order to obtain the best single-determinant wavefunction 'P, the Hartree-Fock method is used. 31 - 34 The Hartree-Fock method is a variational method used to determine a set of molecular orbitals (MOs) in terms of a linear combination of atomic orbitals (the LCAO approximation). The LCAO approximation can be written as p llr. = >. c . c.p '+'1 1l';;)).Il~ (12) where 1)ri is the molecular orbital,C~iare the expansion coefficients, and 'PI> 'Pz ,..., 'P p are the atomic basis functions. The expansion coefficientsC~iare determined by the variational method, which minimizes the energy E with respect to the coefficientsC~i. d ---=0 dC~i (13) In carrying out a Hartree-Fock calculation, an initial set of coefficients is chosen and used to construct the Fock matrix. This matrix is diagonalized, and a new set of coefficients is obtained. This process is repeated until the coefficients from two consecutive iterations are identical to within a given tolerance. The Hartree-Fock method is a self-consistent field (SCF) method, which means that the molecular orbitals are derived from their own effective potential. The Hartree-Fock energy is the lowest energy that can be obtained from a single determinant wavefunction with a given nuclear configuration and basis set. 31 - 34 It is guaranteed to be an upper bound to the exact energy. C~iC~i 14 Moreover, the Hartree-Fock wavefunction is size-consistent, which means that if a system of several molecules is infinitely separated, the energy computed at infinite separation is equal to the sum of these energies computed for the isolated molecules.22,31 ,32,34 The most severe limitation of the Hartree-Fock method is that it describes an average (SCF) potential for the electrons. 31 - 34 The model does not properly consider instantaneous interaction between electrons. This is a particular problem for two electrons in the same orbital, which on average are too close to each other. As a result, the Hartree-Fock energy is always too high. The difference between the Hartree-Fock energy and the true energy of the system is referred to as the electron correlation error. There has been a great effort in quantum chemistry to improve the Hartree-Fock wavefunction and energy by explicitly treating the electron correlation problem. The most rigorous and straightforward way of doing this is through configuration interaction. In a configuration interaction (CI) treatment, the wavefunction <1> is written as 32 ,34 ~= Co 'pO +~Cit 'Pt + ,2::, cijab 'P ij ab + 2:: Cijk abc 'Pijkabc + "'," l.a lS.J is.j<:k ts.b as.b is a full CI wavefunction, which is the exact solution of the Schrbdinger equation with a given basis ~ ~ ~ ~ 15 set. Unfortunately, full CI calculations are feasible only for extremely small molecules or molecular ions?2,31,32,34 Therefore, in practice, it is necessary to approximate the full CI wavefunction in some way. The most commonly used methods for obtaining correlated wavefunctions are CI truncated to singles and doubles (CISD), coupled cluster (CC) theory, and M\Zlller-Plesset perturbation theory truncated to some order n (MPn). CISD is a computationally feasible method for moderate size systems, is variational, but is not size consistent. 22 ,31,32,34 For hydrogen-bonded complexes this is a significant problem since it yields incorrect binding energies. In fact, the computed binding energy is often positive, which means that the complex is not bound at all. Therefore, the CISD method is not recommended for hydrogen-bonded systems?2 The second method is the coupled cluster (CC) method in which the Hamiltonian is written in exponential form. 32 There are various levels of CC theory, including coupled cluster singles and doubles (CCSD); CCSD(T), which includes non-iterative triples; CCSDT which includes full triples, and CCSDTQ, which includes triples and quadruples. Because of the exponential ansatz, CC methods are size consistent, but they are not variational. CC methods are most reliable, but are computationally demanding and therefore not feasible for large systems such as those that will be investigated in this work. 22 ,32 In M\Zlller-Plesset (MP) theory, correlation effects are treated as a perturbation using many-body perturbation theory. M\Zlller-Plesset theory is widely used to investigate hydrogen-bonded complexes, since it is size consistent. 22 ,31,32,34 However, MP calculations are not variational, and the MP expansion may converge slowly. 16 In M¢ller-Plesset perturbation theory the generalized electronic Hamiltonian is written as 3 ! 1\ 1\ (15) where Ho is the unperturbed Hamiltonian, and the perturbation AV, is defined by 1V=1(H-Ho) (16) H is the correct Hamiltonian and Ais a dimensionless parameter. The unperturbed or 1\ zero Hamiltonian H o is taken to be the sum of the one-electron Fock operators. The exact or full CI wavefunction ('Fl) and energy (E l ) may be written in powers ofA. In practical applications the parameter Ais set equal to 1, and the series truncated at various orders. The method used is referred to by the highest term considered. For instance, if the series is truncated after the second order correction, it is referred to as MP2, if it is truncated after third order it is MP3, and so forth. The first terms in the expansion are 3 ! (17) (18) (19) 17 (20) (21) where 'Po is the Hartree-Fock wavefunction, E(O) is the Hartree-Fock energy, and Ci are the one-electron energies of the occupied molecular orbitals (1JrD at Hartree-Fock. Therefore, the MP energy to first order is the Hartree-Fock energy. The first-order contribution to the wavefunction is (22) where E s is the eigenvalue of a particular doubly-substituted determinant 'P s , while V so 1\ are matrix elements of the perturbation operator V, represented by (23) The integration is over all space and spin coordinates of the electrons. By Brillouin's theorem, the first-order energy corrections do not change the Hartree-Fock energy, since the Hamiltonian matrix elements between the Hartree-Fock wavefunction and singly- excited determinants are zero. The second-order M¢ller-Plesset energy can be expressed as (24) where (ij I Iab) is a two-electron integral over spin-orbitals defined by 18 and the integration is over all space and spin coordinates of the electrons. Level of Theory for Studies of Hydrogen-bonded Complexes: There have been many studies carried out on hydrogen-bonded complexes to establish what level of theory gives reliable structures and binding energies at minimal computational expense. 22 ,35-40 It has been shown that for hydrogen-bonded complexes, the MP energy expansion is dominated by the second-order term, while the third and fourth order contributions are small and may even be of opposite sign. Thus, MP2 with an appropriate basis set adequately describes the energy of hydrogen-bonded systems with minimal computational effort. The smallest basis set required is a split-valence basis set augmented with polarization functions on all atoms and diffuse functions on nonhydrogen atoms?2,35-38 The MP2/6-31+G(d,p) level of theory has also been found to produce reliable structures and vibrational frequency shifts of the proton-stretching band in good agreement with experimental data, provided that anharmonicity corrections are not large. 7 ,22 Binding energies at this level are reasonable, but are usually too high. Improved energies require a larger basis set, and in some cases a better wavefunction.22.35-40 In the current study, the MP2/6-31+G(d,p) level of theory will be used to determine structures, vibrational frequencies and binding energies. Although the binding energies may be too high, trends in binding energies in a closely related series of complexes have been shown to be reliable. 5 ,6,22 Geometry Optimization: The geometries of the monomers were fully optimized using the Gaussian 98 41 program at the MP2/6-31+G(d,p) level of theory. The nitrogen bases HCN, LiCN, 19 NaCN, SCN- and OCN- are linear, with C oov symmetry, and the bases NH 3 and N(CH 3 )3 have C 3v symmetry. Pyrrole and the 2,5- and 3,4-disustituted pyrroles have C 2v symmetry. The complexes of pyrrole and the 2,5- and 3,4-disubstituted pyrroles with HCN, LiCN, NaCN, SCN- and OCN- were optimized at the MP2/6-31+G(d,p) level of theory under the constraint of C 2v symmetry. The complexes with NH 3 and N(CH 3 )3 have C s symmetry, but local C 2v symmetry was imposed on pyrrole or the disubstituted pyrrole, and local C 3v symmetry on NH 3 and N(CH 3 )3. These restrictions maintain the linearity of the hydrogen bond. Vibrational frequencies were computed to verify that these optimized structures are equilibrium structures on their respective potential energy surfaces. For selected complexes of pyrrole with NH 3 and N(CH 3 h, two rotamers of C s symmetry were optimized. In one rotamer, one hydrogen of NH 3 or one carbon of N(CH 3 )3 was placed in the plane of the pyrrole ring. In the second rotamer, the NH 3 or N(CH 3 )3 molecule was rotated 90 degrees from the previous conformation, so that one hydrogen or one carbon lies in the symmetry plane perpendicular to the plane of pyrrole. The rotational energy barrier for interconversion of the rotamers was found to be less than .01 kcal/mol, signifying that there is free rotation about the hydrogen bonding N-N axis. For all complexes except those with NaCN as the proton acceptor, and all monomers except NaCN, the standard frozen core approximation was used. This approximation freezes all Hartree-Fock orbitals below the valence shell, which are then omitted from the correlation calculation. However, for complexes with NaCN as a base, 20 the standard option for freezing inner shell orbitals split degeneracies, or led to small energy gaps between frozen and active orbitals. To avoid these problems, only Is orbitals were frozen in complexes with NaCN. This necessitated freezing only Is orbitals in the NaCN monomer. Calculation of Harmonic Vibrational Spectra: Harmonic vibrational frequencies were calculated at the MP/6-31+G(d,p) level of theory to confirm equilibrium structures, to simulate IR spectra, and to obtain the zero point vibrational energies necessary to evaluate binding enthalpies at 10 K. The harmonic vibrational calculations were done using the standard algorithms for computing analytical first and second-derivatives implemented in Gaussian 98. 41 Harmonic vibrational calculations can be used to probe the vicinity of the minimum on the potential energy surface to determine whether a structure is a true minimum or a saddle point. If no imaginary frequencies exist, then the structure is a true minimum. If there are imaginary vibrational frequencies, then the optimized structure is a saddle point of order 1 if there is only one imaginary frequency, two if there are two imaginary frequencies, three if there are three imaginary frequencies, and so on. However, if the imaginary frequencies are small, then for the complexes investigated in this study, the optimized structures must be transition structures between two equivalent equilibrium structures, and the optimized structure is the vibrationally averaged structure. On the other hand, if the imaginary frequency is large, then the optimized structure is not close to the equilibrium structure, and a full optimization can yield significant geometry and energy changes. 21 Reaction Energies: The reaction for the formation of a hydrogen-bonded complex can be written as A-H(g) + B(g) ---+ A-H:B(g) Since the vibrational spectra of hydrogen-bonded complexes are usually obtained in low temperature matrices (-10K), the reaction enthalpy at this temperature can be approximated as the enthalpy at OK, and~Homay be written as ~HI0~~Eeo+~Ev0 where~Eeois the electronic binding energy and~Ev°is zero-point energy contribution to ~Ho.The electronic binding energy(~Eeo)is the difference between the electronic energy of the hydrogen-bonded complex [E e o (A-H:B)], and the electronic energies of the two monomers, E e o (A-H) and E e o (B). ~Eeo=E e o (A-H:B) - E e o (A) - E e o (B) (29) The zero point energy contribution~Ev°,is evaluated as the difference between the zero point energy of the hydrogen-bonded complex a.o (A-H:B) and the zero-point energies of the isolated monomers, Ev° (A-H) and Ev° (B) ~Ev°=Ev° (A-H:B) - Ev° (A) - Ev° (B) (30) To determine the basicities of the proton-acceptor molecules, the electronic proton affinity (PA) was evaluated at OK. The protonation reaction can be written as The electronic energy of this reaction is ~Eeo(PA) = Eeo(B-W) - Eeo(B) (31) (32) where Eeo(B-W) is the electronic energy ofthe protonated base and Eeo(B) is the electronic energy ofthe base B.~Eeo(PA) for equation 32 is always negative. However, ~HI0~~Eeo~Ev0 ~Ho. ~Eeo= ~Ev°= ~Eeo ~Eeo ~HI0~~Eeo~Ev0 ~Ho. ~Eeo= ~Ev°= ~Eeo ~Eeo 22 since the proton affinity at 298K is defined as the negative energy (-i1E PA O) for reaction 31, the computed electronic proton affinities will be reported as positive numbers. III. RESULTS and DISCUSSION: The MP2/6-31+G(d,p) electronic energy (Beo), zero-point vibrational energy (Eyo), equilibrium Na-H distance [Re(Na-H)], proton-stretching frequency (v) and intensity of the proton-stretching band (I) for pyrrole and disubstituted pyrroles are presented in Table 1. The Na-H distances, frequencies, and intensities are given as reference data so that changes in these quantities due to hydrogen bonding can be seen. Pyrrole and the disubstituted pyrroles are listed in order of increasing acidity as determined by their binding energy to HCN. Table 2 presents electronic energies (E e o), zero-point vibrational energies (By), and electronic proton affinities (-i1E pA 0) of the nitrogen bases. The bases, HCN and its derivatives, are arranged in order of increasing basicity as evaluated from their electronic proton affinities. The bases NH 3 and N(CH 3 )3 are also listed in Table 2. Table 3 presents the electronic energies, zero-point vibrational energies and imaginary frequency data for all of the complexes of pyrrole and disubstituted pyrroles with HCN and its derivatives that have been investigated in this study. An entry of "0" in the last column indicates that there are no imaginary frequencies. This means that the optimized structure is an equilibrium structure on the potential energy surface. There are some complexes listed in Table 3 that have small imaginary frequencies. The complexes pyrrole:NCS-, pyrrole:NCO-, 3,4-difluoropyrrole:NCS-, 3,4-difluoropyrrole:NCO-, 2,5 difluoropyrrole:NCS-and 2,5-difluoropyrrole:NCO- have a small imaginary frequency, 23 Table 1. MP2/6-31+G(d,p) electronic energies (Eeo, amu), zero-point vibrational energies (Evo, kcal/mol), Na-H distances (R e , A), harmonic proton-stretching frequencies ('0, em-I) and band intensities (I, km/mol) for pyrrole and disubstituted pyrroles Monomer E e o E v o Re(Na-H) '0 I Pyrrole -209.53768 52.0 1.007 3735 80 3,4-difluoropyrrole -407.55477 41.8 1.006 3745 110 2,5-difluoropyrrole -407.55947 41.6 1.008 3730 149 2,5-diberylliumpyrrole+ 2 -236.92122 40.8 1.012 3673 141 3,4-diberylliumpyrrole+2 -236.92778 41.0 1.017 3620 295 24 Table 2. MP2/6-31+G(d,p) electronic energies (E e o, amu), zero-point vibrational energies (Evo, kcal/mol), and electronic proton affinities(~EpAO,kcal/mol) for the nitrogen bases Monomer E e o E v o -~EpA ° HCN - 93.17212 9.9 174.3 LiCN -100.07104 4.3 228.5 NaCN -254.46613 4.0 236.7 SCN- -490.29666 5.1 326.4 OCN- -167.69373 6.4 343.7 NH 3 - 56.39205 -173.91095 22.1 77.6 214.7 236.5 -~EpA-~EpA Table 3. MP2/6-31+G(d,p) electronic energies (E e o, amu), zero-point vibrational energies (E v o, kcal/mol), and frequency data for complexes of pyrrole and substituted pyrroles with HCN and its derivatives 25 Donor a Acceptor E e o E v o Imaginary frequencies b Py HCN -302.71818 62.7 0 LiCN -309.62486 57.2 0 NaCN -464.02228 57.0 0 SCN- 1 -699.86388 57.6 _ 30 c , - 14 d OCN- 1 -377.26720 58.9 _20 c 3,4-diFPy HCN -500.73713 52.5 0 LiCN -507.64592 47.0 0 NaCN -662.04392 46.8 0 SCN- 1 -897.88985 47.3 _ 31 c _ 14 d , OCN- 1 -575.29456 48.6 _21 c 2,5-diFPy HCN -500.74210 52.4 0 LiCN -507.65034 46.8 0 NaCN -662.04836 46.6 0 SCN- 1 -897.89082 46.8 _31 c OCN- 1 -575.29598 47.7 _24 c 3,4-diBePy+2 HCN -330.13341 51.7 0 LiCN -337.07294 44.7 0 NaCN -491.48087 43.8 _42 e SCN- l -727.48058 43.8 -252 f -214 g _105 h , , OCN- l -404.89957 45.5 -244 f , -238 g , _108 h 2,5-diBePy+2 HCN -330.12815 51.7 _ 31 i LiCN -337.07249 45.5 _ 83 i a) Py = pyrrole b) An entry of 0 means that there are no imaginary frequencies; frequencies in cm,l. c) Change ofhybridization at the proton acceptor N coupled with pyrrole ring puckering out of the plane of pyrrole d) Change ofhybridization at the proton acceptor N coupled with an in-plane bending mode of the pyrrole ring e) Ring puckering of the substituted pyrrole out of the plane f) Change ofhybridization at the proton acceptor N coupled with pyrrole ring puckering out of the plane of pyrrole g) Change ofhybridization at the proton acceptor N coupled with an in-plane bending mode of the pyrrole ring. h) Ring puckering mode out ofthe plane of pyrrole i) In-plane rotation of the substituted pyrrole due to strong interactions of Be+ with the electron pair on the nitrogen of the proton acceptor 26 which corresponds to bending the proton acceptor out of the plane of the pyrrole ring, indicating a hybridization change of the nitrogen of NCO· and NCS·. This bending is coupled to a slight ring puckering of the pyrrole ring. The imaginary frequencies range from -31 cm- 1 for 2,5-difluoropyrrole:NCS· and 3,4-difluoropyrrole:NCS-, to -20 cm· 1 for pyrrole:NCO·. The imaginary frequencies of -14 cm- I for pyrrole:NCS· and 3,4 difluoropyrrole:NCS· correspond to an in-plane proton acceptor bend, also signifying a hybridization change of the NCS· nitrogen. This motion is coupled to in-plane bending of the pyrrole ring. The imaginary frequency of-42 cm- 1 for 3,4 diberylliumpyrrole:NCNa+ 2 corresponds to an out-of-plane ring puckering of the pyrrole ring. As noted previously, structures with imaginary frequencies correspond to transition structures on the surface, but if the imaginary frequency is small, the optimized planar structure is the vibrationally averaged structure. Thus, complexes that have small imaginary frequencies « 50 cm· l ) have been included in this work for comparative purposes. The complexes of 3,4-diberylliumpyrrole:NCS+ 1 and 3,4 diberylliumpyrrole:NCO+ 1 have large imaginary frequencies, and these complexes have not been included in this study. The complexes 2,5-diberylliumpyrrole:NCH+ 2 and 2,5 diberylliumpyrrole:NCLi+ 2 have imaginary frequencies corresponding to rotations that break the hydrogen bond and make Be+the electron pair acceptor. Thus, these two complexes have not been included in this work. Table 4 reports the electronic energies, zero-point vibrational energies, and imaginary frequency data for complexes of pyrrole and disubstituted pyrroles with NH 3 and N(CH 3 h. Except for 2,5-diberylliumpyrrole:N(CH 3 )3+2, the single imaginary Table 4: MP2/6-31+G(d,p) electronic energies (Eeo, amu), zero-point vibrational energies (E v o, kcal/mol), and frequency data for complexes of pyrrole and disubstituted pyrroles with NH 3 and N(CH 3 )3. Donor a Acceptor E e o E v o Imaginary frequencies b Py NH 3 c -265.94341 75.8 0 NH 3 d -265.94341 3,4-diFPy NH 3 c -463.96290 65.6 - 18 e NH 3 d -463.96290 Py N(CH3h C -383.46481 130.4 _ 14 e N(CH3h d -383.46481 2,5-diFPy NH 3 c -463.96952 65.4 - 7 e NH 3 d -463.96951 3,4-diFPy N(CH3)3 C -581.48445 120.2 0 2,5-diBepy+2 NH 3 c -293.34999 64.5 _ Be NH 3 d -293.34999 2,5-diFPy N(CH3)3 C -581.49261 119.9 0 3,4-diBepy+2 NH 3 c -293.36161 64.0 0 NH 3 d -293.36161 2,5-diBePl 2 N(CH3)3C -410.88485 118.8 _40 f -12 e , N(CH3h d -410.88483 3,4-diBePy+2 N(CH3)3 C -410.91210 119.9 0 N(CH3)3 d -410.91209 a) Py =pyrrole b) An entry of0 means that there are no imaginary frequencies; frequencies in em-I. c) Proton acceptor with one H or one C in the plane of the pyrrole d) Proton acceptor with one H or one C in the plane perpendicular to the plane of the pyrrole e) Rotation of proton acceptor f) Ring puckering of the substituted pyrrole out of the plane of pyrrole ring 27 28 frequency in some of these complexes corresponds to rotation of the proton acceptor molecule about the hydrogen-bonding axis. However, the energies of the two rotamers, one with an H of NH 3 or a C of N(CH 3 h in the plane of the pyrrole ring, and the other with an H of NH 3 or a C of N(CH 3 )3 in the plane perpendicular to the plane of the pyrrole ring, are identical, as evident from Table 4. This indicates that there is free rotation of the proton acceptor molecule about the hydrogen bonding N-N axis. The complexes with the N-H or N-C bonds in the plane of the pyrrole ring will be discussed below. The complex 2,5-diberylliumpyrrole:N(CH 3 )3+ 2 has two imaginary frequencies. The frequency of -12 cm- 1 corresponds to rotation of N(CH 3 )3 about the hydrogen-bonding axis. The frequency of -40 cm- 1 corresponds to a ring puckering vibration. However, since these frequencies are small, this complex has been included in this study. The energies in Tables 1-4 are given as raw data from which binding energies and enthalpies can be computed. The z-matrices for the optimized monomers and complexes are given in Appendix 1 and 2, respectively. The binding enthalpies for the complexes are reported in Appendix 3. Table 5 reports equilibrium Na-N b distances [Re(Na-N b )], Na-H distances [Re(Na-H)], electronic binding energies(~Ee),harmonic proton-stretching frequencies (u) and intensities of the proton-stretching band (I) for complexes of pyrrole and disubstituted pyrroles with HCN and its derivatives. The complexes are arranged in order of increasing acidity of pyrrole and disubstituted pyrroles as determined by their binding energy with HCN. For a given proton donor, the complexes are arranged in order of increasing base strength as determined by the electronic proton affinity of the proton acceptor. Table 5. MP2/6-31+G(d,p) Na-N b and Na-H distances (R e , A), electronic binding energies(~Ee,kcal/mol), harmonic proton-stretching frequencies (u, em-I) and band intensities (I, km/mol) for complexes of pyrrole and substituted pyrroles with HCN and its derivatives. Donor a Acceptor Re(Na-N b ) Re(Na-H)b~EeIv Py HCN 3.164 1.011 - 5.3 3671 466 LiCN 3.010 1.019 -10.1 3527 989 NaCN 2.971 1.021 -11.6 3483 1171 SCN- I 2.835 1.037 -18.5 3198 2653 OCN- I 2.762 1.048 -22.5 2993 2761 3,4-diFPy HCN 3.115 1.012 - 6.4 3656 591 LiCN 2.956 1.022 -12.6 3472 1249 NaCN 2.916 1.025 -14.4 3414 1477 SCN- I 2.770 1.045 -24.1 3034 3388 OCN- I 2.699 1.061 -28.9 2767 3442 2,5-diFPy HCN 3.068 1.016 - 6.6 3594 777 LiCN 2.906 1.029 -12.4 3343 1662 NaCN 2.863 1.033 -14.3 3260 2023 SCN- I 2.713 1.065 -21.8 2710 4750 OCN- I 2.629 1.097 -26.9 2202 6340 3,4-diBePl 2 HCN 2.828 1.043 -21.0 3127 2706 LiCN 2.603 1.119 -46.5 1924 6469 NaCN c 2.666 1.557 -54.6 1871 2836 2473 4208 a) Py = pyrrole b) The Na-H distance is measured from the pyrrole nitrogen. c) There are two strong bands associated with the Na-H stretching mode in this complex. 29 ~Ee~Ee 30 The first set of complexes in Table 5 are those with pyrrole as the proton donor. Table 5 shows that for these complexes as the base strength increases, the binding energy increases. The binding energies range from -5.3 kcal/mol for the weakest hydrogen bonded complex pyrrole:NCH, to -22.5 kcal/mol for the strongest, pyrrole:NCO-. As evident from Table 5, as the binding energy increases the Na-H distance also increases from 1.011 Ain pyrrole:NCH to 1.048 Ain pyrro1e:NCO-. These Na-H bond distances are all greater than the Na-H distance in isolated pyrrole (1.007 A). The lengthening of the Na-H bond is a consequence of hydrogen bonding. Moreover, the lengthening of the Na-H distance is also accompanied by a decrease in the Na-N b distance, which ranges from 3.164 Ain pyrrole:NCH to 2.762 Ain pyrrole:NCO-. Figure 2 illustrates the variation of the Na-H and Na-N b distances with binding energy. These data show that as the binding energy increases, the hydrogen moves away from N a towards N b . This is the beginning of proton transfer, which is facilitated by a decrease in the Na-N b distance. The IR properties of the complexes of pyrrole with HCN and its derivatives may also be related to their structural and energetic properties. As the base strength increases, the proton-stretching frequency decreases, and the intensity of the proton-stretching band increases. The proton-stretching frequency and the intensity of the proton-stretching band for isolated pyrrole are 3735 cm- 1 and 80 km/mol, respectively. The proton stretching frequency decreases from 3671 cm- 1 in pyrrole:NCH to 2993 cm- 1 in pyrrole:NCO-. At the same time, the intensity of the proton-stretching band increases from 466 km/mol in pyrrole:NCH to 2761 km/mol in pyrrole:NCO-. The frequency 2.85 3.15 3.25 -0- Na-H -l:r- Na-Nb -23-21-19-11 -13 -15 -17 LlE e 0 (kcal/mol) -9-7 Figure 2: Na-N b and Na-H distances versus MP2/6-31+G(d,p) binding 1---- , , , , , , , , - , I 2.75 -25-5 1.02 1.01 1.04 1.05 l3.05~ ~j 0« 0« -- ~1.03 .0 z , co 2.95 z energies of complexes with pyrrole as the proton donor to HeN and its derivatives w ..... ~--..-------.-----r---_--_--..---r------r--.........--~2.75 3.05 ..- ..- -- I 1.03 I I ('(l ('(l z derivatives --~ 32 shifts and the increased intensities of the proton-stretching band in the complexes relative to pyrrole are typical of hydrogen-bonded complexes, and are the IR signature of the hydrogen bond. The variation in the frequency and intensity of the proton-stretching band for the complexes with pyrrole as the proton donor is illustrated in Figure 3. The vibrational spectra are arranged in order of increasing base strength. The intensity of the proton stretching band is much greater compared to the other fundamental bands, which cannot be seen on the scale shown in Figure 3. For the charged complexes of pyrrole:NCS- and pyrrole:NCO-, one additional relatively intense band at 2015 cm- 1 and at 2206 cm- 1 , respectively, can be seen. These correspond to monomer NCO- and NCS stretching vibrational modes that are essentially unchanged in the complexes. The spectra in figure 3 illustrate that as the Na-H distance increases, the Na-H bond becomes weaker. This implies a decrease of the force constant for the Na-H stretch, which leads to a decrease in the proton-stretching frequency. In addition, as the proton moves towards the base, the dipole moment of the complex increases. As a result, the intensity of the proton-stretching band also increases. The data in Table 5 indicate that complexes with pyrrole as the proton donor to HCN and its derivatives are stabilized by traditional hydrogen bonds. 42 In these complexes the Na-H covalent bond remains intact, and the Na-N b distances are typical of complexes stabilized by traditional hydrogen bonds. 42 The spectra of these hydrogen bonded complexes are characterized by a single intense proton-stretching band that is shifted to lower energy relative to the monomer stretching frequency. 33 pyrrole:NCH 4000 r 3000 2300 0 1600~ 900-'" ,--__ff-36_7_ 1 r-,-------.-----,-----,-----,-----,-----,-------j 200 3500 3000 2500 2000 1500 1000 500 0 v (cm-1) pyrrole:NCU 4000 [ 3000 2300 0 160012 ,-- 1--r,35_4_2 ,--- --,-- -----. ,--- --,- -----. :~ 3500 3000 2500 2000 1500 1000 500 0 V (cm-1) pyrrole: NCNa 50010002000 1500 v (cm-1) 25003500 3000 r 3000 2300- ,-- J_ 3483 ---.- -----r -,----- -,--- -r,---;-_---.--_,----;--__ :1 o4000 50010002000 1500 v (cm-1) 25003500 3000 1 31 ", _,NCS t~i ,-------,----.--,..-.---,------!-------.------.-----r-----Il200 o4000 50010002000 1500 v (cm-1) 250030003500 pyrrole:NCO· I ~f=-1600~ 1 900~ ,-------,----- -------r-- -----,--------,-----,----------,-------1 200 o4000 Figure 3: MP2/6-31+G(d,p) vibrational spectra ofpyrrole:NCH, pyrrole:NCLi, pyrrole:NCNa, pyrrole:NCS· and pyrrole:NCO-. Only bands with intensities greater than 250 km/mol are shown. 1600~ ~ ~ I ~~ ~ 1600~ ~ ~ I ~~ ~ 34 The next set of complexes in Table 5 are those with 3,4-difluoropyrrole as the proton donor to HCN and its derivatives. For this set of complexes, it is again seen that as the base strength increases the binding energy increases. The binding energies range from -6.4 kcal/mol or 3,4-difluoropyrrole:NCH to -28.9 kcal/mol for 3,4 difluoropyrrole:NCO-. As the binding energy increases, the Na-H distance increases from 1.012 Ain 3,4-difluoropyrrole:NCH to 1.061 Ain 3,4-difluoropyrrole:NCO-. Moreover, as the Na-H distance increases the Na-N b distance decreases from 3.115 Afor 3,4 difluoropyrrole:NCH to 2.699 Afor 3,4-difluoropyrrole:NCO-. Thus, the same trends in binding energies, as well as Na-H and Na-N b distances, seen previously in the complexes with pyrrole, are observed in complexes with 3,4-difluoropyrrole. The IR properties of the complexes of 3,4-difluoropyrrole with HCN and its derivatives are also related to the structural and energetic properties. As the base strength increases, the proton-stretching frequency decreases, and the intensity of the proton stretching band increases. The 3,4-difluoropyrrole monomer has an Na-H stretching frequency of 3745 cm- l , and the band intensity is 110 km/mol. In the complexes, the proton-stretching frequency decreases from 3656 cm- l in 3,4-difluoropyrrole:NCH to 2767 cm- l in 3,4-difluoropyrrole:NCO-. The intensity of the proton-stretching band increases from 591 km/mol in 3,4-difluoropyrrole:NCH to 3442 km/mol in 3,4 difluoropyrrole:NCO-. The frequency shifts and increased intensities of the proton stretching band in the complexes relative to the monomer are due to hydrogen bonding. The complexes with 3,4-difluoropyrrole as the proton donor with HCN and its derivatives are also stabilized by traditional hydrogen bonds. The Na-H distances in these complexes are characteristic of a perturbed covalent Na-H bond, and the Na-N b distances 35 are typical of traditional hydrogen bonds. 42 Furthermore, the vibrational spectra of these complexes are characterized by a single proton-stretching band which is shifted to lower energy relative to 3,4-difluoropyrrole. Data for complexes of 2,5-difluoropyrrole with HCN and its derivatives are also reported in Table 5. Once again as the base strength increases, the binding energy increases, the Na-H distance increases, and the Na-N b distance decreases. The binding energies vary from -6.6 kcal/mol for 2,5-difluoropyrrole:NCH to -26.9 kcal/mol for 2,5 difluoropyrrole:NCO-. The Na-H distances increases from 1.016 Afor 2,5 difluoropyrrole:NCH to 1.097 Afor 2,5-difluoropyrrole:NCO-. The Na-N b distance decreases from 3.068 Afor 2,5-difluoropyrrole:NCH to 2.629 Afor 2,5 difluoropyrrole:NCO-. Thus, the same trends in binding energies and Na-H and Na-N b distances for complexes of pyrrole and 3,4-difluoropyrrole are observed for the complexes with 2,5-difluoropyrrole. The IR properties of the complexes of 2,5-difuoropyrrole with HCN and its derivatives are again related to the structural and energetic properties. As the binding energy increases, the proton-stretching frequency decreases, and the intensity of the proton-stretching band increases. The proton-stretching frequency and the intensity of the proton-stretching band for 2,5-difluoropyrrole are 3730 cm- 1 and 149 km/mol, respectively. The proton-stretching frequency decreases from 3594 cm- 1 for 2,5 difluoropyrrole:NCH to 2202 cm- 1 for 2,5-difluoropyrrole:NCO-. The intensity of the proton-stretching band increases from 777 km/mol for 2,5-difluoropyrrole:NCH to 6340 km/mol for 2,5-difluoropyrrole:NCO-. 36 All the complexes of 2,5-difluoropyrrole as the proton donor to HCN and its derivatives, except 2,5-difluoropyrrole:NCO-, are stabilized by traditional hydrogen bonds. The Na-H distances and Na-N b distances of these complexes are typical of traditional hydrogen bonds,42 and their vibrational spectra are characterized by a single intense proton-stretching band that is shifted to lower energy relative to the Na-H stretch of the 3,4-difluoropyrrole monomer. The complex 2,5-difluoropyrrole:NCO' has an Na-H distance of 1.097 A, and a very short Na-N b distance of 2.629 A. These distances are approaching distances found for proton-shared N-H-N hydrogen bonds. 42 The Na-H distance of this complex cannot be simply described as a perturbed Na-H distance. Furthermore, the proton-stretching band has dramatically shifted to lower energy by 1528 cm'} relative to the 2,5-difluoropyrrole monomer. This spectral property also indicates that the hydrogen bond in 2,5 difluoropyrrole:NCO' has proton-shared character. Thus, a hydrogen bond with proton shared character is found in this negatively charged complex. Only three complexes with 3,4-diberylliumpyrrole+ 2 as the proton donor are listed in Table 5. The first is 3,4-diberylliumpyrrole:NCW 2 , which has a binding energy of -21.0 kcal/mol. The Na-H and Na-N b distances for this complex are 1.043 Aand 2.828 A, respectively, which are comparable to distance seen in complexes with traditional hydrogen bonds. Furthermore, the vibrational spectrum of this complex is characterized by a single intense proton-stretching band at 3127 cm'} with a band intensity of 2706 km/mol. These data are typical for a complex stabilized by a traditional hydrogen bond. The structural, energetic and spectroscopic properties of the complex 3,4 diberylliumpyrrole:NCLi+ 2 are different. The binding energy for this complex has 37 dramatically increased to -46.5 kcal/mol. Furthermore, the Na-H distance has increased to 1.119 A, and the Na-N b distance has decreased to 2.603 A. Thus, this complex has the longest Na-H distance and shortest Na-N b distance observed so far, and is stabilized by a proton-shared hydrogen bond. For the complex of 3,4-diberylliumpyrrole:NCLi+ 2 the proton-stretching frequency has decreased to 1924 cm- I and the intensity of the proton stretching band has increased to 6469 km/mol. This frequency is the lowest thus far, and corresponds to a shift of 1696 cm- I relative to 3,4-diberylliumpyrrole+ 2 . Thus, both structural and spectral data for this cationic complex lead to the characterization of the hydrogen bond as a proton-shared hydrogen bond. The last complex in Table 5 is 3,4-diberylliumpyrrole:NCNa+ 2 . The binding energy of this complex is -54.6 kcal/mol, the largest in Table 5. The Na-H distance for this complex has dramatically increased to 1.557 A, and the Na-N b distance has increased to 2.666 Arelative to 3,4-diberylliumpyrrole:NCLi+ 2 . Thus, the Nb-H distance is 1.109 A, indicating a perturbed covalent Nb-W bond. Therefore, the hydrogen bond in this complex is on the ion-pair side of proton-shared. That is, if 3,4 diberylliumpyrrole:NCLi+ 2 and 2,5-difluoropyrrole:NCO- have proton-shared hydrogen bonds, then 3,4-diberylliumpyrrole:NCNa+ 2 must have a hydrogen bond with ion-pair character. This is also supported by the spectrum of this complex, which is characterized by two strong proton-stretching bands appearing at 1871 and 2473 em-I, with intensities of 2836 and 4208 km/mol, respectively. The strongest band at 2473 cm- l is at a higher frequency compared to 3,4-diberylliumpyrrole:NCLi+ 2 , further indicating the ion-pair character of this complex. The proton-stretching band is due to a perturbed Nb-H stretch, shifted to lower frequency relative to HNCNa+ I . The fact that there are two bands in the 38 spectrum is due to coupling of the N b -H stretch to the N-C stretch of HCNNa+1. The local N-C stretching band increases in intensity due to intensity borrowing from the Nb-H stretch. Complexes of pyrrole and disubstituted pyrroles with NH 3 and trimethylamine have C s symmetry, and binding energies, Na-N b and Na-H distances, proton-stretching frequencies, and intensities of the proton-stretching bands are reported in Table 6. These complexes are arranged in order of increasing Na-H distance so that changes in hydrogen bond type can be observed. The first seven complexes listed in Table 6, namely, pyrrole:NH 3 , 3,4 difluoropyrrole:NH 3 , pyrrole:N(CH 3 )3, 2,5-difluoropyrrole:NH 3 , 3,4 difluoropyrrole:N(CH 3 )3, 2,5-diberylliumpyrrole:NH 3 +2 and 2,5-difluoropyrrole:N(CH 3 )3 are stabilized by traditional hydrogen bonds. The Na-H distances for these complexes increase from 1.021 Afor pyrrole:NH 3 to 1.053 Afor 2,5-difluoropyrrole:N(CH 3 )3, while the Na-N b distances decrease from 3.034 Afor pyrrole:NH 3 to 2.785 Afor 2,5 difluoropyrrole:N(CH 3 h. These distances correspond to Na-H and Na-N b distances already seen in Table 5 for complexes with traditional hydrogen bonds. The proton stretching frequencies for these complexes decrease from 3484 cm- 1 for pyrrole:NH 3 to 2848 cm- 1 for 2,5-difluoropyrrole:N(CH 3 )3, and the intensities of the proton-stretching bands increase from 862 km/mol for pyrrole:NH 3 to 3237 km/mol for 2,5 difluoropyrrole:N(CH 3 )3' The vibrational spectra are characterized by a single strong proton-stretching band consistent with the spectra of complexes with traditional hydrogen bonds seen in Table 5. It is expected that the binding energies for these complexes should increase from pyrrole:NH 3 to 2,5-difluoropyrrole:N(CH 3 h. However, this trend is 39 Table 6. MP2/6-31+G(d,p) Na-N b and Na-H distances (R e , A), electronic binding energies(.6.E e , kcal/mol), harmonic proton-stretching frequencies (v, em-I) and band intensities (I, km/mol) for complexes of pyrrole and substituted pyrroles with NH 3 and N(CH 3 )3 Donora Acceptor ReCN-N) Re(Na-H)b .6.E e v I Py NH 3 3.034 1.021 - 8.6 3484 862 3,4-diFPy NH 3 2.992 1.023 -10.1 3429 1086 Py N(CH 3 )3 2.931 1.029 -10.1 3305 1453 2,5-diFPy NH 3 2.924 1.032 -11.3 3272 1560 3,4-diFPy N(CH 3 )3 2.881 1.035 -11.8 3196 1773 2,5-diBep/ 2 NH 3 2.870 1.050 -23.0 c 2964 1857 2,5-diFPy N(CH 3 )3 2.785 1.053 -13.9 2848 3237 3,4-diBePy+2 NH 3 2.698 1.107 -26.2 2071 5305 2,5-diBep/ 2 N(CH 3 )3 2.837 1.773 -33.1c 2772 1972 3,4-diBePy+2 N(CH 3 )3 2.875 1.822 -46.0 2961 2264 a) Py = pyrrole b) The Na-H distance is measured from the pyrrole nitrogen. c) Strong electrostatic interactions of Be+ with the nitrogen of the proton acceptor contribute to the large stabilization energy ofthis complex 40 only observed if 2,5-diberylliumpyrrole:NH 3 is excluded. The binding energy for the complex 2,5-diberylliumpyrrole:NH 3 + 2 is -23.0 kcal/mol and is greater than the binding energy of the complex immediately below it in Table 6. The increased stabilization is due to strong electrostatic interactions of the Be+ atoms with the nitrogen of the proton acceptor. The next complex in Table 6 is 3,4-diberylliumpyrrole:NH 3 +2. This complex is stabilized by a proton-shared hydrogen bond and has a binding energy of -26.2 kcal/mol. In this complex, the Na-N b distance has decreased to 2.698 A, and the Na-H distance has increased to 1.107 A. Furthermore, the proton-stretching frequency ofthis complex is 2071 cm- 1 and the intensity of the proton-stretching band is 5305 km/mol. The shift relative to 3,4-diberylliumpyrrole+ 2 is 1549 em-I. The Na-H and Na-N b distances and proton stretching frequency are similar to those observed for complexes with proton shared hydrogen bonds in Table 5. Therefore, this cationic complex is stabilized by a proton-shared hydrogen bond. The last two complexes in Table 6, 2,5-diberylliumpyrrole:N(CH 3 h+ 2 and 3,4 diberylliumpyrrole:N(CH 3 )3+2, are stabilized by ion-pair hydrogen bonds. For these two complexes the Na-N b distances have increased relative to 3,4-diberylliumpyrrole:NH 3 +2 to 2.837 Aand 2.875 A. The Na-H distances are very long at 1.773 Aand 1.822 A, respectively. Thus, the Nb-W distances are 1.064 Ain 2,5-diberylliumpyrrole:N(CH 3 )3+2 and 1.053 Ain 3,4-diberylliumpyrrole:N(CH 3 )3+2. Thus, the Nb-H+ bond is a perturbed Nb-H+ bond of protonated trimethylamine. The Nb-H distance in isolated HN(CH 3 )3+1 is 1.024 A. Furthermore, the IR spectra of these complexes are characterized by a single strong proton-stretching band at 2772 cm- 1 for 2,5-diberylliumpyrrole:N(CH 3 )3+2 and 41 2961 cm- 1 for 3,4-diberylliumpyrrole:N(CH 3 )/2, with band intensities of 1972 kmImol and 2264 kmImol, respectively. These proton-stretching bands are best described as arising from perturbed Nb-H+ stretches. These bands are shifted to lower frequency compared to the proton-stretching frequency of 3501 cm- 1 for HN(CH 3 )3+1. Therefore, the structural and spectroscopic data for these two complexes indicate that they are stabilized by ion-pair hydrogen bonds. The binding energies for these complexes are -33.1 kcal/mol for 2,5-diberylliumpyrrole:N(CH 3 h+ 2 and -46.0 kcal/mol for 3,4 diberylliumpyrrole:N(CH 3 h +2. Thus, ion-pair hydrogen-bonded complexes occur when the strongest cationic proton donors, 3,4-diberylliumpyrrole+ 2 and 2,5 diberylliumpyrrole+ 2 are combined with the strong proton acceptor, N(CH 3 h. The characteristic changes in the Na-N b and Na-H distances and proton-stretching frequencies that accompany changes in hydrogen bond type are illustrated in Figure 4. In this figure, the Na-N b distance and the proton-stretching frequency are plotted against the Na-H distance for the series of complexes of pyrrole and disubstituted pyrroles with ammonia and trimethylamine. At short Na-H distances, traditional hydrogen bonds are found, and the Na-N b distance and the proton-stretching frequency change almost linearly with the Na-H distance. The points in this region of the graph correspond to the first seven complexes in Table 6, which are stabilized by traditional hydrogen bonds. Figure 4 suggests that the change from a traditional to a proton-shared hydrogen bond is not a dramatic one. Nevertheless, the two points which correspond to the shortest Na-N b distance and lowest proton-stretching frequency, are found for the complex with a proton-shared hydrogen bond. Subsequently, there is a rather dramatic change in the km/mol km/mol, 3600 3400 3200 3000 ...- ..- 'E 2800 () "'-"" ;> 2600 2400 2200 .. .. ... ,.. .. .. .. .. .. .. .. .. .. .. .. III .. 'Ill 3.10 3.00 2.90 ... 0« "'-"" .0 Z I ell 2.80 z 2.70 2000 I I I I I I 2.60 1.00 1.20 1.40 1.60 1.80 2.00 Na-H (A) Figure 4: Na-N b distances and proton-stretching frequencies versus Na-H distances in complexes of pyrrole and disubstituted pyrroles with NH 3 and N(CH 3 )3 --0-- Na-N b -----i:c- v ~ ['J .. , E .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. +------_-----__-------.-------r--------+ 43 slope of the two lines which signals the formation of ion-pair hydrogen bonds. It would be interesting to have had at least one more point for a complex in this series in which the Na-H distance had a value of about 1.2 A. Such a point would indicate whether proton shared hydrogen bonds span a relatively narrow range of Na-H distances or not. Unfortunately, no such point exists for the complexes investigated in this work. IV. CONCLUSIONS: Optimized MP2/6-31+G(d,p) structures have been determined for the complexes of pyrrole and disubstituted pyrroles with the nitrogen bases HCN, LiCN, NaCN, SCN-, OCN-, NH 3 and N(CH 3 )3. The harmonic vibrational spectra of these complexes have also been calculated at MP2/6-31+G(d,p). The following statements are supported by these calculations. 1. By systematically varying the proton-donating ability of pyrrole and substituted pyrroles and the proton-accepting ability of nitrogen bases, complexes stabilized by traditional N-H...N, proton-shared N...H...N, and ion-pair N- ...+H-N hydrogen bonds have been produced. 2. Proton-shared and ion-pair hydrogen bonds are not found in neutral complexes. They occur only in charged complexes. 3. Most of the complexes investigated in this study are stabilized by traditional hydrogen bonds as evident from the Na-H distances, Na-N b distances, and proton-stretching frequencies. In series of closely related complexes stabilized by traditional hydrogen bonds, as the base strength increases, the binding energy increases, the Na-N b distance 44 decreases, the Na-H distance increases, the proton-stretching frequency decreases, and the intensity of the proton-stretching band increases. 4. Among the complexes with HCN and its derivatives as proton acceptors, two charged complexes are stabilized by proton-shared hydrogen bonds, and a third is stabilized by a hydrogen bond which is on the ion-pair side of proton-shared. Proton-shared hydrogen bonds have short intermolecular Na-N b distances, long Na-H and Nb-H distances, and proton-stretching frequencies that are dramatically shifted to lower energy compared to traditional and ion-pair hydrogen bonds. 5. Among complexes with NH 3 as the proton acceptor, only traditional hydrogen bonds are formed. When the stronger base N(CH 3 h is the proton acceptor, one charged complex has a proton-shared hydrogen bond, and two charged complexes have ion-pair hydrogen bonds. 6. This study suggests that proton-shared and ion-pair hydrogen bonds probably do not exist in neutral complexes stabilized by N-H-N hydrogen bonds. 45 v. REFERENCES: 1. Hanessian, S.; Yang, H.; Schaum, R J. Am. Chem. Soc. 1996,118,2507. 2. Cantor, C. R, Schimmel, P. P. Biophysical Chemistry; Freeman: San Francisco, CA, 1980; Vall. 3. Baker, E. N.; Hubbard, R E. Prog. Biophys. Mol. BioI. 1984,44,97. 4. Schramm, V. L.; Purich, D. L. Methods in Enzymology; Academic Press: San Diego, CA, 1999; Vol. 308, pp 219-297. 5. Del Bene, J. E.; Person, W. B.; Szczepaniak, K Chem. Phys. Lett. 1995,247,89. 6. Del Bene, J. E.; Person, W. B.; Szczepaniak, K Mol. Phys. 1996,89,47. 7. Del Bene, J. E.; Jordan, M. J. T. Int. Rev. Phys. Chem. 1999,18,119. 8. Morokuma, K; Pedersen, L. J. Chem Phys. 1968,49, 3275. 9. Kollman, P. A; Allen, L. C. J. Chem Phys. 1969,51,3286. 10. Morokuma, K; Winick, J. 1. Chem Phys. 1970,52,1301. 11. Del Bene, J.; Pople, J. A J. Chem Phys. 1970,52,4858. 12. Hankins, D.; Moskowitz, J. W.; Stillinger, F. H. Chem. Phys. Lett. 1970,4,527. 13. Pople, J. A; Krishnan, R; Schlegel, H. B.; Binkley, J. S. Int. J. Quantum Chem. 1979, 13S, 225. 14. Schlegel, H. B., J. Comput. Chem. 1982,3,214. 15. Schlegel, H. B.; Brinkely, J. S.; Pople, J. A J. Chem. Phys. 1984,80, 1979. 16. Fogarasi, G.; Pulay, P. Ann. Rev. phys. Chem. 1984,35, 191. 17. Pulay, P. Adv. Chem. Phys. 1987,69,241. 46 18. Dykstra, C. E.; Augspurger, J. D.; Kirtman, B.; Malik, D. J. Reviews of Computational Chemistry, Lipkowitz, K. B.; Boyd, D. B., Eds.; VCH: NY, 1990; Vol. 1, pp 83-118. 19. Bartlett, R. J.; Stanton, J. F.; Watts, 1. D. Reviews ofComputational Chemistry, Lipkowitz, K. B.; Boyd, D. B., Eds.; VCH: NY, 1991; Vol V., pp 65-169. 20. Gauss, J.; Cremer, D. Adv. Quant. Chem. 1992,23,206. 21. Pimentel, G. C.; McClellan, A. L. The Hydrogen Bond, Freeman: San Francisco, CA,1960. 22. Del Bene, J. E. Hydrogen Bonding:1. Encyclopedia ofComputational Chemistry; Schyleyer, P. V.; Allinger, N. L.; Clark, T.; Gasteiger, J.; Kollman, P. A.; Schaefer, H. F. , III; Schreiner, P. R., Eds.; Wiley & son: Chichester, UK, 1998; Vol. 2, pp 1263-1271. 23. Beckenham, K. Hydrogen Bonding and Other Physicochemical Interactions Studied By IR and Raman Spectroscopy. Encyclopedia ofSpectroscopy and Spectrometry, London, J. C.; Tranter, G. E.; Holmes, J. L., Eds.; Academic Press: UK, 1999, Vol. 1, pp 837-843. 24. Scheiner, S. Hydrogen Bonding: A Theoretical Perspective, Oxford University Press: NY, 1997. 25. Scheiner, S. Calculating the Properties of Hydrogen Bonds by ab Initio Methods. Reviews ofComputational Chemistry, Lipkowitz, K. B.; Boyd, D. B., Eds.; VCH: NY, 1991, Vol 2, pp 165-264. 26. Del Bene, J. E.; Jordan, M. J. T. 1. Chem. Phys. 1998,108,3205. 27. Del Bene, J. E.; Jordan, M. J. T. 1. Am. Chem. Soc. 2000, 122, 2101. 47 28. Tai, J. C.; Yang, L.; Allinger N. L. 1. Am. Chern. Soc. 1993,115, 11906. 29. Lee, S. Y.; Boo, B. H. J. Phys. Chern. 1996,100, 15073. 30. Jiang, J. c.; Tsai M. H. J. Phys. Chern. 1997,101,1982. 31. Hehre W. J.;Radom, L.; Schleyer, P. v.R; Pople, J. A. Ab initio Molecular Orbital Theory; John Wiley & Son: Toronto, Canada, 1986. 32. Bartlett, R J.; Stanton, J. F. Applications of Post-Hartree-Fock Methods: A Tutorial. Reviews o/Computational Chemistry, Lipkowitz, K. B.; Boyd, D. B., Eds.; VCH: NY, 1994, Vol. V, pp 65-169. 33. Pople, J. A.; Beveridge, D. L. Approximate Molecular Orbital Theory; McGraw Hill Book Company: NY, 1970. 34. Atkins, P. W.; Friedman, R S. Molecular Quantum Mechanics, 3rd edition; Oxford University Press: NY, 1997. 35. Del Bene, J. E. J. Comput. Chern. 1987,8,810. 36. Del Bene, J. E J. Chern. Phys. 1987, 86, 2110. 37. Del Bene, J. E International Journal o/Quantum Chemistry: Quantum Biology Symposium 1987, 14,27. 38. Del Bene, J. E. J. Phys. Chern. 1988, 92, 2874. 39. Del Bene, J. E; Shavitt, I. J. Mol. Struct. 1994,307,27. 40. Tzeli, D.; Mavridis, A; Xantheas, S. S. J. Chern. Phys. 2000, 112,6178. 41. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E; Robb, M. A; Cheeseman, J. R; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R E. Jr.; Burant, J. c.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, 0.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R; Mennucci, B.; 48 Pomelli, c.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A; Ayala, P. Y; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. v.; Baboul, A G.; Stefanov, B. B.; Liu, G.; Liashenko, A; Piskorz, P.; Komaromi, L; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; AI-Laham, M. A.; Peng, C. Y; Nanayakkara, A; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A Gaussian 98; Gaussian, Inc.: Pittsburgh, PA, 1998. 42. Del Bene, J. E.; Perera, A; Bartlett, R. J. 1. Phys. Chern. A. 2000, 105, 930. Appendix 1 Z-matrices for optimized monomers reported in Tables 1 and 2, taken directly from Gaussian 98 output HCN MP2/6-31+G(D,P) OPTIMIZED STRUCTURE Symbolic Z-matrix: Charge =0 Multiplicity =1 C N 1 eN x 1 1. 2 90. H 1 CH 3 90. 2 180. 0 Variables: CH 1.06652 CN 1.17845 Al-l LICN MP2/6-31+G(D,P) OPTIMIZED STRUCTURE Symbolic Z-matrix: Charge = 0 Multiplicity = 1 C N 1 CN X 1 1. 2 90. Li 1 CLI 3 90. 2 180. 0 Variables: CN 1.19257 CLI 1.94586 AI-2 NCNa MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH ONLY THE Is ORBITALS FROZEN Symbolic Z-matrix: Charge =0 Multiplicity =1 X NIL C 2 CN 1 90. X 3 1. 2 90. 1 O. a Na 3 CNa 4 90. 1 180. 0 Variables: CN 1.192947 CNa 2.228593 AI-3 NCS' MP2/6-31+G(D,P) OPTIMIZED STRUCTURE Symbolic Z-matrix: Charge =-1 Multiplicity = 1 X NIL C 2 CN 1 90. X 3 1. 2 90. 1 O. 0 S 3 CS 4 90. 1 180. 0 Variables: CN 1.2022 CS 1.66173 AI-4 NCO- MP2/6-31+G(D,P) OPTIMIZED STRUCTURE Symbolic Z-matrix: Charge =-1 Multiplicity = 1 N C 1 CN X 2 1. 1 90. o 2 CO 3 90. 1 180. 0 Variables: CN 1.21407 CO 1.24452 AI-5 NH 3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE Symbolic Z-matrix: Charge = 0 Multiplicity = 1 N X 1 1. H 1 R 2 A H 1 R 2 A 3 120. 0 H 1 R 2 A 3 -120. 0 Variables: R 1.0117 A 110.82744 Al-6 ------------------------------------------ N(CH 3 )3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ------------------------------------------ Symbolic Z-matrix: Charge =0 Multiplicity =1 X N 1 1. X 1 1. 2 90. X 1 1. 3 90. 2 O. 0 C 2 CN 1 ANG 3 O. 0 H 5 CH6 2 ANG6 1 180. 0 H 5 CH7 2 ANG7 6 120. 0 H 5 CH7 2 ANG7 6 -120. 0 X 2 1. 1 90. 3 120. 0 C 2 CN 1 ANG 3 120. 0 H 10 CH6 2 ANG6 9 180. 0 H 10 CH7 2 ANG7 11 120. 0 H 10 CH7 2 ANG7 11 -120. 0 X 2 1. 1 90. 3 -120. 0 C 2 CN 1 ANG 3 -120. 0 H 15 CH6 2 ANG6 14 180. 0 H 15 CH7 2 ANG7 16 120. 0 H 15 CH7 2 ANG7 16 -120. 0 Variables: CN 1.45524 ANG 108.31107 CH6 1.10391 ANG6 112.12778 CH7 1.08927 ANG7 109.58619 Al-7 PYRROLE MP2/6-31+G(D,P) OPTIMIZED STRUCTURE -------------------------------------- Symbolic Z-matrix: Charge =0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 Variables: NH 1.00719 NC3 1.37454 ANG3 124.91284 CC5 1.38571 ANG5 107.39955 CH7 1.07699 ANG7 121.2349 CH9 1.07789 ANG9 125.5103 AI-8 3,4-DIFLUOROPYRROLE MP2/6-31+GCD,P) OPTIMIZED STRUCTURE ------------------------------- Symbolic Z-matrix: Charge =0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 o. 0 H 4 CH7 1 ANG7 2 o. 0 F 5 CF9 3 ANG9 7 o. 0 F 6 CF9 4 ANG9 8 o. 0 Variables: NH 1.00642 NC3 1.37537 ANG3 124.35981 CC5 1.38271 ANG5 106.17897 CH7 1.07515 ANG7 122.98626 CF9 1.353 ANG9 126.09995 AI-9 2,5-DIFLUOROPYRROLE MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ------------------------ Symbolic Z-matrix: Charge =0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 F 3 CH7 1 ANG7 2 o. 0 F 4 CH7 1 ANG7 2 o. 0 H 5 CF9 3 ANG9 7 O. 0 H 6 CF9 4 ANG9 8 O. 0 Variables: NH 1.00822 NC3 1.37245 ANG3 126.57082 CC5 1.36933 ANG5 1102977 CH7 1.34964 ANG7 118.43637 CF9 1.07588 ANG9 126.21828 Al-lO Al-11 2,5-DlBERYLLIUMPYRROLE+ 2 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ------------------------------------------------------------------ Symbolic Z-matrix: Charge = 2 Multiplicity =1 N R 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 Be 3 CB7 1 ANG7 2 O. 0 Be 4 CB7 1 ANG7 2 O. 0 R 5 CR9 3 ANG9 7 O. 0 R 6 CR9 4 ANG9 8 O. 0 Variables: NH 1.01205 NC3 1.37986 ANG3 124.70633 CC5 1.41761 ANG5 106.80951 CB7 1.63453 ANG7 131.01899 CR9 1.08059 ANG9 126.42836 Al-12 3,4-DIBERYLLIUMPYRROLE+ 2 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ------------------------------------------------------------------ Symbolic Z-matrix: Charge = 2 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 Be 5 CB9 3 ANG9 7 O. 0 Be 6 CB9 4 ANG9 8 O. 0 Variables: NH 1.01704 NC3 1.35921 ANG3 124.71613 CC5 1.40565 ANG5 108.75509 CH7 1.08075 ANG7 120.15547 CB9 1.62595 ANG9 115.79211 Appendix 2 Z-matrices for optimized complexes reported in Tables 5 and 6, taken directly from Gaussian 98 output PYRROLE:NCH MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ------------------------------------------ Symbolic Z-matrix: Charge = 0 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CNC 13 90. 1 180. 0 X 14 1. 12 90. 13 O. 0 H 14 CHC 15 90. 12 180. 0 Variables: NH 1.01145 NC3 1.37274 ANG3 125.0453 CC5 1.38729 ANG5 107.69096 CH7 1.0773 ANG7 121.10689 CH9 1.07817 ANG9 125.63111 R 3.16425 CNC 1.17684 CHC 1.06718 A2-1 -------------------------------- PYRROLE:NCLi MP2/6-31+G(D,P) OPTIMIZED STRUCTURE -------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 H 5 eH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 Li 14 CLi 15 90. 12 180. 0 Variables: NH 1.01918 NC3 1.37077 ANG3 125.1347 CC5 1.38902 ANG5 107.92298 CH7 1.07748 ANG7 120.94505 CH9 1.0785 ANG9 125.73911 R 3.00996 CN 1.19037 CLi 1.95756 A2-2 A2-3 PYRROLE:NCNa MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH ONLY THE 1s ORBITALS FROZEN ----------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 Na 14 CNa 15 90. 12 180. 0 Variables: NH 1.02136 NC3 1.37017 ANG3 125.15207 CC5 1.38941 ANG5 107.97728 CH7 1.07754 ANG7 120.90459 CH9 1.07859 ANG9 125.76743 R 2.97113 CN 1.19061 CNa 2.2352 PYRROLE:NCS- MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ---------------------------------------- Symbolic Z-matrix: Charge =-1 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 S 14 CS 15 90. 12 180. 0 Variables: NH 1.03697 NC3 1.36755 ANG3 125.23316 CC5 1.39197 ANG5 108.24055 CH7 1.07777 ANG7 120.65796 CH9 1.07931 ANG9 125.86253 R 2.83533 CN 1.20192 CS 1.64653 A2-4 -------------------------------- PYRROLE:NCO- MP2/6-31+G(D,P) OPTIMIZED STRUCTURE --------------------~----------- Symbolic Z-matrix: Charge =-1 Multiplicity =1 N R 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 R 3 CR7 1 ANG7 2 O. 0 R 4 CR7 1 ANG7 2 O. 0 R 5 CR9 3 ANG9 7 O. 0 R 6 CR9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 90. 0 0 14 CO 15 90. 12 180. 0 Variables: NH 1.04842 NC3 1.36662 ANG3 125.32201 CC5 1.39319 ANG5 108.43029 CR7 1.078 ANG7 120.5426 CR9 1.07963 ANG9 125.95387 R 2.76188 CN 1.21103 CO 1.23307 A2-5 3,4-DIFLUOROPYRROLE:NCH MP2/6-31+G(D,P) OPTIMIZED STRUCTURE -------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 F 5 CF9 3 ANG9 7 O. 0 F 6 CF9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 H 14 CH 15 90. 12 180. 0 Variables: NH 1.01197 NC3 1.37354 ANG3 124.47348 CC5 1.38394 ANG5 106.42129 CH7 1.07534 ANG7 122.7922 CF9 1.35565 ANG9 126.22311 R 3.11482 CN 1.17644 CH 1.06735 A2-6 3,4-DIFLUOROPYRROLE:NCLi MP2/6-31+G(D,P) OPTIMIZED STRUCTURE --------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 F 5 CF9 3 ANG9 7 O. 0 F 6 CF9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 Li 14 CLi 15 90. 12 180. 0 Variables: NH 1.02162 NC3 1.37164 ANG3 124.561 CC5 1.38518 ANG5 106.624 CH7 1.07546 ANG7 122.603 CF9 1.35875 ANG9 126.302 R 2.95558 CN 1.18991 CLi 1.96118 A2-7 3,4-DIFLUOROPYRROLE:NCNa MP2/6-31 +G(D,P) OPTIMIZED STRUCTURE WITH ONLY THE Is ORBITALS FROZEN ---------------------------------------.--_.- Symbolic Z-matrix: Charge = 0 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 F 5 CF9 3 ANG9 7 O. 0 F 6 CF9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 Na 14 CNa 15 90. 12 180. 0 Variables: NH 1.02453 NC3 1.37101 ANG3 124.58235 CC5 1.38554 ANG5 106.68114 CH7 1.0755 ANG7 122.54787 CF9 1.35956 ANG9 126.32501 R 2.91624 CN 1.19013 CNa 2.23883 A2-8 3,4-DIFLUOROPYRROLE:NCS- MP2/6-31+G(D,P) OPTIMIZED STRUCTURE --------------------------------------------------- Symbolic Z-matrix: Charge = -1 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 F 5 CF9 3 ANG9 7 O. 0 F 6 CF9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 90. 0 S 14 CS 15 90. 12 180. 0 Variables: NH 1.04549 NC3 1.36861 ANG3 124.6613 CC5 1.38743 ANG5 106.88917 CH7 1.07576 ANG7 122.24548 CF9 1.36603 ANG9 126.50156 R 2.77007 CN 1.20183 CS 1.64283 A2-9 A2-1O 3,4-DIFLUOROPYRROLE:NCO- MP2/6-31+G(D,P) OPTIMIZED STRUCTURE --------------------------------------------------- Symbolic Z-matrix: Charge = -1 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 II 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 F 5 CF9 3 ANG9 7 O. 0 F 6 CF9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 90. 0 0 14 CO 15 90. 12 180. 0 Variables: NH 1.06116 NC3 1.36779 ANG3 124.74926 CC5 1.38839 ANG5 107.05874 CH7 1.07595 ANG7 122.11971 CF9 1.36825 ANG9 126.58183 R 2.69898 CN 1.21034 CO 1.2305 2,5-DIFLUOROPYRROLE:NCH MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ----------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 F 3 CF7 1 ANG7 2 O. 0 F 4 CF7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 H 14 CH 15 90. 12 180. 0 Variables: NH 1.01597 NC3 1.36978 ANG3 126.75778 CC5 1.3709 ANG5 110.70464 CF7 1.35217 ANG7 118.57975 CH9 1.07602 ANG9 126.3571 R 3.06836 CN 1.17607 CH 1.06723 A2-11 A2-12 2,5-DIFLUOROPYRROLE:NCLi MP2/6-31+GCD,P) OPTIMIZED STRUCTURE --------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 F 3 CF7 1 ANG7 2 O. 0 F 4 CF7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 Li 14 CLi 15 90. 12 180. 0 Variables: NH 1.02908 NC3 1.36752 ANG3 126.896 CC5 1.37259 ANG5 111.015 CF7 1.35387 ANG7 118.831 CH9 1.07619 ANG9 126.437 R 2.90552 CN 1.18942 CLi 1.9598 A2-13 2,5-DIFLUOROPYRROLE:NCNa MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH ONLY THE Is ORBITALS FROZEN ------------------------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 F 3 CF7 1 ANG7 2 O. 0 F 4 CF7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 Na 14 CNa 15 90. 12 180. 0 Variables: NH 1.03329 NC3 1.36672 ANG3 126.93282 CC5 1.37298 ANG5 111.10681 CF7 1.35445 ANG7 118.88337 CH9 1.07624 ANG9 126.46172 R 2.86275 CN 1.18956 CNa 2.236 2,5-DIFLUOROPYRROLE:NCS· MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ---------------------------------------------------- Symbolic Z-matrix: Charge =-1 Multiplicity =1 N R 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 F 3 CF7 1 ANG7 2 o. 0 F 4 CF7 1 ANG7 2 o. 0 R 5 CR9 3 ANG9 7 o. 0 R 6 CR9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 S 14 CS 15 90. 12 180. 0 Variables: NH 1.06469 NC3 1.36385 ANG3 127.16137 CC5 1.37605 ANG5 111.59723 CF7 1.35582 ANG7 119.32725 CR9 1.07672 ANG9 126.49287 R 2.71255 CN 1.20091 CS 1.64206 A2-14 A2-15 2,5-DIFLUOROPYRROLE:NCO- MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ---------------------------------------------------- Symbolic Z-matrix: Charge =-1 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 F 3 CF7 1 ANG7 2 O. 0 F 4 CF7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 90. 0 0 14 CO 15 90. 12 180. 0 Variables: NH 1.09727 NC3 1.36218 ANG3 127.3628 CC5 1.37767 ANG5 111.99689 CF7 1.35866 ANG7 119.41544 CH9 1.07702 ANG9 126.60676 R 2.62714 CN 1.20816 CO 1.22899 A2-16 3,4-DIBERYLLIUMPYRROLE+ 2 :NCH MP2/6-31+G(D,P) OPTIMIZED STRUCTURE ---------------------------------------------------------------------- Symbolic Z-matrix: Charge = 2 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 Be 5 CB9 3 ANG9 7 O. 0 Be 6 CB9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 90. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 H 14 CH 15 90. 12 180. 0 Variables: NH 1.04261 NC3 1.3562 ANG3 124.97802 CC5 1.40873 ANG5 109.27251 CH7 1.08096 ANG7 120.01148 CB9 1.61987 ANG9 115.86886 R 2.82797 CN 1.17367 CH 1.07127 A2-17 3,4-DIBERYLLIUMPYRROLE+ 2 :NCLi MP2/6-31+G(D,P) OPTIMIZED STRUCTURE o o o o o o o 3 180. 2 180. 2 180. 2 o. 2 o. 7 o. 8 o. o. 0 o. 0 o. 0 180. 0 o. 0 180. 0 2 ANG3 2 ANG3 1 ANG5 1 ANG5 1 ANG7 1 ANG7 3 ANG9 4 ANG9 2 90. 3 11 90. 2 1 90. 11 13 90. 2 12 90. 13 15 90. 12 1 NH 1 NC3 1 NC3 3 CC5 4 CC5 3 CH7 4 CH7 5 CB9 6 CB9 1 1. 1 R 12 1. 12 CN 14 1. 14 CLi Symbolic Z-matrix: Charge = 2 Multiplicity = 1 N H C C C C H H Be Be X N X C X Li Variables: NH NC3 ANG3 CC5 ANG5 CH7 ANG7 CB9 ANG9 R CN CLi 1.119 1.35389 125.36541 1.41281 109.9593 1.08143 119.86253 1.61305 115.84257 2.60336 1.1865 2.03465 3,4-DIBERYLLIUMPYRROLE+ 2 :NCNaMP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH ONLY THE Is ORBITALS FROZEN ----------------------------------------------------------------- Symbolic Z-matrix: Charge = 2 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 o. 0 H 4 CH7 1 ANG7 2 o. 0 Be 5 CB9 3 ANG9 7 O. 0 Be 6 CB9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 X 12 1. 1 90. 11 O. 0 C 12 CN 13 90. 2 180. 0 X 14 1. 12 90. 13 O. 0 Na 14 CNa 15 90. 12 180. 0 Variables: NH 1.57739 NC3 1.35632 ANG3 126.58773 CC5 1.41964 ANG5 111.81214 CH7 1.08261 ANG7 120.05359 CB9 1.60544 ANG9 116.69726 R 2.66562 CN 1.17928 CNa 2.37425 A2-18 A2-19 PYRROLE:NH 3 MP2/6-31+0(D,P) OPTIMIZED STRUCTURE WITH N-H IN THE PLANE OF PYRROLE ---------------------------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 AN03 C 1 NC3 2 AN03 3 180. 0 C 3 CC5 1 AN05 2 180. 0 C 4 CC5 1 AN05 2 180. 0 H 3 CH7 1 AN07 2 O. 0 H 4 CH7 1 AN07 2 O. 0 H 5 CH9 3 AN09 7 O. 0 H 6 CH9 4 AN09 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 H 12 NH13 1 AN013 11 O. 0 H 12 NH13 1 AN013 13 120. 0 H 12 NH13 1 AN013 13 -120. 0 Variables: NH 1.02051 NC3 1.37221 AN03 125.17681 CC5 1.38816 AN05 107.92035 CH7 1.07762 AN07 121.09502 CH9 1.07829 AN09 125.71059 R 3.03747 NH13 1.01419 AN013 111.87013 A2-20 3,4-DIFLUOROPYRRROLE:NH 3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH N-H IN THE PLANE OF PYRROLE ---------------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 F 5 CF9 3 ANG9 7 O. 0 F 6 CF9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 H 12 NH13 1 ANG13 11 O. 0 X 12 1. 1 ANG14 13 180. 0 H 12 NH15 14 HALF 2 90. 0 H 12 NH15 14 HALF 2 -90. 0 Variables: NH 1.02315 NC3 1.37312 ANG3 124.61556 CC5 1.3846 ANG5 106.65671 CH7 1.0756 ANG7 122.7817 CF9 1.35641 ANG9 126.26573 R 2.99166 NH13 1.01442 ANG13 111.0618 ANG14 129.79508 NH15 1.01426 HALF 53.52968 A2-21 PYRROLE:N(CH 3 )3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH C-HIN THE PLANE OF PYRROLE ----------------------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 C 12 CN13 1 ANG13 11 O. 0 H 13 CH14 12 ANG14 2 180. 0 H 13 CH15 12 ANG15 14 120. 0 H 13 CH15 12 ANG15 14 -120. 0 X 12 1. 1 90. 11 120. 0 C 12 CN13 1 ANG13 11 120. 0 H 18 CH14 12 ANG14 17 180. 0 H 18 CH15 12 ANG15 19 120. 0 H 18 CH15 12 ANG15 19 -120. 0 X 12 1. 1 90. 11 -120. 0 C 12 CN13 1 ANG13 11 -120. 0 H 23 CH14 12 ANG14 22 180. 0 H 23 CH15 12 ANG15 24 120. 0 H 23 CH15 12 ANG15 24 -120. 0 Variables: NH 1.02932 NC3 1.37262 ANG3 125.23777 CC5 1.38864 ANG5 108.00819 CH7 1.07797 ANG7 121.09147 CH9 1.0784 ANG9 125.74811 R 2.93063 CN13 1.46104 ANG13 CH14 ANG14 CH15 ANG15 108.27578 1.10083 111.72914 1.08932 109.51702 A2-22 A2-23 2,5-DIFLUOROPYRROLE:NH 3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH N-H IN THE PLANE OF PYRROLE -------------------------------------------------- Symbolic Z-matrix: Charge = aMultiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. a C 3 CC5 1 ANG5 2 180. a C 4 CC5 1 ANG5 2 180. a F 3 CF7 1 ANG7 2 o. a F 4 CF7 1 ANG7 2 o. a H 5 CH9 3 ANG9 7 o. a H 6 CH9 4 ANG9 8 o. a x 1 1. 2 90. 3 o. a N 1 R 11 90. 2 o. a H 12 NH13 1 ANG13 11 o. a x 12 1. 1 ANG14 13 180. a H 12 NH15 14 HALF 2 90. a H 12 NH15 14 HALF 2 -90. a Variables: NH 1.03199 NC3 1.36824 ANG3 126.91867 CC5 1.37175 ANG5 111.02601 CF7 1.35493 ANG7 118.46533 CH9 1.07611 ANG9 126.49143 R 2.92369 NH13 1.01461 ANG13 110.77965 ANG14 129.57577 NH15 1.01432 HALF 53.52718 A2-24 3,4-DIFLUOROPYRROLE:N(CH 3 )3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH C-H IN THE PLANE OF PYRROLE -----------------------------------------------~-------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 F 5 CF9 3 ANG9 7 O. 0 F 6 CF9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 C 12 CN13 1 ANG13 11 O. 0 H 13 CH14 12 ANG14 2 180. 0 H 13 CH15 12 ANG15 14 120. 0 H 13 CH15 12 ANG15 14 -120. 0 X 12 1. 1 90. 11 120. 0 C 12 CN13 1 ANG13 11 120. 0 H 18 CH14 12 ANG14 17 180. 0 H 18 CH15 12 ANG15 19 120. 0 H 18 CH15 12 ANG15 19 -120. 0 X 12 1. 1 90. 11 -120. 0 C 12 CN13 1 ANG13 11 -120. 0 H 23 CH14 12 ANG14 22 180. 0 H 23 CH15 12 ANG15 24 120. 0 H 23 CH15 12 ANG15 24 -120. 0 Variables: NH 1.03483 NC3 1.37355 ANG3 124.68895 CC5 1.38506 ANG5 106.76203 CH7 1.07593 ANG7 122.76846 CF9 1.35665 ANG9 126.31675 R 2.8813 CN13 1.4626 ANG13 CH14 ANG14 CH15 ANG15 108.36704 1.10006 111.64753 1.08929 109.51161 A2-25 A2-26 2,5-DIBERYLLIUMPYRROLE+ 2 :NH 3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH N-H IN THE PLANE OF PYRROLE ---------------------------------------------------------------------- Symbolic Z-matrix: Charge = 2 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 Be 3 CB7 1 ANG7 2 O. 0 Be 4 CB7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 H 12 NH13 1 ANG13 11 O. 0 X 12 1. 1 ANG14 13 180. 0 H 12 NH15 14 HALF 2 90. 0 H 12 NH15 14 HALF 2 -90. 0 Variables: NH 1.04998 NC3 1.37678 ANG3 125.29748 CC5 1.42237 ANG5 107.8302 CB7 1.62835 ANG7 130.43711 CH9 1.08061 ANG9 126.61663 R 2.86999 NH13 1.0177 ANG13 114.61623 ANG14 131.06211 NH15 1.01801 HALF 52.84008 A2-27 2,5-DIFLUOROPYRROLE:N(CH 3 )3 MP2/6-31+GCD,P) OPTIMIZED STRUCTURE WITH C-H IN THE PLANE OF PYRROLE -------------------------------------------------------------- Symbolic Z-matrix: Charge = 0 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 F 3 CF7 1 ANG7 2 O. 0 F 4 CF7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 C 12 CN13 1 ANG13 11 O. 0 H 13 CH14 12 ANG14 2 180. 0 H 13 CH15 12 ANG15 14 120. 0 H 13 CH15 12 ANG15 14 -120. 0 X 12 1. 1 90. 11 120. 0 C 12 CN13 1 ANG13 11 120. 0 H 18 CH14 12 ANG14 17 180. 0 H 18 CH15 12 ANG15 19 120. 0 H 18 CH15 12 ANG15 19 -120. 0 X 12 1. 1 90. 11 -120. 0 C 12 CN13 1 ANG13 11 -120. 0 H 23 CH14 12 ANG14 22 180. 0 H 23 CH15 12 ANG15 24 120. 0 H 23 CH15 12 ANG15 24 -120. 0 Variables: NH 1.05382 NC3 1.36713 ANG3 127.03289 CC5 1.37264 ANG5 111.26767 CF7 1.35739 ANG7 118.25659 CH9 1.07629 ANG9 126.63561 R 2.78489 CN13 1.46348 ANG13 CH14 ANG14 CH15 ANG15 108.09165 1.09938 111.45819 1.08913 109.45257 A2-28 A2-29 3,4-DIBERYLLIUMPYRROLE+ 2 :NH 3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH N-H IN THE PLANE OF PYRROLE ---------------------------------------------------------------------- Symbolic Z-matrix: Charge = 2 Multiplicity = 1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 Be 5 CB9 3 ANG9 7 O. 0 Be 6 CB9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 H 12 NH13 1 ANG13 11 O. 0 X 12 1. 1 ANG14 13 180. 0 H 12 NH15 14 HALF 2 90. 0 H 12 NH15 14 HALF 2 -90. 0 Variables: NH 1.10749 NC3 1.35665 ANG3 125.38834 CC5 1.4108 ANG5 109.89971 CH7 1.08129 ANG7 120.10222 CB9 1.61813 ANG9 116.27728 R 2.69776 NH13 1.01801 ANG13 112.89233 ANG14 130.52483 NH15 1.01797 HALF 53.01118 ---------------------------------------------------------------------- 2,5-DIBERYLLIUMPYRROLE+ 2 :N(CH 3 )3 MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH C-H IN THE PLANE OF PYRROLE ---------------------------------------------------------------------- Symbolic Z-matrix: Charge = 2 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 Be 3 CB7 1 ANG7 2 O. 0 Be 4 CB7 1 ANG7 2 O. 0 H 5 CH9 3 ANG9 7 O. 0 H 6 CH9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 C 12 CN13 1 ANG13 11 O. 0 H 13 CH14 12 ANG14 2 180. 0 H 13 CH15 12 ANG15 14 120. 0 H 13 CH15 12 ANG15 14 -120. 0 X 12 1. 1 90. 11 120. 0 C 12 CN13 1 ANG13 11 120. 0 H 18 CH14 12 ANG14 17 180. 0 H 18 CH15 12 ANG15 19 120. 0 H 18 CH15 12 ANG15 19 -120. 0 X 12 1. 1 90. 11 -120. 0 C 12 CN13 1 ANG13 11 -120. 0 H 23 CH14 12 ANG14 22 180. 0 H 23 CH15 12 ANG15 24 120. 0 H 23 CH15 12 ANG15 24 -120. 0 Variables: NH 1.7733 NC3 1.3807 ANG3 127.90116 CC5 1.4362 ANG5 111.91469 CB7 1.61848 ANG7 136.72016 CH9 1.08299 ANG9 127.8354 R 2.83734 CN13 1.49175 A2-30 AN013 CH14 AN014 CH15 AN015 107.80266 1.08759 109.19568 1.08662 108.75663 A2-31 3,4-DIBERYLLIUMPYRROLE+ 2 :N(CH 3 h MP2/6-31+G(D,P) OPTIMIZED STRUCTURE WITH C-H IN THE PLANE OF PYRROLE --------------------------------------~------------------------------- Symbolic Z-matrix: Charge = 2 Multiplicity =1 N H 1 NH C 1 NC3 2 ANG3 C 1 NC3 2 ANG3 3 180. 0 C 3 CC5 1 ANG5 2 180. 0 C 4 CC5 1 ANG5 2 180. 0 H 3 CH7 1 ANG7 2 O. 0 H 4 CH7 1 ANG7 2 O. 0 Be 5 CB9 3 ANG9 7 O. 0 Be 6 CB9 4 ANG9 8 O. 0 X 1 1. 2 90. 3 O. 0 N 1 R 11 90. 2 O. 0 C 12 CN13 1 ANG13 11 O. 0 H 13 CH14 12 ANG14 2 180. 0 H 13 CH15 12 ANG15 14 120. 0 H 13 CH15 12 ANG15 14 -120. 0 X 12 1. 1 90. 11 120. 0 C 12 CN13 1 ANG13 11 120. 0 H 18 CH14 12 ANG14 17 180. 0 H 18 CH15 12 ANG15 19 120. 0 H 18 CH15 12 ANG15 19 -120. 0 X 12 1. 1 90. 11 -120. 0 C 12 CN13 1 ANG13 11 -120. 0 H 23 CH14 12 ANG14 22 180. 0 H 23 CH15 12 ANG15 24 120. 0 H 23 CH15 12 ANG15 24 -120. 0 Variables: NH 1.82163 NC3 1.35924 ANG3 126.9271 CC5 1.42016 ANG5 112.26277 CH7 1.08281 ANG7 120.23648 CB9 1.60752 ANG9 117.53224 R 2.87458 CN13 1.49239 A2-32 ANG13 CH14 ANG14 CH15 ANG15 107.51908 1.08738 108.92436 1.0864 108.73134 A2-33 Appendix 3 Binding Enthalpies(~Ho)for complexes in Tables 5 and 6 A3-1 Binding Enthalpies (kcal/mol) Pl 3,4-diFPy 2,5-diFPy 2,5-diBePy+2 3,4-diBePy+2 NCR -4.5 -5.6 -5.7 -20.2 NCLi -9.3 -11.7 -11.6 -47.2 NCNa -10.7 -13.5 -13.3 -55.8 NCS- -18.1 -23.7 -21.7 NCO- -22.0 -28.6 -27.2 NH 3 -7.0 -8.4 -9.6 -21.4 -25.4 N(CR 3 )3 -9.4 -11.0 -13.3 -32.6 -44.8 a) Py=pyrrole