This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
) and ¥£
(A.3.11) Here we have denoted the matrix elements of the dipole moment operator p for the a2-b and a r b pairs by p^b and paib, respectively. A.3.2 Effective Hamiltonian approach In the considered example, the operator H' takes the form: H'= where
7talb E.s al s b +7t a2b E.s a2 s b
' " ^ ' " [ ^ ( o , ->b)
K*2b=-4h2bPJ
1
(A.3.12)
AE(b-*a,)]
+
a2bya2b[AE(a2^b)
l-
1
(A.3.13)
AE(b^a2)\
When applying the operator H' on the zero-order order wave functions eq A.3.2 we substitute the saisb and s^Sb spin-operators making use of the Dirac permutation operator, which replaces SjSj by (1/2)(P;J-1/2), where Py interchanges spins of electrons occupying orbitals i and j . Here we omit the constant term -(1/4) (of relevance for diagonal matrix elements only): H
' ¥ £ = H'.j=(ata-2b+ -a^a+2b+) =
Modelling of Anisotropic Exchange Coupling 193
-l( 1 - J V JtalbE.Palb +7ta2bE.Pa2b).-Wfll+a2-&+ -araJA*)^[*, ! i I.( f l , + a ! -!> t )-^,£Ka 2 t A-) + f j 2 t i . ( a I t a 2 T ) - ^ I . ( a - i ! 2 t i t ) ] (A.3.14) (*£ \H\V&) = -J3XMJZ + V s ^ J f
(A.3.15)
This is identical with eq A.3.11 obtained using perturbation theory.
References 1. R.Valiente, O.S.Wenger and H.U.Gudel (2000) Chem.Phys.Lett. 320,639. 2. R.Valiente, O.S.Wenger and H.U.Gudel (200l)Phys.Rev.B 63,165102. 3. P.Gerner, O.S.Wenger, R.Valiente and H.U.Gudel (2001) Inorg.Chem. 40, 4534. 4. CReinhard, R.Valiente and H.U.Gudel (2002) J.Phys.Chem.B 106, 10051. 5. Y.Tanabe, LMoriya and S.Sugano, (1965) Phys.Rev.Lett. 15,1023. 6. J.Ferguson, HJ. Guggenheim and Y.Tanabe (1966) J.Phys.Soc.Japan 21, 692. 7. K.-I.Gondaira and Y. Tanabe (1966) LPhys.Soc. Japan 21, 1527. 8. N.Fuchikami and Y.Tanabe (1978) J.Phys.SocJapan 45, 1559. 9. N.Fuchikami and Y. Tanabe (1979) J.Phys.Soc.Japan 47,505. 10. H.Shigi and Y.Tanabe (1982) J.Phys.SocJapan 51, 1415. 11. J.Ferguson, H.J.Guggenheim and Y.Tanabe (1965) J.Appl.Phys. 36, 1046. 12. J.Ferguson, HJ.Guggenheim and Y.Tanabe (1965) Phys.Rev.Lett. 14, 737. 13. J.Ferguson, H.J.Guggenheim and Y.Tanabe (1966) J.Chem.Phys. 45, 1134. 14. J.Ferguson, HJ.Guggenheim and Y.Tanabe (1967) Phys.Rev. 161,207. 15. J.Ferguson, H.U.Gudel and E.R.Krausz, Mol.Phys. 30,1139. 16. H.U.Gudel (1975) Chem.Phys.Lett. 36, 328. 17. H.U.Gudel (1984) Comments Inorg.Chem. 3,189. 18. H.Riesen and H.U.Gudel (1987) MoLPhys. 60, 1221. 19. M.Tachiki (1968) J.Phys.SocJapan 25, 686.
194 M. Atanasov, C. Daul and H. U. Giidel
20. P.W.Anderson (1959) Phys.Rev. 115, 2 21.0.Kahn, Molecular Magnetism (VCH Publishers, Weinheim, 1993) 22. T.Oguchi (1965) J.Phys.Soc Japan 20, 2236. 23. J.Torrance, Jr. and M.Tinkham (1968) J.appl. Phys. 39, 822. 24. W.Urland (1976) Chem.Phys. 14, 393 25. R.W.Schwartz (1977) Inorg.Chem. 16, 1694. 26J.S.Griffith, The Theory of Transition-Metal Ions (Cambridge University press, Cambridge, 1971) 27.M.Atanasov, C.A.Daul and C.Rauzy (2003) Chem.Phys.Lett. 367, 737. 28.M.Atanasov, C.A.Daul and C.Rauzy (2003) Struct.and Bonding, 106, 97. 29. We made use of the Amsterdam Density Functional (ADF) program; E.LBaerends, D.E.Ellis and P.Ros (1973) Chem.Phys. 2,42.; P.M.Boerrigter, G.te Velde and E.J.Baerends (1988) IntJ.Qunatum Chem., 33, 87; G.te Velde and EJ.Baerends (1992) ComputPhys. 99, 84. 30.S.H.Vosko, L.Wilk and M.Nusair (1980) CanJ.Phys. 58, 1200. 31J.P.Perdew, J.A.Chevary, S.H.Vosko, KAJackson, M.R.Pederson, D.J.Singh and CFiolhais (1992) Phys.Rev.B, 46,6671. 32. M.Wolfsberg and L.Hehnholz (1952) J.Chem.Phys. 20, 827. 33. P.Alemany and R.Hoffmann (1993) J.Am.Chem.Soc. 115, 8290. 34. J.des Cloizeaux (1960) Nucl.Phys. 20, 321. 35. M.A.Aebersold, H.U.Gudel, A.Hauser, A.Furrer, H.Blank and R.Kahn (1993)Phys.Rev.B,48, 12723. 36.V.S.Mironov, L.F.Chibotaru and A.Ceulemans (2003) Phys.Rev. B, 67, 14424.
Chapter 6: Is a Dihydrogen Bond a Unique Phenomenon?
Slawomir J. Grabowskiab and Jerzy Leszczynskib a Department
of Crystallography and Crystal Chemistry, University of Lodz, 90-236-todz, ul.Pomorska 149/153, Poland
b Computational
Center for Molecular Structure and Interactions, Department of Chemistry, Jackson State University, Jackson, MS 39217, USA
Abstract The dihydrogen bonds (DHBs) are considered. The examples of such interactions existing in crystal structures are presented. These interactions are compared with conventional hydrogen bonds. The comparison shows that DHBs are often similar in nature to O-H...O and other typical H-bonds. One can observe for DHBs the elongation of the proton donating bond due to complexation and the correlations between geometrical, energetic and topological parameters. There are different types of DHBs, such as those with 7t-electron delocalization, similar to resonance assisted hydrogen bonds (RAHBs). The insight into the nature of DHBs shows that in principle the interaction energy decomposition scheme for them is similar to that one for conventional H-bonds showing that the electrostatic energy term is the most important attractive term. The case of o-bonds as the proton acceptors is also considered. The relationships derived from the bond valence model are in agreement with the experimental neutron diffraction results as well as with ab initio and DFT calculations. The spin-spin coupling constants for dihydrogen bonded systems are analyzed.
195
196 S. J. Grabowski and J. Leszczynski
1. Introduction For a long time the nature of hydrogen bonding has been the subject of many investigations.1'2'3 This is due to the importance of H-bond interactions in chemical, physical and bio-chemical processes, especially life processes.3'4'5 Among the most pronounced examples are catalysis processes, proton transfer reactions, enzymatic reactions, cell membrane transport processes and many others which take place in the gas phase, in liquids and in solids. In addition, this is an important interaction for phase transitions and the driving force in crystal engineering.6 The criteria for the existence of hydrogen bonds are not strictly defined because this interaction is understood as a very broad phenomenon.4'7 Early works on hydrogen bonds define them as interactions between X-H donating bonds and acceptor centers Y; hence, the H-bond is usually designated as X-H...Y where X and Y should be electronegative atoms such as O, N, Cl, etc.2 Hence early works also consider H-bonds as mainly electrostatic interactions where there is excess positive charge on the H-atom and excess negative charge on the X and Y atoms. In principle, there was no problem to detect hydrogen bonds if the geometries of the complexes or interatomic contacts were known. The situation was changed drastically when it was pointed out that the C-H bond may act as a proton donor. The CH...O and C-H...N bonds were analyzed by Suttor in the crystal organic structures.8 The subject of controversy in the 60s and 70s was the issue of whether or not such interactions fulfil the criteria for the existence of hydrogen bonding. However it was commonly accepted that there are such Hbonds after the appearance of the work of Taylor and Kennard.9 The authors have performed refined statistical analyses justifying the existence of C-H... Y H-bonds in crystals. Afterwards also the other kinds of H-bonds were analysed and those with C-atom acceptors (X-H...C) or 7t-electron acceptors (X-H...71).4 Even the C-H...C hydrogen bonds were detected, for example in one of the polymorphic forms of acetylene crystals (with the Acam space group).10 Accorging to Jeffrey,5 H-bonds may be classified in the following way: those for which H-bond energies are about 15-40 kcal/mol are known as strong ones; the range of 4-15 kcal/mol is for moderate H-bonds and 1-4 kcal/mol for weak ones. The H-bonds mentioned here with C-donors or/and C-acceptors usually belong to the last group. However there are exceptions; for example the C-H...C interactions of moderate strength were analyzed theoretically, and the binding energy for H3N+-CH2'...HCCH at the MP2/6311++G(3d,3p) level of theory (BSSE included) was calculated to be 8.16 kcal/mol.11 One can mention the other kinds of unconventional H-bonds
Is a Dihydrogen Bond a Unique Phenomenon?
197
which are recently the subject of investigations: for example, blue-shifting Hbonds for which the blue shift of the stretching mode for the donating bond is observed.12 For conventional H-bonds the stretching vibration of the X-H proton donating bond changes significantly upon complexation with the acceptor molecule. The stretching mode can be shifted to the red of the spectrum by hundreds of cm"1 due to the X-H bond elongation, and the band is intensified several fold. The red shift may be treated as a measure of the hydrogen bonding strength since it correlates with the other typical descriptors such as H-bond energy as well as the proton - acceptor (H...Y) distance.13 Hobza and Havlas have studied intensively the reverse situation for which the shortening of the X-H bond is observed due to complexation. It is accompanied by a shift to the blue of the stretching mode and most often by a decrease in the intensity of the band. Such interactions with mainly C-H bonds as donors were named as "improper" or "inverse" hydrogen bonds, but currently the name blue-shifting H-bonds is commonly accepted.12 In the middle of the 90s, the X-H...H-Y bond (another kind of H-bond interaction) was reported and named as a dihydrogen bond since for such a system there is the typical conventional X-H donor, and another hydrogen atom with excess negative charge which acts as a proton acceptor and is attached to the heavier non-hydrogen atom Y.14 Recently X-H....H-Y interactions have been the subject of an increasing number of studies, both experimental as well as theoretical.15 There are few reviews of dihydrogen bonds; however, they mainly consider these interactions from an experimental point of view.15'1647 The aim of this chapter is to briefly summarise the theoretical studies of different kinds of dihydrogen bonds (DHBs) and the nature of such interactions. It seems that deeper insight into the details of DHBs will be useful for the analysis of a number of chemical and biochemical reactions.18 It was pointed out that DHBs play an important role in many biochemical processes, similar to the involvement of typical conventional X-H...Y bonds.18
2. First examples of dihydrogen bonds The first examples of the existence of dihydrogen bonds (DHBs) were described in detail by Crabtree and co-workers.16 Two such examples which prove that M-H bonds (M-metal) may act as proton acceptors were reported in 1990.19'20 The results of neutron diffraction measurements of the structure of [Ir-(PMe)3(H)(OH)]+ indicate an H...H distance of 2.4 A between the OH
198 S. J. Grabowski and J. Leszczynski
proton and the Ir-H hydride, and the authors have claimed that this probably represents a weak dipole-dipole interaction.19 The neutron diffraction crystal structure of [Fe(H)2(H2)(PEt2-Ph)4] displays the molecular hydrogen acting as a ligand and closely attached to the Fe-H group. This Fe-H...H2 interaction was termed as a "cis-effect,"20 but Morris and co-workers claimed that this may be considered to be intramolecular hydrogen bonding.21 In the middle of the 90s two groups, Morris and co-workers 21'22 and Crabtree's group,23'24'25'26 reported several examples of the metal organic crystal structures containing DHBs. For the structure of [IrH2(Me2CO)2(PPh3)2]+ the syn-orientation of the iminol OH was found which allows the OH proton to closely approach one of the iridium hydrides. This accounts for coupling between the OH and IrH protons (JHH = 2-4 Hz).23 Spectroscopic studies confirm direct H...H interactions since the v(OH) IR band of the iminol group is in the range 3310-3409 cm"1 which corresponds to lower energy than expected for free iminol OH. Similarly, the ! H NMR spectrum confirms the attractive H.. .H intramolecular interction. The cases of intermolecular DHBs were also studied in the middle of the 90s; for example the neutron diffraction solid state studies of the ReH5(PPh3)3 • indole complex14 reveal N-H...H-Re bifurcated intermolecular interactions. Fig. 1 presents the geometrical details of this type of H-bond.
/H\1.734(8)
l 6 8 3 X 1 1 8.i)\ 1 6 3 - 1 ( 6 ) /H
RB
\
/
^N 1.007(6)
\97.2(3)/ 1.683(6)\ / 2212 ( 9 ) H Fig. 1. The geometrical parameters of Re-H...H-N interactions obtained from neutron diffraction measurements, distances in A, angles in degrees.
Is a Dihydrogen Bond a Unique Phenomenon?
199
One can see that one of the H...H contacts is significantly shorter (1.73 A) than the corresponding sum of van der Waals radii, and the N-H.. .H angle for this interaction is equal to 163°, close to linearity as for the conventional intermolecular H-bonds. Fig. 2 shows the packing of molecules for another crystal structure ([ReH5(PPh3)2(imidazole)]) for which intermolecular bifurcated DHBs were detected.27 Spectacular evidence for the existence of DHBs was presented for boraneamine (H3BNH3) by Crabtree and co-workers.16 A comparison of melting points for H3BNH3 (104°C) and isoelectronic species C2H6 (-181°C) show that stronger interactions, not typical van der Waals interactions, should occur for boraneamine. H3BNH3 is a polar compound which may explain such a situation; but for non H-bonding CH3F species isoelectronic with ethane, the melting temperature amounts to -141 °C. The lack of lone electron pairs for H3BNH3 suggests the existence of DHBs. It is worth mentioning that the crystal structure of H3BNH3 was determined by the neutron diffraction method, and the existence of H...H interactions was confirmed.28 Fig. 3 displays the closest contacts between three neighbouring boraneamines. As one can see there are intermolecular bifurcated dihydrogen bonds for this crystal structure. The two H...H contacts are symmetrically equivalent; the H.. .H distance is equal to 2.02(3) A.
H2A
Fig. 2. The neighbouring molecules of [ReH5(PPh3)2(imidazole)] in the X-ray crystal structure. Reprinted with permission from ref 27. © 1995 Royal Society of Chemistry. In the initial studies of DHBs the complexes for which X-H...H-M interactions exist were considered. Hence it was pointed out that DHBs occur
200 5. J. Grabowski and J. Leszczynski
if there is the typical X-H proton donor, such as O-H or N-H, where typically there is excess positive charge on the hydrogen atom. H-M is an accepting group (M designates transition metal or boron), and the H-atom is negatively charged. However, further studies have shown that other kinds of dihydrogen bonds may exist, such as C-H...H-M (M-transition metal) 29 and even CH...H-C.30 The term "dihydrogen bond" was proposed for X-H...H-Y interactions in 1995 by Crabtree and co-workers.31 The authors of the article stated that "a referee points out that these unconventional hydrogen bonds require a new name, for which we suggest the term 'dihydrogen bond.' " This term is currently commonly accepted. Alkorta and co-workers7 claim that the "protic" or "hydride" nomenclature may be applied for hydrogen bonds. There are protic X-H^.-.Y HBs, hydric X-H^.-.Y HBs (designated by Alkorta and co-workers as inverse hydrogen bonds) and protic-hydric X-H1"8 ...'^-Y DHBs. The authors pointed out that protic HBs cover more than 99.9% of the studies of H-bonding, but the situation will change in the future since unconventional hydrogen bonds, in spite of being weak, play an important role in some cases in many chemical and biological processes. m
% I
Fig. 3. The closest N-H.. .H-B contacts for the neutron diffraction crystal structure of boraneamine. Reprinted with permission from ref 28. © 1999 American Chemical Society.
Is a Dihydrogen Bond a Unique Phenomenon?
201
3. First theoretical studies of DHBs The first theoretical calculations of dihydrogen-bonded systems were performed by Liu and Hoffmann.32 The authors considered the structure of [Ir{H(Ti1-SC5H4NH)(PCy3)}2]+ [BF4]" (where Cy = cyclohexyl) synthesized and experimentally investigated by Morris and co-workers.22 To simplify calculations Liu and Hoffman neglected the counter ion [BF4]' as not influencing the region with dihydrogen bonds. They also replaced the PCy3 group with PH3 and slightly changed the geometry to finally obtain the [Ir{H(Ti1-SC5H4NH)(PH3)}2]+ ion with C2v symmetry. The authors have applied the extended Huckel (EH) method in their study.33 The geometrical parameters for the calculations have been taken from crystallographic data since the EH method is not suitable for geometry optimizations. The considered model ion has two intramolecular symmetry equaivalent N-H.. .HIr dihydrogen bonds. The H...H distances are equal to 1.75 A, less than the corresponding sum of van der Waals radii. The calculated Mulliken net atomic charges amount to +0.22 and -0.27 for the H(N) and H(Ir) atoms, respectively, showing that an attractive electrostatic interaction should appear. The authors claimed that a positive Mulliken overlap population (OP) of 0.016 also indicates stabilization of the H.. .H interaction. Liu and Hoffman32 have also considered two model complexes (FH.. .HLi and FH...HMn(CO)5) using the RHF method. The first structure after the optimization (the 6-3 lG(d) basis set was used) has an H.. .H distance of 1.658 A with Cs symmetry and nearly linear F-H...H and Li-H...H angles. The binding energy for this complex without BSSE calculations is equal to 9.3 kcal/mol. An approximation of the electrostatic contribution using Mulliken atomic charges allowed the authors to obtain the electrostatic term of 6.7 kcal/mol. Hence they claimed that this is mainly an electrostatic interaction as observed for the other typical H-bonds. For the second FH.. .HMn(CO)5 complex calculated at the RHF/3-21G(d) level of theory, a minimum energy structure was found for the H.. .H distance of 1.683 A and the binding energy of 6.6 kcal/mol. The FH and H-Mn bonds were elongated due to complexation. Taking into account the features of these two model complexes discussed here, Liu and Hoffman claimed that they are similar in nature to the other H-bonds. Almost at the same time as the appearance of the work of Liu and Hoffman, a theoretical work of Crabtree and co-workers31 was published. In this study the H3BNH3 dimer was investigated. The authors studied the cyclic dimer of C2 symmetry with equivalent B-H.. .H-N bonds. The PCI-80/B3LYP method was applied, and the following geometrical parameters were found for
202 S. J. Grabowski and J. Leszczynski
dihydrogen bonds: the H.. .H contact is equal to 1.82 A; the N-H.. .H angle is 158.7°; and B-H..H is 98.8°. It is worth mentioning that in the scope of the PCI-80/B3LYP method the geometries were obtained after optimization with the B3LYP functional34 and the standard double X, plus polarization basis sets. The energies for the optimized structures were predicted with the use of the PCI-80 parametrized method35 which gives greatly improved results compared to unparametrized methods. The binding energy calculated for the H3BNH3 dimer amounts to 12.1 kcal/mol which corresponds to 6.1 kcal/mol for each dihydrogen bond. The Mulliken net atomic charges are equal to +0.27e and -0.09e for the H(N) and H(B) atoms, respectively. The authors claim that the existence of opposite charges at the hydrogen atoms of the H...H contact indicates that this interaction may be classified as hydrogen bonding. The same conclusion may be derived from the H...H contact energy which corresponds to the range of the typical H-bond energies (3-8 kcal/mol). The angles of the B-H...H-N system are in agreement with those found from the Cambridge Structural Database (CSD).36 Crabtree and co-workers searched CSD for B-H...H-N contacts and found 26 such species for which the H.. .H distance is less than 2.2 A. This search filter was applied since the authors expected that such a value (less than the corresponding sum of van der Waals radii at 2.4 A) is evidence of the attractive interaction. It is of course only a rough assumption since the crystal packing may be constrained more than the sum of van der Waals radii interatomic contacts. It was found that the N-H...H(B) angles vary (an average of 149°, a range of 117-171°, o=17°); they are larger than the B-H.. .H(N) angles (an average of 120°, a range of 90-171°, o=26°). Hence the results for the H3BNH3 dimer are in line with the tendencies found for the crystal structure complexes. The authors claim that the strong bending of the B-H...H(N) angle is evidence that o-bonds are acceptors of protons, not Hatoms with a negative net charge. In further studies37 they explain that the hydrogen bond concept has been extended in recent years from the typical XH.ione pair interaction to X-H...71 electron interactions, and the last case involves the interaction between the proton donating X-H bond and o-bond acceptor, the X-H...G interaction commonly known as the dihydrogen bond. Such a concept was also suggested by Cramer and Gladfelter.38 Theoretical calculations up to the MP2/6-31G(d,p) level with the BSSE correction (counterpoise method of Boys and Bernardi39) were carried out on a series of A-H...H-X dihydrogen bonds (A= B, Li, Be; X= N, C).40 It was found that the considered complexes fulfilled the commonly accepted criteria for the existence of hydrogen bonds. The systems studied were BH4...HCN,
Is a Dihydrogen Bond a Unique Phenomenon?
203
BH4...CH4, LiH...NH4+, LiH...HCN, LiH...HCCH,BeH2... NH,+, BeH2...HCN and CH4... NHt*. It is worth mentioning that Alkorta and co-workers were the first to apply the Bader theory41 to analyse dihydrogen bonded systems. Systematic studies of simple dihydrogen-bonded complexes have been performed by Remko42 who considered electron-rich hydrogen atoms as proton acceptors (for LiH and BeH2 molecules) and as proton donors with positive H-atoms: HF, HCN, and HCCH. For these complexes full geometry optimization was performed at the MP2(full)/6-31G(d) level. Such obtained geometries were used to carry out the Gl and G2 energetic calculations. Generally, the obtained results show that the dihydrogen bonds are energetically similar to the typical, conventional H-bonds. The H...H distances are in the range of 1.575 A (for LiH...HF) to 2.271 A (for H-BeH...HF). Orlova and Scheiner have studied43 intermolecular H...H interactions between HF, H2O and H3O+ donating species and metalloorganic compounds such as Mo(H)(CO)2(L)2(L')2 (where cis-ligand L=PH3 and NH3, while L' trans-ligand =NO,C1,H) and W(H)(CO)2(NO)(PH3)2 using the HF and DFT methods. The results show that the energy of the H...H contact for these complexes surpasses 10 kcal/mol. The same authors investigated the specific role of the H...H bonded complexes in reactions of transition metal hydrides with the proton donors. The ruthenium and rhenium compounds were considered, and the calculations were performed at the DFT level using nonlocal correction. The B3PW91 and LANL2DZ basis sets were used.44 Rozas and co-workers have investigated the problem of the influence of field effects on dihydrogen bonded systems.18 They found, using the HF/631G(d,p) level of theory, that a coordinated H2 molecule within the Li+...H2...F" system may react yielding the dihydrogen bonded system LiH...H-F. And a dihydrogen bonded complex may yield two molecules coordinated to the H2 molecule as in the case of the following reaction: H3lSrH...H-BH3" = NH3...H2...BH3. The other early calculations of dihydrogen-bonded complexes include particularly small systems where acceptor molecules involve main group elements, eg., LiH, BeH2, BH3 and A1H3.3M5'46'47
4. Geometrical and energetic features of DHBs The equilibrium geometry of molecules may be treated as a source of chemical information. Especially, geometrical parameters describe the nature of intra- and intermolecular interactions.3'4 Some of them are sensitive to
204 S. J. Grabowski and J. Leszczynski
environmental effect; terminal bonds are usually elongated due to the H-bond formation. For example this was observed for O-H...O=C systems.48 For the neutron diffraction results taken from the Cambridge Structural Database,36'49 it was revealed that not only the proton donating bonds but also the accepting C=O bonds are elongated.48 The statistical approach has showed that for shorter H...O contacts the elongation of O-H and C=O bonds is greater. The proton donating bond elongation is usually considered as strong evidence of the existence of hydrogen bonding3'4 and it was many times detected for different types of hydrogen bonds. There are also good correlations between the H...Y distance and the X-H proton donating bond length for X-H...Y Hbonds.3'5 Such correlations were observed for O-H...O systems,50'51 for NH...O, N-H...N52 and even for species with C-H...O.53 The only criterion that should be fulfilled for the investigated samples is that they must consist of related compounds; in other words the samples have to be homogeneous.54 Fig. 4 shows the relationship between the H...O distance and the O-H bond length for O-H...O bonds taken from accurate crystal structure neutron diffraction results.55 The open circles relate to the O-H...O systems, while the full ones correspond to deuterated crystals (O-D...O systems). The solid line represents the relationship between H...0 and O-H obtained from the Bond Valence Model (BV model - see Appendix). It is worth mentioning that Fig. 4 shows the correlation obtained for the sample of O-H...O systems taken from different crystal structures, and in spite of the heterogenity of the sample the relationship is fulfilled. Similar relationships for O-H...O species were obtained from the results of ab initio and DFT calculations.55'56 The presented dependencies are well fulfilled for typical H-bonds. For example the elongation of the proton donating bond is not a criterion for the existence of such interactions for blueshifting H-bonds for which the shortening of the X-H bond is observed due to the H-bond formation.12 Similarities between dihydrogen bonds and conventional O-H.. .O bonds were investigated,57 and a relationship between the elongation of the proton donating bond and the H...H distance was found for samples of simple complexes (Fig. 5). The following samples were taken into account. The first one consists of XH4..HF complexes (where X designates the elements of the fourth group of the periodic table of elements: C, Si, Ge, Sn and hydrogen fluoride was chosen as relatively strong proton donor). The CrH2...HF complex is also included in the first sample to show the proton accepting abilities of transition metal hydrate since the first crystal structures for which DHBs were observed represent metal organic compounds of transition metals. This sample is designated by circles in Fig. 5, and the related complexes were
Is a Dihydrogen Bond a Unique Phenomenon?
205
calculated within the B3LYP/Lanl2DZ level of theory. For two remaining samples there is the same proton donor molecule (HF) and the following acceptors: CH4, S1H4, BeH2, MgH2, LiH and NaH. For one of the samples the optimization of complexes was performed at the MP2/6-31G(d,p) level of theory (triangles in Fig. 5); for another sample optimization was performed at the MP2/6-311++G(d,p) level (squares). All calculations were carried out with the use of the Gaussian98 suite of programs.58 One can see good correlation between the H...H distance and the proton donating bond length, similarly as for typical H-bonds. The greatest elongation of the H-F bond and the shortest H...H contact for the first sample is revealed for the CrH2...HF complex; for the remaining samples the greatest elongations occur for LiH...HF and NaH...HF. Fig. 5 also displays the relationship between the H...H distance and the elongation of the H-F bond obtained from the Bond Valence model (continuous line - see Appendix). The results of a simple BV model are in good agreement with the MP2 and DFT calculations. The presented analysis of DHBs was based on the samples for which there is the same proton donor (hydrogen fluoride molecule) and different acceptors. , i
1.26 -j
o
°#
|
1.020.94-I 1.1
i
°^L 1 1.4
1 1.7
1 2
i 2.3
H..0 distance
Fig. 4. The relationship between the H...O distance and the O-H bond length (both parameters in A). Reprinted with permission from ref 55. © 2000 Elsevier Science B.V.
206 S. J. Grabowski and J. Leszczynski 2.8 I
— — —
~
2.4 -1
i
\ 1.6-
*
1.2 J
0
^ ^
•
1
•
0.01
0.02
0.03
•
1
~
0.04
0.05
elongation of H-F Fig. 5. The dependence between the elongation of the H-F bond and the H...H distance (both values in A). Reprinted with permission from ref 57. © 1999 Elsevier Science B.V.
A similar analysis based on MP2 results of DHBs was performed for the sample where there is the same proton acceptor (the BeH2 molecule), and there are different donors such as HF, HC1, HBr, HCN, HCCH, HNNN, HBF2 and HOF.59 For such a sample, there is no correlation either between the elongation of the Be-H donating bond and the H.. .H distance or between this elongation and the H-bond energy. Practically the Be-H bond length is constant; there is only one exception: the BeH2.. .HOF complex for which the H-bond is strongest within the sample (4.8 kcal/mol) and for which a slight elongation of Be-H is observed due to the H-bond formation (of about 0.03
A).
The following geometrical criteria for the existence of hydrogen bonds are commonly applied and accepted by crystallographers:9 the distance between the proton and the acceptor atom is shorter than the sum of their van der Waals radii;
Is a Dihydrogen Bond a Unique Phenomenon?
207
the conformation of the X-H... Y system should be close to linearity, at least for intermolecular H-bonds because for intramolecular H-bonds the strain effects may force greater deviations from linearity; commonly the systems for which the X-H.. .Y angles are less than 90° are not considered as Hbonds; - sometimes the next geometrical criterion is included which requires that the proton donating bond is elongated due to the H-bond formation.3'53 However these criteria are controversial since the hydrogen bonding interaction is mainly electrostatic, and there is the electrostatic attraction far beyond the van der Waals cut-off.4'51 It is also difficult to estimate accurate positions of hydrogen atoms from X-ray crystal structure measurements. Hence the neutron diffraction results should be applied for a geometrical analysis of H-bonds in crystals.3 However the number of neutron diffraction crystal structures collected in CSD is less than 0.5% (about 1000 crystal structures)60 of the whole database (the remaining are X-ray crystal structures), and sometimes for the statistical analysis of a required class of compounds there is not a sufficient number of neutron diffraction results. For the weaker C-H...Y bonds the elongation of the proton donating bond is not detectable or is negligible.60 Additionally, as was mentioned before, there are blue-shifting H-bonds for which the third criterion cannot be applied. Recently few crystal structures were analyzed where there are repulsive interactions existing due to crystal packing and fulfilling the H-bond geometrical criteria.61 In spite of the reservations given above these criteria may be applied for DHBs shown in Fig. 5. These ab initio results accurately reveal the H-atom positions, and elongation of HF bonds is detectable. For some cases the H.. .H distance is much smaller than the sum of van der Waals radii. For example for the NaH...HF complex the H...H distance is equal to 1.371 A, and the H-F bond length amounts to 0.958 A (the MP2/6-311++G(d,p) level), while for HF not involved in H-bond interactions the bond length is equal to 0.917 A. The analyzed systems have features typical of other conventional H-bonds, a meaningful difference between the energy of the complex on one hand and the energies of isolated monomers on the other hand. The binding energy is often treated as the H-bond energy of intermolecular connections. It amounts to -13.8 and -12.6 kcal/mol for the NaH...HF and LiR.HF complexes (the MP2-6-311++G(d,p) level). One can see that selected DHBs are stronger than the typical conventional H-bond of the linear-trans water dimer (about -5 kcal/mol). And for all DHB systems presented here there are the opposite net
-
208 S. J. Grabowski and J. Leszczynski
atomic charges on connected H-atoms, positive on H(F) and negative on the second H-atom. The only exception is the CH4...HF dimer for which both charges are positive and which is not energetically stable; the binding energy is positive (repulsive) at the MP2/6-311++G(d,p) level (0.39 kcal/mol). All XH.. .H systems included in Fig. 5 are linear or are close to linearity. The relationships mentioned above were confirmed by the use of high level ab initio calculations performed at the MP4(SDQ)/6-311++G(d,p), OCISD/6-311++G(d,p) and QCISD(T)/6-311++G(d,p) levels of theory.62 For example, the H-bond energy of the NaH...HF complex calculated at the latter level of theory amounts to -12.66 kcal/mol (correction for BSSE included), and the corresponding H.. .H distance is equal to 1.426 A. In this high level ab initio study the Bader theory41 was also applied to analyse DHBs (see next sections).
5. Application of the "atoms in molecules" theory to study DHBs The electron density, p(r), defines a structure as well as ground-state properties of many-electron systems.63 The topological "atoms in molecules" (AIM) theory of Bader41 allows quantitative structural information from electron density to be obtained. The AIM theory is often applied to locate and characterize critical points. The following relation, which requires that the gradient of p(r) vanish, stands for the definition of the critical points (CPs): (1)
Vp(rJ = 0
CPs are classified according to the number of negative eigenvalues of the matrix of partial second derivatives of p with respect to {x,y,z} (Hessian matrix). h
=
_a^n
( r
)
P.q=x,y,z
(2)
There are four types of stable critical points: maxima in p(r) correspond to attractors usually attributed to nuclei, and minima correspond to cage, bond or ring critical points. The characteristics of bond critical points (BCPs) are often used in studies of inter-atomic interactions. For example, the negative value of Laplacian for BCP means that there is an increase in electron density in the inter-atomic region; such a situation corresponds to the covalent bond. The
Is a Dihydrogen Bond a Unique Phenomenon?
209
positive Laplacian value shows that there is the depletion of the electron density, and this corresponds to the interaction of closed-shell systems: ionic interaction, hydrogen bonding, and the van der Waals interaction.64 The characteristics of the bond critical points may be useful to estimate the H-bond strength. ' ' The correlations between different topological parameters and H-bond energy were found for various samples of complexes. It is worth mentioning that the AIM theory provides new measures of H-bond strength. From a geometrical point of view the proton.. ..acceptor distance is a rough descriptor of H-bond strength, although it does not always correlate with H-bond energy.3 The elongation of the proton donating bond may also reveal the H-bond strength, and for stronger H-bonds the X-H...Y system should be linear. There are three geometrical parameters often used by crystallographers; however, they are not accurate indicators of H-bond strength. The characteristics of BCPs such as the electron density at H...Y (Y designates acceptor) BCP (PH...Y), its Laplacian (V2pH...vX the electron density at the proton donating bond length (PXH) and its Laplacian (V2pXH) are often applied to characterize H-bond interactions. All of these descriptors reveal the H-bond strength and correlate with the H-bond energy if the homogeneous sample of complexes is taken into account.54 One can observe that for typical H-bonds there is the elongation of the proton donating bond due to complexation; the pXH value decreases in comparison with the free donating molecule; V2pXH increases. For stronger H-bonds H...Y contacts are shorter, and hence the PH...Y and V2pH...y values are greater.68 Similar observations were revealed for dihydrogen bonds. Fig. 6 shows the relief map of the electron density for the Li-H...H-F system. One can see the differences between the electron densities of hydrogen atoms. For both of them the densities are much lower than those of the F and Li atoms. However, the hydrogen atom connected with Li is separated from Li (left side of the figure), while the second hydrogen atom is strongly connected with the fluorine atom (right side of the figure).
210 S. J. Grabowski and J. Leszczynski
Fig. 6. The relief map of the electron density of the Li-H...H-F complex; the large "mounds" correspond to the Li (left) and F (right) attractors, small ones to hydrogen atoms. The first rough correlation between H-bond energy and pH...y for DHBs was found by Alkorta and co-workers (Y in such a case denotes an H-atom acceptor). Table 1 presents the topological parameters obtained in subsequent studies from MP2/6-311++G(d,p) wave functions for homogeneous samples of dihydrogen bonded complexes.57'62 For all DHB complexes hydrogen fluoride represents the proton donating molecule. Hence other systems with an HF donor, with conventional H-bonds and calculated at the same level of theory, are included in Table 1 for comparison. The translinear conformation of the water dimer is also included as the most often applied reference system in studies of hydrogen bonding.69
Is a Dihydrogen Bond a Unique Phenomenon? 211
Table 1. The topological parameters (in au) and H-bond energies (EHB, corrected for BSSE and in kcal/mol) for dihydrogen-bonded systems and for other conventional Hbonds at the MP2/6-311++G(d,p) level of theory. For all cases (except for the translinear dimer of water) the hydrogen fluoride molecule is a donor. pXH
V 2 p XH
PH...Y
V 2 p H ... Y
EHB
0.323
-2.323
0.042
0.057
-12.62
NaHa
0.314
-2.205
0.046
0.048
-13.81
BeH2a
0.361
-2.763
0.016
0.049
3!94
MgH 2a
0.348
-2.624
0.026
0.060
^02
CH/
0.370
-2.830
0.002
0.005
+0.39
SiH,"
0.367
-2.810
0.009
0.028
^085
CH 2 O a a
0.359
-2.787
0.022
0.107
-5.43
H2Oaa
0.347
-2.652
0.037
0.142
^54
NH 3 a a
0.325
-2.365
0.050
0.120
-11.18
F&aa
0.174
-0.349
0.174
-0.349
-39.87
H 2 O" & * aa
0.356
-2.512
0.023
0.091
A45
Acceptor LiH
a
&(F...H...F)"
system liner-trans water dimer arefs 57,59 aaref68 &&
The results presented in Table 1 show the tendencies discussed above that the topological parameters correlate with H-bond energy. For the DHB complexes the linear correlation coefficient between P H . H and H-bond energy is equal to 0.995, while it decreases to 0.991 for all systems presented in the table. This indicates that P H . Y topological parameter may be applied not only for homogeneous samples but also for heterogeneous samples. This was also revealed for the other H-bonded systems. Fig. 7 shows the details of the aforementioned relationships.
212
S. J. Grabowski and J. Leszczynski
5-
.
0*b
c 0 * n 0.05 -5<*o |-15-
,
,
,
0.1
0.15
0.2
*
•S - 2 0 | X
-25
-30 -35-40-
O
-45electron density at H...Y BCP
Fig. 7. The relationship between PH...Y and the H-bond energy for the complexes collected in Table 1: foil circles correspond to DHBs, while empty circles correspond to the other conventional H-bonds. There is also good linear correlation for DHBs between PXH and EHB (R, the linear correlation coefficient, amounts to 0.997). For the relation between V2pXH and EHB, R is equal to 0.991; only the correlation between V2pH...y and EHB is poor (R = 0.687). Similar relationships between parameters describing DHB systems were observed in other studies. For example, for the neutral and charged dihydrogen-bonded complexes with the LiH molecule as a proton acceptor, good correlations between H-bond energy and H...H distance and H-bond energy vs. PH...H were found.71 The authors also observed that there is a slight shortening of the Li-H proton accepting bond for neutral complexes, while for positive charged complexes there is the elongation of the Li-H bond. A detailed analysis of the (BH3NH3)2 dimer based on the AIM theory was performed by Popelier.72 The optimization of the geometry of the dimer was carried out at the HF and MP2 levels of theory using the 6-31G(d,p) basis set. In this case a topological analysis based on appropriate wave functions was performed using the program MORPHY97.73 The positions of the critical
Is a Dihydrogen Bond a Unique Phenomenon?
213
points were found using the eigenvector method.74 There are three B-H...H-N contacts for the optimized (BH3NH3)2 dimer; because of its symmetry (Cs symmetry) two of them are equivalent due to the presence of a mirror plane. The geometrical parameters of these DHB connections for both levels of theory are presented in Table 2. One can see that the N-H...H-B connections of the (BH3NH3)2 dimer roughly fulfil the geometrical criteria for the existence of H-bonds; H...H distances are less than the corresponding sum of van der Waals radii; N-H and B-H bonds are elongated due to complexation; N-H...H and B-H...H angles are greater than 90°. Also the latter angles are smaller than N-H...H angles which is in accordance with the results known for crystal structures where DHBs exist. Table 2. The geometrical parameters of B-H...H-N connections (in A and degrees) for the (BH3NH3)2 dimer (resultsfromref. 72). Parameter
First N-H...H-B bond
I
HF/6-31G(d,p) I MP2/6-31G(d,p)
&
Second N-H...H-B bond & HF/6-31G(d,p)
I MP2/6-3 lG(d,p)
H...H
L914
1/726
2.324
2.149
N-H...H
1631
168?7
128/7
BOO
B-H...H
144.6
139.2
113.8
112.6
AN-H &&
0.004
0.008
0.001
0.002
AB-H**
0.011
0.011
0.007
0.007
there are two such symmetrically equivalent systems the elongation of the bond due to complexation
44
For the (BH3NH3)2 complex Popelier has also investigated the topological criteria for the existence of hydrogen bonds72 which were proposed earlier.75 There are eight topological criteria based on the AIM theory: - there is the bond critical point and the bond path for the H...Y interaction; - the electron density at this BCP ranges from 0.002 to 0.035 au; - the Laplacian of this electron density ranges from 0.024 to 0.139 au; - there is the mutual penetration of the hydrogen and the acceptor atoms; - an increase in net charge of the hydrogen atom is observed; - energetic destabilization of the hydrogen atom; - decrease of dipolar polarization of the hydrogen atom;
214
S. J. Grabowski and J. Leszczynski
- decrease of the hydrogen atomic volume. For the H.. .H contacts of (BH3NH3)2 there are BCPs and bond paths. The HF/6-31G(d,p) level pH...H values are equal to 0.0126 and 0.0067 au (0.0190 and 0.0096 calculated at the MP2/6-31G(d,p) level). The corresponding values of Laplacians (V2pH...H) amount to 0.0353 and 0.0247 au (the HF level) and 0.0464 and 0.0347 au (the MP2 level). One can see that the electron densities and Laplacians are within the ranges proposed by Koch and Popelier. Usually for studies of hydrogen bonds with the use of the AIM theory only the first three criteria are investigated. However, for the (BH3NH3)2 dimer all eight criteria are fulfilled.
6.rc-electron delocalization-assisted intramolecular dihydrogen bonds First experimental works on DHBs have included the metalloorganic crystal structures for which mainly the intramolecular interactions of this type exist. Similarly the first theoretical calculations have been often performed for such systems. For example in the work of Liu and Hoffman32 the intramolecular NH...H-Ir dihydrogen bonds have been analyzed. In this section the cases of intramolecular DHBs for which the effect of TT-electron delocalization occurs are analyzed. The influence of this effect on the strength and other characteristics of dihydrogen bonds are presented. The resonance-assisted hydrogen bonds (RAHBs) are very well known as those whose properties are strongly influenced by 7t-electron delocalization.51'76 For example, malonaldehyde (Fig. 8) is an example of a simple system in which intramolecular resonance-assisted hydrogen bond exists.77 For malonaldehyde and its simple derivatives (different R-substituents Fig. 8) there is equalization of dl and d4 CO bonds on one hand and equalization of the d2 and d3 bonds on the other hand. For such intramolecular RAHBs the Q-parameter was introduced as a descriptor of the delocalization.51 Q = dl-d4 + d3-d2
(3)
Is a Dihydrogen Bond a Unique Phenomenon?
R1
d2
215
C
I R2 Fig. 8. Malonaldehyde (R1=R2=R3=H) is an example of a system with intramolecular H-bonds assisted by Ji-electron delocalization. Reprinted with permissionfromref 77. © 2001 Elsevier Science B.V. The greater is the Jt-electron delocalization, the smaller is the Qparameter, and also the H-bonds are stronger. Such dependencies were observed for experimental RAHBs in organic crystal structures51 as well as were confirmed theoretically for malonaldehyde and its fluoro- and chloroderivatives77 at the MP2/6-311++G(d,p) level of theory. For the border case of delocalization, when there is full equalization of conjugated CO and CC bonds, the proton is exactly in the middle of the 0....0 distance. H-bonds in such a case are very strong, and the H-bond energy amounts to about 30-40 kcal/mol.51 Such systems are often known as low barrier hydrogen bonds.78 It was also pointed out that the electron withdrawing Rl-substituents and the electron donating R3-substituents (Fig. 8) should cause greater 7C-electron delocalization, greater bond length equalization and stronger RAHBs.51 This conclusion was confirmed by MP2/6-311++G(d,p) calculations.77 The influence of 7t-electron delocalization on intramolecular dihydrogen bonds was also studied for (lZ)-2-borylethen-l-ol (Fig. 9)79 and its derivative containing simple R-substituents causing an increase in H-bond energy (R1=C1, R3=Na). For these molecules ab initio calculations were performed up to the MP2/6-311++G(3d,3p) level of theory. The H-bond energies were roughly estimated as the difference between the closed conformations (such as the one
216 S. J. Grabowski and J. Leszczynski
presented in Fig. 8) and the open conformations. The open conformation may be obtained from the closed form after the rotation of the O-H bond by 180° around the dl (C-O) bond. The molecular geometries are optimized before and after rotation. Such an approach in the evaluation of intramolecular Hbond energy is often applied.77'80 However, it is a rough estimation of H-bond energy since other effects come into play such as the unfavourable approach of the lone pairs of the two oxygen atoms after the changing of conformations.80
T<->
TH
d1l R1
^
J B<M [
C •
<*
R3
R2
Fig. 9. (lZ)-2-borylethen-l-ol (R1=R2=R3=H) as an example of intramolecular dihydrogen bonding. Reprinted with permission from ref 79. © 2000 Elsevier Science B.V.
It is worth mentioning that in the case of typical intramolecular RAHBs such as malonaldehyde the 7t-electron delocalization occurs not only for the closed conformation but also for the open one.77 However, for the closed conformation the delocalization and bond equalization should be greater. For (lZ)-2-borylethen-l-ol an elongation of the C=C (d2) and B-H (d4) bonds and a shortening of the C-O (dl) and C-B (d3) bonds can be observed due to the transformation from the open to the closed conformation. Table 3 partly supports this, showing the meaningful influence of the formation of an intramolecular dihydrogen bond on the geometrical parameters of the molecules. The only exception is for the d2 bond length of the A molecule.
Is a Dihydrogen Bond a Unique Phenomenon?
217
Table 3. Geometrical parameters (in A) of (lZ)-2-borylethen-l-ol (A) and its derivative, R1=C1, R2=Na (B), at the MP2/6-311++G(3d,3p) level of theory (results taken from ref. 79). Molecule
I
dl
I
d2
I
d3
I
d4
A closed conformation
1.340
1.346
1.516
1.200
A open conformation
1.352
1.355
1.528
1.189
B closed conformation
1.333
1.367
1.534
1.225
B open conformation
1.343
1.356
1.547
1.202
The results collected in Table 3 also indicate that the delocalization is greater for the B molecule where there are Cl and Na substituents which make this effect stronger. The H...H intramolecular distances amount to 1.933 and 1.647 A for the A and B closed conformations, respectively (the MP2/6311++G(3d,3p) level), while the H-bond energies roughly estimated in the manner described previously are equal to 4.6 and 6.5 kcal/mol. Similarly, the values of the topological parameters show that the intramolecular DHBs presented here are of medium strength. The PH.H values are equal to 0.016 and 0.027 au for A and B, respectively, and the V2pH...H values are equal to 0.049 and 0.059 au for A and B (the MP2/6-311++G(2d,2p) level). A more extended sample of intramolecular dihydrogen bonds was also considered to investigate if the dependencies known for typical H-bonds are fulfilled for DHBs.81 Systems similar to (lZ)-2-borylethen-l-ol were considered, but the -BH2 group was replaced by -BH3\ All possible fluoro derivatives were included in this study. The calculations have been performed at the MP2/6-311++G(d,p) level of theory. Again, the well known approach of roughly estimating H-bond strength as a difference in energy between the closed and open conformations was applied. However, in this case this difference does not correlate with the other parameters describing H-bond strength, not with H...H distance or with the PH...H values. It seems that for this sample with negative charged species, other effects than H...H interactions strongly disturb the H-bond interaction. Hence another approach was applied81 to estimate the H-bond strength. It was found that the local potential energy density at the proton...acceptor contact, V(rcp), for H-bonds correlates with the H-bond energy.67 For H-bonds taken from crystal structures and for their electron densities, the local energy densities, V(rCp), were calculated, and the following relationship was proposed: E H B = V* V(rcp).67 The H-bond energies were calculated from the local energy densities
218 S. J. Grabowski and J. Leszczynski
for the intramolecular DHBs mentioned above, and good correlations between such energy and the other parameters were found. The linear correlation coefficient for the relationship between the calculated H-bond energy and H.. .H distance is equal to 0.995 (Fig. 10 shows this dependence). _
•on
16
>. -6
2> g-7J
HLHdstsnoe
1
1
1
1.65
1.7
1.75
1
1.8
R=Q995
Fig. 10. The relationship between the H-bond energy (kcal/mol) calculated from the local potential energy density and H...H distance (in A) for intramolecular DHBs. Reprinted with permissionfromref 81. © 2003 Elsevier Science B.V. Similarly, the linear correlation coefficient between this energy and the topological parameter pH...H is equal to 0.988. It is worth mentioning that the linear correlation coefficient for the relationship between the energy calculated from the local energy density, and the difference in energy between the open and closed conformations amounts to 0.226. For this sample of ionized intramolecular DHBs, the properties of the ring critical point were also investigated.81 The ring critical point (RCP) is a point of minimum electron density within the ring surface and a maximum on the ring line.64 For O-H...O intramolecular H-bonds it was found previously that electron density at RCP correlates with H-bond strength.82 Similar relationships were found for O-H.. .N and N-H.. .O intramolecular H-bonds83 and also for double intermolecular H-bonds for which the closed rings of the covalent bonds and the H...Y contacts are formed.84 For the sample of BH3'-CH-CH-OH and its fluoro derivatives a good polynomial correlation (the polynomial of the
Is a Dihydrogen Bond a Unique Phenomenon?
219
second degree) was found with a correlation coefficient of 0.973. Fig. 11 displays this relationship which confirms that, for intramolecular dihydrogen bonds, the topological properties of RCP may be used as descriptors of Hbond strength as for the other conventional intramolecular hydrogen bonds. These molecules with intramolecular DHBs were chosen as model systems because of their simplicity and the fact that they are convenient species for higher level ab initio calculations. However, similar but usually much more complex molecules exist in the liquid or solid phase. For example, for crystals of 2,2,2-tris((Boranato(diphenyl)phosphonio)methyl)ethanol (YIBTAE refcode in the Cambridge Crystal Structural Database36) intramolecular B-H'5... +5H-O hydrogen bonds were detected (the BH3" group as a proton acceptor) with an H...H distance of 2.230 A. Single point ab initio calculations of the moiety of 2,2,2-tris((Boranato(diphenyl)phosphonio)methyl)ethanol (geometry taken from the crystal structure) have been performed.81 Because of the complexity and the large size of the moiety the HF/3-21G* level of theory was chosen. The AIM calculations yielded the value of the electron density of the H"8../5!! bond critical point (pH...H = 0.0060 au) and the value of the electron density at the ring critical point (pRCp = 0.0055 au). The corresponding Laplacian values, V2pH..H and V2pRCp, amount to 0.0340 and 0.0282 au, respectively. These results suggest that for this crystal structure intramolecular dihydrogen bonds may exist since the topological criteria are fulfilled.75 0.030-I
0.028 r
0.026-
1=1
0.0240.0220.020 I +^—r 0.015 0.016
Jt *
/
j(* ^ ^ ^ 1 0.017
1 0.018
1 0.019
1 0.020
PRCP
Fig. 11. The relationship between electron density at H.. .H BCP and electron density at RCP (values in au). Reprinted with permission from ref 81. © 2003 Elsevier Science B.V.
220 S. J. Grabowski and J. Leszczynski
7. The nature of DHBs - energy decomposition analysis It is possible to obtain more detailed insight into the nature of interactions by decomposition of the total interaction energy into different components. There are different approaches to perform such a decomposition, but the scheme which is the most often applied is that of Morokuma and co-workers.85 Briefly speaking, the electrostatic (ES) term represents the Coulombic interaction between the charge distributions of the two subunits of the dimer, and the exchange energy (EX) corresponds roughly to steric repulsion between the two charge clouds. When the monomers are allowed to influence the charge clouds between themselves, it is possible to extract polarization (POL) energy connected with the shifts of electron density occurring within monomers and charge transfer (CT) energy connected with the density shifts from one monomer to the other. The effects of electron correlation are included in the CORR term. Such a decomposition scheme was applied recently to different complexes containing dihydrogen bonds.86 Among them the (H3BNH3)2 and (H2BNH2)2 complexes were considered where for the former four equivalent H...H contacts exist, and for the latter there are two equivalent H....H interactions. Their binding energies calculated at the MP2/aug-cc-pVDZ level of theory (BSSE included) are equal to 14.1 and 2.1 kcal/mol, respectively. For the (H3BNH3)2 complex for each equivalent B(sp3)-H...H-N(sp3) dihydrogen bond, the H-bond energy amounts to 3.5 kcal/mol, and for each B(sp2)-H.. .HN(sp2) interaction the energy is 1.1 kcal/mol for (H2BNH2)2. This is indicative of the influence of hybridization on the strength of the dihydrogen bond. For these and other complexes the Morokuma partitioning scheme was applied.85 The results are presented in Table 4 and compared with the other complexes connected through conventional and unconventional H-bonds.87'88'89 The results presented in Table 4 show that for conventional H-bonds in trans-linear dimers of water, the electrostatic attractive term is the most important; it overweighs the repulsive exchange term. The polarization and charge-transfer terms are much smaller than the electrostatic term, similar to the electron correlation energy term. For the unconventional C-H.. .0 bond of the CH4...OH2 complex, the electrostatic energy is compensated by the exchange energy, but due to the small attractive remaining terms, the total interaction for this complex is slightly attractive. The substitution of a methane-donating molecule by electronegative fluorine atoms causes an increase in the donating characteristics of the C-H bond. Hence, the percentage contributions of the energy terms for the F3CH.. .OH2 complex are similar to those of the water dimer. For two complexes with 7t-electrons of the
Is a Dihydrogen Bond a Unique Phenomenon? 221
acetylene molecule as proton acceptors (FH...C2H2 and HCCH...C2H2) the electrostatic term is compensated by the exchange term, and the remaining attractive contributions cause stabilization of the complexes. The polarization and charge-transfer contributions are greater in the case of the FH...C2H2 complex. There is a similar situation for HOH...C6H5OH; however, in such a case the correlation term is significant. Table 4. Morokuma's partitioning of Interaction Energy (kcal/mol). Complex/ref
I
ES
I
EX
I
POL
I
CT
I CORRd
(H 2 O) 2 /87 a
^6
42
4J7
4$
^03
CH 4 ...OH 2 /87 a
~^0A
O4
50
~4A
^Ol
F 3 CH...OH 2 /87 a
I7l
4~\
^O7
To
^03
FH...C2H2/88b
4A
63
T5
22
^O4
HCCH...C2H2/88b
^2
Ti
^03
^O5
^O4
HOH...C 6 H 5 OH/89 a
^
32
^09
^07
2~A
(H3BNH3)2 /86 c
^9
4\4
3?7
T6
4l
(H2BNH2)2 /86 c
A3
12
^5
^O8
4 l
"6-31+G** basis set b 6-311++G** basis set c aug-cc-pVDZ basis set d CORR=AE(MP2)- AE(HF)
What is a unique feature of the two dihydrogen bonded complexes presented in Table 4? For (H2BNH2)2 the binding energy and H-bond interactions are weaker than for the (H3BNH3)2 complex. For the first complex the exchange energy overweighs the electrostatic term, but finally the total interaction is attractive due to the remaining attractive terms, the correlation term being the most important. For the second complex, (H3BNH3)2, the electrostatic term is slightly greater than the exchange term, and the remaining terms contribute significantly to the total interaction energy. Hence one can see that for stronger dihydrogen bonds such as in the (H3BNH3)2 complex the decomposition components are slightly similar to typical H-bonds; the electrostatic contribution is slightly greater than the exchange contribution, and the other attractive terms play an important role.
222 S. J. Grabowski and J. Leszczynski
Another decomposition scheme was applied recently for the sample of complexes with dihydrogen bonds.90 The interaction energy for the IJH...H2, LiH.. .CH4, LiH.. .C2H6 and LiH.. .C2H2 complexes has been partitioned using the IMPPT scheme.91'92'93'94 The IMPPT energy terms are designated by e(ij) where i and j are the orders of the intermolecular interaction operator and the intramolecular correction operator, respectively. All e(i j ) terms are calculated in the basis of full complex to avoid the basis set superposition error. The total interaction energy at the all-electron MP2 level is decomposed into a HartreeFock (SCF) contribution and a correlation term: AEMP2 = AE SCF +AE CORR
(4)
The Hartree-Fock term is further decomposed: AESCF = A E SCF d e f + A E HL
(5)
Where AESCFdef is the deformation energy, and AEHL is a Heitler-London term. The first term describes an effect due to the relaxation of orbitals in the Coulomb field of the partner with the restriction imposed by the Pauli principle. The second-order approximation to AESCFdef considers this term as the induction response term e ^ O ) ^ . The Heitler-London contribution is further decomposed into electrostatic and exchange energies: AEHL = e(l,0) els + eHLexch (6) The correlation energy term is also decomposed into: AECORR = AE(2)exch + 8(2,0)^+ e(l,2)els,r
(7)
Where AE(2)exCh is the second-order exchange correlation correction; e(2,0)disP is the dispersion correlation correction; and E(l,2)eisr is the secondorder electrostatic correlation correction. The complexes mentioned above have been studied at the MP2 level of theory using the aug-cc-pVDZ basis set. For smaller complexes the aug-ccpVTZ basis set was also used. Table 5 shows the energy contributions for the complexes. Only results of MP2/aug-cc-pVDZ are given in order to compare all complexes. The results obtained within the same level and the same scheme of energy decomposition for the water dimer are also presented for comparison. Emphasis is put only on the selected terms.
Is a Dihydrogen Bond a Unique Phenomenon?
223
Table 5. Decomposition of the MP2 interaction energy (in kcal/mol) of the LiH...H2, L1H...CH4, LiH...C2H2, LiH...C2H6 and H2O...H2O complexes; the aug-cc-pVDZ basis set was used (results taken from ref. 90). Term 8(1,0)*
I LiH...H2 I
H
I LiH...CH4 I LiH...C2H6 I LiH...C2H2 I H2O...H2O Tl
A2
^2
TS
O
T3
22
6M
7^
AE^drf
: O6
5?7
5H
5?7
53
lOOW
5?7
TI
T5
53
42
£(2,1)^
51
51
51
51
51
e(l,2)elv
51
00
00
O6
51
~^~
A brief analysis of the results presented in Table 5 indicates that only for one of dihydrogen bonded complexes (LiH.. .C2H2) the first order electrostatic term (e(l,0) e i s ) is greater than the exchange term, similarly as for the water dimer. For the remaining dihydrogen bonded complexes the dispersion term (£(2,0YjjsP) is comparable with the electrostatic term. The authors of this paper explain that, for LiH...C 2 H 2 , the dihydrogen bond may be classified as hydrogen bond interaction; however, the remaining species should be characterized as van der Waals complexes.
8. Spin-spin coupling constants across dihydrogen bonds Vibrational frequencies and NMR shielding constants have been applied for a long time to describe hydrogen bonds and other intermolecular interactions. However, in recent years data concerning spin-spin coupling constants has become a new tool in the analysis of interactions, especially hydrogen bonds. A review of ab initio and DFT scalar coupling constants has appeared recently, and the usefulness of this new method is herein explained in detail. The scalar coupling can be defined as the mixed second derivative of the energy of the molecule regarding the magnetic moments of two nuclei. It can be divided into four components: Fermi contact (FC), paramagnetic spin-orbit (PSO), diamagnetic spin-orbit (DSO) and spin-dipolar (SD). The nJ designation is usually used for nuclear spin-spin coupling constants through n bonds, where bond means not only the typical covalent bond but also the
224 S. J. Grabowski and J. Leszczynski
inter- or intramolecular interaction such as, for example, the H-bond. Parentheses or subscripts may be also used to designate the coupled nuclei. Couplings through hydrogen bonds are usually designated as ***]. Hence the hydrogen-hydrogen coupling between atoms of two different water molecules connected through the O-H...O hydrogen bond is designated as 2hJ if one of the hydrogens belongs to the proton donating O-H bond and as 4hJ if none of the coupled hydrogen atoms belongs to the O-H.. .O system. A number of studies dealing with calculations of J-values have been performed in recent years. For example, the J-values for the water monomer and for the water dimer were calculated and compared with experimental results. The scalar coupling constants were analyzed for X-H.. .0 H-bonds in the following complexes: CH2O...H2O, C2H2...H2O, CH3OH...H2O and (HCOOH)2.97 The calculations were performed at the MCSCF level, and the results show that lhJoH intermolecular coupling correlates with the H-bond strength and is dominated by the FC term. Similarly the 2hJxy couplings through the X-H...Y hydrogen bonds are dominated by the FC component. Complexes with N-H...O=C and N-H...N=C systems were also studied, and exponential decay of 2hJxY upon increasing X... Y distance was observed. Different kinds of other H-bonds and van der Waals complexes95 were analyzed in terms of1*! values. However in the latter case only the theoretical results are available, and the experimental results are not available for verification. Among ^J values related to this review those concerning dihydrogen bonds are the most interesting. Recently ab initio calculations have been performed to determine onebond IhJHH and three-bond 3hJHH spin-spin couplings across X-H...H-M dihydrogen bonds for complexes with 13C-'H, 15N-[H and 17O-!H protondonor groups and 7Li-!H and 23Na-!H proton-acceptor metal hydrides. Coupling constants have been calculated using the equation-of-motion coupled cluster singles and doubles method (EOM-CCSD) employing Cl-like approximation.99>10"ll01'102The lhJm couplings containing the PSO, DSO, FC and SD components mentioned above have been calculated for linear CTOV dihydrogen bonded complexes of NCH...HLi, NCH...HNa, CNH...HLi, CNH...HNa and LiNCH+...HLi and for nonlinear Q complexes of HOH.. .HLi and HOH.. .HNa. The authors have observed that the SD term is negligible (ranging from -0.04 to -0.20 Hz). The relationships between the remaining terms and the H...H distance for each of the complexes were investigated. The FC term decreases exponentially with an increase in the distance of H.. .H. PSO and DSO depend linearly on the H.. .H intermolecular distance but have opposite signs and tend to cancel each other. Hence the
Is a Dihydrogen Bond a Unique Phenomenon?
225
relationship between the total coupling and H...H distance is practically the same as between FC and H.. .H distance. And hence the FC value may be an indication of the strength of the hydrogen bond. Similar dependencies were observed between the 3hJHH spin-spin coupling and the C...M (M = Li, Na) distance for each of the complexes. However, in the case of three-bond spinspin couplings, three terms, PSO, DSO and SD, are negligible and this is the reason why almost the whole contribution to the total coupling comes from the FC term. An analysis of the intermolecular coupling constants have been performed very recently for the following complexes: LiH...H2, LiH.-.CtL,, LiH...C2H6 and LiH...C2H2.90 The intermolecular lhJHH and 3hJXY (3hJuc) constants have been plotted as functions of internuclear distance. The ^JRH values have been calculated at the DFT (B3LYP) and CCSD levels of theory using the aug-ccpVTZ basis set. The obtained DFT and CCSD curves are similar, and the following conclusions were derived. 3hJLjC decreases exponentially with an increase in the Li...C distance for all four investigated complexes. For three weaker complexes (LiH...H2, L1H...CH4 and LiH...C2H6) lhJHH is negative and tends to approach zero when the H...H distance increases. A different behaviour was noticed for the LiH...C2H2 complex. In this case lhJHH is negative for larger distances, is positive for shorter H...H contacts and is similar to the other typical covalent bonds. It is worth mentioning that the authors have classified the LiH...C2H2 complex as dihydrogen-bonded, and the three remaining complexes as van der Waals complexes since the exchange energy terms overweight the electrostatic contributions (see previous section).
9. Borders and limitations - summary It is difficult to conclude that there is something specific for dihydrogen bonds in comparison with the other conventional H-bonds. It is certain that at least one feature is unique: in DHB complexes another hydrogen atom is the proton acceptor. It was pointed out that such an atom should be negatively charged as was revealed for some metal or boron hydrides. However even in early studies of DHBs it was indicated that such interactions may be treated in the another way, ' namely that a bonds are the proton acceptors following the trend: conventional X-H....Y hydrogen bonds, H-bonds with Tt-electrons as acceptor (X-H...Jt) and X-H...0 H-bonds. However Crabtree et al. also claim that early examples of interactions termed as dihydrogen bonds were
226 S. J. Grabowski and J. Leszczynski
studied for transition metal hydrides. For such cases one cannot be sure if the M-H a bond is an acceptor since transition metals also possess nonbonding d^-electrons that could act as alternative H-bond acceptors. The strength of DHBs is usually similar as predicted for typical H-bonds. DHBs are usually considered to be medium to strong H-bonds. The results of the decomposition energy analysis do not indicate unique properties of DHBs. For stronger DHBs such as for LiH...C2H2 the energy terms are similar to those of the typical H-bonds. The most important is the electrostatic term on which is greater than the exchange energy term (Table 5). For other systems such as LiH...H2, IJH...CH4 and LiH...C2H2, the exchange term overweighs the electrostatic contribution. These complexes are weak, and the dispersion term is important for them (Table 5). Such systems may be classified as van der Waals complexes. The clusters of ammonia with hydrogen molecules (NH4+(H2)n) could be considered an important example of the continuously changing characteristic of the DHB complex. These clusters have been studied where the number of H2 ligands ranges from one to eight. Geometry optimizations have been performed at the MP2/6-311G(d,p) level of theory, and the dissociation energies were calculated within the MP2 theory using the single point calculations in the extended aug-cc-pVDZ basis set. ' 5 Corrections for the basis set superposition error according to the scheme of Boys and Bernardi were applied. For the studied systems no symmetry constraints were imposed during the geometry optimization. The energetically lowest structure for the NH»+ + H2 complex is the vertex bond moiety with the H2 molecule perpendicular to the line of the N-H bond (Fig. 12). The binding energy for such a complex amounts to 2.56 kcal/mol, and the molecular arrangement supports the idea of the proton acceptor abilities of o bonds. It is worth mentioning that for the other arrangement of molecules for N H / + H2 (for the face bonded molecules where the H2 molecule is parallel to the HHH plane of NH/) the transition state is predicted, and such a complex is higher in energy (by 0.98 kcal/mol) than the equilibrium complex. It was indicated, using the interaction energy decomposition scheme, that for NH4+(H2)n complexes the exchange energy usually overweighs the electrostatic term, and the dispersion energy significantly contributes to the total energy. The situation is similar to that observed for the LiH...H2, IJH...CH4 and LiH...C2H2 complexes; however, for NH41" H2 the binding energy is greater. Therefore, it seems reasonable for the latter case to be considered the charge-assisted (N-H... o)+ interaction.
Is a Dihydrogen Bond a Unique Phenomenon?
227
Fig. 12. The structure of the NH/ complex of the lowest energy (distances in A, angles in degrees). Reprinted with permissionfromref 103. © 2001 Elsevier Science B.V. Very recently, a comprehensive study of C-H.. .H-C dihydrogen bonds existing in crystal structures was published. This kind of interaction in crystals is very important since for a large number of metal organic and organic crystals the H.. .H contacts are common. Until now they were usually treated as typical van der Waals contacts. Robertson et al.30 considered the crystal structures of the tetraphenylborates for which DHBs exist as well as the other H-bonds such as N-H.. .N, C-H.. .N, N-H.. .Ph and C-H.. .Ph. These interactions have been characterized using the Bader theory after multipole refinement of the structures. Different dependencies were studied; the authors found that the difference between the Mulliken charges on H-atoms in the uncomplexed X-H and H-Y components correlates well with some of the parameters of the X-H...H-Y dihydrogen-bonded system. They also found that the H...H intermolecular distance is within the range of 2.18 - 2.57 A; the electron density at H...H BCP is within 0.05-0.08 au, and the binding energy is about 1.5 kcal/mol. The range of the electron density values shows that the systems analyzed are within the criterion of Koch and Popelier.75 The authors also pointed out that there is a detectable border between the H2
228 S. J. Grabowski and J. Leszczynski
connections and the H...H contacts. The border is connected with the acute change of the geometrical and topological parameters. The correlation curves have characteristic gaps indicating these borders. And there are no discontinuities in the considered features between the dihydrogen bonds and the van der Waals complexes. Interestingly, early studies of the methane dimer show that the H.. .H interaction is repulsive and that this system is not stable. However, the later, more accurate post-HF calculations indicate that the methane dimer is bound in all six examined configurations: H3CH...H2CH2 (Cs), H2CH2...H2CH2 (D2h and D2d), HCH3...H3CH (D3h and D3d) and H3CH...HCH3 (D3h). It is worth mentioning that for such weak H...H interactions the analyzed structures are very sensitive to the level of the calculations. For example, the minimum (no imaginary frequencies) D3<j symmetry H3CH...HCH3 dimer not included in the analysis of Novoa et al.106 was obtained at the MP2/6-311++G(3d,3p) level of theory.107 The other lower basis sets indicate that such a configuration, H3CH...HCH3 (D3d), is a transition state. Fig. 13 shows a molecular graph of this D3d symmetry methane dimer obtained within the AIM theory.
Fig. 13. Molecular graph of the H3CH...HCH3 (D3(i) methane dimer (the MP2/6311++G(3d,3p) level); big circles correspond to attractors attributed to nuclei, and small circles correspond to bond critical points. The electron density at H...H BCP for this dimer is equal to 0.0035 au and is within the range of Koch and Popelier which suggests that this interaction may be classified as an H-bond. The examples presented here show that there are problems with distinguishing between van der Waals interactions and DHBs. The only not-
Is a Dihydrogen Bond a Unique Phenomenon?
229
so-sharp difference is in the nature of the interactions; for the former species the dispersion term is the most important; for the latter it is the electrostatic term that should overweight the exchange contribution.
10. Appendix - the Bond Valence Model Previously, in 1947, for metal crystal structures, L. Pauling introduced the concept of the bond number n (Pauling, 1947). rn-r0 = Ar = -clogn
(Al)
In the above equation, rn is the considered interatomic distance for which the bond number is equal to n; c is the constant; and r0 is the reference distance for which the bond number is usually equal to unity. The bond number may be understood as the number of electron pairs attributed to the considered pair of atoms, and it correlates with the strength of the interaction. The number of applications for this idea exists such as for example the analysis of chemical reactions and the construction of reaction paths (Burgi, 1975; Dunitz, 1979). This example concerns the proton transfer reactions. This concept is also useful for studies of intra- and intermolecular interactions (Grabowski, Krygowski, 2002). This idea could be apply to studies of hydrogen bond interactions. For example, for the O-H bond not involved in any intermolecular interactions, the bond number should be equal to unity. When the O-H bond is involved in the O-H...O interaction, the O-H bond is elongated, and n-value decreases according to eq. (Al). But the decrease of n is compensated by the H...0 contact. The sum of bond numbers of O-H and H...0 should be equal to unity. It is known as the bond number conservation rule (Dunitz, 1979). A similar and more often applied idea of bond valence was developed by Brown (Brown, 1978; Brown, 1992) within the consistent and simple model known as the Bond Valence (BV) model. According to the BV model, the bond valences s are usually related to bond distances in the following way: siJ = expf(ro-riJ)/BJ
(A2)
Sy = (ro/riJf
(A3)
or
230 5. J. Grabowski and J. Leszczynski
Where, sg designates that the valence bond concerns the pair of i-th and j th atoms; r0 is the reference (usually single) bond length. B and N are constants which may be easily determined if, for the partial bond, both values S;J and ry are known. For example, for the symmetrical linear system of O-H.. . 0 with the hydrogen atom in the middle of the O.. .0 distance, the sy value of H.. .O is equal to 0.5, and the H.. .O distance is equal to 1.22 A (Dunitz, 1979) or even less, 1.2 A (Gilli et al., 1994). Such an attitude is often applied to evaluations of constant values. However, Brown proposed to treat all constant values, B, N and r0, iteratively. This means that all constants may be determined from the large sets of systems taken from the results of crystal structures (Brown and Altermat, 1985) and by applying the least squares or one of the other procedures. One of the most important statements of the BV model is that the atomic valence may be expressed in terms of the bond valences. This is the so-called valence sum rule (VSR):
r.-zs,
The valence V; of the i-th atom is assumed to be shared between the bonds that it forms. According to the Bond Valence Model the bond between the i-th and j-th atoms may be the typical covalent bond or the intermolecular contact. The atomic valences usually correspond to the oxidation states. For the hydrogen atom within O-H...O system the valence sum rule takes the form: exp[(r0OH - rOH)/BoHl + exp[(r0OH - rH...o)/BOHI
=1
(A5)
This equation simply corresponds to the bond number conservation rule mentioned above. In other words the valence sum rule is more general than the bond number conservation rule. If the constant values r0 and B are known it is possible to derive the relationship between the rOH bond length and the H...0 distance rH...o (Grabowski, 1988). This is a theoretical relationship based on the simple BV model, and it agrees excellently with the experimental results (Gilli et al., 1994). It is worth mentioning that the r0 and B constant values are attributed to pairs of atoms; thus they are the same for O-H covalent bonds and the H...0 contacts.
Is a Dihydrogen Bond a Unique Phenomenon?
231
Similar relationships as the one expressed by eq. (A5) may be constructed for any atom in the solid, liquid or gas phase. For example, for the proton acceptor oxygen atom in C=O...H-O systems, the following valence sum rule equation may be written: exp[(roc'° - rc=o)/Bc=o] + exp[(r0OH - rK,.o)/Bo«]
=2
(A6)
The relationship between the C=O bond length and H.. .O contact distance obtained from eq. (A6) corresponds well to the experimental neutron diffraction results (Grabowski, 1998). The VSR for the +8H-atom within the F-lT8...'5!! dihydrogen bonded system may be written in the following way: VH = exp[(ro"-F - rH.F)/BHF]
+ exp[(roH"H - rH_jJ/BH..jJ
(A7)
Where, rH.F and rH...H are the H-F bond length and the H...H distance within F-H...H system; roHF ,BHF , r0H'H and BH...H are constants. The dependencies mentioned above obtained from the VSR often correspond to experimental results and to the relationships obtained from ab initio or DFT calculations (see the main text of this chapter). They also show the proper dependencies between the geometrical parameters and are indicative of the changes in the geometrical parameters during chemical reactions. Brown, I.D. Chem. Soc. Rev. 1978,7,359. Brown, I.D. Acta Cryst. 1992, B48,553. Brown, I.D. Altermat, D. Acta Cryst. 1985, B41,244. Burgi, H.-B. Angew. Chem. Internat. Edit. 1975,14,460. Dunitz, J.D. X-ray Analysis and the Structure of Organic Molecules, Cornell University Press, Ithaca, 1979. Gilli, P., Bertolasi, V., Ferretti, V., Gilli, G. J. Am. Chem. Soc. 1994,116,909. Grabowski, S.J. Croat. Chem. Acta 1988,61, 815. Grabowski, S.J. Tetrahedron 1998,54,10153. Grabowski, S.J., Krygowski, T.M. Molecular Geometry as a Source of Chemical Information - Application of the Bond Valence-Bond Number Models, in Advances in Quantitative Structure Property Relationships, Eds. B.Charton & M.Charton, Vol. 3, pages 27-66, Elsevier Science B.V. 2002. Pauling, L. J. Am. Chem. Soc. 1947, 69, 542.
Acknowledgement Support for this research is provided by grant 505/675 2003 (University of L6dz), grant No. 3 T09A 138 26 (State Committee for Scientific Research),
232 S. J. Grabowski and J. Leszczynski
NSF CREST HRD-0318519, NSF EPSCoR 02-01-0067-08/MSU, and ONR Grant N00034-03-1-0116.
11. References 1 Pauling, L., The nature of the chemical bond. Cornell University Press, Ithaca NY,
1939. Pimentel, G.C., McClellan, A.L. The Hydrogen Bond. Freeman, San Francisco, CA, 1960. 3 Jeffrey, G.A., Saenger, W. Hydrogen Bonding in Biological Structures. SpringerVerlag: Berlin, 1991. 4 Desiraju, G.R., Steiner, T. The weak hydrogen bond in structural chemistry and biology. Oxford University Press Inc., New York, 1999. 5 Jeffrey, G.A. An Introduction to Hydrogen Bonding. Oxford University Press, New York, 1997. 6 Desiraju, G.R. Crystal Engineering - The Design of Organic Solids. Elsevier, Amsterdam, 1989. 7 Alkorta, I., Rozas, I., Elguero, J. Chem.Soc.Rev. 1998, 27, 163 (and references cited therein). 8 Suttor, D. J. J. Chem. Soc. 1963, 1105. * Taylor, R., Kennard, O. J. Am. Chem. Soc. 1982, 104, 5063. Grabowski, S.J. Polish J. Chem. 1995, 69,223. 11 Platts, J. A., Howard, S.T., Wozniak, K. Chem.Commun. 1996, 63. 12 Hobza, P., Havlas, Z. Chem.Rev. 2000, 100,4253 (and references cited therein). 13 Scheiner, S. Hydrogen Bonding: A Theoretical Perspective. Oxford University Press: New York, 1997. 14 Wessel, J., Lee, J. C , Peris, E., Yap, G. P. A., Fortin, J. B., Ricci, J. S., Sini, G., Albinati, A., Koetzle, T. F., Eisenstein, O., Rheingold, A. L., Crabtree, R. H. Angew. Chem. Int. Ed. Engl. 1995,34,2507. 15 Custelcean, R., Jackson, J. E. Chem. Rev. 2001, 101,1963. 16 Crabtree, R. H., Siegbahn, P. E. M., Eisenstein, O., Rheingold, A. L., Koetzle, T. F. Ace. Chem. Res. 1996,29, 348. 17 Epstein, L. M., Shubina, E. S. Coord. Chem. Rev. 2002, 231, 165 (and references cited therein). 18 Rozas, I., Alkorta, I., Elguero, J. Chem.Phys.Lett. 1997, 275, 423 (and references cited therein). 19 Stevens, R. C , Bau, R., Milstein, D., Blum, O., Koetzle, T. F. J.Chem.Soc, Dalton 2Q Trans. 1990, 1429. Van der Sluys, L. S., Eckert, J., Eisenstein, O., Hall, J. H., Huffman, J. C , Jackson, S. A., Koetzle, T. F., Kubas, G. J., Vergamini, P. J., Caulton, K. G. J. Am. Chem. 21 Soc. 1990,112,4831. Park, S., Ramachandran, R., Lough, A. J., Morris, R. H. J. Chem. Soc. Chem. Commun. 1994,2201. 2
Is a Dihydrogen Bond a Unique Phenomenon?
233
Lough, A. J., Park, S., Ramachandran, R., Morris, R. H. J. Am. Chem. Soc. 1994, 116,8356. Lee, J. C , Jr., Rheingold, A. L., Miiller, B., Pregosin, P. S., Crabtree, R. H. J. Chem. Soc. Chem. Commun. 1994,1021. 24Lee, Jr., J. C , Peris, E., Rheingold, A. L., Crabtree, R. H. J. Am. Chem. Soc. 1994, 116,11014. " Peris, E., Lee, Jr., J. C , Crabtree, R. H. J. Chem. Soc. Chem. Commun. 1994,2573. Peris, E., Lee, Jr., J. C , Rambo, J., Eisenstein, O., Crabtree, R. H. J. Am. Chem. Soc. 1995,117,3485. 27 Peris, E., Wessel, J., Patel, B. P., Crabtree, R. H. J. Chem. Soc. Chem. Commun. 1995,2175. 28 Kloosrer, W. T., Koetzle, T. F., Siegbahn, P. E. M., Richardson, T. B., Crabtree, R. H. J. Am. Chem. Soc. 1999,121, 6337. Z* Richardson, T. B., Koetzle, T. F., Crabtree, R. H. Inorg. Chim. Ada 1996,250,69. Robertson, K. N., Knop, O., Cameron, T. S. Can. J. Chem. 2003,81,727. 31 Richardson, T. B., de Gala, S., Crabtree, R. H., Siegbahn, P. E. M. J. Am. Chem. Soc. 1995,117, 12875. I Liu, Q., Hoffinan, R. J. Am. Chem. Soc; 1995, 117,10108. Hoffinan, R. J. Chem. Phys. 1963,39,1397. 34 Becke, A. D. Phys. Rev. 1988, A38,3098. 35 Siegbahn, P. E. M., Blomberg, M. R. A., Svensson, M. Chem. Phys. Lett. 1994, ™ 223,35. Allen, F. H., Kennard, O. Chem. Des. Autom. News 1993, 8,31. 37 Klooster, W. T., Koetzle, T. F., Siegbahn, P. E. M., Richardson, T. B., Crabtree, R. 38 H. J. Am. Chem. Soc. 1999,121, 6337. Cramer, C. J., Gladfelter, W. L. Inorg. Chem. 1997,36, 5358. 39 Boys, S. F., Bernardi, F. Mol. Phys. 1970,19, 553. 40 Alkorta, I., Elguero, J., Foces-Foces, C. Chem. Commun., 1996,1633 41 Bader, R. F. W., (1990) Atoms in Molecules. A Quantum Theory. Oxford University Press, New York. f Remko, M. Molecular Physics, 1998,94,839. Orlova, G., Scheiner, S. /. Phys. Chem. 1998, 102, 260. */ Orlova, G., Scheiner, S. J. Phys. Chem. 1998,102,4813. * Kulkarni, S. A. J. Phys. Chem. A 1998,102, 7704. *! Kulkami, S. A. J. Phys. Chem. A 1999,103,9330. Kulkarni, S. A., Srivastava, A. K. /. Phys. Chem. A 1999,103, 2836. 48 Grabowski, S. J., Tetrahedron 1998, 54, 10153 (and references cited therein). 49 Allen, F. H., Davies, J. E., Galloy, J. E., Johnson, J. J., Kennard, O., Macrave, C. F., Mitchel, E.M., Smith, J. M., Watson, D. G. J. Chem. Inf. Comput. Sci. 1991, 22
,,
50
31, 187. 3 1 ' 1 8 7 '
51 Ichikawa, M. Acta Cryst. 1978, B34,2074. 52 Gilli, P., Bertolasi, V., Ferretti, V., Gilli, G. J. Am. Chem. Soc. 1994,116,909. " Steiner, T. J. Phys. Chem. 1998, 102,7041. 54 Steiner, T. Chem. Soc. Perkin Trans. 2, 1996, 1315. Grabowski, S. J. J. Phys. Org. Chem. 17,2004,18.
234 5. J. Grabowski and J. Leszczynski
Grabowski, S. J. J. Mol. Struct. 2000, 552, 153. Grabowski, S. J., Krygowski, T. M. Chem. Phys. Letters, 1999,305,247. "Grabowski, S. J., Chem. Phys. Lett. 1999,312, 542. 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., Burant, J. C , Dapprich, S., Millam, J. M., Daniels, A. D., Kudin, K. N., Strain, M. C , Farkas, O., Tomasi, J., Barone, V., Cossi, M., Cammi, R., Mennucci, B., 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., Stefanov, B. B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Gomperts, R., Martin, L. R., Fox, D. J., Keith, T., Al.-Laham, M. A., Peng, C. Y., Nanayakkara, A., Gonzalez, G., 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; Revision A.6; Gaussian, Inc.: Pittsburgh, PA, 1998. 5 Grabowski, S. J. J. Mol. Struct. 2000, 553,151. 60 Grabowski, S. J. "Simulations of Hydrogen Bonds in Crystals and Their Comparison with Neutron Diffraction Results, in Neutrons and Numerical Methods - N2M," Grenoble, France December 1998, AIP Conference Proceedings 479, (eds.) M. R. Johnson, G. J. Kearley and H. G. Buttner, American Institute of Physics 1999. 61 Novoa, J. J., Nobeli, I., Grepioni, F., Braga, D. NewJ. Chem. 2000,24, 5. "Grabowski, S. J. J. Phys. Chem. A2000,104, 5551. £ Hohenberg P., Kohn, W. Phys. Rev. 1964, B864, 136. Popelier, P. Atoms in Molecules. An Introduction, Prentice Hall, Pearson Education Limited 2000. 65 M6,0., Yanez, M. Elguero, J. /. Chem. Phys. 1992,97,6628. ^ Mo, O., Yanez, M. Elguero, J. J. Mol. Struct. (Theochem) 1994,314,73. Espinosa, E., Molins, E., Lecomte, C. Chem. Phys. Lett. 1998,285,170. * Grabowski, S. J. J. Phys. Chem. A, 2001, 105,10739. Scheiner, S. Annu. Rev. Phys. Chem. 1994,45,23. 70Grabowski, S. J. J. Mol. Struct. 2002,615,239. 71 Alkorta, I., Elguero, J., M6, O., Yanez, M., del Bene, J. E. J. Phys. Chem. 2002, „ 106,9325. Popelier, P. L. A. J. Phys. Chem. A 1998,102, 1873. 73 MORPHY97, a program written by P. L. A. Popelier with a contribution from R. G.A. Bone, UMIST, Manchester, U.K., EU, 1997. 74Popelier, P. L. A. Chem. Phys. Lett. 1994,228,160. "Koch, U. Popelier, P. L. A. J. Phys. Chem. A 1995,99,9747. Madsen, G. K. H., Iversen, B. B., Larsen, F. K., Kapon, M., Reisner, G. M., Herbstein, F. H. J. Am. Chem. Soc. 1998,120,10040. 77 Grabowski, S. J. J. Mol. Struct. 2001, 562,137. 78 Garcia-Viloca, M., Gelabert, R., Gonzalez-Lafont, A., Moreno, M., Lluch, J. M. J. Phys. Chem. A 1997,101,8727 (and references cited therein). 79 Grabowski, S. J. Chem. Phys. Lett. 2000,327,203. 55
Is a Dihydrogen Bond a Unique Phenomenon?
235
", Cuma, M., Scheiner, S., Kar, T. J. Mol. Struct. (Theochem), 1999,467, 137. Wojtulewski, S., Grabowski, S. J. J. Mol. Struct. 2003,645,287. 82 Grabowski, S. J, Monatsh.Jur Chetnie, 2002,133,1373. 83 Rybarczyk-Pirek, A. J., Grabowski, S. J., Malecka, M., Nawrot-Modranka, J. J. 84 Phys. Chem. A 2002,106,11956. Dubis, A., Grabowski, S. J, Romanowska, D. B., Misiaszek, T., Leszczynski, J. /. g5 Phys. Chem. A 2002,106,10613. Morokuma, K., Kitaura, K. in Molecular Interactions, vol. 1, W. J. Orville-Thomas (ed.), Wiley, New York, 1980, page 21. I Kar, T., Scheiner, S. J. Chem. Phys. 2003, 119,1473. Gu, Y., Kar, S., Scheiner, S. J. Am. Chem. Soc. 1999, 121,9411. * Scheiner, S., Grabowski, S. J. J. Mol. Struct. 2002,615,209. f Scheiner, S., Kar, T., Pattanayak, J. J. Am. Chem. Soc. 2002,124, 13257. Cybulski, H., Pecul, M., Sadlej, J. J. Chem. Phys. 2003,119, 5094. I* Chalasinski, G., Sczeiniak, M. Mol. Phys. 1988,63,205. Moszyhski, R., Rybak, S., Cybulski, S. M., Chalasinski, G. Chem. Phys. Lett. 1990, „ 166,609. Cybulski, S. M., Chaiasinski, G., Moszyhski, R. J. Chem. Phys. 1990,92,4357. 9 Chalasinski, G., Sczfiniak, M. Chem. Rev. 1994,94, 1723. Alkorta, I., Elguero, J. Int. J. Mol. Sc. 2003,4, 64. : ! Pecul, M., Sadlej, J. Chem. Phys. Lett. 1999,308,486. 9J Pecul, M., Leszczynski, J., Sadlej, J. J. Chem. Phys. 2000,112, 7930. * Pecul, M., Leszczynski, J., Sadlej, J. J. Chem. Phys. 2000,104, 8105. i Perera, S. A., Sekino, H., Bartlett, R. J. J. Chem. Phys. 1994,101,2186. °, Perera, S.A., Nooijen, M., Bartlett, R. J. J. Chem. Phys. 1996,104,3290. j°, Perera, S. A., Bartlett, R. J. J. Am. Chem. Soc. 1995, 117,8476. ~T Perera, S. A., Bartlett, R. J. J. Am. Chem. Soc. 1996, 118, 7849. !°: Urban, J., Roszak, S., Leszczynski, J. Chem. Phys. Lett. 2001, 346,512. r* Dunning, Jr., T. H. J. Chem. Phys. 1989, 90, 1007. Kendall, R. A., Dunning, Jr., T. H., Harrison, R. J. J. Chem. Phys. 1992, 96, 6769. ^ Novoa, J. J., Whangbo, M.-H., Williams, J. M. J. Chem. Phys. 1991,94,4835. Grabowski, S. J., Leszczynski, J. unpublished results.
INDEX cr-bond acceptor 202, 225, 226 7r-electron delocalization 214-219 ab initio 85, 88, 89, 105 adhesion 61 aglycone bond 68 amylose 67, 70, 71 Anderson's theory 154-157 angle bending 105-107 angular overlap model 166 anisotropic exchange 154, 155, 157, 163, 165, 187 atomic force microscope (AFM) 47-51, 53-62, 64, 66-71 atomic valence 230 atoms in molecules (AIM) theory 208-214 ATPase 56, 65, 66 basis set superposition error 222, 226 Bethe lattice 7, 8, 11, 38 biased molecular dynamics 54, 62, 64, 71 bifurcated dihydrogen bonds 198, 199 binding energy 196, 201, 202, 207, 208, 220, 221, 226, 227 blue-shifting H-bonds 197, 204, 207 bond critical point (BCP) 208, 209, 213, 219, 228 bond number 229-231 bond number conservation rule 229, 230 bond valence 204, 205, 229, 230 bond valence model (BV model) 204, 205, 229-231 Cambridge structural database (CSD) 202, 204 carbon nanotube (CNT) 31-33, 35, 37 chair-boat transition 50, 67 charge transfer 157, 182, 190 charge transfer energy 220 CHARMM 67 conduction channel 3, 4, 19 conventional H-bonds 197, 199, 203, 204, 207, 210-212, 219, 220, 225 correlation energy term 220, 222 coupled electron-pair approximation (CEPA) 121-124, 126-132, 136, 138, 139, 148
237
238
Index
coupling constant 153, 156, 166, 170, 173, 183-185, 188-191 coupling integral 189 criteria for the existence of hydrogen bonds 196, 202, 206, 207, 213, 214, 219 crystal (ligand) field 158, 160 crystal engineering 196 density functional theory (DFT) 5, 6, 9, 13, 14, 27, 28, 53, 70, 169 dextran 67, 68 dihedral 109, 110, 112-115 dihydrogen bond 195-235 dipole moment 88, 114, 116, 179, 180, 192 Dzyaloshinski-Moriya 163 elastin 59 electric field 179, 180 electronic hopping 159 electronic transition 178 electrostatic energy term 201, 207, 220-223, 225, 226, 229 energy decomposition analysis 220-223 enzymatic catalysis 53 EPR-spectra 163, 165 exchange energy 220-223, 225, 226, 229 exchange interaction 154, 155, 157, 158, 163 exchange operator 153, 156, 180, 186 excited state 121, 125, 126, 133, 134, 136, 138-140, 142-144, 145, 148 exclusion principle violating (EPV) terms 123, 124, 129, 130, 132 ferromagnetic 154, 163, 164, 177, 178, 186 fibronectin 50, 56, 58 force probe molecular dynamics 60 force probe spectroscopy 49 free energy 53, 55, 59, 62, 63 frequency spectrum 92, 93, 101, 103, 110 fullerene 1, 24, 26-28, 30 Gaussian Embedded Cluster Method (GECM) GAUSSIAN98/03 1, 69 Go-like models 65 Green's function 7, 9, 10, 40, 41 g-tensor 163 H... H interactions 202, 203, 212
6, 25, 26, 31
Index 239
H-bond energy 197, 206-212, 215-218, 220 H-bond strength 209, 217-219, 224 Heisenberg-Dirac-van Vleck Hamiltonian 157 heterogeneity of the sample 204 HF calculations 67 homogeneity of the sample 204, 209-211 hydrogen bonding 195-235 implicit solvent 59, 62 IMPPT energy terms 222 intramolecular dihydrogen bonds 198, 201, 214-219 intruders 121, 124, 125, 127, 131-133, 139, 143, 145, 148 Jarzynski 63, 64 Keldysh formalism 13 kinetic exchange 153, 156, 182, 185 Landauer's formalism, formula 2-5, 25, 37 Laplacian of the electron density 208, 209, 213, 214, 219 LDA 169 Lennard-Jones 105 Li2 121, 143 linear response theory (LRT) 125, 126, 133 local energy density 217, 218 localized magnetic orbitals 154, 156 low barrier hydrogen bonds 215 magnetic exchange 153, 166 magnetic field 163, 165 magnetic orbitals 154-156, 186 MBPT 124 mechanically controllable break junction (MCBJ) 2, 17 methane dimer 228 molecular bridge 2, 7, 22 molecular device 86, 88, 89, 91, 101, 103, 115 molecular dynamics 47, 52, 54, 56, 60, 69, 85, 88, 89, 92, 93, 114, 115 molecular junction 2, 6 MRCC 124-126 MRCEPA 124-126, 138-140 MR-CISD 123, 124, 128, 130 MscL channel 60
240
Index
Mulliken atomic charges 201, 202, 227 NAMD2 code 68 nanocell 85, 115 nanocontact 2, 4, 6, 10-12, 14-25 NDR 86, 89, 91 negative differential resistance 86, 89, 91 optomechanical 66 orbital momentum 168 P4 121, 139-143, 145, 146, 148 pectin 67 perturbation theory 156, 164, 182, 190, 193 polarization energy 220, 221 polysaccharides 50, 51, 66, 67 potential energy surface (PES) 121-125, 139-141, 143-148 potential exchange 156, 185 potential of mean force 55, 56, 59, 62 protein folding 51-53 proton acceptors 197, 202-204, 206, 212, 219, 221, 224-226, 231 proton donors 196, 197, 200, 202-205, 207, 209, 210, 224 pustulan 67-69 PW96 169 pyranose ring 67, 68 quantum molecular dynamics 70 radiation field 179 random expulsion molecular dynamics (REMD) rare earth 153, 186 reaction coordinate 55, 61, 62 reaction kinetics 53 resonance assisted hydrogen bonds 214-219 resonant bond 93, 98, 99, 101, 112-114 ring critical point (RCP) 208, 218, 219 scanning tunneling microscope (STM) SCC-DFTB 70 Schrodinger equation 3, 4 size-extensivity 123-125, 127 size-intensivity 138
64, 65
2, 17, 24, 25, 31
Index 241
spectrin 60 spin operator 159, 160, 168 spin-forbidden transition 179 spin-orbit coupling 153, 157, 160, 166, 167, 170, 172, 185, 187, 188 spin-orbit interaction 158 spin-spin coupling constants 223-225 S-S bridge 58 SS-MRCC 122, 125-128, 133, 134, 136, 146 streptavidin-biotin 54 targeted molecular dynamics (TMD) 54, 60, 66, 72 temperature 85, 89, 96, 98, 103, 105, 107, 110, 112, 113, 115, 116 the electron density 208-210, 213, 218-220, 227, 228 titin 50, 51, 56-59, 61, 65 topological parameters 209-211, 217, 218, 228 transition metal 153, 154, 181 unconventional H-bonds 196, 200, 220 up-conversion 153-155, 178, 183, 185 valence sum rule (VSR) 230, 231 van der Waals interaction 199, 209, 225-228 van der Waals radii 199, 201, 202, 206, 207, 213 van der Waals 93 vertical excitation energies 145 vibrational modes 87, 88, 92, 98, 114, 115 water dimer 207, 210, 211, 220, 222-224 wave function 210, 212 WHAM 56, 64
CONTENT INDEX Volume 1 Relativistic Many-Body Calculations on Atoms and Molecules (Y. Ishikawa & U. Kaldor) Modern Developments in Hartree-Fock Theory: Fast Methods for Computing the Coulomb Matrix (M. Challacombe, E. Schwegler & J. Almlof) Local Shape Analysis of Macromolecular Electron Densities (P. G. Mezey) Liquid-State Quantum Chemistry: Computational Applications of the Polarizable Continuum Models (J.-L. Rivail & D. Rinaldi) Elemental Boron Route to Stuffed Fullerenes (E. D. Jemmis & B. Kiran) Interactions of DNA Bases and the Structure of DNA. A Nonempirical Ab Initio Study with Inclusion of Electron Correlation (J. Sponer, P. Hobza & J. Leszczynski) Computational Approaches to the Design of Safer Drugs and Their Molecular Properties (N. Bodor & M.-J. Huang) Volume 2 The Electron Propagator Picture of Molecular Electronic Structure (J. V. Ortiz) SAC-CI Method: Theoretical Aspects and Some Recent Topics (H. Nakatsuji) Quantum Monte Carlo and Electronic Structure (R. N. Barnett & W. A. Lester, Jr.) Molecular Structure and Infrared Spectra of the DNA Bases and Their Derivatives: Theory and Experiment (M. J. Nowak, L. Lapinski, J. S. Kwiatkowski & J. Leszczynski) Derivation and Assessment of a New Set of Ab Initio Potentials and Its Application to Molecular Dynamics Simulations of Biological Molecules in vacuo, in Crystal and in Aqueous Solution (M. Aida) Practical Exercises in Ab Initio Quantum Chemistry — the World Wide Web as a Teaching Environment (H. P. Lu'thi, G. Vacek, A. Hilger & W. Klopper) Volume 3 Multireference Brillouin-Wigner Coupled-Cluster Theory (/. Hubac, J. Masik, P. Mach, J. Urban & P. Babinec)
243
244 Content Index Model Core Potentials: Theory and Applications (M. Klobukowski, S. Huzinaga & Y. Sakai) Stratospheric Bromine Chemistry: Insights from Computational Studies (5. Guha & J. S. Francisco) Bonding in Gas-Phase Sulfur Radicals (D. C. Young & M. L. McKee) Chemistry of the Liquid State: Current Trends in Quantum-Chemical Modeling (L. Gorb & J. Leszczynski) Chemistry and the Internet (K. Flurchick, M. Hurley, J. K. Labanowski, G. H. Lushington & T. L. Windus) Volume 4 Topography of Atomic and Molecular Scalar Fields (5. R. Gadre) The Ab Initio Model Potential Method: A Common Strategy for Effective Core Potential and Embedded Cluster Calculations (L. Seijo & Z. Barandiarari) Continuum Models of Macromolecular Association in Aqueous Solution (M. A. Olson) Interactions of Nucleic Acid Bases: The Role of Solvent (M. Orozco, E. Cubero, B. Hernandez, J. M. Lopez & F. J. Luque) Recent Advances in Multireference M0ller-Plesset K. Nakayama, T. Nakajima & H. Nakano)
Method
(K. Hirao,
Detonation Initiation and Sensitivity in Energetic Compounds: Some Computational Treatments (P. Politzer & H. E. Alper) Volume 5 In Search of the Relationship between Multiple Solutions Characterizing CoupledCluster Theories (P. Piecuch & K. Kowalski) Computational Time-Dependent Two-Electron Theory and Long-Time Propagators (C. A. Weatherford) Self-Consistent Field Theory of Weakly Bonded Systems (E. Gianinetti, I. Vandoni, A. Famulari & M. Raimondi) Aromatic DNA Base Stacking and H-Bonding (J. Sponer, P. Hobza & J. Leszczynski) Direct Ab Initio Dynamics Methodology for Modeling Kinetics of Biological Systems (T. N. Truong & D. K. Maity)
Content Index 245
Molecular Structure and Vibrational IR Spectra of Fluoro, Chloro and Bromosubstituted Methanes, Silanes and Germanes: An Ab Anitio Approach (J. S. Kwiatkowski & J. Leszczynski) Volume 6 Relativistic Multireference M0ller-Plesset Perturbation Theory (Y. Ishikawa & M. J. Vilkas)
15 Years of Car-Parrinello Simulations in Physics, Chemistry and Biology (U. Rothlisberger)
Methods of Combined Quantum/Classical (QM/MM) Modeling for Large Organometallic and Metallobiochemical Systems (/. B. Bersuker) A Review of Ab Initio Calculations on Proton Transfer in Zeolites (M. Allavena & D. White) Ionic Clusters with Weakly Interacting Components-Magic Numbers Rationalized by the Shell Structure (S. Roszak & J. Leszczynski) Turning Point Quantization and Scalet-Wavelet Analysis ( C R. Handy) Volume 7 Molecules as Components in Electronic Devices: A First-Principles Study (M. Di Ventra) Tackling DNA with Density Functional Theory: Development and Application of Parallel and Order-N DFT Methods (C. F. Guerra, F. M. Bickelhaupt, E. J. Baerends & J. G. Snijders) Low-Scaling Methods for Electron Correlation (5. Saeb0) Iterative and Non-Iterative Inclusion of Connected Triple Excitations in CoupledCluster Methods: Theory and Numerical Comparisons for Some Difficult Examples (J. D. Watts) Explicitly Correlated Coupled Cluster R12 Calculations (J. Noga & P. Valiron) Ab Initio Direct Molecular Dynamics Studies of Atmospheric Reactions: Interconversion of Nitronium Ions and Nitric Acid in Small Clusters (Y. Ishikawa & R. C. Binning, Jr.) Volume 8 Computational Modelling of the Solvent Effects on Molecular Properties: An Overview of the Polarizable Continuum Model Approach (R. Cammi, B. Mennucci & J. Tomasi)
246
Content Index
Electronic and Nonlinear Optical Properties of 2-Methyl-4-Nitroaniline Clusters (M. Guillaume, B. Champagne, F. Castet & L. Ducasse) Ab Initio Calculations of the Intermolecular Nuclear Spin-Spin Coupling Constants (M. Pecul & J. Sadlej) Base Polyad Motifs in Nucleic Acids - Biological Significance, Occurrence in Three-Dimensional Experimental Structures and Computational Studies (M. Meyer & J. Siihnel) Model Calculations of Radiation Induced Damage in DNA Constituents Using Density Functional Theory (D. M. Close) Excited States of Nucleic Acid Bases (M. K. Shukla & J. Leszczynski)