SciELO - Scientific Electronic Library Online

 
vol.49 número1GROWTH AND ACCUMULATION OF TOTAL CAROTENOIDS IN TWO STRAINS OF Dunaliella salina Teod. (CHLOROPHYCEAE) FROM THE NORTHERN AND CENTRAL COAST OF PERÚTuning the Excited States in -fac--[Re(X2dppz)(CO)3(L)]: Intraligand, Charge Transfer or both? índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Journal of the Chilean Chemical Society

versión On-line ISSN 0717-9707

J. Chil. Chem. Soc. v.49 n.1 Concepción mar. 2004

http://dx.doi.org/10.4067/S0717-97072004000100013 

Molecular Orbital Calculations of Hydrogen Bonding in Ammonia -
Formic Acid System in the Presence of Electric Fields

J. E. Parra - Mouchet,*, 1 G. Zapata - Torres1, W. H. Fink,2 and C. P. Nash.2

1Facultad de Ciencias, Universidad de Chile. Casilla 653, Santiago, Chile.
2 Department of Chemistry, University of California, Davis, California 95616.U.S.A.

(Received: January 16, 2003 -Accepted: October 14, 2003)

ABSTRACT

The effects of electric fields on hydrogen bonding in ammonia - formic acid system, are examined with STO-3G and 6-31 G(d) wavefunctions. This system was used in a previous work to model hydrogen bonding in crystalline amino acids and the calculations were performed at HF / STO-3G level. The results on the relative position of the tautomeric equilibrium between the neutral and zwitterionic forms were explained in terms of the relative stabilization of the ionic partners as a function of their placement in positive and negative wells created by the external electric field.

In order to rationalize those results at electronic structure level, in this paper we analyze the response of the molecular orbitals implicated in the hydrogen bridge, N··H··O, to various imposed external fields. It is found that the stabilization of the zwitterionic structure occurs due to the destabilization of the MO localized essentially at the nitrogen electron lone pair, n-orbital, and concomitant with the stabilization of the MO essentially localized at the functional oxygen electron lone pair, o-orbital. The stabilization of the neutral structure occurs in the opposite situation. In addition, these eigenvalues are shown to be adequate regional molecular descriptors of the base reactivity of amines and conjugated bases of carboxylic acids, in gas phase and in the presence of electric fields.

Keywords: hydrogen bonding; electric field; molecular orbitals; amino acids; reactivity descriptors.

RESUMEN

Se estudian los efectos producidos por campos eléctricos sobre el enlace por hidrógeno en el sistema amoníaco - ácido fórmico, utilizando funciones de onda STO-3G y 6-31G(d). Este sistema fue utilizado en un trabajo previo, como modelo de aminoácidos en estado cristalino. Los resultados obtenidos acerca de la posición del equilibrio tautomérico entre las formas neutra y zwiteriónica, fueron explicados a través de la estabilización relativa de las formas iónicas en pozos positivos y negativos de potencial electrostático, creados por el campo externo.

Con el propósito de racionalizar estos resultados, a nivel de estructura electrónica, en este artículo se analiza la respuesta de los orbitales moleculares implicados en el puente hidrógeno, N··H··O, frente a una serie de campos eléctricos aplicados al complejo molecular. Se encuentra que la estabilización del zwitterión se produce debido a la desestabilización del orbital molecular localizado esencialmente en el par electrónico libre del N, n-orbital, concomitante con la estabilización del orbital molecular localizado esencialmente en el par electrónico libre del oxígeno funcional, o-orbital. La estabilización de la forma neutra ocurre en la situación opuesta. Se muestra, además, que la energía de los orbitales moleculares mencionados, constituye un descriptor molecular adecuado para la reactividad ácido base de aminas y ácidos carboxílicos, en fase gaseosa y en la presencia de campos eléctricos.

Palabras clave: enlace por H; campo eléctrico; orbitales moleculares; aminoácidos; descriptores de reactividad

INTRODUCTION

The nature of H bonding has been profusely studied. From Morokuma Analysis [1], Kollman reported that for water dimmer, about half of the interaction is of electrostatic nature at the equilibrium distance [2]. Based upon crystal structures, Gilli suggested that some hydrogen bonds can be stabilized by resonance (resonance - assisted hydrogen bonds) which implicates appreciable covalent character; while in certain crystals short H bonding interactions could be explained in terms of electrostatic effects produced by the polarization of its neighbors [3]. More recently, Mo and Gao have reported that Lewis acid - base complexes, with a monomer separation larger than 2.5 Ǻ are primarily electrostatic in nature [4]. In amino acids the monomers separation is larger than 2.6 Ǻ, being 2.8 Ǻ the most representative one [5]. Moreover, in amino acid crystals, the preferred structure is a zwitterions; implying that H - bonding interactions are essentially of electrostatic nature.

On the other hand, the influence of electric fields on hydrogen bonding, has lately received major attention [6 -7]. Also, in theoretical descriptions of H-bonding in crystals, besides the crystal orbital model, point charges have been used at various levels of approximation [8 -9].

In the past we have reported STO-3G calculations on ammonia - formic acid system in the presence of electric fields [10], as a model of hydrogen bonding in some crystalline amino acids reported by Nash et al. [11 -12]. The effects of the electric field created by several arrays of point charges on the proton transfer potential of the molecular system were calculated. The results obtained were rationalized in terms of electrostatic stabilization of ionic forms, as a function of their placement in positive and negative electric wells, created by the external field.

In order to rationalize these results al electronic structure level, we analyze the response of the molecular orbitals implicated in the H - bridge, i.e., the n-orbital and o-orbital, to the applied electric field.

Correlation between base strength and the n-orbital energy have been reported for amines. In gas phase, Del Bene has found STO-3G linear relationships between the base strength of substituted pyridines and the n-orbital energy [13]. In solution, electrostatic solvent effects on histamine bacisity have been explained in terms of the heterocyclic and aliphatic n-orbital energies inversion upon salvation [14]. In the present work, we use these parameters to rationalize the protonation processes induced by electric fields, in the ammonia - formic acid system.

In addition, we calculate the electron density on N and O in the n-orbital and o-orbital, as an estimate of the electron localization degree of N and O in the corresponding molecular orbital. It turns out that this parameter, measured only in the HOMO, has been defined, in the Density Functional Theory framework, as the regional nucleophilic Fukui function, fX---, [15 - 16], which governs the site selectivity for nucleophilic attack. Due to the proton hardness, Li and Evans [17] have proposed that protonation occurs at the site with minimum Fukui function, with successful results in many systems [18]. Recently, in an extensive analysis of these reactions, Chataraj [19] has shown that this criterion is valid only for system presenting a unique protonation site in different chemical environments; otherwise the Frontier Orbital Theory [20] is not valid. Moreover, this author conclude that "hard - hard interaction are charge controlled and predominantly ionic in nature, and for these reactions the preferred site is that which contains maximum net charge that may coincide, in certain cases, with the site associated with the minimum value of the Fukui function".-

On the other hand, the hard-soft acid-base, HSAB, principle [21] states that "among potential partners of a given electronegativity, hard likes hard and soft likes soft". In gas phase both partners; ammonia and formate are classified by Pearson [22] as hard bases.-

Finally, we show that the n-orbital and o-orbital eigenvalues constitute adequate regional reactivity descriptors of the bases involved respect to protonation, in gas phase and in the presence of external fields.-

MODEL HAMILTONIAN:

For an amino acid crystal, electrons will be highly localized on individual molecules contained within unit cells. Therefore, the electronic Hamiltonian, after invoking the Born - Oppenheimer separation, may be written in the form

=

Here and are position vectors identifying the origin of different cells within the lattice. VINT corresponds to the attractive interaction potential between electrons and nuclei in different cells (u and v) plus interactions between electrons in different unit cells (u and v).

If we now seek a solution to the Schroedinger equation of the simple

antisymmetrized product form:

where is a symbol standing for all electronic coordinates at unit cell u, and apply the calculus of variations, optimizing:

subject to the orthonormality constraints:

=

we will get a set of equations parallel in form to the Hartree Fock equations:

with

Here ρ is an index running over effective point charge centres, such that Zρ might be either a positive or a negative charge. VEXT corresponds to the sum of the interactions of nuclei in v with electrons and nuclei in u, plus coulombic and exchange interactions of electrons in u with electrons in v.

Again, because of the high degree of electron localization in the crystal and because of the short range of exchange effects, we neglected exchange effects between different unit cells, and the coulombic interactions between electrons of different cells were approximated as a superposition of purely coulombic point charge interactions.

Now, a further specialization of this form in which the sums on v are restricted to near neighbours only, and charge distributions are replaced by effective point charges on the near neighbours, gives us the approximations used in the model for H bonding in solid amino acids.

RESULTS AND DISCUSSION

The model system consists of a linearly hydrogen bonded complex between formic acid and ammonia representing one of the intramolecularly hydrogen bonded amino acid units in the crystal, surrounded by a nearest neighbor array of point charge dipoles which will generate an electric field similar to that expected in a zwitterion crystal lattice.

All the HF/STO-3G calculations were performed using a version of the HONDO quantum chemistry program [23] which we adapted to a DCG Eclipse S/230.

To build up the molecular complex in vacuum, we adopted the frozen geometry approach, but since the geometry of donor formic acid is different from that of the formate, a nuclear relaxation of this fragment was performed. This procedure was performed on the linearly hydrogen bonded complex, at a fixed N-O distance of 2.8 Å.

The geometry of ammonia was taken from HF/STO - 3 G calculations reported by Lathan et al. [24], corresponding to a N-H distance of 1.033 Å and to the angle (H-N-H) of 104.2 °.

The initial formic acid structure, which was taken from HF / STO-3G calculations by Peterson and Csizmadia for the syn conformer [25], was permitted to relax at three different values of the H-bonding hydrogen position, 1.154, 1.455 and 1.741 angstroms from the nitrogen atom, and the corresponding structures were denominated as Z (zwitterions), M (medium) and N (neutral), respectively. Geometries involved in proton transfer were obtained by linear interpolation of these structures.

As expected, the calculated proton transfer potential is a single well curve [10] with the minimum at the neutral structure. Also, the energy difference between Z and N was 118 mH. No doubt that this extremely high value is due to the low flexibility of the minimum basis set. In a later section we show that calculations performed with 6-31G(d) wavefunctions, give a significant smaller value for this energy. However; since we are interested in rationalize the results obtained at HF / STO-3G level, as a function of the MO implicated in the H-bridge, these parameters are examined at the same accuracy level and at 6-31G(d) level.

The geometry obtained in vacuum was used in the calculations with the field on.

Based on the crystal structure of the benzene inclusion compound of N, N-diethyl-β-alanine reported by Nash et al. [11-12], the system was constructed in the following way: the structural unit H3N .. H .. OCOH was situated in the crystal ab plane, with two of the amine hydrogen atoms placed above and below the plane. The four nearest neighbors lying in the ab plane containing the origin were considered. The charges on bare dipoles located on the b-axis are defined as +A and -A and those on a axis are +B and -B. The distances between the centers of the two dipoles on the a and b axes are 11.25 Å and 12.0 Å, respectively, while the "length" of each dipole is 1.65 Å. The crystal b axis makes angles of -a with dipoles +B - B while with +A...-A dipoles it makes angles of +a. The internal dipole should make the angle of - a with the b axis.

A Cartesian coordinate system was defined such that the yz plane was coplanar with the ab plane; the z axis was chosen as the hydrogen bond axis (N..H..O), with the functional oxygen of formic acid at the origin. The external field position then is such that the y axis makes an angle .q with the crystal b axis. The origin of the ab plane lies on the center of charges of the ammonia formic acid system.

The external field and the molecular complex are shown in Figure 1. Notice that the dipoles orientation are such that +A-A pairs will produce attractive interactions with the central unit, while the +B-B pairs produce repulsive interactions with the central unit.

Fig. 1. Positioning of the molecular complex in the imposed external field.
H1, 2 indicates two hydrogen atoms lying above and below the molecular plane.
Parameters explained in the text.

HF / STO-3G calculations on the system were performed provided that crystal symmetry was preserved. In this respect, the center of the field was made to coincide with the Mulliken [26] center of charges, and the α angle was made to follow the crystal pattern (details in ref [10]).

Relevant to the purposes of this work was the study concerning the effects of the point charges field parameters on the proton transfer potential function (PTPF). To accomplish it, various external fields were imposed to the molecular complex. In all of those calculations the position and the geometry of the complex were fixed as in Figure 1, and the STO-3G energies of the N, M and Z structures were calculated in the presence of various fields. To account for the influence of the field in the PTPF, the results are reported in terms of Z-N, Z-M and N-M energies. When Z-N is positive, it means that N is more stable than Z and vice versa. If M is larger than both Z and N, then we are in the presence of a double well PTPF. In this way, these three parameters give the information about the equilibrium position between the two tautomeric forms of the complex. Table 1 records the effect of field parameters in the complex PTPF.

Table 1: Energy profiles as a function of field parameters


qa
angle

A b
Charge

B b
charge

(Z-N) c
energy

(Z-M) c
energy

N-M c
energy


-- 53

3.0

0.6

47.2

17.6

--29.6

-- 53

4.0

0.8

22.1

3.1

--19.0

-- 53

5.0

1.0

-- 3.6

--11.2

--7.6

-- 53

6.0

1.2

-- 29.7

--25.3

4.3

0

5.0

1.0

51.4

14.6

-- 36.9-

--40

5.0

1.0

7.8

-- 4.7

--12.5

-- 53

5.0

1.0

-- 3.6

--11.2

-- 7.6

-- 60

5.0

1.0

-- 9.1

--14.3

-- 5.2

-- 53

1.0

1.0

103.5

51.7

-- 51.8

-- 53

5.0

5.0

40.0

13.3

-- 26.7

-- 53

8.0

8.0

--11.9

--13.6

-- 4.7


aIn degrees. bIn atomic units. cIn mH.

The effects of the field strength in the PTPF were examined at the fixed Q value of -53° and, maintaining the charges ratio in A/B = 5 we changed the magnitude of the dipoles charge, creating the field 1/0.2; 2/0.4; 3/0.6; 4/0.8 and 5/1. The results are shown in the upper part of Table 1.

The effects of the external field orientation in the PTPF were calculated at the bare dipoles charges fixed in: A=5 and B=1 (Field 5/1) while Q was varied. He middle part of Table 1 show the results obtained.

The effects of the charges ratio in the PTPF, were analyzed at Θ = 53 with the charges ratio of A/B =1. The results are shown in the lower part of Table 1.

In order to rationalize the results obtained (Tables1 -3), contour plots of the nuclear - electronic potential created by the dipolar array were calculated. Twenty contours were calculated for each external field imposed. The values of the electric external potential in the calculated contours were chosen to be in the range -0.5 a.u. to +0.45 a.u., with increments of 0.05 a.u.

Table 2. HF/6-31G(d). Molecular Orbital Parameters for Neutral Form.


Field

(Z-N)a
energy

MO-position

MOb
eigenvalue

MOSy
mmetry

rNc
electron density


0.0/0.0

25.2

HOMO

--0.43129

p

0.041761

HOMO-1

--0.45509

s

0.000059

HOMO-2

--0.46491

s

0.751586

0.2/1.0

16.6

HOMO

--0.44427

s

0.753581

HOMO-1

--0.47522

s

0.114147

HOMO-2

--0.48218

p

0.000111

0.5/2.5

0.3

HOMO

--0.42853

s

0.872548

HOMO-1

--0.52286

p

0.000366

HOMO-2

--0.52522

s

0.008406

0.6/3.0

-- 6.5

HOMO

--0.42255

s

0.875044

HOMO-1

--0.53647

p

0.000661

HOMO-2

--0.54281

s

0.00562

0.8/4.0

-- 22.8-

HOMO

--0.41088

s

0.871635

HOMO-1

--0.56382

p

0.004304

HOMO-2

--0.57819

s

0.013947

1.0/5.0

-- 44.5-

HOMO-

--0.40019

s

0.861221

HOMO-1

--0.58143

s

0.658032

HOMO-2

--0.58526

s

0.646285


aIn kcal/mol. bIn Hartrees. cIn atomic units. n-orbital parameters.

The contours corresponding to negative values of the external field were drawn as dashed lines, while those corresponding to positive values or zero were drawn as solid lines; this last one is denoted by a 0, the contour at the external potential value 0f -0.05 au. is labeled with an a, and that of the value of +0.05 is labeled with an x. The values of successive contours are easily inferred. The molecular complex was drawn in each plot at the coordinates corresponding to the zwitterionic structure, at the N-O distance of 2.8

-Table 3. HF/6-31G(d) Molecular Orbital Parameters for Z Form.


Field

E( Z-N)
(Kcal/mol)

zwitter

E OM

Symmetry

ro


0.0/0.0-

25.2

HOMO

--0.36066

p

0.463697

 

 

HOMO-1

--0.36636

s

0.122224

 

 

HOMO-2

--0.38071

s

0.71071

0.2/1.0

16.6

HOMO

--0.3799

s

0.49431

 

 

HOMO-1

--0.39337

s

0.143073

 

 

HOMO-2

--0.40175

s

0.667596

0.5/2.5

0.3

HOMO

--0.41406

s

0.503519

 

 

HOMO-1

--0.43674

s

0.161357

 

 

HOMO-2

--0.4383-

s

0.652431

0.6/3.0

6.5

HOMO

--0.42737-

p

0.505893

 

 

HOMO-1

--0.45074-

s

0.584108

 

 

HOMO-2

--0.45362-

s

0.325539

-0.8/4.0-

-- 22.8

HOMO

--0.4578-

p

0.508045

 

 

HOMO-1

--0.48034-

s

0.568374

 

 

HOMO-2

--0.48733-

s

0.220604

1.0/5.0

-- 44.5

HOMO

--0.49507-

p

0.509732

 

 

HOMO-1

--0.51377-

s

0.601828

 

 

HOMO-2

--0.52504-

s

0.147286


aIn kcal/mol. bIn Hartrees. cIn atomic units. o-orbital parameters.

All the calculated contours show that the dipolar array created positive potentials in the formate region, while in the ammonium region it created negative potentials. Figure 2 shows the results for field 5/1 at a Θ value of -53 degrees. Notice that the formic acid/formate end lies in a positive electric potential well with the ammonia / ammonium end placed in a negative well.


Fig. 2. Contours of the electrostatic energy from the imposed field 5/1 at an orientation angle of 53°. Solid contours at are positive values of the potential while dashed contours are negative values.

From the plot, it can be observed that when the field strength decreases (Fields 4/0.8; 3/0.6; 2/0.4), the formate / formic acid lies in a less positive well and the ammonia/ ammonium fragment lies in a less negative well. Since the molecular complex is a polar molecule with the ammonium / ammonia fragment positively charged, and the formate / formic acid fragment is negatively charged, the field should stabilize more the ionic structures than the neutral ones. Also, an increase of the field strength produces an increase of Z stabilization. The effects of the field orientation can be observed by rotating the field around the molecular complex; it can be observed that as Θ value goes from 0Ί to 60º, the potential wells where the molecular fragments are placed, increase in magnitude. The correlation with the calculated Z -N energies is apparent. The results for the fields of ratio 1/1 must follow the same pattern as those of 5/5, but it takes a larger bare charge set to induce the Z form because the dipoles in a axis exert repulsive interactions with the unit cell dipole.

Molecular orbital diagrams.

Since in our model all interactions between unit cells are approximated by those of effective point charges, the external electric field will modify the one electron electronic Hamiltonian whose eigenfunctions are the molecular orbitals. It is then expected that both, the MO eigenvalues and its composition, will reflect the effects of the field on the complex electronic structure.

In order to study the response of the MOs for the three different structures to the external field, MO energy - level diagrams for the last three occupied orbitals were constructed and compared with that calculated in the absence of any external field. For the sake of consistency, the first set of calculations was performed at STO-3G level, using the same geometries and same external fields corresponding to Tables 1 to 3.

Figure 3 shows the diagram in the absence of the field for the three highest occupied MO, which have been labeled with the letters a, b, and c, in order to correlate them with those in the presence of the field.


Fig. 3. Molecular orbital energy-levels diagram for the isolated system. Energy in Hartrees. Q= - 53°.

Examination of the MO expansion atomic orbital coefficients for the orbitals labeled a and b show the same qualitative nature of the orbitals across all three structures. The orbital labeled a can be identified as a delocalized π (Cs symmetry a’’) on the formic acid / formate fragment; i.e., major contributions of the orbital are from 2px atomic orbitals on the oxygen atoms. The orbital labeled b can be identified as a delocalized _one pair orbital (Cs symmetry a’) on both oxygen atoms. The orbital labeled c (Cs symmetry a’) changes in character in progressing from zwitterionic to midpoint to neutral structure. In the zwitterion it can be identified as a second _one pair largely localized in the 2pz functional oxygen, o-orbital. The designated midpoint structure largely retains the same characterization. By the neutral structure, however, the nature of the orbital is clearly a _one pair largely localized in the 2pz orbital of the nitrogen atom, n-orbital. The adiabatic characterization of orbital c thus moves from one fragment to the other end, and can be called the lone pair involved as the hydrogen bond receptor. A non-adiabatic description of the orbitals would indoubtly identify an orbital crossing.

Figure 4 shows the three HOMOs for fields 3/0.6; 4/0.8; 5/1and 6/1.2 at a Θ value of -53 degrees, labeled with the letter corresponding to the MO in the absence of the field.


Fig. 4. Moleculdar orbital energy-levels diagram, for fields of ratio 5/1. Energy in Hartrees. Q= - 53°.

Analysis of the MO diagram shows that the presence of the field stabilizes the a delocalized p (a’’) MO, antibonding between the two oxygen atoms. This result may be correlated with the electric potential contour maps since both oxygen atoms are placed in a positive hole; therefore, they must be stabilized by the field.

Orbital b, which corresponds to the σ delocalized MO between the two oxygen atoms, is also stabilized as the field strength increases. Correlation with the contour maps shows the agreement for the same reason that orbital a is stabilized with the increasing of field strength.

The MO c in the zwitterion and midpoint structures is observed to be stabilized with the increasing of the field strength; in agreement with our analysis, since this orbital is the o-orbital located in a positive potential.

The MO c in the neutral structure, however, corresponds to the n-orbital, and it is observed to be destabilized with the increasing of the field strength. Correlation with the contour map show that N is always in a negative potential, so the lone pair should be destabilized when the external potential on it becomes more negative, increasing its reactivity as a Lewis base and also as a Brönsted base, in this case.

The MO diagrams obtained for fields of charge ratio 1 at a _value of -53 as well as that of fields 5/1, for various _value, showed the same trend, i.e., as the field strength increases on the molecular complex, the neutral structure n-orbital eigenvalue increases (destabilizes) and the zwitterion o-orbital eigenvalue decreases (stabilizes).

In summary, the negative potential on N stabilizes the ionic ammonium ion, but at MO level the formation of ammonium occurs because of the destabilization of the n-orbital, which increases the ammonia base reactivity (and proton affinity). The same argument holds for the stabilization of formate on the positive potential, which stabilizes the o-orbital decreasing the formate reactivity as a base.

Due to the restrictions that the minimum basis set imposes on the results, we have performed HF / 6-31G(d) calculations on the complex, to establish that the fields effects on the hydrogen transfer profile that we have described above, are not artifacts of that particular basis set.

-One way to judge the adequacy with which a given basis set might represent the hydrogen transfer profile in the complex, would be to compare the calculated and experimental proton affinities of the infinitely separated bases, i.e., ammonia and formate ion, in particular for the Z and N energies. The calculated proton affinity of ammonia (including optimum geometries) with STO-3G and 6-31G(d) basis set are 259 kcal/mol [24] and 207.4 [27], respectively. The experimental value is 204.2 [28]. For formate, the calculated STO- 3G, and 6 31G(d) proton affinity are 482 [25] kcal/mol and 359 [27], respectively. The experimental value is 345 kcal/mol [29]. These results show the adequacy of the extended basis to describe the relative energy between Z and N in vacuum, and we expect the hydrogen transfer profile calculated with this basis to be more reliable. The next set of calculation was performed at HF / 6-31G(d) level using the GAUSSIAN98 package [30].

In order to compare the influence of the basis set on the complex structure, we have chosen the set of 5/1 4/0.8; 3/0.6; 2/0.4 fields, at the Θ value of -53°.

The MO diagrams for neutral and zwitterionic structures are shown in Table 2 and Table 3, respectively. In both tables, the first column indicates the applied field, and the second records the energy difference between the Z and N forms. These data show same qualitative dependence of well depths on field strength as obtained with the STO-3G, but the 6-31G(d) basis set, being more flexible, produces a true double well potential with much more smaller external fields than was required for the STO-3G. The three highest occupied molecular orbitals, HOMO, HOMO-1 and HOMO-2 are examined and identified in column 3. The 4th column gives the MO eigenvalue, the 5th shows it symmetry. In order to emphasize the localization of the MO, the N electron densities at the n-orbital, N, and the functional oxygen electron density at the o-orbital, O, are recorded in 6th column of both Tables. The large electron density on nitrogen easily identifies the n-orbital for the neutral form, while that on functional oxygen identifies the o-orbital.

Table 2 shows that as the field strength increases, the n-orbital eigenvalue become more positive, in the neutral form. This destabilization of n-orbital increases the proton affinity of ammonia, and therefore the stabilization of Z respect to N, as shown in (Z-N) energy decrease. In turn, Table 3 shows the opposite trend for o-orbital.

The results show that, regardless the basis set used, the fields (of ratio 5/1 and value of -53°) imposed stabilize the zwitterionic structure respect to the neutral form, via increases of the n-orbital energy and decreases of the o-orbital one. In turn, the destabilization of the lone pair electrons of N can be rationalized by its repulsive interaction with the negative field created by the dipolar array, and the attractive interaction of the O lone pair electrons with the positive field where they are immersed. Therefore, the electrostatic effects of the field on the electronic structure of the complex in the polar medium are explained ion terms of MO energies. It sounds paradoxical that the charge controlled protonation reactions are in turn controlled by MO eigenvalues.

The systematic study presented here strongly suggests that, at least for the actual hard bases (amines and conjugated bases of carboxylic acids), the protonation site is determined the by x-MO eigenvalue, in gas phase and in the presence of external fields.

The electron density on MOs, deserves some comments. When the MO is the HOMO, it represents the nucleophilic regional Fukui function [15-16]. Inspection of this parameter in both Tables, show that as the field strength increases, fN- increases for neutral structure and fO- also increases for the zwitterionic structure. However, in Z form, the HOMO belongs to the p system; therefore, it cannot be reflecting properties of the σ electron lone pair of functional oxygen atom. Same argument is valid for N form in vacuum.

For N structure, --N, which actually represents the nitrogen electron lone pair involved in the H-bridge, increases with the field strength. For large fields, it is observed to decrease; however, HOMO-1 and HOMO-2 are localized on N and its bonded H atoms, revealing a large concentration of charge on nitrogen atom. For Z form, O shows a decrease with the increasing field strength, with an inversion for large fields. These results suggest the possibility of incorporating X in the description of regional reactivity descriptors.

In summary, the x-orbital energy and its associated electron density, --X, which we present in this article, could be adequate parameters to take into account in the construction of local molecular descriptors for protonation processes, in vacuum as well as in the presence of electric fields. We are presently working on this problem.

REFERENCES

1. Morokuma, K.; Kitacura, K. Molecular Interactions; Ratajczak, H. Orville - Thomas, W.J., Eds.; Wiley: New York ; 1, p21 -27, (1980).        [ Links ]

2. Singh, U. C.; Kollman, P. A. J. Chem. Phys.; 85, 364, (1983).         [ Links ]

3. Gilli, P.; Ferreti, V. Gilli, G. Acta Crystallogr. Sect B Struct. Sci., B, 54, 50-65, (1998).         [ Links ]

4. Mo,Y.; and Gao, J. J. Phys. Chem. A, 105, 6530-6536, (2001).         [ Links ]

5. Ramakrishnan, C.; Prasad, C. Int. Protein Res., 3, 209, (1971).         [ Links ]

6. Danenberg, J. J.; Haskasp, L.; Masunov, A. J. Phys. Chem. A., 103, 7083 - 7086, (1999).         [ Links ]

7. J. Yin, M. Green, J. Chem. Phys. A, 102, 7181 -7190, (1998).         [ Links ]

8. Masunov A., Dannenberg J. J.; Contreras R. H. J. Phys. Chem. A, 105 (19), 4737 - 4740, (2001).        [ Links ]

9. Nesspolo, Massino, Ferraris Giovanni; Ivaldi, Gabriela; Hoppe, Rudolf; Acta Crystallografica b, 57, 652 - 664, (2001).-        [ Links ]

10. Parra - Mouchet, J; W.H. Fink and C.P. Nash. J. Phys. Chem. 89, 524 --532, (1985).        [ Links ]

11. Peterson, M. A.; Hope, H. and Nash, C. P. J. Am. Chem. Soc. 101, 946 - 950, (1979).        [ Links ]

12. Peterson, M. A.; Nash, C. P. J. Phys. Chem., 89, 521 - 524, (1985).-        [ Links ]

13. Del Bene Janet E. J. Am. Chem. Soc. 102, 5191 - 5195 (and references there in), (1980).        [ Links ]

14. C.A. Olea-Azar and J. Parra-Mouchet. Bol. Soc. Chil. Quím. 43, 189-200, (1998).        [ Links ]

15. Parr, R. G.; Density Functional Theory of Atoms and Molecules. Oxford University Press. New York, (1989).        [ Links ]

16. Parr, R.G.; Yang, W. J. J. Am. Chem. Soc., 106, 4049, (1984).         [ Links ] Contreras, R.; Fuentealba, P.; Galván, M.; Pérez, P. Chem. Phys. Letts. 304, 405-413, (1999).        [ Links ]

17. Li, Y.; Evans, J. N. S. J. Am. Chem. Soc., 117, 7756, (1995).        [ Links ]

18. Fuentealba, P. and Contreras, R: "Fukui Function in Chemistry". Review in Modern Quantum Chemistry. A celebration of the contributions of Robert G. Parr. Vol II p.1013 -1052 , (2002), (and references there in).        [ Links ]

19. P. K. Chattaraj, J. Phys. Chem. A. 105, 511 - 513, (2001).-        [ Links ]

20. Fukui, K. Theory of Orientation and Stereoselection; Springer-Verlag:-Berlin, p.134, (1973);         [ Links ] Science (Washington D.C.), 218, 747, (1982).        [ Links ]

21. Pearson, R. G: Chemical Hardness: Application from molecules to solids; Wiley-VCH Verlag GMBH; (1997).        [ Links ]

22. Pearson, R. G. J. Am Chem. Soc., 110, 7684, (1988).        [ Links ]

23. Dupois, M.; King, H.; Rys, J. QCPE, 13, 338, (1977).        [ Links ]

24. W.A. Lathan, W.J. Hehre, L.A. Curtis, J. A. Pople. J. Am. Chem. Soc. 93,-6377 - 6387,(1971).        [ Links ]

25. M. R. Peterson; J. G. Czismadia, J. Am. Chem. Soc., 10, 1076, (1979).         [ Links ]

26. R. S. Mulliken; J. Chem. Phys., 23, 1835, (1955).         [ Links ]

27. Pérez, P. ; Zapata - Torres, G.; Parra-Mouchet, J.; Contreras, R. -Int. J. Quant. Chem., 74, 387 - 394. (1999).         [ Links ]

28. LIas, S. G.; Bartmess, J. F.; Liebman, J. L.; Holmes, J. L.; Mallatrd. W.G. J.-Phys. Chem. Ref. Data., 27, No.3, (1988).        [ Links ]

29. Ceyer, S. T.; Tiedemann, P. W.; Mahan, B. H.; Lee, Y. T. J. Chem. Phys. 14 - 27 (1979).        [ Links ]

30. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle, and J. A. Pople. Gaussian98. Gaussian Inc. Pittsburgh, PA. (1998).        [ Links ]

* To whom correspondence should be addresse