Water Solvent Effect on Theoretical Evaluation of ... - ACS Publications

Subscriber access provided by University of Newcastle, Australia

Article

The Water Solvent Effect on Theoretical Evaluation of 1h Nmr Chemical Shifts: The O-Methyl-Inositol Isomer Helio F. Dos Santos, Marcelo Chagas, Leonardo De Souza, Willian Rocha, Mauro V. De Almeida, Cleber P. A. Anconi, and Wagner Batista De Almeida J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.7b01067 • Publication Date (Web): 22 Mar 2017 Downloaded from http://pubs.acs.org on March 24, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry A is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The Water Solvent Effect on Theoretical Evaluation of 1H NMR Chemical Shifts: The oMethyl-Inositol Isomer Hélio F. Dos Santos[a], Marcelo A. Chagas[c], Leonardo A. De Souza[e], Willian R. Rocha[c], Mauro V. De Almeida[b], Cleber P. A. Anconi[d], Wagner B. De Almeida*[e] [a]

Núcleo de Estudos em Química Computacional (NEQC), [b] Núcleo Multifuncional de Pesquisas Químicas (NUPEQ), Departamento de Química, ICE, Universidade Federal de Juiz de Fora, Campus Martelos, 36036-330, Juiz de Fora, MG, Brazil. [c] Laboratório de Química Computacional e Modelagem Molecular (LQC-MM), Departamento de Química, ICEx, Universidade Federal de Minas Gerais, Campus Pampulha, 31270-901, Belo Horizonte, MG, Brazil. [d] Laboratório de Química Fundamental (LQF), Departamento de Química, Universidade Federal de Lavras, 37200-000, Lavras, MG, Brazil. [e] Laboratório de Química Computacional (LQC), Departamento de Química Inorgânica, Instituto de Química, Universidade Federal Fluminense (UFF), 24020-150, Niterói, RJ, Brazil. Abstract In this paper, density functional theory calculations of nuclear magnetic resonance (NMR) chemical shifts for L-quebrachitol isomer, previously studied in our group are reported aiming to investigate in more details the water solvent effect on the prediction of 1H NMR spectra. In order to include explicit water molecules, twenty water-L-quebrachitol configurations obtained from Monte Carlo simulation were selected to perform geometry optimizations using the effective fragment potential method encompassing 60 water molecules around the solute. The solvated solute optimized geometries were then used in B3LYP/6-311+G(2d,p) NMR calculations with PCM-water. The inclusion of explicit solvent in the B3LYP NMR calculations resulted in large changes in the 1H NMR profiles. We found a remarkable improvement in the agreement with experimental NMR profile when the explicit hydrated L-quebrachitol structure is used in B3LYP 1H NMR calculations, yielding mean absolute error (MAE) of only 0.07 ppm, much lower than reported previously for gas phase optimized structure (MAE = 0.11 ppm). In addition, a very improved match between theoretical and experimental 1H NMR spectrum measured in D2O was achieved with the new hydrated optimized L-quebrachitol structure, showing that a fine tuning of the theoretical NMR spectra can be accomplished once solvent effects are properly considered.

*Corresponding author: [email protected] (W. B. De Almeida)

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Introduction

Nuclear resonance magnetic (NMR) spectroscopy have largely been used in conformational analysis studies and also as a useful guide for structural determination, in particular in the area of supramolecular chemistry involving inclusion compounds with biological activity1-5. The NMR measurements are conducted for a macroscopic sample in a given solvent, usually deuterated water (D2O) and the NMR assignments are made based on pre-established criteria. The theoretical calculations are made for selected local minima equilibrium structures obtained from geometry optimization for a single molecule in the vacuum using a wisely chosen level of theory. From a theoretical point of view many structures can be used to generate NMR spectra and a comparison with experiment provide a strong indication of the precise molecular structure present in the experimental sample. That is precisely what we have been done in recent works, showing the usefulness of the quantum chemical calculations to experimentalists, being of great help for example in structural determination6,7. In a previous combined experimental and theoretical work7 with L-quebrachitol (Scheme 1) we successfully used density functional theory (DFT) with the B3LYP functional and 6311++G(2d,p) basis set for the calculation of NMR spectra, which enabled the elucidation of the molecular structure through comparison with experimental data, showing the significance of the theoretical calculation of 1H NMR chemical shift in conformational analysis studies. As the experimental job was conducted using D2O as solvent, we used the polarisable continuum model (PCM) approach to account for the solvent effect on the calculated NMR spectra. It is worth mentioning that previous B3LYP NMR studies for D-glucose in water8, through comparison with experimental data consistent reaction field models was suggested to reasonably account for solvent effects. In addition, quantum chemical methods using continuum solvent models have been for some time currently used in the prediction of NMR chemical shifts of organic compounds9. However, we wondered whether the changed in the L-quebrachitol geometry in the presence of explicit water molecules would result in a better agreement with experimental data. Therefore, we make use of the effective fragment potential (EFP) model to include explicitly the solvent effect on the geometry of the L-quebrachitol solute and then to evaluate the NMR spectra of the solvated optimized structure. In Ref. 7 the molecular structure of L-quebrachitol was determined as isomer 2a based on the comparison between experimental (in D2O) and theoretical PCM-water 1H NMR and infrared spectra for a series of five possible candidate structures. While the agreement with experimental 1H NMR pattern was not perfect it was satisfactory enough and the accordance between experimental and theoretical IR spectrum for structure 2a was a very useful complementary information which left no doubt that structure 2a was the correct molecular ACS Paragon Plus Environment

Page 2 of 18

Page 3 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


structure present in the 1H NMR experiment conducted in D2O. However, we thought that it could possible to improve the agreement with experimental 1H NMR spectrum, and that the solvent effect can play an important role in this study. As NMR chemical shifts are very sensitive to the functional groups relative position in a given molecular structure we expect that the optimization of L-quebrachitol geometry in the presence of explicit water solvent molecules would adjust the ∠HCCC and ∠HOCC dihedral angles closest to the observed experimental conformation and then produce more realistic theoretical NMR spectra. Therefore, two approaches were used in this work: (i) B3LYP PCM-water calculation of NMR chemical shifts with the bare L-quebrachitol solute (with no explicit water included) but with the geometry optimized in the presence of 60 water molecules using the EPF model (ii) the same B3LYP NMR calculation including the 60 water molecules surrounding the optimized solute geometry. Through direct comparison with experimental NMR spectra measured in D2O we hope to provide a clear understanding of the real need of including explicit water molecules in DFT calculations of NMR spectra for organic molecules, that is much more computational expensive procedure than single point DFT PCM calculations, using gas phase optimized geometries which have been commonly utilized in many application (see Refs. 2,3).

Methodology The level of theory employed in this work was DFT10 with the B3LYP functional11,12 using the 6-31G(d,p) and triple-zeta quality 6-311++G(2d,p) basis set13. The starting point in our study is the B3LYP/6-311++G(2d,p) gas phase 2a structure of L-quebrachitol reported in Ref. 7, shown in Figure 1a. In order to simulate the solvent effect on the NMR spectra we used the EFP1 method14 as implemented in the Gamess package15. The DFT-based EFP method was used to optimize the structure of the L-quebrachitol isomer 2a in the presence of the solvent at the B3LYP/6-31G(d,p) level. Initially the solute molecules were placed in a cluster of 60 EFP water molecules, generated by Monte Carlo (MC) statistical mechanics simulations carried out in the NVT ensemble at T = 298 K, according to the procedure described previously16. The MC simulations were performed with the DICE program developed by Canuto and Coutinho17. In the simulation, a total of 140000 steps were used for the evolution of the structural and thermodynamic averages in the equilibration step. Every 400 steps of the simulation a lower energy molecular configuration for L-quebrachitol relative to the previous step was saved. At the end, a total of 350 configurations were generated for the solvated system. After an analysis of the structures with uncorrelated energies, we obtained a sample space of 175 lower energy configurations. From these structures, we selected 20 lower energy configurations. Selected structures of the cluster were then used in B3LYP-EFP geometry ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

optimization calculations. Twenty optimized structures (named I to XX) were used in B3LYP/6311++G(2d,p)-PCM-water NMR chemical shifts calculations with the Gaussian09 package18 employing the gauge-independent atomic orbital (GIAO) method implemented by Wolinski, Hilton and Pulay19 for calculation of 1H magnetic shielding constants (σ), with chemical shifts (δ), obtained on a δ-scale relative to the TMS, taken as reference. Solvent effect on the calculation of NMR chemical shifts was also evaluated using the polarizable continuum model (PCM)21 by performing B3LYP/6-311+G(2d,p) single point NMR calculations, with the corresponding fully optimized geometries in vacuum.

Results and Discussion In the EFP method of Gordon et al.22,23 the system is partitioned into a reactive or quantum mechanical region (QM), composed of the solute, and the solvent molecules are treated explicitly as fragments, described as a set of one electron potentials that are added to the Hamiltonian of the quantum mechanical region. These potentials contain terms describing: (i) coulombic interactions (including charge penetration) (ii) polarization or induction interaction between solvent-solvent and solvent-solute molecules and (iii) exchange repulsion, charge transfer and other terms that are not taken into account in previous two terms23,24. One very attractive characteristic of this QM/EFP method is that we do not need to parameterize intermolecular potential functions for the solute, whatever system is studied. This fact is particularly important if we are interested in studying transition metal compounds in solution, for which there is a lack of available intermolecular interaction potentials16,21-23. The relevant dihedral angles for the twenty B3LYP/6-31G(d,p) optimized structures of the L-quebratchitol, inside the EFP water cluster, are given as Supplementary Material (Table S1-S3), along with the corresponding values of the gas phase and PCM fully optimized geometries for the structure 2a previously reported7. Relative DFT energy values for twenty hydrated (and also non-hydrated) structures are given as Supplementary Material (Table S7). A snapshot of the optimized VIII structure is shown in Figure 1b to visualize the solute hydrated configuration from the MC simulation. Table 1 reports optimized ∠HOCC, ∠HCCC and ∠CCCC dihedral angles for structure VIII along with corresponding values of the gas phase and PCM optimized structures for L-quebrachitol 2a. Results for the optimized supermolecule geometry including the 60 water molecules explicitly are also included in the last row of Table 1, to make a comparison with the EFP geometry optimization. It can be seen from Table 1 that the PCM dihedral angles do not differ significantly from the gas phase values having a maximum deviation of about 5⁰ , showing that the PCM method produces a smooth change in the gas phase structure. On the other hand a large variation in both ACS Paragon Plus Environment

Page 4 of 18

Page 5 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


∠HOCC, and ∠HCCC dihedral angles is observed due to the presence of water molecules during to the geometry optimization. Using the gas phase optimized 2a structure as reference, it can be observed large deviations for the ∠HOCC dihedral angles (about ±90⁰) for some structures. The ∠HCCC dihedral angles deviations are smaller but sizeable (about ±15⁰) for some structures. Figure 2 show the deviations of dihedral angles with respect to the gas phase optimized 2a structure (having a null deviation symbolizing by horizontal line) and their PCM optimized geometry and two selected EFP structures, VIII and XVII. Two additional sets of geometry optimization calculations were carried out: (i) re-optimizing a supermolecule hydrated structure including 60 water molecules without using the EFP approach and (ii) re-optimizing the EFP structure but removing the 60 water molecules, like gas phase structure. It can be seen from Figure 2 that, the ∠HOCC dihedral angles were significantly changed due to the solvent effect, as could be expected. A distribution deviation of ±15º is observed for the other dihedral angles. Therefore, regarding the structure, results indicate that the explicit solvent effect can alter considerably the solute geometry, as could be expected, and it is interesting to evaluate how this can affect the calculation of NMR chemical shifts. There has been significant development in the ab initio theory of NMR spectroscopy8,9,24-26, encompassing theoretical studies describing a method for calculation of isotropic shielding constants in solution using the EFP method coupled with London's GIAO theory27,28. In the EFP waters, the electron density of the ab initio region will be perturbed, and this information is carried in the quantum mechanical region, including the effect of an external magnetic field and nuclear magnetic moments27,29. Comparison between experimentally observed and theoretical spectral pattern has been proved very useful for identification of chemical behaviour and elucidation of molecular structure of a given compound. In this connection, a match between experimental and theoretical NMR chemical shifts has been found very useful for the assignment of the most probable isomer or conformation present in an experimental sample usually in aqueous solution. The ab initio calculation of NMR spectrum is made for a single gas phase molecule and so to make realistic comparison with experimental data the solvent effect must be considered. Bagno et al.8 in a theoretical-experimental work contribute considerably for the importance of considering the solvent effect in the determination of the

13

C and 1H NMR spectra of organic molecules. The authors

observed from the experiments and PCM-B3LYP/6-31G(d,p) calculations a substantial correlation of NMR spectra in polar solvent environment for strychnine, corianlactone, daphnipaxinin and boletunone B molecules found in natural products. The authors also report the need for a broader sampling of the molecules conformations studied that are quite flexible and a better treatment of specific solvent effects.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The experimental 1H NMR spectrum (in D2O) and the B3LYP/6-311++G(2d,p)-PCM-water chemical shifts for selected structures VIII and XVII, along with the corresponding results for gas phase and PCM optimized 2a structure spectra are shown in Figure 3. Structures VIII and XVII were selected among all twenty EFP optimized geometries based on the accordance with experimental 1H NMR spectrum (see Supplementary Material, Figures S1 and S2). Five sets of 1H NMR B3LYP/6-311++G(2d,p)-PCM-water calculations were performed: (i) using the B3LYP/631G(d,p)-EFP optimized structure; (ii) re-optimizing the EFP optimized structure including explicitly the 60 water molecules like a supermolecule calculation; (iii) removing the 60 water molecules from the EFP optimized structure; (iv) removing the 60 water molecules from the B3LYP/6-311++G(2d,p) supermolecule optimized structure; and (v) re-optimizing EFP optimized structure, removing the EFP solvent water molecules. MAE (median absolute deviation) values (in ppm) relative to experimental data (∆δExp-Aver) are quoted. The labels (a)-(m) of each spectrum is indicated in the caption of Figure 3. The DFT PCM-water chemical shifts for all twenty Lquebrachitol structures (I to XX), with and without including explicitly the 60 water molecules in the NMR calculation, are given as Supplementary Material (Figures S1 and S2). MAE and mean absolute percentage error (MAPE) with respect to the experimental NMR data for all twenty EFP optimized structures are given as Supplementary Material (Table S2), along with B3LYP/6311++G(2d,p)-PCM-water relative energy values. The agreement with experimental NMR spectra can be analysed in two distinct ways: (i) a visual comparison of theoretical and experimental 1H NMR spectra profiles reported in Figure 3 and (ii) a straight line fit as shown in Figure 4. It should be clarified that to make easy the comparison with experimental spectrum, the H1 chemical shift was fixed in the experimental value and used as reference for the others hydrogen nuclei. So, both experimental and translated theoretical spectra will be in a comparable scale. For L-quebrachitol, structure VIII gives the 1H NMR spectrum in best accordance to the experimental data in D2O according to Figures 3 and 4. As can be seen from Figure 4 the previously reported 2a optimized gas phase structure7 show a larger deviation from the experimental straight line than the new structure VIII, where only proton H6 shows a sizeable deviation of approx. 0.1 ppm comparing to the corresponding experimental observed 1H NMR profile. The lowest MAE value, with respect to experimental 1H NMR chemical shift (about 0.07 ppm) was obtained for a structure VIII including 60 water molecules surrounding the L-quebrachitol solute. This structure also reproduces very well the experimental 1H NMR pattern yielding the best straight line correlation with experimental data, as shown in Figure 4. Our results reveal that it is possible to reach an ideal agreement with experimental NMR pattern through inclusion of explicit water molecules, representing the solvent effect in the DFT calculation of NMR chemical shifts using the PCM model. The PCM optimized 2a structure alone does not improve the agreement with 1H NMR profile, but in conjunction with ACS Paragon Plus Environment

Page 6 of 18

Page 7 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


inclusion of explicit water molecules the PCM model makes a good contribution for an efficient description of the solvent effect in the calculation of NMR chemical shifts.

Conclusions

In this work, we showed that the use of the EFP model for the geometry optimization in the presence of 60 solvent water molecules leads to large variations in the L-quebrachitol equilibrium structure and consequently in the B3LYP calculated 1H NMR spectra, affecting the agreement with experimental chemical shift data measured in D2O. Our results indicate that a comparison between theoretical and experimental 1H NMR profile, using the least shielded hydrogen atom as a reference, can be considered a better criterion for structural determination than only analysis of the MAE and MAPE values. Summarizing, the B3LYP/6-311+G(2d,p)-PCM-water NMR calculation for the gas phase fully optimized 2a structure provided a satisfactory agreement with experimental 1H NMR spectrum enabling the determination of the L-quebrachitol molecular structure. However, a very good match with experimental spectrum is attained with DFT-PCM calculation with optimized hydrated solute structure, revealing the important role played by solvent effects. Thus, we have shown that due to the high sensitivity of 1H NMR chemical shifts to local chemical molecular environment, an almost perfect match between experimental and theoretical (DFT) spectra can be achieved through an extensive conformational search followed by calculations of NMR chemical shifts taking solvent effect into account properly, what however can be a long and time consuming computational task. Such procedure can be recommended when agreement with experimental NMR chemical shift pattern is not good enough to allow an assignment of the preferred molecular structure. It appears to us that in many cases the DFT/PCM NMR calculations using DFT gas phase optimized geometries would be reasonable enough to enable a fingerprint like comparison between experimental and theoretical NMR spectra to elucidate the preferred conformation. However, our results strongly indicate that changes in the solute structure due to intermolecular interactions with solvent molecules can be fundamental to reach a reliable agreement with experimental 1H NMR spectra measured in solution.

Supporting Information

Table S1. B3LYP optimized ∠HOCC (and methoxy) dihedral angles for L-quebrachitol 2a structure and B3LYP-EFP.60 H2O optimized geometries (I to XX).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table S2. B3LYP optimized ∠HCCC dihedral angles for L-quebrachitol 2a structure and B3LYPEFP.60 H2O optimized geometries (I to XX). Table S3. B3LYP optimized CCCC dihedral angles for L-quebrachitol 2a structure and B3LYPEFP.60 H2O optimized geometries (I to XX).

Table S4. B3LYP/6-311++G(2d,p)-PCM-water NMR chemical shifts deviation with respect to experimental values for L-quebrachitol 2a optimized hydrated structures (I to XX): Mean Absolute Error (MAE), in ppm, and Mean Absolute Percentage Error (MAPE), in %. Results for a gas phase 2a structure and also PCM optimized geometry are also quoted.

Table S5. B3LYP/6-311++G(2d,p)-PCM-water NMR chemical shifts values, along with experimental data, for L-quebrachitol 2a optimized hydrated structures (I to XX). Results for a gas phase 2a structure and also PCM optimized geometry are also quoted.

Table S6. B3LYP/6-311++G(2d,p)-PCM-water NMR chemical shifts values, along with experimental data, for L-quebrachitol 2a optimized hydrated structures (I to XX). Results for a gas phase 2a structure and also PCM optimized geometry are also quoted.

Table S7. Relative energy values (in kcal mol-1) for L-quebrachitol structures.

Figure S1. Experimental (a) and B3LYP/6-311++G(2d,p)-PCM-water (b-m) 1H NMR chemical shifts for L-quebrachitol. No solvent water molecule was explicitly included in the B3LP NMR calculations, but the solute hydrated optimized geometry was used. Only the solute geometrical changes in the presence of explicit water solvent molecules was take into account, with NO water molecules added in the DFT 1H NMR calculations. Figure S2. Experimental (in D2O) (a) and B3LYP/6-311++G(2d,p)-PCM-water (b-g) 1H NMR chemical shifts for L-quebrachitol. In Figs (d-g) 60 solvent water molecules as included in the B3LYP NMR calculations. (b) Str. 2a B3LYP/6-311++G(2d,p) optimized in gas phase from Ref. 7. (c) Str. 2a B3LYP/6-31G(d,p)-PCM optimized structure. (d-g) B3LYP/6-31G(d,p) EFP model with 60 water molecules optimized structures.


Page 8 of 18

Page 9 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Acknowledgments We would like to thank the Brazilian Agencies CNPq, CAPES and FAPEMIG for their continuing support of our laboratories. L. A De Souza thanks CAPES for a Post-Doctoral scholarship.

References (1) Teixeira, M. G.; Soares, C. G. P.; De Assis, J. V.; De Almeida, M. V.; Lopes, J. F.; De Almeida, W. B.; Fernandes, S. A. Experimental and Theoretical NMR Determination of Isoniazid and Sodium p-Sulfonatocalix[n]arenes Inclusion Complexes. Eur. J. Pharm. Sci. 2012, 47, 539-548. (2) De Souza, L. A.; Nogueira, C. A. S.; Lopes, J. F.; Dos Santos, H. F.; De Almeida, W. B. DFT Study of Cisplatin@Carbon Nanohorns Complexes. J. Inorg. Biochem. 2013, 129, 71-83. (3) Teixeira, M. G.; De Assis, J. V.; Soares, C. G. P.; Venâncio, M. F.; Lopes, J. F.; Nascimento Jr, C. S.; Anconi, C. P. A.; Carvalho, G. S. L.; Lourenço, C. S.; De Almeida, M. V. et al. Theoretical and Experimental Study of Inclusion Complexes Formed by Isoniazid and Modified βCyclodextrins: 1H NMR Structural Determination and Antibacterial Activity Evaluation. J. Phys. Chem. B 2014, 118, 81-93. (4) De Souza, L. A.; Nogueira, C. A. S.; Lopes, J. F.; Dos Santos, H. F.; De Almeida, W. B. Theoretical Study of the Formation of Inclusion Complex between Cisplatin and Single-Wall Carbon Nanotube. J. Phys. Chem. C, 2015, 119, 8394-8401. (5) De Sousa, S. M. R.; Guimarães, L.; Ferrari, J. L.; De Almeida, W. B.; Nascimento Jr., C. S. A DFT Investigation on the Host/Guest Inclusion Process of Prilocaine Into β-Cyclodextrin. Chem. Phys. Lett. 2016, 652, 123-129. (6) De Almeida, M. V.; De Assis, J. V.; Couri, M. R. C.; Anconi, C. P. A.; Guerreiro, M. C.; Dos Santos, H. F.; De Almeida, W. B. Experimental and Theoretical Investigation of Epoxide Quebrachitol Derivatives Through Spectroscopic Analysis. Org. Lett. 2010, 12, 5458-5461. (7) De Almeida, M. V.; Couri, M. R. C.; De Assis, J. V.; Anconi, C. P. A.; Dos Santos, H. F.; De Almeida, W. B. 1H NMR Analysis of O-Methyl-Inositol Isomers: a Joint Experimental and Theoretical Study. Magn. Reson. Chem. 2012, 50, 608-614. (8) Bagno, A.; Rastrelli, F.; Saielli, G. Toward the Complete Prediction of the 1H and

13

C NMR

Spectra of Complex Organic Molecules by DFT Methods: Application to Natural Substances. Chem. Eur. J. 2006, 12, 5514-5525. (9) Barone, G.; Gomez-Paloma, L.; Duca, D.; Silvestri, A.; Riccio, R.; Bifulco, G. Structure Validation of Natural Products by Quantum-Mechanical GIAO Calculations of 13C NMR Chemical Shifts. Chem. Eur. J. 2002, 8, 3233-3239. ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(10) Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and Molecules, Oxford University Press: Oxford, 1989. (11) Becke, A. D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648-5652. (12) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti Correlation-Energy Formula Into a Functional of the Electron Density. Phys. Rev. B 1988, 37, 785-789. (13) Hehre, W. J.; Radom, L.; Schleyer, P. V. R.; Pople, J. A. Ab initio Molecular Orbital Theory, Wiley, New York, 1986. (14) Adamovic, I.; Freitag, M. A.; Gordon, M. S. Density Functional Theory Based Effective Fragment Potential Method. J. Chem. Phys. 2003, 118, 6725-6732. (15) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su, S. et al. General Atomic and Molecular Electronic Structure System. J. Comput. Chem. 1993, 14,1347-1363. (16) Da Silva, J. C. S.; Rocha, W. R. C-H Bond Activation of Methane in Aqueous Solution: a Hybrid Quantum Mechanical/Effective Fragment Potential Study. J. Comput. Chem. 2011, 32, 3383-3392. (17) Coutinho, K.; Canuto, S. DICE: A Monte Carlo Program for Liquid Simulation; University of São Paulo, Brazil, 1997. (18) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A. et al. Gaussian, Inc., Wallingford CT, 2009. (19) Wolinski, K.; Hilton, J. F.; Pulay, P. Efficient Implementation of the Gauge-Independent Atomic Orbital Method for NMR Chemical Shift Calculations. J. Am. Chem. Soc. 1990, 112, 82518260. (20) Cances, E.; Mennucci, B.; Tomasi, J. Evaluation of Solvent Effects in Isotropic and Anisotropic Dielectrics, and in Ionic Solutions with a Unified Integral Equation Method: Theoretical Bases, Computational Implementation and Numerical Applications. J. Chem. Phys. 1997, 107, 3032-3041. (21) Day, P. N.; Jensen, J. H.; Gordon, M. S.; Webb, S. B.; Stevens, W. J.; Krauss, M.; Garmer, D.; Basch, H.; Cohen, D. An Effective Fragment Method for Modeling Solvent Effects in Quantum Mechanical Calculations. J. Chem. Phys. 1996, 105, 1968-1986. (22) Gordon, M. S.; Freitag, M. A.; Bandyopadhyay, P.; Kairys, V.; Jensen, J. H.; Stevens, W. J. The Effective Fragment Potential Method: A QM-Based MM Approach to Modeling Environmental Effects in Chemistry. J. Phys. Chem. A 2001, 105, 293-303.


Page 10 of 18

Page 11 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(23) Aguilar, C. M.; Rocha, W. R. Ligand Exchange Reaction Involving Ru(III) Compounds in Aqueous Solution: A Hybrid Quantum Mechanical/Effective Fragment Potential Study. J. Phys. Chem. B 2011, 115, 2030-2037. (24) Chagas, M. A.; Rocha, W. R. Solvent Effects on the Metal-to-Ligand Charge Transfer Transition of the Complex [Ru(NH3)5(Pyrazine)]2+. Chem. Phys. Lett. 2014, 612, 78-83. (25) Bagno, A. Complete Prediction of the 1H NMR Spectrum of Organic Molecules by DFT Calculations of Chemical Shifts and Spin–Spin Coupling Constants. Chem. Eur. J. 2001, 7, 16521661. (26) Grimblat, N.; Sarotti, A. M. Computational Chemistry to the Rescue: Modern Toolboxes for the Assignment of Complex Molecules by GIAO NMR Calculations. Chem. Eur. J. 2016, 22, 12246-12261. (27) Navarro-Vázquez, A. State of the Art and Perspectives in the Application of Quantum Chemical Prediction of 1H and 13C Chemical Shifts and Scalar Couplings for Structural Elucidation of Organic Compounds. Magn. Reson. Chem. 2017, 55, 29-32. (28) London, F.; Théorie Quantique des Courants Interatomiques Dans les Combinaisons Aromatiques. J. Phys. Radium 1937, 8, 397-409. (29) Freitag, M. A.; Hillman, B.; Agrawal, A.; Gordon, M. S. Predicting Shielding Constants in Solution Using Gauge Invariant Atomic Orbital Theory and the Effective Fragment Potential Method. Chem. Phys. 2004, 120, 1197-1202.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

HO

4 HO

OH

OH OMe

6

5 3

2

1

OH

Scheme 1. L-quebrachitol structure.

3

5

4 1

2

6 7

(a)

3

5

4 2

1 6 7

(b)

Figure 1. B3LYP/6-311+G(2d,p) gas phase optimized geometry structure 2a from Ref. [7] (hydrogen atoms relevant for NMR analysis are highlighted) (a) and B3LYP/6-31G(d,p)-EFP optimized geometry for structure VIII (hydrogen atoms relevant for NMR analysis are highlighted) (b). ACS Paragon Plus Environment

Page 12 of 18

Page 13 of 18

70 60 50 40 30 20 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100

Dihedral Angle Deviation (degrees)

H,O,C2,C3 H,O,C3,C4 H,O,C4,C5 H,O,C5,C6 H,O,C6,C1 CH3,O,C1,C2

1

2

3

4

5

6

7

8

Molecular Structure

(a) 70 60 50 40 30 20 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100


H,C1,C2,C3 H,C2,C3,C4 H,C3,C4,C5 H,C4,C5,C6 H,C5,C6,C1 H,C6,C1,C2

1

2

3

4

5

6

7

8

Molecular Structure

(b) 70 60 50 40 30 20 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100

C1,C2,C3,C4 C2.C3.C4.C5 C3,C4,C5,C6 C4,C5,C6,C1 C5,C6,C1,C2 C6,C1,C2,C3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

(c)

2

3

4

5

6

7

8

Molecular Structure

Figure 2. Deviation between ∠HOCC, ∠HCCC and ∠CCCC dihedral angles (a-c, respectively) calculated for VIII and XVII hydrated structures (named Strs. 3-8) and PCM optimized structure (named Str. 2) with respect to the gas phase optimized structure 2a from ref. [7], named Str. 1 (null deviation, horizontal line). The molecular structures are indicated below: ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 18

(1) Str. 2a - gas phase optimized from ref. 7 (2) Str. 2a - PCM optimized (3) Str. VIII - EFP (4) Str. XVII - EFP (5) Str. VIII - EFP re-optimized (supermolecule calculation) (6) Str. XVII - EFP re-optimized (supermolecule calculation) (7) Str. VIII (gas phase) - re-optimized removing the EFP solvent water molecules (8) str. XVII (gas phase) - re-optimized removing the EFP solvent water molecules Expt. data (in D2O) 2 3 4 5,6 7 1

4 HO

OH

HO 3 5.0

4.8

4.6

4.4

1

(a)

4.2

4.0

3.8

3.6

3.4

3.2

3.0

2.8

OH OMe

6

5 2

1

OH

2.6

H NMR Chemical Shifts (ppm)

Str. 2a: Full Opt. PCM W ater 2 36 4 75 1

Str. 2a: Full Opt. Gas Phase 2 36 4 5 71

∆δExp-Aver= 0.11ppm

5.0

4.8

4.6

4.4

(b)

1

4.2

4.0

3.8

3.6

3.4

3.2

3.0

2.8

2.6



5.0

4.8

3

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

4 5 671

2

3

4 5

67

4.8

4.6

4.4

1

(d)

4.2

4.0

3.8

3.6

3.4

3.2

3.0

2.8

2.6


2

5.0

4.8

4.6

4.4

1

(e)

4.2

4.0

3.8

3.6

3.4

3.2

4.8

4.6

4.4

4.2 1

(f)

4.0

3.6

3.4

3.2

3.0

3 2

2.8

2.6

5.0

4.8

6

74

51

5 6 74

4.6

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

1

3 2

5 4 6 7

4.8

4.6

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

3.0

2.8

2.6


(h)

5.0

4.8

4.6

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

4.8

4.6

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

3.0

2

2.8

2.6

6475

643 7 5

1

5.0

4.8

4.6

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

(l)

4.8

4.6

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

3.0

2.8

2.8

2.6

Str. XVII EFP.60H2O Reoptimized with NO water included

1


3.0


2364 7 5

1

∆δExp-Aver= 0.12ppm 5.0

2.6


(k)

Str. VIII EFP.60H2O Reoptimized with NO water included

32

2.8

Str. XVII 60H2O Supermolecule Opt.(NO water included)

6 4375 1


(j)

3.0


(i)


2.6


Str. XVII EFP.60H2O (NO water included)

2

2.8

1


3.0


Str. VIII 60H2O Supermolecule Opt.(NO water included)

Str. VIII EFP.60H2O (NO water included)

2

2.6


(g)


3

2.8

Str. XVII 60H2O Supermolecule Opt.(60 water included)

6 4 7 51

3.8

3.0



2.6


Str. XVII EFP.60H2O (60 water included)

3

2.8

1


3.0


Str. VIII 60H2O Supermolecule Opt.(60 water included)

Str. VIII EFP.60H2O (60 water included)

2

4.6

(c)

2.6


(m)

4.8

4.6

4.4

1

4.2

4.0

3.8

3.6

3.4

3.2

3.0

2.8

2.6


Figure 3. Experimental (in D2O) and theoretical (DFT-PCM) 1H NMR spectra for selected optimized structure for O-methyl-inositol isomer (2a structure from Ref. 7). MAE (Median ACS Paragon Plus Environment

Page 15 of 18

Absolute Deviation) value (in ppm) com respect to experimental data (∆δExp-Aver) is quoted. The molecular structures are indicated below: (a) Experimental 1H NMR spectrum (in D2O) from Ref. 7 (b) Str. 2a - gas phase optimized from Ref. 7. (c) Str. 2a - PCM optimized (d) Str. VIII - EFP (e) Str. VIII - EFP re-optimized (supermolecule calculation) (f) Str. XVII - EFP (g) Str. XVII - EFP re-optimized (supermolecule calculation) (h) Str. VIII (gas phase) - single point calculation removing the EFP solvent water molecules (i) Str. VIII (gas phase) - single point calculation removing the 60 water molecules re-optimized (supermolecule calculation) (j) Str. XVII (gas phase) - single point calculation removing the EFP solvent water molecules (k)Str. XVII (gas phase) - single point calculation removing the 60 water molecules re-optimized (supermolecule calculation) (l) Str. VIII (gas phase) - re-optimized removing the EFP solvent water molecules (m) Str. XVII (gas phase) - re-optimized removing the EFP solvent water molecules

3.3

1

Theoretical H NMR Chemical Shift (in ppm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.4 3.5 3.6

(a) Expt (in D2O)

3.7

(b) Str. 2a Gas Phase Opt. (c) Str. 2a PCM Fully Opt. (d) VIII-EFP.60H2O Opt. (e) Str. VIII.60H2O Re-opt. (f) Str. XVII-EFP.60H2O Opt. (g) Str. XVII.60H2ORe-opt. (h) Str. VIII-EFP NO Water Included (j) Str. XVII-EFP NO Water Included (i) Str. VIII.60H2O Re-Opt NO Water (k) Str. XVII.60H2O Re-Opt NO Water (l) Str. VIII-EFP NO Water Re-optimized (m) Str. XVII-EFP NO Water Re-optimized

3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.4

4.3

4.2

4.1

4.0

1

3.9

3.8

3.7

3.6

3.5

3.4

3.3

Expt. (in D2O) H NMR Chemical Shift (in ppm)

Figure 4. A comparative analysis of experimental (in D2O) spectrum and theoretical B3LYP/6311+G(2d,p)-PCM-water 1H NMR calculations for selected structures. The experimental values are indicated by squares on a straight line. (a) Experimental 1H NMR spectrum (in D2O) from Ref. 7 (b) Str. 2a - gas phase optimized from Ref. 7. (c) Str. 2a - PCM optimized (d) Str. VIII - EFP (e) Str. VIII - EFP re-optimized (supermolecule calculation) (f) Str. XVII - EFP (g) Str. XVII - EFP re-optimized (supermolecule calculation) (h) Str. VIII (gas phase) - single point calculation removing the EFP solvent water molecules (i) Str. VIII (gas phase) - single point calculation removing the 60 water molecules re-optimized (supermolecule calculation) ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(j) Str. XVII (gas phase) - single point calculation removing the EFP solvent water molecules (k)Str. XVII (gas phase) - single point calculation removing the 60 water molecules re-optimized (supermolecule calculation) (l) Str. VIII (gas phase) - re-optimized removing the EFP solvent water molecules (m) Str. XVII (gas phase) - re-optimized removing the EFP solvent water molecules


Page 16 of 18

Page 17 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Table 1. B3LYP optimized dihedral angles for L-quebrachitol 2a gas phase structure and EFP selected optimized geometries (structure VIII -Figure 1b). Structures (1) Str. 2a - gas phasea (2) Str. 2a - PCM Optb (3) Str. VIII - EFPc (5) Str. VIII – EFP Re-Optd

∠HOC2C3 ∠HOC3C4 174.7 46.8 170.7 44.4 163.1 83.4 168.6 82.3 ∠HC1C2C3 ∠HC2C3C4 a (1) Str. 2a - gas phase 68.3 170.9 (2) Str. 2a - PCM Optb 69.5 169.2 (3) Str. VIII - EFPc 77.0 175.3 (5) Str. VIII – EFP Re-Optd 76.2 178.1 ∠C1C2C3C4 ∠C2C3C4C5 a (1) Str. 2a - gas phase 51.8 -53.9 (2) Str. 2a - PCM Optb 49.6 -53.0 (3) Str. VIII - EFPc 55.6 -61.8 (5) Str. VIII – EFP Re-Optd 57.8 -64.3 a B3LYP/6-311+G(2d,p) gas phase optimized structure from Ref. [7]. b

∠HOC4C5 151.4 152.8 108.2 116.3 ∠HC3C4C5 -172.4 -172.1 -179.3 178.2 ∠C3C4C5C6 56.0 58.3 54.1 55.1

∠HOC5C6 -175.9 177.0 -112.5 -109.5 ∠HC4C5C6 -64.3 -62.4 -63.8 -64.0 ∠C4C5C6C1 -55.4 -59.4 -42.2 -41.6

∠HOC6C1 -176.3 -171.6 146.6 136.9 ∠HC5C6C1 65.2 60.4 76.6 77.5 ∠C5C6C1C2 53.2 55.2 39.4 38.6

B3LYP/6-311+G(2d,p)-PCM fully optimized 2a structure.

c

This work: B3LYP/6-31G(d,p) optimized geometry in the presence of 60 water molecules (EFP Model).

d

This work: Using B3LYP/6-31G(d,p) fully optimized supermolecule calculation including explicit 60 water molecules.


∠CH3OC1C2 -142.6 -143.8 -117.0 -111.7 ∠HC6C1C2 -66.4 -63.7 -77.0 -79.3 ∠C6C1C2C3 -51.9 -51.0 -45.7 -46.2


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Graphic


Page 18 of 18

Water Solvent Effect on Theoretical Evaluation of ... - ACS Publications

Recommend Documents