Computational Study of Urea and Its Homologue Glycinamide

Apr 13, 2007 - Revealing the parameters to design the habit modifiers for rock-salt crystals: empirical to ... Journal of Molecular Graphics and Model...
0 downloads 0 Views 240KB Size
Download PDF
5406

Langmuir 2007, 23, 5406-5411

Computational Study of Urea and Its Homologue Glycinamide: Conformations, Rotational Barriers, and Relative Interactions with Sodium Chloride Ajeet Singh, Shampa Chakraborty, and Bishwajit Ganguly* Analytical Science Discipline, Central Salt & Marine Chemicals Research Institute, BhaVnagar, Gujarat-364002, India ReceiVed August 15, 2006. In Final Form: February 6, 2007 Conformational behaviors of urea and glycinamide have been investigated using the B3LYP functional with the 6-311+G* and 6-311+G** basis sets. Urea monomers have nonplanar minima at all the levels studied, even in the aqueous phase. In the case of glycinamide, the intramolecular hydrogen bond formed from the amide to the amine is important for stabilizing the global minimum. Bond rotations and nitrogen inversion barriers for glycinamide conformations have also been reported. The DFT calculated results suggest that urea conformers interact preferentially with the {111} surface of sodium chloride and such interactions can be responsible for the change in the habit of sodium chloride. Glycinamide conformers have a lower affinity toward the {111} surface of sodium chloride in water. The pyramidality of nitrogens in urea conformers does not influence the relative trends of interaction energies with sodium chloride surfaces. The mode of interactions predicted at the LDA/PWC/DND level for urea and glycinamide with sodium chloride for both slab and cluster models shows that the amide functionality (-CONH2) interacts with both Na+ and Cl- ions on the {100} surface; however, the carbonyl oxygen of these additives predominantly interacts with the sodium ions on the {111} surface.

Introduction Urea and its higher homologue glycinamide are among the simplest amides and are of interest in organic, inorganic, and biological chemistry because of their capacity to form complexes.1,2 Urea and glycinamide provide interesting and contrasting examples of how small conformational changes (-CH2-) in molecular structures can have very different effects on crystal habit modification of alkali halides such as sodium chloride. Urea is known to be a habit modifier for sodium chloride, which changes the morphology of NaCl from cubic to octahedral.3 However, the higher homologue glycinamide does not influence the morphology of sodium chloride.4 The application of habit modification to salt technology has been an active area of research for decades. The effects of sodium chloride crystal size and shape on solid processes are far-reaching. They influence the rate at which material can be processed (such as filtering, washing and drying), as well as physical properties such as the bulk density, mechanical strength, and storage and handling characteristics. However, not enough is yet known about the precise relationship between crystals and the growth poison to do very much in the way of tailor-made molecules to produce specified effects. The intelligent application of analogy helps to increase the chance of finding a better material than that already known, but the discovery of active agents remains largely empirical. Therefore, an understanding at the molecular level is required with the known additives, which can further help to design new additives with the desired morphology. In this regard, urea is one of the simplest and most well-known additives for sodium chloride, and hence it becomes an attractive candidate for recent studies. Radenovic´ * Corresponding author. E-mail: [email protected]. (1) Kuharski, R. A.; Rossky, P. J. J. Am. Chem. Soc. 1984, 106, 5786. (2) Lavrich, R. J.; Farrar, J. O.; Tubergen, M. J. J. Phys. Chem. A 1999, 103, 4659. (3) (a) Rome de l’lsle, J. B. L. Cristallographie, 2nd ed.; Imprimerie de Monsieur: Paris, 1783. (b) Fenimori, C. P.; Thrailkill, A. J. Am. Chem. Soc. 1949, 71, 2714. (4) Sarig, S.; Glasner, A.; Epstein, J. A. J. Cryst. Growth 1975, 28, 295.

et al. have experimentally studied the habit change of sodium chloride from cubic to octahedral in further presence of smaller amides.5 The observed results have been rationalized on the basis of charge distributions and strong interaction between the more exposed carbonyl oxygen of amides such as formamide and urea with the sodium ions, which stabilize the {111} surface of sodium chloride and lead to octahedral morphology of sodium chloride.5 In earlier observations, Speidel and Cabrera have also considered that the additive strongly interacts with certain crystal faces to influence the morphology of the crystal.6 Bunn explained that the habit modification of NaCl by urea in aqueous solutions is due to the adsorption of the impurity on certain crystal faces during crystal growth.6c There are many proposals that have been reported to explain the habit of sodium chloride; however, studies on interactions at the molecular level are limited. Therefore, in this article, we have investigated computationally the mode and relative interactions of urea and glycinamide conformers with sodium chloride surfaces. The modeling studies were performed by mimicking the surfaces of sodium chloride with slab models and with the Na9Cl9 cluster, where urea is known to change the habit of sodium chloride from cubic to octahedral,5 but glycinamide is not though it has similar amide (-CONH2) functionality.4 It has been demonstrated by different groups that molecular modeling techniques predict and rationalize the morphology of crystalline solids reasonably well.7 We have employed density functional methods to examine the interaction of urea and glycinamide conformers with the surfaces of sodium chloride in an isolated state and in water. The appearance of (5) (a) Radenovic´, N.; Enckevort, W. van; Verwer, P.; Vlieg, E. Surf. Sci. 2003, 523, 307. (b) Radenovic´, N.; Enckevort, W. van; Vlieg, E. J. Cryst. Growth 2004, 263, 544. (6) (a) Speidel, R. Neues Jahrb. Mineral. 1961, 4, 81. (b) Cabrera, N.; Vermilyea, D. A. Growth and Perfection of Crystals; John Wiley & Sons: 1958; p 393. (c) Bunn, C. W. Proc. R. Soc. London, Ser. A 1933, 141, 567. (7) (a) Coveney, P. C.; Humphries. W. J. Chem. Soc., Faraday Trans. 1996, 92, 831. (b) Coveney, P. C.; Davey, R; Griffin, J. L. W.; He, Y.; Hamlin, J. D.; Stackhouse, S.; Whiting, A. J. Am. Chem Soc. 2000, 122, 11557. (c) Wierzbicki, A.; Cheung, H. S. J. J. Mol. Struct.: THEOCHEM 1998, 287. (d) Coveney, P. V.; Davey, R. J.; Griffinc, J. L. W.; Whiting, A. J. Chem. Soc., Chem. Commun.

10.1021/la062405o CCC: $37.00 © 2007 American Chemical Society Published on Web 04/13/2007

Study of Urea and Its Homologue Glycinamide

octahedral sodium chloride crystals from cubic in the presence of urea indicates that the {111} face of sodium chloride survives because of preferential adsorption of urea on {111} rather than that of the {100} face of sodium chloride. In other words, the preferential interaction of urea with only Na+ present on the top surface of {111} would retard the growth of this face. Such retardation in the growth of the {111} face would lead to a change in the habit of sodium chloride from cubic to octahedron.5,8 Hence, the difference in adsorption energies for urea and glycinamide conformers on [{100} versus {111}] surfaces of NaCl would indicate the morphological change in sodium chloride, as speculated.5,6,8 To model the interactions of urea and glycinamide conformers with sodium chloride, a conformational search was performed for urea and glycinamide at the density functional level. Our calculated results for urea were found to be in agreement with earlier reports,9 but the global minimum (IV) predicted for the glycinamide conformer was not in agreement with Sulzbach et al.’s report.10

Computational Methodology The geometries for the different conformers of urea and glycinamide were fully optimized without any constraints at the B3LYP level with 6-311+G* and 6-311+G**.11 The transitionstate and ground-state geometries were fully characterized by harmonic vibrational frequencies. Furthermore, to evaluate the influence of the basis set, single-point calculations have been performed at the B3LYP/6-311++G** and MP2/6-311++G** levels of theory. Single-point solvent calculations have been performed with the IPCM model unless otherwise stated.12 The interaction study of urea and glycinamide with slabs and cluster models of sodium chloride has been performed by employing density functional program DMol3 in Material Studio (version 3.2, Accelrys Inc.), and the physical wave functions are expanded in terms of numerical basis sets.13 We used the DND double numerical basis set, which is comparable to the 6-31G* basis set. A local spin density approximation with the Perdew-Wang correlational (LDA/PWC) and the generalized gradient-corrected functional (GGA/PW91) has been employed for the adsorption study of urea and glycinamide conformers on the {100} and {111} surfaces of sodium chloride.13,14 Three-dimensional (3D) slabs of sodium chloride [{100} and {111}] were generated employing periodic boundary conditions. The optimized PerdewWang correlational (LDA/PWC) {100} slab model contains seven layers of sodium chloride (112 ions) with a vaccum thickness 1998, 1467. (e) Wierzbicki, A.; Sikes, C. S.; Sallis, J. D.; Madura, J. D.; Stevens, E. D.; Martin. K. L. Calcif. Tissue Int. 1995, 56, 297. (f) Wierzbicki, A.; Sikes, C. S.; Sallis, J. D.; Madura, J. D.; Drakee, B. Calcif. Tissue Int. 1994, 54, 133. (g) Lu, J. J.; Ulrich, J. Cryst. Res. Technol. 2003, 38, 63. (8) Mullin, J. W. Crystallization, 3rd ed.; Butterworth-Heinmann: Boston, 1993; p 248. (9) (a) Ha, T.-K.: Puebla, C. Chem. Phys. 1994, 181, 47. (b) Dixon, D. A.; Matsuzawa, N. J. Phys. Chem. 1994, 98, 3967. (c) Masunov, A.; Dannenberg, J. J. J. Phys. Chem. A 1999, 103, 178. (d) Ishida, T.; Rossky, P. J.; Castner, E. W., Jr. J. Phys. Chem. B 2004, 108, 17583. (e) Åstrand, P.-O.; Wallqvist, A.; Karlstro¨m, G.; Linse, P. J. Chem. Phys. 1991, 95, 8419. (f) Mennucci, B.; Cammi, R.; Cossi, M.; Tomasi, J. J. Mol. Struct.: THEOCHEM 1998, 426, 191. (g) Kontoyianni, M.; Bowen, P. J. Comput. Chem. 1992, 13, 657. (h) Meier, R. J.; Coussens, B. J. Mol. Struct. 1992, 253, 25. (i) Gobbi, A.; Frenking, G. J. Am. Chem. Soc. 1993, 115, 2362. (10) Sulzbach, H. M.; Schleyer, P. v. R.; Schaefer, H. F., III J. Am. Chem. Soc. 1994, 116, 3967. (11) (a) TITAN; Wavefunction, Inc.: Irvine, CA; Schrodinger, Inc.: Portland, OR. (b) Jaguar, 5.5 ed.; Schrodinger, Inc.: Portland, OR, 2004. (c) 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.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993, 14, 1347. (12) Wiberg, K. B.; Rablen, P. R. J. Comput. Chem. 1993, 14, 1504. (13) (a) Delley, B. J. Chem. Phys. 1990, 92, 508. (b) Delley, B. J . Chem. Phys. 1996, 100, 6107. (c) Delley, B. J. Chem. Phys. 2000, 113, 7756. (14) (a) Perdew, J. P.; Wang, Y. Phys. ReV. B 1986, 33, 8800. (b) Perdew, J. P.; Wang, Y. Phys. ReV. B 1992, 45, 13244.

Langmuir, Vol. 23, No. 10, 2007 5407

Figure 1. Relative energies and imaginary frequencies calculated at the B3LYP/-311+G* level for urea conformers I-III.

of 12 Å. The corresponding k points used were medium and the separations were 0.051/Å. For {111}, the optimized slab contains eight layers of sodium and chloride ions (128 ions) with a 14 Å vaccum thickness. To estimate the relaxation effects on additive interaction energies, the top first and second layers of the slabs were relaxed while keeping the bottom layers fixed at the bulk position. The tolerances of the energy, gradient, and displacement convergence were 2 × 10-5 Ha, 4 × 10-3 Ha/Å, and 5 × 10-3 Å, respectively. The maximum gradient for most of the optimized structures was less than 2 × 103 Ha/ Å. Cluster model calculations were performed with two-layer Na9Cl9 constructed by choosing the appropriate distances between Na+ and Cl- ions from the available crystal data. The optimization of additives on the cluster surface of sodium chloride was performed in the gas and solvent phases. The conductor-like screening model (COSMO) has been employed for aqueous-phase calculations.15 Adsorption energies were computed by subtracting the energies of the gas- and solventphase additive molecules and surface from the energy of the adsorption system as shown in eq 1.

Ead ) E(additive/surface) - E(additive) - E(surface)

(1)

Results and Discussion Urea Conformers. It has generally been assumed that urea is a planar symmetrical molecule, and the crystal structure reinforced this assumption.16 However, the theoretical calculations reported on urea showed that urea is nonplanar.9b,g-i We have examined the conformations of urea in the gas and aqueous phases at the B3LYP/6-311+G* level. Figure 1 depicts the three possible conformers Cs I, C2 II and C2V III for urea. The calculated results show that the C2 II conformer of urea is the most stable conformer and a true minimum (no imaginary frequency), in agreement with previous reports (Figure 1).9a-d,17-19 Glycinamide Conformers. The relative structures and energies of urea conformers have been predicted well at the DFT level of theory and have suggested that the methods can be further extended for the conformational search of its homologue glycinamide. The conformational behavior of glycinamide has been investigated at the B3LYP/6-311+G* and B3LYP/6311+G** levels of theory. Four stable conformers IV-VII have been predicted at these levels of theory with C1 symmetry (Figure 2). The calculated results show that conformer IV is the global minimum for glycinamide in the gas and aqueous phases, in agreement with Li et al. study20 (Figure 2). Furthermore, we have also examined the rotational barriers for glycinamide (15) (a) Klamt, A.; Schu¨u¨rmann, G. J. Chem. Soc., Perkin Trans. 2 1993, 799. (b) Tomais, J.; Persico, M. Chem. ReV. 1994, 94, 2027. (16) Worsham, J. E., Jr.; Levy, H. A.; Peterson, S. W. Acta Crystallogr. 1957, 10, 319. (17) Godfrey, P. D.; Brown, R. D.; Hunter, A. N. J. Mol. Struct. 1997, 413, 405. (18) Brown, R. D.; Godfrey, D.; Storey, J. J. Mol. Spectrosc. 1975, 58, 445. (19) Gilkerson, W.; Srivastava, K. J. Phys. Chem. 1960, 64, 1485. (20) (a) Li, P.; Bu, Y.; Ai, H. J. Phys. Chem. A 2003, 107, 6419. (b) Li, P.; Bu, Y.; Ai, H. J. Phys. Chem. B 2004, 108, 1405.

5408 Langmuir, Vol. 23, No. 10, 2007

Singh et al.

Figure 2. Relative energies calculated for stable conformers of glycinamide IV-VII calculated at the B3LYP/6-311+G** level. Solvent calculated results are in parentheses. Hydrogen bond distances are shown by dashed lines (---).

conformers, which suggest that conformers are separated by ∼6.0 kcal/mol free energies of activation and hence they may not easily interconvert at room temperature (Supporting Information). Interaction of Urea and Glycinamide with Sodium Chloride. By identifying the most stable forms of urea and glycinamide conformers, the possible interaction of these conformers has been modeled with the 3D slabs and cluster models of sodium chloride. For simulations of surfaces of crystalline solids, slab and cluster models are nevertheless by far more popular because they are feasible from a computational point of view.21 To investigate the interactions of urea and glycinamide conformers with the surfaces of sodium chloride, a conventional array of Na+ and Cl- ions has been employed in both slab (constructed using periodic boundary conditions) and two-layered Na9Cl9 cluster models. To examine the reliability of DFT models employed in this study toward the mode of adsorption of additives and their relative binding energies on NaCl surface, water has been considered to be a test case at the GGA/PWC/DND//LDA/PWC/DND level using a Na9Cl9 cluster. It is noteworthy that the water adsorption on the NaCl surface has been well studied in the literature.22-26 Our DFT calculated results have reproduced the experimental binding energy (∼40 kJ/mol) of a water molecule on the {100} surface of a NaCl cluster reasonably well. The calculated binding energy at the GGA/PWC/DND//LDA/PWC/DND level has been found to be 44.0 kJ/mol. Importantly, the relative orientation of the water molecule on the sodium chloride surface was found to be similar to that observed in the infrared study.26 Furthermore, the stability trend predicted for urea and glycinamide conformers at the B3LYP level has been found to be in excellent agreement with the predicted trend at the LDA/PWC/DND level (Supporting Information). The stable {100} surface of sodium chloride was modeled with an alternating arrangement of Na+ and Cl- ions. However, modeling the electrostatically polar {111} surfaces of the sodium chloride crystal structure was considered to be a mystery in surface science because it is difficult to investigate (21) Deak, P. Phys. Status Solidi B 2000, 9, 217. (22) (a) Jug, K.; Geudtner, G. Surf. Sci. 1997, 371, 95. (b) Jug, K.; Geudtner, G.; Bredow, T. J. Mol. Catal. 1993, 82, 171. (23) (a) Allouche, A. Surf. Sci. 1998, 406, 279. (b) Alloche, A. J. Phys. Chem. B 1998, 102, 10223. (24) Pramanik, A.; Kalagi, R. P.; Barge, V. J.; Gadre, S. R. Theor. Chem. Acc. 2005, 114, 129. (25) (a) Nachtigall, P.; Jordan, K. D.; Janda, K. C. J. Chem. Phys. 1991, 95, 8652. (b) Raghavachari, K. J. Chem. Phys. 1986, 84, 5672. (c) Jing, Z.; Whitten, J. L. J. Chem. Phys. 1993, 98, 7466. (d) Wu, C. J.; Ionova, I. V.; Carter, E. A. Surf. Sci. 1993, 295, 64. (e) Radeke, M.; Carter, E. A. Phys. ReV. B 1997, 55, 4649. (f) Shoemaker, J.; Burggraf, L. W.; Gordon, M. S. J. Chem. Phys. 2000, 112, 2994. (g) Steckel, J. A.; Phung, T.; Jordon, K. D.; Nachtigall, P. J. Phys. Chem. B 2001, 105, 4031. (26) Foster, M. C.; Ewing, G. E. J. Chem. Phys. 2000, 112, 6817.

both experimentally and theoretically. Because the bulk structure consists of alternating cationic and anionic sheets stacked along the 〈111〉 directions, the {111} polar surfaces must have a very high divergent electrostatic energy, which makes them theoretically highly unstable.27 It has been shown in earlier studies that the adsorption of additives would be preferred with the positive ions on top of the surface of alkali halides.5 Therefore, in the present study, the polar {111} surface has been modeled with Na+ ions on top of the surface. Recently, Radenovic´ et al. have presented a surface X-ray diffraction determination of the {111} NaCl-liquid interface structure in the presence of aqueous solution and formamide.28 It was determined that the rock salt {111} surface is Na+-terminated for both of the environmental conditions. Furthermore, they have suggested that a quarter to a half monolayer of disordered Cl- ions is located on top of a fully ordered Na+ crystal surface that can lead to the stabilization of the polar surface through the formation of an electrochemical double layer. Furthermore, authors have mentioned that the large changes in morphology for NaCl as a function of additives are not apparent from the interface structures. The interactions of urea II and the glycinamide conformers (IV and VI) with the 3D slabs of sodium chloride have been shown in Figure 3. For glycinamide conformers, the free-energy differences (∆G298) show that conformer IV is slightly preferred (0.8 kcal/mol) over IV. Therefore, we have considered both conformers for interactions with sodium chloride surfaces. Optimizations of these additives on the surface of NaCl have been performed at the LDA/PWC/DND level of theory. Three different calculations have been performed for each conformer to examine the relaxation effects of surfaces on the interaction energies of additives. First, the slab was kept fixed, while optimizing the urea and glycinamide conformers on the {100} and {111} surfaces of sodium chloride. In another set of calculations, the first and second layers of NaCl surfaces were relaxed while optimizing the geometries of additives on the surfaces. In this case, the rest of the bottom layers were kept fixed in the bulk position. The interaction energies calculated for urea and glycinamide conformers with sodium chloride surfaces are shown in Table 1. The computed results suggest that the interactions of II, IV, and VI are energetically preferred with {111} by ∼15.0 kcal/mol over those with the {100} surface of sodium chloride. A larger slab of 294 ions constructed for the {100} surface of sodium chloride as a test case showed a similar binding energy (-22.8 kcal/mol) for the urea conformer II (Table 1). The binding energies calculated for conformers II, IV, and (27) (a) Tasker, P. W. Philos. Mag. A 1975, 39, 119. (b) Tasker, P. W. J. Phys. C: Solid State Phys. 1979, 12, 4977. (28) Radenovic´, N.; Kaminski, D.; Enckevort, W. van .; Graswinckel, S.; Shah, I.; Veld, M. i.; Alga, R.; Vlieg, E. J. Chem. Phys. 2006, 124, 164706.

Study of Urea and Its Homologue Glycinamide

Langmuir, Vol. 23, No. 10, 2007 5409

Figure 3. LDA/PWC/DND calculated geometries of urea II (c, d) and glycinamide conformers IV (e, f) and VI (g, h) with the {100}/{111} surface of sodium chloride in slab models (purple, sodium; green, chlorine; red, oxygen; blue, nitrogen; and white, hydrogen). Table 1. Interaction Energies Calculated at the LDA/PWC/DND Level for Urea II and Glycinamide (IV and VI) Conformers with the {100} and {111} Surfaces of Slabs in kcal/mola II

IV {111} {100}

VI

surface

{100}

{111} {100}

{111}

fixed

-22.6 -36.7 -25.1 -41.2 -26.4 -35.3 (-15.7) (-30.9) (-18.5) (-38.6) (-19.5) (-32.4)

one layer relaxed -23.6 -37.3 -27.4 -39.4 -27.2 -42.6 (-16.7) (-31.7) (-19.9) (-36.2) (-20.4) (-36.3) two layers relaxed -23.5 -35.3 -25.7 -39.4 -27.5 -43.2 (-18.1) (-32.7) (-21.4) (-37.6) (-20.0) (-32.4) a

GGA/PW91/DND calculated results are in parentheses.

VI with the fixed and relaxed surface of sodium chloride are not very different. Importantly, the relative trends are the same in all cases. The modes and orientations of these additives were found to be similar on fixed (Figure 3) and relaxed surfaces (Supporting Information). The single-point calculations performed at the GGA/PW91/ DND level using the optimized geometries at the LDA/PWC/ DND level also predict the preference for the binding of additives with the {111} surface of sodium chloride (Table 1). The amide functionality (-CONH2) of urea and glycinamide conformers interacts with the {100} plane of sodium chloride, whereas the carbonyl group interacts with the {111} plane of sodium chloride (Figure 3). Urea and glycinamide conformers fit in a similar fashion to {100} and {111} of sodium chloride even in relaxed surfaces. It is noteworthy that glycinamide conformers IV and VI are different in terms of the orientation of their functional groups in space. Conformer IV is intramolecularly hydrogen bonded (Figure 2) and hence the amide (-CONH2) group is exposed to interact with the surface of sodium chloride. However, in the case of conformer VI the amino group hydrogens are also free to interact with the surface of NaCl. We have examined the binding energies involving the amide NH2 hydrogen and the amino NH2 hydrogens of VI with the {100} surface of sodium chloride. The computed results suggest that the amide NH2 binds more strongly than that of the amino NH2 by 3.8 kcal/mol. These results indicate that the urea and glycinamide conformers would prefer to interact in a similar fashion with the surfaces of sodium chloride, and that has been reflected in the calculated binding energies. It is noteworthy that additives interact with the sodium chloride surface in water, and hence the role of the medium should also be examined. DMol3 (version 3.2) is not supported for solvent calculations with the 3D periodic surfaces, and thus the adsorption of urea and glycinamide conformers on the sodium chloride surface has been extended with the Na9Cl9 cluster model to examine the role of the solvent in these interaction studies.

Figure 4. {100} and {111} Na9Cl9 cluster models (purple, sodium; and green, chlorine).

The interaction of urea conformers I-III (Figure 1) has been examined with the {100} and {111} surfaces of the Na9Cl9 cluster, which consists of 3 × 3 ions in each layer in a two-layered model (Figure 4). The choice of considering all three urea conformers was to examine the effect of the pyramidality of nitrogen atoms on the interaction energies with the sodium chloride surface. These calculations with the Na9Cl9 cluster were performed simply because they are computationally less intensive than slab models employed in this study. Conformers have been placed at the midpoint of the surface in each case and optimized at the LDA/PWC/DND level. On the basis of our slab model results and earlier reports, the clusters of NaCl have been kept fixed during the optimization of additives on the surface.24,29 The interactions of urea and glycinamide conformers in the aqueous phase have been optimized with the COSMO model at the LDA/PWC/DND level. Single-point calculations have been performed at the GGA/PW91/DND level on LDA/PWC/DND optimized geometries in the gas and aqueous phases to further estimate the interaction energies at gradientcorrected potentials. Urea conformers Cs I, C2 II, and C2V III preserve their symmetry, while interacting with the {100} surface of Na9Cl9 (Figure 5a-c, respectively). The interactions with {111} in the gas-phase calculations show that Cs I converges to C2 II (Figure 5d), whereas in the case of C2V III one of the amino nitrogens deviates from planarity (Figure 5e). Urea conformers prefer to interact with Na+ and Cl- ions on the {100} surface of sodium chloride both in gas- and solvent-phase calculations (Figure 5). The carbonyl group of the urea moiety interacts with the sodium ion, whereas one of the amino hydrogens interacts with the chloride ion on the NaCl surface. However, at the {111} surface, the carbonyl group of urea interacts with the layer of sodium ions. In the solvent-phase calculations, Cs I preserves its symmetry on the {111} surface (Figure 5f), but C2 II tends toward planarity (Figure 5 g). Importantly, in the case of the {111} surface, urea conformers prefer to sit in the middle of lattice points even in aqueous-phase calculations (Figure 5f,g). (29) (a) Shih, H. D.; Jone, F.; Jepin, D. W.; Marais, D. Phys. ReV. Lett. 1976, 37, 162. (b) Julg, A.; Allouche, A. Int. J. Quantum Chem. 1982, 22, 739.

5410 Langmuir, Vol. 23, No. 10, 2007

Singh et al.

Figure 5. LDA/PWC/DND calculated geometries of urea and glycinamide conformers with the {100} and {111} surfaces of the Na9Cl9 cluster (purple, sodium; green, chlorine; red, oxygen; blue, nitrogen; and white, hydrogen). Table 2. Interaction Energies Calculated at the LDA/PWC/DND and GGA/PW91/DND Levels for Urea Conformers (I-III) and Glycinamide Conformers (IV and VI) with the {100} and {111} Surfaces of Na9Cl9 in kcal/mola LDA/PWC/DND

(I) (II) (III) (IV) (VI)

{100}

{111}

-21.6 (-16.3) -21.2 (-16.2) -22.0 (-16.6) -23.3 (-17.9) -23.8 (-17.4)

-39.6 (-22.2) -38.4 (-21.2) -40.8 (-23.1) -27.2 (-13.8) -33.0 (-11.5)b

GGA/PW91/DND

(I) (II) (III) (IV) (VI)

{100}

{111}

-15.1 (-10.6) -15.6 (-10.7) -15.3 (-10.5) -17.6 (-12.7) -17.4 (-11.3)

-26.2 (-18.5) -25.5 (-17.8) -26.8 (-19.5) -26.4 (-12.9) -24.2 (-8.80)

a

Solvent calculations are in parentheses. b Single-point calculation performed using gas-phase-optimized geometry.

Glycinamide conformers IV and VI show similar modes of interactions as observed in the case of urea conformers. The amide functionality of the glycinamide unit interacts with both Na+ and Cl- ions present on the {100} surface of sodium chloride in the gas and aqueous phases (Figure 5h,i). The intramolecular hydrogen-bonding remains intact (2.2 Å) in IV while adsorbing on both the {100} and {111} surfaces (Figure 5h,j). Conformer VI moves in between the lattice points on the {111} surface of sodium chloride in the gas-phase optimization and prefers to interact with more than one Na+ ion that is present on the surface (Figure 5k). However, the solvent optimization of VI on the {111} NaCl surface failed to converge in this conformation. The interaction energies calculated for urea and glycinamide conformers with with the {100} and {111} planes of Na9Cl9 is shown in Table 2. The adsorption of urea conformers I-III shows a clear preference toward the {111} surface of NaCl both in the gas and aqueous phases at the LDA/PWC/DND level of theory. Furthermore, the single-point calculations performed at the GGA/PW91/DND level using LDA/PWC/DND optimized geometries show a preference for the {111} plane for all of the three-urea conformers (Table 2). These results suggest that thtowarde degree of pyramidality of nitrogen atoms of urea conformers does not seem to be important while interacting with

sodium chloride surfaces. Furthermore, the interaction of glycinamide conformer IV with the {100} and {111} surfaces of sodium chloride at the LDA/PWC/DND level showed a preference for the {111} surface in the gas phase. However, in the aqueous phase, the {100} surface is preferred by 4.1 kcal/ mol over that of the {111} plane (Table 2). Glycinamide conformer VI also showed a preference for {100} in the aqueousphase calculations (Table 2). The slab and cluster models predicted similar trends in interactions for urea conformers with sodium chloride surfaces in the gas phase, with the relative energy preference for the {111} surface being reasonably larger in both models. For glycinamide conformers, the slab and cluster models also show similar trend of interactions with sodium chloride surfaces with slight variations that are likely due to differences in the models. The mode of interaction of additives for the {111} surface does vary in cluster and slab models (in the middle vs at the lattice points), affecting the relative energy differences in some cases. In the absence of any experimental evidence toward the modes of adsorption of urea and glycinamide on the sodium chloride surface, it is difficult to compare the mode of adsorption of these additives with the actual state. However, the qualitative agreement of the results of the cluster with more accurate slab model calculations suggests that the difference in the sites of adsorption of these additives (in the middle vs at the lattice points) does not affect the overall trends in energy in these cases. Earlier reports on concerted cluster and slab models in solids have also shown similar concurrence as observed in the present study.25g,30 Moving from the gas phase to the aqueous phase, the trends for urea were found to be different from those for glycinamide, while interacting with the surfaces of sodium chloride. The aqueous-phase calculations performed with cluster models predicted the preference for the adsorption of urea conformers (30) (a) Pogosov, V. V.; Kurbatsky, V. P.; Vasyutin, E. V. Phys. ReV. B 2005, 71, 195410.( b) Ferro, Y.; Marinelli, F.; Allouche, A. J. Chem. Phys. 2002, 116. 8124. (c) lllas, F.; Rubio, J.; Ricart, J. M. Phys. ReV. B 1985, 31, 8068. (d) Batra, I. P.; Bagus, P. S.; Hermann, K. Phys. ReV. Lett. 1984, 52, 1384.

Study of Urea and Its Homologue Glycinamide

Langmuir, Vol. 23, No. 10, 2007 5411

on the {111} surface even in the aqueous phase (Table 2). The relative energy preference for urea conformers to interact with the {111} surface has been found to be reduced (∼7.0 kcal/mol) in the aqueous phase compared to the gas-phase results (∼17.0 kcal/mol). Importantly, these calculated results qualitatively support the explanation that the adsorption of urea on the {111} growing face of sodium chloride can influence the habit of NaCl.5,6c,31 However, calculations of glycinamide conformers IV and VI in the aqueous phase indicate the preferential interaction with the {100} surface of sodium chloride (Table 2). This observation is opposite to the predicted gas-phase behavior of glycinamide conformers. Hence, the calculated results suggest that the affinity of glycinamide conformers toward the stable {100} surface of sodium chloride in the aqueous phase is one of the factors responsible for not influencing the habit of sodium chloride.4 Whereas our DFT calculated results have shown the modes of adsorption of urea and glycinamide conformers on important surfaces of sodium chloride and the role of the medium in the morphological changes in sodium chloride in the presence of these additives, they do not address the relative influences of supersaturation, temperature, pressure on the habit of sodium chloride, which can also be important in actual sytems.8 More studies are in progress to understand further the nature of interactions of impurities with alkali halide surfaces at the molecular level and their relative importance toward the morphological changes in actual systems.

Conclusions

(31) Cooke, E. G. Second Symposium on Salt; The Northern Ohio Geological Society, Inc.: Cleveland, OH, 1961; Vol. II, p 259.

In the present study, we applied DFT methods to examine the urea and glycinamide conformers in the gas and solvent phases for interaction with the important surfaces of sodium chloride. The slab and cluster model calculations show that the amide functionality (-CONH2) interacts with both Na+ and Cl- ions on the {100} surface; however, the carbonyl oxygen of these additives interacts with the sodium ions on the {111} surface. Urea prefers to interact with the {111} surface of sodium chloride in both the gas and aqueous phases, which supports the observation5,6,31 that it can inhibit the growth of that plane resulting the change in habit of NaCl from cubes to octahedrons. However, glycinamide conformers show a preference for the stable {100} surface of sodium chloride in water and hence do not seem to influence the morphology of sodium chloride. Acknowledgment. This work was supported by the Department of Science and Technology, New Delhi, India. We are thankful to the reviewers for their suggestions and comments that have helped us to improve the article. Supporting Information Available: Optimized geometries for conformers, transition states, and complexes of sodium chloride with urea and glycinamide. This material is available free of charge via the Internet at http://pubs.acs.org. LA062405O