Crystallographic and Theoretical Investigation on the Nature and

Publication Date (Web): November 3, 2016. Copyright © 2016 American .... Chalcogen bonding in synthesis, catalysis and design of materials. Kamran T...
0 downloads 0 Views 2MB Size
Communication

Subscriber access provided by RYERSON UNIVERSITY

Crystallographic and Theoretical investigation on the nature and characteristics of Type I C=S…S=C interactions. Rahul Shukla, and Deepak Chopra Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.6b01530 • Publication Date (Web): 03 Nov 2016 Downloaded from http://pubs.acs.org on November 3, 2016

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.

Crystal Growth & Design 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 27

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

Crystal Growth & Design

Crystallographic and Theoretical investigation characteristics of Type I C=S…S=C interactions.

on

the

nature

and

Rahul Shuklaa and Deepak Chopraa* a

Crystallography and Crystal Chemistry Laboratory Department of Chemistry,

Indian Institute of Science Education and Research Bhopal, Bhopal 462066, Madhya Pradesh, India Email: [email protected] Fax: +91-755-6692392 Abstract In this study, we have performed an extensive crystallographic and theoretical analysis to explore the nature and characteristics of C=S…S=C interactions. A Cambridge Structural Database study revealed the abundance of C=S…S=C interactions wherein more than 70% of the crystal structures can be categorized as Type I chalcogen-chalcogen interactions. The binding energies for these contacts range in magnitudes from +2.02 kcal/mol (highly destabilized) to -1.67 kcal/mol (stabilized). Ab initio studies on (X2CS)2 models systems where X=-H,-NH2,-OH,-F,-Cl reveals that C=S…S=C are governed by the presence of negative σholes for X=-NH2, -OH while the presence of a positive electrostatic region on sulphur is observed for the halogen substituted complexes. These interactions are of dispersive nature with electrostatics contributing towards the destabilization in some cases. Introduction During the formation of a covalent bond, the electron density of the interacting atoms gets redistributed depending on the electronic environment and it leads to the anisotropic distribution of the electron density. This anisotropic distribution can lead to the presence of a positive electrostatic region on the surface of the atom, opposite to the covalent bond. This region of the positive electrostatic region is termed as a σ-hole1-2. The σ-hole was first studied in the case of halogens where it was observed that the lone pair (lp) of chlorine, iodine, and bromine forms a belt of negative electrostatic regions leaving a formation of σ-hole in the outer region3-4. As discussed in the recent review5, the anisotropic electron density distribution on the halogen atom was known since 1970s6-8 but it only received the prominence it deserved after the introduction of the term “σ-hole”. Also in some cases, the presence of the σ-hole is described by the presence

1 ACS Paragon Plus Environment

Crystal Growth & Design

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

of a negative electrostatic region whose magnitude is less negative than the surrounding regions1,3. The interaction of this “σ-hole” in halogens with a negative region of an electron donor results in the formation of σ-hole bonding (halogen bonds)2,9-13. Since σ-hole is the result of the anisotropic electronic environment around the covalently bonded atoms, the concept of σhole bonding has been expanded beyond the realm of halogen bonding and has been employed in classifications of other non-covalent interactions such as chalcogen bonds14-20, pnicogen bonds21-25, tetral bonds26-28, aerogen bonding29-31. Chalcogens are an important group of elements because of their important role in chemistry and biology32-37. Chalcogen bonds which involve Group-VI elements (O, S, Se, Te, Po) are defined as noncovalent bonds where the positive electrostatic region on the chalcogen atoms interacts with an electron donor atom14. The donor atom can be the other chalcogen atom38-45 or any other electron rich heteroatom46-52. The role of chalcogen bonding in molecular crystals is also well known53-57. The ability of the chalcogen atom to act as a donor as well as an acceptor atom is of interest to evaluate their role in the formation of non-covalent interactions. Several studies have shown that the chalcogen bonds are stabilized interactions, their stability is on certain occasions comparable to other non-covalent interactions59-62. A detailed investigation of the literature discussing the strength of the chalcogen bonds reveals that the hybridization of the sulphur atom involved in the interactions also plays a very important role. It has been observed that chalcogen bonds involving sp3 hybridized chalcogen atom are mainly governed by electrostatic and exchange energy with relatively less contribution from dispersion46,47,49,51,58. However in the case of an sp2 hybridized sulphur atom, a dispersive contribution can become a more dominant factor than electrostatics14,48. Amongst the chalcogen-chalcogen interactions, the S…S contact has garnered immense attention due to their wide applications such as in organic conductors63, and nanotube formation40. S…S interactions have also been observed to play a very important role in the processes of molecular self-assmeblies40,64-66. One of the earliest studies on S…S contacts revealed that there is a preferred orientation of electrophilic and nucleophilic regions in divalent sulphur atoms67. In another study, a CSD analysis and a theoretical investigation discussed the stabilizing characteristics of S…S contact68. A recent study concluded that S…S prefers parallel orientations69. Again, S…S interactions have been mainly studied for a sp3 hybridized sulphur 2 ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27

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

Crystal Growth & Design

atom while the understanding of S…S contact involving sp2 hybridized sulphur atom is relatively scarce70. In this study we have analyzed the nature and characteristics of C=S…S=C chalcogenchalcogen interactions by screening the Cambridge Structural Database 71-72. In addition to this, we have also performed a topological analysis and binding energy calculations on full scale dimers extracted from the Cambridge Structure Database in addition to analyzing C=S…S=C interactions in (X2CS)2 model systems where X =-H, -NH2, -OH, -F, -Cl by means of the electrostatic potential, energy decomposition analysis, orbital interactions, and an electron density difference mapping. Methodology CSD Search We have first analyzed the Cambridge Structure Database (CSD)71-72 [version 5.37] using the ConQuest73 module [version 1.18] for C=S…S=C interactions. Figure 1 shows the Scheme for the search. All contacts with S…S distance ≤ 4.5Å were involved in the search. The C=S…S angularity was restricted to be greater than 90° in all cases to avoid the presence of any competitive intermolecular interactions.

Figure 1: Scheme of the CSD Search performed for C=S…S=C contacts. In addition to geometrical parameters, we have imposed certain search parameters also. Only structures whose 3D coordinates are determined and are error free were involved in the search 3 ACS Paragon Plus Environment

Crystal Growth & Design

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

query. In addition to this disordered structures, polymers, ionic and powder structures were also excluded from the search. To ensure that we only have high-quality structures, R-factor which represents the agreement between the obtained crystallographic model and the experimental diffraction data was kept at ≤ 0.1for the search. Theoretical Calculations. Geometry optimization on the model complexes was performed at second order Møller-Plesset theory (MP2) and aug-cc-pVTZ basis set using Gaussian0974. Counterpoise-corrected binding energy for all the molecular pairs extracted from CSD was performed at MP2/aug-cc-pVDZ while those of model systems were obtained at MP2/aug-cc-pV(x)Z level (x=D, T, Q) level by taking into account the basis set superposition error (BSSE)75. BSSE corrected binding energies for C=S…S=C contacts present in model systems were also evaluated at CCSD(T) level using aug-cc-pVDZ basis set. Binding energies were further extrapolated to Complete Basis Set limit (CBS)76-77 at MP2 level which is based on the idea that correlation energy is roughly proportional to X-3 for basis sets of the aug-cc-pV(x)Z type. This method is employed in recent studies also78-81. ∆EMP2/CBS = (64 ∆EMP2/aug-cc-pVQZ – 27 ∆EMP2/aug-cc-pVTZ)/37 The variation between MP2 and CCSD(T) limit was addressed by using the following equation. ∆ECCSD(T)/CBS = EMP2/CBS + (∆ECCSD(T)/aug-cc-pVDZ – ∆EMP2/aug-cc-pVDZ). GaussView (Version 5)82 was utilized to plot molecular electrostatic potential maps for the monomers participating in noncovalent interactions. Energy decomposition analysis (EDA) at MP2/aug-cc-pVDZ level was also performed using the LMOEDA module present in GAMESSUS83-84 to obtain the total binding energies of the complexes partitioned into the corresponding electrostatic (Eelec), exchange-repulsion (Eex-rep), polarization (Epol) and dispersion (Edispe) energy components. The basis set for EDA analysis was obtained from the EMSL basis set library85-86. The topological properties such as the electron densities (ρ), Laplacian of the electron density (∇2ρ) at the bond critical point (BCP) were obtained for all the S…S contacts by using AIMALL87 which is based on the Quantum Theory of Atoms in Molecules88. The topological calculations were performed at MP2/aug-cc-pVDZ level for molecular pairs extracted from the crystal structure while for model systems the calculations were performed at MP2/aug-cc-pVTZ level. Natural Bond Orbital (NBO)89-90 analysis was performed using NBO6.091 at B33LYP 4 ACS Paragon Plus Environment

Page 4 of 27

Page 5 of 27

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

Crystal Growth & Design

functional using aug-cc-pVTZ basis set to obtain second-order perturbation energy E(2). To get further insight into the nature of the interaction, the electron density difference maps were obtained for dimers using Multiwfn92 and plotted using VMD93 at MP2/aug-cc-pVTZ level. Results and Discussion CSD search reveals that a total of 1054 crystal structures were present having a total of 1224 unique C=S…S=C intermolecular interactions. Out of these, 1115 interactions (~91%) were present in organic structures while the remaining 109 interactions (~9%) were present in organometallic compounds. The S…S distance ranged from 3.128 Å (REFCODE: THBARB02) to 4.500 Å (REFCODE: HEYJUR) [Figure 2] with a mean value of 3.950Å. Figure 3a shows that the frequency of C=S…S=C contacts increases steadily up to ~4 Å and then decrease afterwards and as a consequence the distance maxima was observed between 3.9-4.2Å range (372 unique interactions).

Figure 2: Shortest and longest C=S…S=C contact observed in the crystal structure.

5 ACS Paragon Plus Environment

Crystal Growth & Design

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

Figure 3: (a) C=S…S=C distance distribution in crystal structures. (b)The angular distribution of θ1. (c) Cone corrected angular distribution of θ1. (d) The angular distribution of θ2. (e) Cone corrected angular distribution of θ2. (f) Scatter plot representing variation of θ1 and θ2. (g) Variation of S…S distance with θ1 (h) Variation of S…S distance with θ2 (i) Torsional distribution of C=S…S=C bond. The analysis of the distribution of C=S…S angle shows that both θ1 and θ2 have frequencies spread across the angularity range with maximum frequencies observed in the region of 140° to 160° [Figure 3b, 3d]. It is also important to note that in case of both θ1 and θ2, frequency of the 6 ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27

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

Crystal Growth & Design

C=S…S=C contact having an angularity of around 180° is relatively less when compared to those observed at other angles. However, the C=S…S angle distribution does not provide a completely true picture and hence a cone-angle correction was employed on both θ1 and θ2 [Figure 3c, 3e]. The cone-corrected angular distribution provides an unbiased distribution and is highly recommended by the scientific community94-96. Cone-corrected angular distribution reveals that C=S…S=C contacts shows preference towards linearity similar to those previously observed for hydrogen bonds97. Figure 3f represents the scatter plot distribution for θ1 and θ2 and it was interesting to note that more than 70% structures were having θ1=θ2, which is similar to halogen bonds and can be termed as Type I chalcogen-chalcogen interactions [Figure 4]. Sub-dividing the angular data further reveals that most of the contacts fall at similar angular ranges which shows that variation between θ1 and θ2 is minimal even if they are not equal in magnitude in most of the cases [Table S1]. The scatterplot of the S…S distance against θ1 and θ2 reveals the absence of short S…S contacts especially at lower angularity [Figure 3g, 3h]. The plots also indicates that most of the C=S…S=C contacts present in crystal structures have S…S distance larger than the sum of the vdW radius (1.80 Å) of two sulphur atoms

98

. An

analysis of C=S…S=C torsional angle (They are dihedral angles, which is defined by 4 points in space ) reveals that the majority of the interactions have a torsion value in the range of |170°| to |180°| [Figure 3i].

Figure 4: Schematic representation of Type I and Type II chalcogen-chalcogen interactions.

7 ACS Paragon Plus Environment

Crystal Growth & Design

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 8 of 27

Since more than 70% of the C=S…S=C interactions observed in the crystal structure falls in the category of Type I chalcogen-chalcogen interactions, it was necessary to evaluate the nature and characteristics of these interactions quantitatively. Hence we performed a topological analysis and binding energies calculations on selected molecular pairs extracted from the crystal structures. We only selected those molecular pairs which exclusively showed the presence of C=S…S=C interactions to avoid the influence of other intermolecular interactions. In total, calculations were performed on 23 molecular pairs [Table 1, Figure S1] expanded across the different ranges in distance and angularity. Table 1: REFCODE, geometrical parameters, topological parameters and binding energies evaluated for the molecular pairs extracted from the database. REFCODE

SUFLEM KUWPEZ CASHOT03 JODHIU GIVHUQ RHODIN01 GIVHEA AZOYUG JODHIU2 XUPFUJ BAFVEJ KAVCAL LAMJAL AKOVOL02 PYRIDS02 SSOXAM02 ZZZGEO04 LIVJEG LIDFOV ETTHUR03 LEZVET SAFGIP KEPLUM

S…S distance (Å)/ ∠C=S…S (°) 3.176/172 3.304/170 3.345/166 3.365/177 3.372/140 3.406/160 3.411/178 3.420/145 3.457/170 3.526/172 3.542/158 3.555/159 3.570/99 3.689/164 3.791/137 3.853/163 3.927/96 4.106/159 4.153/157 4.124/120 4.212/151 4.271/120 4.291/161

BPL (Å)

ρ (e/Å3)

∇2ρ (e/Å3)

3.177 3.305 3.347 3.365 3.390 3.420 3.410 3.425 3.457 3.529 3.560 3.569 3.572 3.692 3.797 3.867 3.930 4.125 4.188 4.131 4.231 4.281 4.303

0.075 0.062 0.057 0.055 0.069 0.055 0.048 0.056 0.047 0.041 0.043 0.043 0.055 0.032 0.033 0.025 0.035 0.016 0.015 0.023 0.014 0.016 0.011

0.968 0.768 0.698 0.677 0.726 0.635 0.602 0.620 0.564 0.487 0.483 0.474 0.494 0.356 0.303 0.261 0.276 0.020 0.147 0.182 0.139 0.132 0.118

8 ACS Paragon Plus Environment

Binding Energies (kcal/mol) 1.82 0.94 1.18 0.89 0.23 -0.80 0.20 1.27 0.53 0.12 1.46 -0.60 -1.67 2.02 1.11 -0.67 -1.16 0.50 1.33 1.89 0.74 -0.23 1.59

Page 9 of 27

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

Crystal Growth & Design

The binding energies of the C=S…S=C interactions present in molecular crystals ranged from 2.02 kcal/mol to -1.67 kcal/mol [Table 1]. Out of 23 dimers extracted from the crystal structures, 17 were observed to destabilize whereas only 6 molecular pairs shows stabilization. This indicate towards the fact that C=S…S=C interactions present in crystal structure can be a consequence of crystal packing also. The most stabilized dimer was observed in case of LAMJAL having a S…S distance of 3.570 Å and directionality of 99° [Table 1, Figure S1]. Also similar to previous studies on halogen-halogen contacts99-101, no correlation was observed between the computed binding energy and S…S distance. ∠C=S…S also did not show any correlation with the binding energies and hence it can be concluded that the magnitude of binding energy in a given dimer is mainly dependent on the anisotropic distribution of the electron density around a sulphur atom in a given molecule rather than the geometrical parameters. The topological analysis performed on all the molecular pairs shows the presence of only one S…S bond critical point (BCP) [Figure S1]. No other type of BCP was observed for any molecular pair under consideration. The magnitude of the bond path length (BPL) ranges from 3.177Å to 4.303Å [Table 1]. It is important to note that although all the magnitudes of bond length and bond path length were similar, the origin of these values are different102. The magnitude of ρ lies in the ranges of 0.075 - 0.011 e/Å3 while the magnitude of the Laplacian of the electron density (∇2ρ) lies in the ranges of 0.118 - 0.968 e/Å5. Both ρ and ∇2ρ in general decreases with increasing BPL and follows an exponential decay which is similar to the trend observed for various other intermolecular interactions99-101 [Figure 5]. It was also interesting to note that ρ does not have any correlation with binding energies and hence the general assumption that higher ρ at the BCP leads to the presence of stronger interactions does not hold good.

9 ACS Paragon Plus Environment

Crystal Growth & Design

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

Figure 5: (a) Variation of ρ with bond path length (BPL). (b) Variation of ∇2ρ with bond path length Since it was clearly evident from the binding energies calculations that the C=S…S=C interactions present in the molecular crystal was stabilized in some molecular pairs while in others it was a mere consequence of crystal packing. Also, the stabilizing or destabilizing nature of the C=S…S=C contact was mainly dependent on the electronic environment around the S 10 ACS Paragon Plus Environment

Page 10 of 27

Page 11 of 27

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

Crystal Growth & Design

atom participating in the interaction in a specific molecular pair rather than S…S distance or C=S…S angularity. Hence, it is of importance to understand the role of different substituent on the electronic environment around the S atom and hence their participation in the formation of C=S…S=C interaction. To understand this, we plotted the molecular electrostatic potential (MESP) maps for X2CS monomers where X =-H, -NH2, -OH, -F, -Cl [Figure 6]. The MESP scale ranged from -19 kcal/mol (red) to 19 kcal/mol(blue).

In case of H2CS, the most

electronegative region on S atom corresponds to the lone pairs of S atom (-20.48 kcal/mol) and there is no presence of σ-hole in this case. In case (NH2)2CS, we observed the presence of a negative σ-hole on the S atom along the C=S where a region of low negative electrostatic potential (-14.70 kcal/mol) was surrounded by a region of high negative electrostatic potential (30.40 kcal/mol). For (OH)2CS, the region of negative σ-hole expands further and having a magnitude of -1.90 kcal/mol. Finally, both F2CS and Cl2CS monomers shows the presence of a positive electrostatic region (σ-hole) on sulphur atom having a magnitude of 13.00 kcal/mol and 11.38 kcal/mol, respectively.

11 ACS Paragon Plus Environment

Crystal Growth & Design

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

Figure 6: Molecular electrostatic potentials of X2CS monomers where a) X =-H; b) –NH2; c) – OH; d) –F; c) -Cl drawn on isosurface of 0.001 au. The scale ranges from -19 kcal/mol (red) to 19 kcal/mol (blue). The next step was to optimize the (X2CS)2 complexes for systematically evaluating the effect of substitution on C=S…S=C interactions. Out of the five complexes we investigated, we observed true minima for X=-NH2,-OH, -F while we were only able to obtain a structure having a saddle point containing exclusively C=S…S=C contact for X=-H,-Cl [Figure 7]. The S…S distance in these complexes ranged from 3.449Å for X =-H to 3.604Å for X=-NH2. The directionality in these complexes ranged from 151° to 180° with the trend being -F