Analysis of non-covalent interactions in the acetone-chloroform system by MP2/CBS quantum chemical calculations

doi:10.21203/rs.3.rs-2493176/v1

We studied the non-covalent interactions present in the acetone-chloroform system at the MP2/Complete Basis Set level of theory. To this end, we did a configurational search for the diverse geometries of (acetone)₂ homodimers, (chloroform)₂ homodimers and acetone-chloroform heterodimers. All the structures we obtained were subjected to full geometry optimization and their vibrational frequencies calculated. The resulting wavefunctions were subjected to Reduced Density Gradient and Non-Covalent Interaction analyses which yielded a variety of critical points. We found that the reported hydrogen bond for the acetone-chloroform dimer likely has a van der Waals component involving interaction between a chlorine in chloroform and a hydrogen in acetone. In addition, we extrapolated energies to the Complete Basis Set limit. The MP2/Complete Basis Set-calculated interaction energies follow the trend (acetone)₂ > acetone-chloroform > (chloroform)₂. It was also found that MP2/cc-pVDZ severely underestimates the interaction energies as well as predicts the wrong trend, when compared to MP2/Complete Basis Set.

interaction energy

azeotrope

weak interaction

hydrogen bond

dimer

The acetone-chloroform system is known for forming an azeotropic mixture. Azeotropic behavior is commonly attributed to a difference in the interaction energies between similar and dissimilar molecules. In the case of our system, hydrogen bonding is a clear candidate as the interaction responsible for formation of a stable species that gives rise to azeotropic behavior. Hydrogen bonding between acetone and chloroform could be due to dimers such as 1 or 2 (or both). Campbell and Kartzmark[1] studied acetone and chloroform in the solid state, assuming species 1, and determined the strength of the hydrogen bond as 2.7 kcal/mol.

Species 1 has been experimentally detected by Kearns[2], by Durov and Shilov[3] and by Domonkos and Ratkovics[4]. After studying IR shifts of the acetone-chloroform azeotrope, Jalilian proposed structure 3[5] for the trimer, that includes an additional interaction between a chlorine atom in chloroform and a hydrogen in acetone. To find out if such an interaction could form in this system we studied the geometries, the energetics and the non-covalent interactions of structure 4.

2.1 Optimized geometries and energetics

After the configurational searches, we found three homodimers of acetone and three homodimers of chloroform (see Fig. 1), and identified structures 5, and 8 as global minima. For the heterodimers, a configurational search uncovered structure 11 (see Fig. 2) as the global minima, Table 1 shows that there is a large energy gap (2.29 kcal/mol) between the global minimum, 5, and the next most stable dimer, 6, so it is expected that both 6 and 7 will be scarcely populated. Among the chloroform dimers, 9, is close in energy (0.23 kcal/mol) to the global minimum, 8, while 10, sits at 2.38 kcal/mol. In the heterodimers as dimers 12 and 13, have energies larger than 3.5 kcal/mol relative to 11, so their populations are estimated to be very low. The geometry of this heterodimer points that geometry 4 is preferable over geometry 1 to describe the acetone-chloroform dimer.

Dimer 5 seems bonded mainly by dispersion interactions. Dimer 8 seems to have a combination of hydrogen bonds, one single, one bifurcated. The single interaction, has a Cl...C distance of 3.66 Å, and a C-H...Cl angle of 127.4°. These values place it in the limits for hydrogen bonding. The bifurcated interaction has C...Cl distances of 3.67 Å and 3.75 Å; and C-H...Cl angles of 130.2° and 140.0°. This could be construed as a need for further optimization, as one would expect perfect symmetry in this dimer, but the potential energy hypersurface seems to be rather shallow in this region, thus, the optimization stops at this point. In dimer 11, there seems to be a hydrogen bond between the carbonylic oxygen and the hydrogen on chloroform. The geometry supports this: C-H...O angle = 167.4°) and the O...C distance = 3.18 Å. A better understanding of all these interactions will have to wait until the section on Non-Covalent Interaction analyses.

Table 1

Energies and zero point corrections calculated at the MP2/cc-pVDZ level for the monomers, homodimers and heterodimers found in the acetone-chloroform system.
Compound	SCF energy /Eh	Corr. energy /Eh	ZPE /Eh	Total energy /Eh
Monomers
Acetone	-191.975553	-0.601614	0.084344	-192.492824
Chloroform	-1416.944359	-0.558694	0.020359	-1417.482694
Acetone homodimers
5	-383.957839	-1.208953	0.170371	-384.996420
6	-383.956714	-1.206290	0.170232	-384.992773
7	-383.955995	-1.205772	0.169931	-384.991836
8	-383.955521	-1.205954	0.169980	-384.991495
9	-383.954716	-1.205941	0.169740	-384.990917
Chloroform homodimers
10	-2833.888787	-1.123447	0.041318	-2834.970916
11	-2833.888084	-1.123643	0.041241	-2834.970486
12	-2833.888914	-1.121636	0.041127	-2834.969423
13	-2833.887020	-1.121174	0.041065	-2834.967129
Heterodimers
14	-1608.926636	-1.164390	0.105646	-1609.985381
15	-1608.926811	-1.163807	0.105758	-1609.984860
16	-1608.924796	-1.164531	0.105481	-1609.983846
17	-1608.918087	-1.165429	0.105351	-1609.978164
18	-1608.920057	-1.162950	0.105065	-1609.977942

Tables 2, 3 and 4 show the energies of all the dimers calculated at higher levels of theory (MP2/cc-pVXZ, X = T, Q, 5). It is worth noting in Table 5 that, increasing the basis set from cc-pVDZ to cc-pVTZ changes the stability of dimer 16, which becomes more stable than dimer 15. Such greater stability is furthered by the calculations at the MP2/cc-pVQZ and MP2/cc-pV5Z levels, where dimer 16 becomes the global minimum.

Table 2

Energies calculated at the MP2/cc-pVTZ level for the monomers, homodimers and heterodimers found in the acetone-chloroform system.
Compound	SCF energy /Eh	Corr. energy /Eh	Total energy /Eh
Acetone	-192.031313	-0.742652	-192.773966
Chloroform	-1417.003193	-0.739249	-1417.742442
Acetone homodimers
5	-384.066505	-1.493314	-385.559819
6	-384.065714	-1.488676	-385.554390
7	-384.065900	-1.488106	-385.554005
8	-384.065406	-1.488238	-385.553644
9	-384.064689	-1.488440	-385.553129
Chloroform homodimers
10	-2834.005139	-1.486367	-2835.491506
11	-2834.004216	-1.486411	-2835.490628
12	-2834.005739	-1.483991	-2835.489730
13	-2834.004121	-1.484338	-2835.488458
Heterodimers
14	-1609.038596	-1.487628	-1610.526225
15	-1609.039198	-1.486280	-1610.525477
16	-1609.037640	-1.488254	-1610.525895
17	-1609.031740	-1.489005	-1610.520745
18	-1609.033266	-1.485946	-1610.519212

Table 3

Energies calculated at the MP2/cc-pVQZ level for the monomers, homodimers and heterodimers found in the acetone-chloroform system.
Compound	SCF energy /Eh	Corr. energy /Eh	Total energy /Eh
Acetone	-192.044999	-0.792675	-192.837673
Chloroform	-1417.016762	-0.808548	-1417.825311
Homodimers acetone
5	-384.092862	-1.593902	-385.686764
6	-384.092353	-1.588772	-385.681125
7	-384.092710	-1.588269	-385.680979
8	-384.092204	-1.588381	-385.680585
9	-384.091619	-1.588594	-385.680213
Homodimers chloroform
10	-2834.031922	-1.625729	-2835.657651
11	-2834.030791	-1.625805	-2835.656596
12	-2834.032558	-1.623225	-2835.655782
13	-2834.030762	-1.623939	-2835.654701
Heterodimers
14	-1609.065243	-1.607287	-1610.672530
15	-1609.065905	-1.605925	-1610.671830
16	-1609.064387	-1.608229	-1610.672616
17	-1609.058468	-1.608896	-1610.667364
18	-1609.060110	-1.605548	-1610.665657

Table 4

Energies calculated at the MP2/cc-pV5Z level for the monomers, homodimers and heterodimers found in the acetone-chloroform system.
Compound	SCF energy /Eh	Corr. energy /Eh	Total energy /Eh
Acetone	-192.048347	-0.812228	-192.860575
Chloroform	-1417.019284	-0.837587	-1417.856872
Acetone homodimers
5	-384.099239	-1.633132	-385.732371
6	-384.098829	-1.627840	-385.726669
7	-384.099232	-1.627385	-385.726617
8	-384.098719	-1.627493	-385.726212
9	-384.098165	-1.627715	-385.725881
Chloroform homodimers
10	-2834.036880	-1.684098	-2835.720978
11	-2834.035703	-1.684173	-2835.719876
12	-2834.037523	-1.681486	-2835.719009
13	-2834.035671	-1.682351	-2835.718022
Heterodimers
14	-1609.070906	-1.655965	-1610.726872
15	-1609.071592	-1.654583	-1610.726175
16	-1609.070105	-1.657089	-1610.727194
17	-1609.064197	-1.657739	-1610.721936
18	-1609.065878	-1.654247	-1610.720126

Table 5

Relative energies (in kcal/mol) calculated at the MP2/cc-pVXZ, X = D, T, Q, 5, level for the homodimers and heterodimers found in the acetone-chloroform system.
Compound	MP2/cc-pVDZ	MP2/cc-pVTZ	MP2/cc-pVQZ	MP2/cc-pV5Z
Acetone homodimers
5	0.00	0.00	0.00	0.00
6	2.29	3.41	3.54	3.58
7	2.88	3.65	3.63	3.61
8	3.09	3.87	3.88	3.86
9	3.45	4.20	4.11	4.07
Chloroform homodimers
10	0.00	0.00	0.00	0.00
11	0.27	0.55	0.66	0.69
12	0.94	1.11	1.17	1.24
13	2.38	1.91	1.85	1.85
Heterodimers
14	0.00	0.00	0.00	0.00
15	0.33	0.47	0.44	0.44
16	0.96	0.21	-0.05	-0.20
17	4.53	3.44	3.24	3.10
18	4.67	4.40	4.31	4.23

2.2 Interaction energies

Table 6 gathers the values of the interaction energies for the lowest-energy homodimers and heterodimers. In this table we observe that: MP2/cc-pVDZ severely underestimates the interaction energies, to the point this model chemistry predicts positive interaction energies for several of these dimers, implying that they are not stable species. In addition, MP2/cc-pVDZ predicts the wrong trend compared to the MP2/CBS values.

Also in Table 6, we observe that in all cases, the strongest interaction energy is present in the global minimum the MP2/CBS interaction energy in 14 is in between the values of homodimers 5 and 10. Our Complete Basis Set extrapolated result differs from the value reported by Campbell[1]: -5.90 vs -2.7 kcal/mol. This discrepancy may be due to the reference point used by Campbell and Kartzmark (the interaction between acetone and carbon tetrachloride) a system which they assumed to behave ideally. This assumption is not warranted, as the non-ideal behavior is evident from the phase diagram[6]. Thompson and Jewell[7] reported that the IR carbonyl peak of acetone shifts 10 cm⁻¹ when dissolved in chloroform, relative to acetone in cyclohexane. Our calculations at the MP2/cc-pVDZ level of theory revealed that the vibration of the carbonyl group shifts from 1790 cm⁻¹ in acetone to 1780 cm⁻¹ in structure 11, in excellent agreement with the experimental results.

Table 6

CBS extrapolated interaction energies, corrected for BSSE, for homodimers and heterodimers.
Compound	MP2/cc-pVDZ /(kcal/mol)	MP2/CBS /(kcal/mol)
Acetone homodimers
5	40.66	-7.08
6	-1.32	-3.44
7	5.89	-3.43
8	-1.37	-3.17
9	-1.19	-2.96
Chloroform homodimers
10	7.18	-4.58
11	-0.98	-3.91
12	0.70	-3.34
13	4.79	-2.83
Heterodimers
14	63.64	-5.90
15	14.92	-5.47
16	5.19	-6.15
17	3.51	-2.85
18	-0.03	-1.67

2.3 Non-covalent interactions

Figure 3 shows the non-covalent interactions found by our calculations. Van der Waals interactions are indicated as light green isosurfaces as well as by (3,−1) critical points. Dimer 5 shows 5 critical points all of (3,−1) nature. The scatter plot shows a spike close to−0.007, indicative of a weak attraction and another spike at + 0.005, which indicates lack of bonding.

Dimer 10 shows 3 critical points, separable in two sets: a solitary point corresponding to the single H...Cl interaction and two corresponding to the bifurcated interaction previously discussed. The scatter plot shows four spikes: two in the negative region between 0 and−0.010 and two others in the positive region between 0 and 0.005. The two spikes in the negative region indicate two types of weak interactions, which mirrors the two sets of (3,−1) critical points. The spikes in the positive region indicate non-bonding interactions, in correspondence to the brown-colored regions of the isosurface. All these are very weak interactions, short from the strength of a hydrogen bond.

Heterodimer 14, shows two (3,−1) critical points, and one (3, + 1) critical point. On the other hand, the scatter plot shows three spikes in the negative region (−0.020 < sign(λ₂)ρ < 0.000) and another in the positive region (0.000 < sign(λ₂)ρ < 0.005). One of the spikes in particular, at the most negative value corresponds to a region on the isosurface with deeper green color containing (also contining a (3,−1) critical point between the carbonylic oxygen and the hydrogen on the chloroform) seems to indicate a hydrogen bond. But the scatter plot of RDG vs sign(λ₂)ρ shows a peak at−0.02, short from the accepted−0.05[8] proper of hydrogen bonding. The other (3,−1) critical point resides on a light green isosurface, indicative of a van der Waals interaction. The brown-colored isosurface around the red (3, + 1) critical point between the chlorine and the oxygen atoms, indicates a very weak, approaching non-bonding, interaction[9]. The (3, + 1) critical point itself reveals a minimum of charge density at that point[10] and could well be that this non-bonding, somewhat repulsive, interaction weakens the quasi-hydrogen bond with the oxygen.

It appears that the hydrogen bonding energy determined by Campbell has a contribution from the van der Waals interaction between the chlorine atom and one of the hydrogens in acetone. There is a previous report of C-Cl as a hydrogen bond acceptor[11], although in a caged environment, which keeps the donor and the acceptor in close proximity.

The critical points found also help us understand the ordering shown by the interaction energies: the interaction between the hydrogen in chloroform and the oxygen in acetone is stronger than the interaction of that same hydrogen with a chlorine atom.

For all the dimers, a configurational search was done building and optimizing several initial geometries to find the lowest energy minimum. All the optimizations and vibrational analyses during the configurational search were done at the MP2/cc-pVDZ level of theory. For extrapolation to the Complete Basis Set limit, single point energies were calculated at the MP2/cc-pVXZ (X = T, Q, 5) levels of theory. Extrapolations to the complete basis set limit were done according to Helgaker[12] and employed the following formulas:

$${E}_{SCF}={a}_{SCF}+{b}_{SCF}{e}^{-cX}$$

1

$${E}_{corr}={a}_{corr}+{b}_{corr}{X}^{-3}$$

2

$${E}_{CBS}={a}_{SCF}+{a}_{corr}$$

3

These extrapolations were carried out for all the components of the BSSE correction, so we could calculate a BSSE-corrected Complete Basis Set energy. NCI analyses followed the method of Johnson et al.[9]. BSSE corrections were done according to Boys and Bernardi[13]. Interaction energies were calculated by

$$\varDelta {E}_{interaction}={E}_{AB}^{AB}\left(AB\right)-{E}_{A}^{A}\left(A\right)-{E}_{B}^{B}\left(B\right)-\left[{E}_{A}^{AB}\left(AB\right)-{E}_{A}^{AB}\left(A\right)-{E}_{B}^{AB}\left(B\right)\right]$$

4

All the structures were built using the facility provided by ChemCompute.org[14],[15]. All optimizations and vibrational analyses were done with GAMESS-US[16],[17],[18] version 30 SEP 2020 (R2) (provided by ChemCompute.org). Single point wavefunctions for NCI analyses were obtained using Orca v.4.2[19],[20]. NCI analyses and critical point determinations were done with Multiwfn v.3.6[21]. Visualizations of molecular structures were generated with Jmol v.14[22]. Visualization of NCI interactions and critical points was done with VMD v.1.9.1[23].

We have found that the interaction energies follow the trend (acetone)₂ > acetone-chloroform > (chloroform)₂ with the energy in 14 being close to the average of the energies of the homodimers. The hydrogen bond between acetone and chloroform is revealed as a composite including two van der Waals interactions, one between a hydrogen on chloroform and the oxygen on acetone, and another between a methyl hydrogen in acetone and a chlorine atom in chloroform. From our results, it appears that 4 is the more adequate to represent the heterodimer.

Acknowledgements

The authors thank Mark Perri for a generous allocation of computer time at ChemCompute.org. This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number ACI-1548562.

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Ethical approval

Not applicable, as no human beings nor animals were subjects in the research.

Competing interests

The authors have no financial or proprietary interests in any material discussed in this article.

Authors’ contributions

Victor M. Rosas-García conceived and designed the study and did some calculations, figures and tables, and most of the RDG and NCI analyses. Cristofer Alanís-Cuevas, did most of the calculations, figures and tables and some RDG and NCI analyses. All authors read and approved the final manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Availability of data and materials

The dataset containing the MP2/cc-pVDZ optimized geometries and their vibrational analyses is available at Figshare: https://doi.org/10.6084/m9.figshare.21867708.v1

Campbell AN, Kartzmark EM (1960) The energy of hydrogen bonding in the system acetone-chloroform. Can J Chem, 38:652–655. https://doi.org/10.1139/v60-094
Kearns ER (1961) Thermodynamic Studies Of The System: Acetone—Chloroform. II. The Relation Of Excess Mixing Functions To Association Complexes, J Phys Chem 65:314–316. https://doi.org/10.1021/j100820a031
Durov VA, Shilov IY (1996) Molecular structure and physicochemical properties of acetone-chloroform mixtures. J Chem Soc Faraday Trans 92:3559–3563. https://doi.org/10.1039/FT9969203559
Domonkos L , Ratkovics F (1985) Determination of the equilibrium constants of association from IR-studies in the acetone-chloroform system. Monatsh Chem 116:437–442. https://doi.org/10.1007/BF00798365
Jalilian MR, Alibabaei L (2005) Spectra and structure of binary azeotropes I. Acetone–chloroform Spectrochim Acta A Mol Biomol 62:322–325 https://doi.org/10.1016/j.saa.2004.12.046
Wyatt WF (1929) Solutions part III. The transition point of carbon tetrachloride, and compounds of carbon tetrachloride or chloroform with acetone, ether, and benzene. Trans Faraday Soc 25:48-53, https://doi.org/10.1039/TF9292500048
Thompson HW, Jewell DJ (1958) Effect of solvent upon vibration bands. Spectrochimica Acta, 13:254–256, https://doi.org/10.1016/0371-1951(58)80085-5
Contreras‐García J, Boto RA, Izquierdo‐Ruiz F, Reva I, Woller T, Alonso M (2016) A benchmark for the non‐covalent interaction (NCI) index or... is it really all in the geometry? Theor Chem Acc 135:242. https://doi.org/10.1007/s00214-016-1977-7
Johnson ER, Keinan S, Mori-Sánchez P, Contreras-García J, Cohen AJ, Yang W (2010) Revealing Non-Covalent Interactions J Am Chem Soc. 132:6498–6506. https://doi.org/10.1021/ja100936w
Bader RFW (1985) Atoms in Molecules. Acc Chem Res 18:9–15. https://doi.org/10.1021/ar00109a003
La Manna P, Talotta C, Floresta G, De Rosa M, Soriente A, Rescifina A, Gaeta C, Neri P (2018) Mild Friedel–Crafts Reactions inside a Hexameric Resorcinarene Capsule: C−Cl Bond Activation through Hydrogen Bonding to Bridging Water Molecules. Angew Chem Int Ed 57:5423–5428 https://doi.org/10.1002/anie.201801642
Helgaker T, Klopper W, Koch H, Noga J (1997) Basis-set convergence of correlated calculations on water. J Chem Phys 106:9639–9646 https://doi.org/10.1063/1.473863
Boys SF, Bernardi F (1970) The calculation of small molecular interactions by the differences of separate total energies. Some procedures with reduced errors. Mol Phys 19:553–566. https://doi.org/10.1080/00268977000101561
Perri MJ, Weber SH (2014) Web-Based Job Submission Interface for the GAMESS Computational Chemistry Program. J Chem Ed 91:2206–2208. https://doi.org/10.1021/ed5004228
Hanson RM, Prilusky J, Renjian Z, Nakane T, Sussman JL (2013) JSmol and the Next-Generation Web-Based Representation of 3D Molecular Structure as Applied to Proteopedia. Isr J Chem 53:207-216. https://doi.org/10.1002/ijch.201300024
Schmidt MW, Baldridge KK, Boatz JA, Elbert ST, Gordon MS, Jensen JH, Koseki S, Matsunaga N, Nguyen KA, Su S, Windus TL, Dupuis M, Montgomery JA (1993) General atomic and molecular electronic structure system J Comput Chem 14:1347–1363. https://doi.org/10.1002/jcc.540141112
Barca GMJ, Bertoni C, Carrington L, Datta D, De Silva N, Deustua JE, Fedorov DG, Gour JR, Gunina AO, Guidez E, Harville T, Irle S, Ivanic J, Kowalski K, Leang SS, Li H, Li W, Lutz JJ, Magoulas I, Mato J, Mironov V, Nakata H, Pham BQ, Piecuch P, Poole D, Pruitt SR, Rendell AP, Roskop LB, Ruedenberg K, Sattasathuchana T, Schmidt MW, Shen J, Slipchenko L, Sosonkina M, Sundriyal V, Tiwari A, Galvez Vallejo JL, Westheimer B, Włoch M, Xu P, Zahariev F, Gordon MS (2020) Recent developments in the general atomic and molecular electronic structure system. J Chem Phys 152:154102. https://doi.org/10.1063/5.0005188
Gordon MS, Schmidt MW (2005) Advances in electronic structure theory: GAMESS a decade later, in: Theory and Applications of Computational Chemistry. The First Forty Years, Elsevier, Amsterdam, pp. 1167–1189. https://doi.org/10.1016/B978-044451719-7/50084-6
Neese F (2012) WIRe: Comput Mol Sci, 2:73–78. https://doi.org/10.1002/wcms.81
Neese F (2017) Software update: the ORCA program system, version 4.0. WIRe: Comp Mol Sci. 8:e1327 https://doi.org/10.1002/wcms.1327
Lu T, Chen F (2012) Multiwfn: A multifunctional wavefunction analyzer. J Comput Chem 33:580–592 https://doi.org/10.1002/jcc.22885
Jmol: an open-source Java viewer for chemical structures in 3D. http://www.jmol.org/ accessed 18 Jan 2023
Humphrey W, Dalke A, Schulten K (1996) VMD - Visual Molecular Dynamics. J Molec Graphics. 14:33-38. https://doi.org/10.1016/0263-7855(96)00018-5

No competing interests reported.

Analysis of non-covalent interactions in the acetone-chloroform system by MP2/CBS quantum chemical calculations

Status:

Version 1

Abstract

Figures

1 Introduction

2 Results

2.1 Optimized geometries and energetics

2.2 Interaction energies

2.3 Non-covalent interactions

3 Computational Methods

4 Conclusions

Declarations

Acknowledgements

Competing interests

Ethical approval

Competing interests

Authors’ contributions

Funding

Availability of data and materials

References

Additional Declarations

Status:

Version 1