Dielectric and conformational studies of methanol + ketone systems
V Madhurima
School of Physics, University of Hyderabad, Hyderabad 500 046 Received 1 November 2004; revised 22 March 2005; accepted 12 May 2005
Semi-empirical conformational analysis of five methanol + ketone systems with a systematic study with the variation of the hydrogen bond angle have been studied. The results of the conformational analysis are compared with experimentally determined dipole moments of the five systems. The ketones chosen for the present study are acetone, 2-butanone, 3- pentanone, cyclohexanone and acetophenone. Conformational analysis is done using the PM3 semi-empirical Hamiltonian and the dipole moments are determined using dielectric relaxation studies at microwave X-band frequency (9.97 GHz).
Further confirmation with previous 13C NMR chemical shift data is also sought.
Keywords: Dielectric studies, Conformational analysis, Dipole moment IPC Code: G01R31/12
1 Introduction
The study of hydrogen bonds, including weak bonds using various physico-chemical tools and computational modelling is well established. These weak interactions are known to alter the physical properties of the complex thus formed, without any chemical changes. That is to say, although various physical properties such as dielectric constant, viscosity, boiling point and absorption spectra of hydrogen bonded complexes undergo changes with respect to their original component; each of the individual component can be re-obtained by fractional distillation at their respective boiling points. Of the many experimental methods mentioned in literature, dielectric relaxation1-3 studies (DRS), ultrasonic interferometry4, infrared spectroscopy5, nuclear magnetic resonance6 and neutron diffraction7 are significant. On the other hand use of computational modelling methods such as molecular dynamics8, Monte Carlo9, semi-empirical10 and ab initio11 techniques are invaluable in the study of structures, bonds and dynamics that are not easily accessible experimentally. Even more advantageous is the use of combination methods12-13 with an experimental method to verify the results obtained. This combination method is particularly useful to determine the geometry of molecular complexes. It is observed from literature that most data available in determining the properties of binary liquid systems are confined to a single experiment, or when an attempt is made to compare the results of two
experiments, data is taken from other research groups, for one of them.
In order to understand the underlying molecular interactions, the binary systems have been studied.
Our previous studies on the binary system of methanol + acetone14 have shown the presence of weak (C-H- -O) hydrogen bond in addition to the main strong (O-H- -O) bond, leading to a ring like conformation for the binary system. Further studies of the re-orientational thermodynamic parameters from dielectric relaxation studies have shown that the methanol molecule undergoes motion about the O-H- -O bond, the motion being restricted by steric hindrance due to the methyl groups of the ketone molecules15. The weak hydrogen bond has energy of the order of kT at room temperature and hence is a dynamic bond, being constantly broken and formed, k is the Boltzmann constant and T is the temperature in kelvin. The conformation of such ring like structure was evaluated by computational semi-empirical and ab-initio considerations.
The assumptions made are that (a) The 1:1 equi- molar system, diluted in non-polar solvent, forms a 1:1 complex. That is, each molecule of ketone interacts with one molecule of methanol (b) since the binary system is dilute in a non-polar solvent, the dipole moment determined experimentally is that of the 1:1 complex (c) no systematic geometrical optimization with respect to the variation of the hydrogen bond angle was performed. That is to say, there is no assurance that the given conformer corresponds to the global minimum.
In this paper an attempt is made to arrive at the geometrically optimized conformation with respect to varying hydrogen bond angle, since it is the only variable, in addition to the hydrogen bond length, if one considers a particular conformation. The bond angle variation is about the C-O- - H bond. The reasoning behind this study is the fact that our previous studies have indicated that the hydrogen bond is likely to be non-linear in methanol + ketone systems. On arriving at the most probable conformation for the system the concurrence of the dipole moment as measured from dielectric studies with that from the computational values as determined from the calculated population analysis is looked at.
2 Experimental Details
All geometric optimizations were performed using the semi-empirical methods using MOPAC-7 software. Geometric optimization was performed for every 10 degrees and the minimum energy conformer is arrived at. Where found necessary, the geometry was optimized for the one parameter and its minimum energy conformer was used as the starting geometry for the next set of optimizations.
The semi-empirical method is found to be sufficient for the present studies since the individual molecular geometries are optimized at higher basis set (typically ab-initio 6-31G) and are used as input for the present studies and the only variation is that of the hydrogen bond angle. Further, no emphasis is given here for absolute values and only the minimum energy structure values are taken. Since two additional experimental methods are used to corroborate these results, semi-empirical methods are found to be adequate.
In order to further ensure the reliability of the semi- empirical methods for the present studies, geometrical optimizations of the chosen six molecules namely methanol, acetone (three conformers namely14, both methyl groups eclipsed, both methyl groups staggered and one eclipsed and one staggered), 2-butanone (two conformers namely16 skew and sync), 3-pentanone, cyclohexanone and acetophenone, all known from our previous studies 17 or literature, were performed. The bond lengths and bond angles thus obtained were compared with those from literature. In addition to these, a similar check was performed for binary systems with respect to the three previously reported conformers of methanol + acetone systems 14. The difference in bond lengths was found to be + 1% and
in bond angles to be + 0.5% in all cases. Semi- empirical studies were performed using the Modified Neglect of Differential Orbitals (MNDO), Modified Intermediate Neglect of Differential Orbitals (MINDO), Austin Model 1(AM1) and Parametric Method 3 (PM3) Hamiltonians and the best correlation was found with those of PM3 in all cases.
Hence PM3 was chosen for all further studies. The dipole moments from the present studies for these systems, along with those determined experimentally, from literature (experiment) and previous semi- empirical PM3 studies are reported in Table 1.
From our previous studies, it is seen that the keto form (both methyl group eclipsed) of acetone is the most favourable configuration. Similarly, it is seen from literature16 that the syn form (alternate alkyl chain, symmetric about C=O) of 2-butanone and 2- pentanone is found to be most stable. Hence, a similar configuration of 2-butanone and 3-pentanone is chosen for the present studies. The geometries of methanol, cyclohexanone and acetophenone are taken from our previous studies.
The experimental methods for DRS and hence the determination of the dipole moment are as published previously6. The dipole moments are determined at X-band frequency (8.97 GHz) using the Higasi19 method for determination of dipole moment. The chemicals used are of Analar grade purity and all chemicals used were distilled prior to use and the mid-fraction was stored over 4Å molecular sieves to avoid moisture contamination.
3 Results and Discussion
The results of the change in total energy with respect to the hydrogen bond angle for the five chosen binary systems namely methanol + (acetone, 2 butanone, 3 pentanone, cyclohexanone and acetophenone) are shown in Figures 1-5 and the corresponding graphical outputs of the respective
Table 1 — Dipole moments of the chosen systems (in debye) Molecule μPM3 μPM3 – ref17 μexpt μexpt-ref18
methanol 1 68 1 87 1 94 1 78
acetone 2 76 3 12 2 85 2 45
2 butanone 2 68 2 70 2 78 2 78
3 pentanone 2 59 2 61 2 75 2 70
cyclohexanone 2 81 2 81 2 90 2 87
acetophenone 2 75 2 76 2 83 3 02
Fig. 1— Total energy versus bond angle and the minimum energy conformation of methanol + acetone
Fig. 2 — Total energy versus bond angle and the minimum energy conformation of methanol + 2-butanone
Fig. 3 — Total energy versus bond angle and the minimum energy conformation of methanol + 3-pentanone
minimum energy conformers are shown in Fig. 1-5.
The dipole moment, hydrogen bond lengths for the minimum energy conformers for each of the systems are given in Tables 2-6.
The results show the presence of a unique minimum energy conformer for all the systems. Since the angle is that of the O-H- -O bond, all the conformers show a marked preference for the oxygen of the OH group to be aligned near the hydrogen of the methyl group of ketone, hence concurring with our previous prediction of ring like conformers for methanol + ketone systems. Methanol + acetone shows a more or less single minimum structure. The
initial low energy configuration corresponds to the rotation of the methanol molecule about the O-H- - O bond. Earlier studies14 have indicated the angle of the H-O- -H (the second hydrogen bond) to be 141 34o. Methanol + 2-butanone has few other lower energy configurations (angles of 110, 240 and 290o) and these correspond to structures wherein the O of the OH group is closer to the methyl group of the ketone molecule or the entire system forms a large O-H- -O bond length, more-or-less linear structure. The minimum energy configuration suggests the possibility of a ring like conformer.
Fig. 4 — Total energy versus bond angle and the minimum energy conformation of methanol + cyclohexanone
Fig. 5 — Total energy versus bond angle and the minimum energy conformation of methanol + acetophenone
In the case of methanol + 3-pentanone, the other minimum energy conformers (angles 100, 230 and 250o) correspond to various positions of the system where the distance between O of the OH group is nearer to the H of the CH3 or making a minimum angle with respect to H of the CH2, hence proving the finite interaction between O of OH and H of
CH2/CH3 groups, thus leading to a ring like conformation. Methanol + cyclohexanone shows a real unique single minimum energy conformation corresponding to the formation of a structure wherein the oxygen of the OH group is closer to the in-plane H of the CH2 group of the carbon closest to the carbonyl carbon of the cyclohexanone molecule.
In methanol + acetophenone, there are two competing mechanisms. Firstly the weak hydrogen bond interaction of the O of the OH group with the H of the methyl group and secondly the repulsion between the oxygen of the OH group and the pi- electron cloud of the benzene like ring of Acetophenone. All other minimum energy conformations (angles 140, 290 and 310o) are at positions where the oxygen molecule is far from the pi-electron cloud.
It is seen that the dipole moment of the minimum energy configurations corresponds to that of the experimental value. The structures seen from figures 1-5 indicate the strong possibility of a ring like conformer for all the systems. Further confirmation of the ring like conformer from the 13C NMR chemical shift data20.
The shift in chemical shift (Δδ) is defined as:
Δδ = δ (Carbon in pure system) -
δ(the same carbon in binary system)
The experimental accuracy of δ is 0.002 ppm. In linear ketones αC refer to the carbonyl carbon (C=O), βC to the carbon next to the αC and γC represents the carbon next to βC. In acetophenone o,m, p refer to the ortho, meta and para positions of the carbon respectively. The Δδ values clearly show the following. The carbon atom of the methyl group of ketones shows a significant shift (of the order of 0.2 ppm) and this is possible only if the methyl group of the ketone participates in a secondary, weak hydrogen bond. The only unclear result here is that of methanol + cyclohexanone wherein the chemical shifts show that all carbons of the ring of the ketone (other than the carbonyl carbon) have an almost equal shift in the chemical shift, meaning that their interaction is almost the same. This could be because of two reasons (a) there is no formation of a second weak hydrogen bond in this system (b) the electron charge density reallocates in the ring to give an almost uniform charge distribution. The former is more likely and
Table 2 — Data for minimum energy conformers System
Methanol +
Hydrogen bond angle of minimum energy configuration (o)
μ (cal) (D)
μ (expt) (D)
Hydrogen Bond Length
(Å)
acetone 240 3 22 3 30 1 8133
2 butanone 180 3 37 3 13 1 8226
3 pentanone 130 3 11 3 02 2 1578
cyclohexanone 190 3 31 3 52 1 8134
acetophenone 230 3 66 3 31 1 8303
Table 3 — 13C Chemical shift data of methanol + ketone systems System
Methanol +
Group δ (ppm) Δδ (ppm) acetone CO
CH3 (M) CH3
208 114 50 055 30 793
1 806 0 395 -0 242
2 butanone CH3
CO CH3 (M)
βCH3
γ CH3
210 133 50 024 370 092
29 411 8 145
1 913 0 364 0 091 -0 107 -0 122
3 pentanone CO
CH3 (M) CH2
CH3
212 288 49 918 35 695 8 024
1 897 0 258 1 060 -0 121 cyclohexanone CO
CH3
oC mC pC
211 226 49 007 41 494 26 983 24 797
1 564 0 653 -0 880 -0 834 -0 956 acetophenone CO
C-CO pC oC mC CH3 (M)
CH3
198 627 137 486 133 524 129 001 128 713 49 888 26 391
1 123 -0 161 0 258 0 076 0 091 0 228 1 230
hence it may be said that despite the proximity of the O of the OH and the H of CH2 groups, it is unlikely that the system forms a ring like conformer. This may be due to the fact that there is no steric hindrance to the methanol molecule’s rotation about the O-H- -O bond thus not giving the system a preferred orientation in order to establish a second weak bond.
In the methanol + acetophenone system, the Δδ of the methyl group is fairly large. This is possible only if this group participates in the bonding. Since both the carbonyl carbon and the methyl carbon of acetophenone show a large Δδ value, it is understood that both these groups participate in hydrogen bonding and thus confirm the presence of a ring like conformer.
4 Conclusions
Minimum energy conformations for methanol + (acetone, 2-butanone, 3-pentanone, cyclohexanone and acetophenone) with a variation of the O-H - -O bond angle are studied. It is found that all the systems show a unique minimum energy conformer, with a few other less energy conformers. The structures of the minimum energy conformers are given here.
Conformational analysis shows the possibility of a ring like conformer for all systems. The accuracy of the minimum energy conformer is sought from the comparison of the experimentally determined dipole moment with that calculated from the population analysis of the semi-empirical PM3 calculation. On further seeking evidence for the ring like conformers for all the systems from 13C NMR chemical shift data, it is seen that all systems except methanol + cyclohexanone show the presence of a ring like structure. The absence of a ring like conformer in methanol + cyclohexanone is attributed to the absence of steric hindrance to the methanol molecule’s rotation about the O-H- - O bond, hence not allowing
the system to be in position for long enough to form a second weak hydrogen bond. This aspect needs further investigation.
References
1 Sengwa R J, Indian J Pure & Appl Phys, 41 (2003) 295.
2 Kalaivani T, Kumar S & Krishnan S, Indian J Pure & Appl Phys, 41(2003) 389.
3 Malathi M, Sabesan R & Krishnan S, J Mol Liq, 109(2004) 11.
4 Awasthi A, Rastogi M & Shukla J P, Fluid Phase Equilibria, 215 (2004) 119.
5 Hawranek J P & Muszyriski, J Mol Struc, 552 (2000) 205.
6 Madhurima V, Sobhanadri J & V R K Murthy, Indian J Pure & Appl Phys, 43A (2004).
7 Tromp R H, Spieser S H & Neilson G W, J Chem Phys, 110 (1998) 2145.
8 Condeiro M A M & Condeiro J M M, J Braz Chem Soc, 15 (2004) 351.
9 Tsuchida E, J Chem Phys, 121 (2004) 4740.
10 Chitra M, Subrahmanyam B & Murthy V R K, Indian J Pure & Appl Phys, 39 (2001) 461.
11 Schulz P W, Leroi G E & Harrison J F, Mol Phys, 88 (1996) 271.
12 Banerjee R, Desiraju G R, Mondal R & Howard J A K, Chem Eur J, 10 (2004) 3373.
13 Radhakrishnamurthy A & Rangacharyulu I, Indian J Pure &
Appl Phys, 42 (2004) 407.
14 Sathyan N, Santhanam V, Madhurima V & Sobhanadri J, J Mol Struc, 342 (1995) 721.
15 Madhurima V, Sobhanadri J, Viswanathan B & Murthy V R K, Communicated to Physics & Chemistry of Liquids.
16 Siam K C, Van Alsenoy C, Klimkowski V J, Ewbank J D et al, J Mol Struc, 110 (1984) 327.
17 Madhurima V, PhD thesis submitted to Indian Institute of Technology, Madras, 1998.
18 Dean J A, Handbook of organic chemistry (McGraw Hill Inc) 1987.
19 Higasi K, Koga Y & Nakamura M, Bull Cem Soc Jpn, 44 (1971) 988.
20 Madhurima V, Moni M S, Sobhanadri J & Murthy V R K, In press J Mol Liq.
