Conference PaperPDF Available

LEARNING PHYSICAL PROPERTIES OF ORGANIC COMPOUNDS USING MOLECULAR MODELING

Authors:

Abstract and Figures

During our courses at the Faculty of Chemistry (Universidad Nacional Autónoma de México) to teach chemist's students the fundamentals of chemical engineering, we are faced with the problem of explaining the different variables involved in a process. For instance, one of the most important purification methods used in the petrochemical industry is distillation. If the students can comprehend how molecules interact between them and with other compounds, they can acquire the basic knowledge to understand the way of separating different products using a distillation tower. The use of Molecular Modelling is an extremely powerful tool to bring key concepts to the students in a very simple way, explaining the complex transformations in everyday phenomena.
Content may be subject to copyright.
LEARNING PHYSICAL PROPERTIES OF ORGANIC COMPOUNDS
USING MOLECULAR MODELING
C.A. Rius-Alonso, Y. González Quezada
Universidad Nacional Autónoma de México (MEXICO)
Abstract
During our courses at the Faculty of Chemistry (Universidad Nacional Autónoma de México) to teach
chemist's students the fundamentals of chemical engineering, we are faced with the problem of
explaining the different variables involved in a process. For instance, one of the most important
purification methods used in the petrochemical industry is distillation. If the students can comprehend
how molecules interact between them and with other compounds, they can acquire the basic
knowledge to understand the way of separating different products using a distillation tower.
The use of Molecular Modelling is an extremely powerful tool to bring key concepts to the students in a
very simple way, explaining the complex transformations in everyday phenomena.
Keywords: Distillation, Molecular Modelling, Petrochemical, 3D visualization.
1 INTRODUCTION
The use of computer programs to perform the Molecular Modeling and the simulations of molecule
clusters provides students with new skills of how to use them to understand the important facts
involved in the chemical transformations. Some of the programs used during the course are; Spartan
and Odyssey. With the first program, calculations of the electrons' density, spectroscopic, shape,
reactivity, polarity, etc. of the molecules, can be done with great accuracy. Odyssey can predict the
macroscopic behaviour of many molecules using molecular dynamics, boiling points, solubility, phase
separations, crystallization, etc. We have found that students have problems in understanding how the
chemical reactions and interactions proceeded in the chemical industry. This course is aimed at
fostering the development of increasing the student's interest in Chemistry. Knowing how a chemical
process takes place in the industry, nurtures a sense of wonder among students about the natural
world. One of the aims is to induce the understanding of the physicochemical interactions of the
compound to be used for purification in large scale.
1.1 Boiling Point
The normal boiling point (i.e. the boiling point at 1 atm) is one of the major physicochemical properties
used to characterize and identify a compound. Besides being an indicator for the physical state (liquid
or gas) of a compound, the boiling point also provides an indication of its volatility. The degree of
volatility is particularly important in industrial applications, as the hazardous nature of a chemical is
usually closely related to its volatility [1]. Intermolecular interactions in the liquid state have a strong
influence in the boiling point of a compound. A large number of methods for estimating boiling points
have been devised. In 1896 James Walker [2] review the prediction methods develop up to that time,
and the most used then, was the increase of the Boiling Point (BP) in a homologous series by 19° for
every addition of a CH2. In order to have a better approach he proposed a simple equation:
T=aMb
Where T is the boiling point expressed in the absolute scale °K, M is the molecular weight and a and b
are constants for each homologous series. With this simple equation it was possible to predict within
, the boiling point in some series. No consideration about the different parameters was done in that
time.
Actually, it is known that the physical properties of the compounds are the result of complex interaction
at the molecular level. The BP is an indicator of the strength of the intermolecular forces which binds
the molecules of a compound together. The stronger the forces, the more densely packed the atoms,
and the higher BP.
Developing predictive models for BP based on its molecular structure is very important for the
determination of new compounds. Gharagheizi did an exhaustive literature review of prior models up
Proceedings of EDULEARN15 Conference
6th-8th July 2015, Barcelona, Spain
ISBN: 978-84-606-8243-1
1331
to 2013 [3] reporting more than 30 very complete works on the subject. Most of them correlate several
parameters and descriptors of the molecules, from few, up to hundreds of them. With such a complex
set of descriptors, students find difficult to understand the key important factors involved in the Boiling
Point of substances.
1.1.1 Intermolecular Forces
Intermolecular forces between molecules are responsible for several molecular properties as melting
and boiling points; the stronger the attractive forces between individual molecules the higher the value
of the property. In general, there are four types of these interactions.
Ionic bonds> Hydrogen Bonds> Van der Waals dipole-dipole> Van der Waals dispersion forces
1.1.2 Ionic Bonds
Ionic Bonds are interactions between charged atoms or molecules. Positively charged ions as Li+, Na+,
K+ are called cations. The negative charged atom’s Cl-, Br-, F-, are called anions. The attractive forces
between oppositely charge ion is very strong and increases with charge and decreases as the
distance between the ions is increases. This strong interaction is reflected in their high melting points,
most of these compounds are solid.( Fig.1)
Fig 1 a) Na+Cl- b) Ca+2 Cl-2
1.1.3 Wan der Waals dispersion forces n-Alkanes.
This is one of the weakest intermolecular forces, also called London dispersion forces. They were
reported at the beginning of 1900 [4, 5]. In 1930 F. London from the Insitute für theoretishe Physik of
the Universität Berlin explained the forces from the theoretical point of view [6,7] The existence of
these forces is related with the movement of the electrons, so that, at any instant of time; the
distribution will be distorted, and small dipoles arise. This momentary dipole will affect the electron
distribution in a second molecule nearby. The negative end of the dipole tends to repeal electrons, and
the positive end tends to attract an electron; these dipoles are constantly changing, resulting in a net
attraction between the molecules. The larger the area, the greater the interaction and this will result in
an increase in the BP of the compound In Fig. 2. We represent the charge distribution of different n-
alkanes from C3 to C22. The blue is more positive and red more negative. With better contact area,
the BP is larger. This is an electronic explanation of why we can observe an increase in the BP of the
n-alkanes as the Molecular weight increases. This trend was observed in the XIX century but not a
clear graphical explanation could be given.
1332
Fig. 2 n-Alkanes
We can observe a close agreement of the molecular weight and the surface area with the BP Fig. 3 if
the range of MW is not very large. The correlation factor is excellent with the MW and the Area.
Fig. 3 n-Alkanes C5-C10
Increasing the length of the chain produces fewer interactions between the molecules, mainly because
now they can have different conformations and not always are in a straight way. Fig. 4, even so, the
correlation is good.
Fig. 4 n-Alkanes C3-C22
Polarizability is how molecules can form these instantaneous dipoles, polarizability increases with
atomic size. If we obtained the correlation with the BP, we can have a better agreement. Fig. 5
1333
Fig. 5 Polarizability-BP, Alkanes C3-C22
1.1.4 Branched Alkenes.
One important aspect to consider is how the molecular shape affects the strength of the dispersion
forces. When the molecules are linear can develop larger temporary dipoles due to electron
movement than short branched ones containing the same number of atoms. Long straight molecular
can lie closer together, and more attractions are developed. When the molecules have branched
atoms, the interaction between the molecules is reduced and fewer attractions are observed. Thus a
reduction in the BP is expected.
We will study the heptane (hydrocarbon with C7) it can exist in nine distinct structures, all of them with
different atom connections.Fig. 6
Fig. 6 Heptanes
Even that all of this molecules has the same Molecular Weight they have different BP and there is no
correlation between MW and BP as can be seen in Fig. 7
Fig. 7 Heptanes
1334
If we consider the area, there is a small correlation, taking into account the volume the correlation is
better, the best correlation is with the heat capacity of the molecules Fig. 8. This is indicative that there
are several factors involved in the process; that is why to have a real prediction of the Boling points
different parameters has to be taken into account.
Fig. 8
1.1.5 Heteroatoms
When different atoms are introduced in the molecule, they tend to have diverse electronegativity this
induces a difference of the charge in the bonds, and a dipole-dipole interaction arises, Deby in 1915
gave an explanation to the interaction [8].
For instance, if we have structures with the same Molecular Weight but with distinct type of atoms, the
Boiling Points will be different.
Let's take several compounds with oxygen and carbon Fig. 9, Ether, aldehyde, ketone, and ester.
Fig. 9
Even that all of them have the same molecular weigh there is no correlation between the area and BP.
When the Dipole is considered a very good correlation exist Fig. 10. This is because in this case, the
most important factor is the formation of a dipole moment in the molecule and the larger, it is, the more
interaction is between molecules and then the BP increases.
1335
Fig. 10
If similar compounds as 1-butanol or 2-butanol are introduced the correlation is lost Fig.11.
This is because a new factor is introduced, in this case one of the stronger intermolecular forces, the
Hydrogen Bond. The alcohols. 1-butanol and 2-butanol, has a lower dipole moment but a higher BP.
Fig.11
If we consider just the alcohols, we can find a good correlation of the area with the BP Fig 12 in this
case we are considering only alcohols with C5.
Fig.12
If we introduce C5 alcohols with steric hindrance like 2,2-dimethyl-1-propanol and 2-methyl-2-butanol,
then a new factor is introduced, and the correlation is not very clear. Fig. 13 because now not only,
the polar nature of the alcohols is taking part but also the size of the methyl groups around the OH.
1336
Fig. 13
2 CONCLUSIONS
The use of Molecular Modelling programs allows students the understanding of the physical properties
of the chemical compounds, the scope for the implementation and limitations of a particular
interaction.
ACKNOWLEDGEMENTS
This work was partially supported by a grant from Dirección General de Asuntos del Personal
Académico (DGAPA) Universidad Nacional Autónoma de México PE205313 and IOCD (International
Organization for Chemical Science in Development).
REFERENCES
[1] Katritzky, Alan R.; Lobanov, Victor S.; Karelson, Mati, Normal boiling points for organic
compounds: Correlation and prediction by a quantitative structure - property relationship.
Journal of Chemical Information and Computer Sciences; vol. 38; nb. 1; (1998); p. 28 41
[2] Walker, J. The boiling points of homologous compounds. Part I. Simple and mixed ethers J.
Chem. Soc. 1894, 65, 193
[3] Farhad Gharagheizi et al, Determination of the normal boiling point of chemical compounds.
Fluid Phase Equilibria, 354 (2013) 250-258.
[4] Johannes Diderik van der Waals, Variation of volume and of pressure in mixing, in: Proceedings
of the Royal Netherlands Academy of Arts and Sciences, 1, 1898-1899, Amsterdam, 1899, pp.
179-191
[5] Johannes Diderik van der Waals, The liquid state and the equation of condition, in: Proceedings
of the Royal Netherlands Academy of Arts and Sciences, 6, 1903-1904, Amsterdam, 1904, pp.
123-15
[6] Eisenschitz,R., London F. Über das Verhältnis der van der Waalsschen Kräfte zu den
Homöopolaren Bindugskräften. Zeitschrift fur Physik 60, 491-527(1930)
[7] London F. Zur theorie und Systematik der Molekularkräfte. Zeitschrift fur Physik 63, 245-279,
(1930)
1337
[8] P. Deby Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen, Mathematisch-
Physikalische Klasse (1915) Volume: 1915, page 70-76
1338
ResearchGate has not been able to resolve any citations for this publication.
Article
In this work, the quantitative structure property relationship (QSPR) strategy is applied to predict the normal boiling point (NBP) of pure chemical compounds. In order to propose a comprehensive, reliable, and predictive model, a large dataset of 17,768 pure chemical compounds was exploited. The sequential search mathematical method has been observed to be the only viable search method capable for selection of appropriate model parameters (molecular descriptors) with regard to a data set as large as is used in this study. To develop the model, a three-layer feed forward artificial neural network has been optimized using the Levenberg-Marquardt (LM) optimization strategy. Using this dedicated strategy, satisfactory results were obtained and are quantified by the following statistical parameters: average absolute relative deviations of the predicted properties from existing literature values: 3.2%, and squared correlation coefficient: 0.94.
Article
Die Molekularkrfte zerlegen sich in drei wesentlich verschiedene Bestandteile, von denen zwei, als Keesomscher Richteffekt und Debyescher Induktions-effekt, mutatis mutandis bereits frher untersucht wurden. Hinzu tritt als dritter Bestandteil der Wechselwirkung die kurzperiodische gegenseitige Strung der inneren Elektronenbewegung der Molekle, welcher Beitrag bei den ein-fachsten nichtpolaren und auch noch bei schwach polaren Moleklen den Haupt-bestandteil der Molekularattraktion darstellt. Er macht insbesondere die frher unvermeidliche Annahme einer Quadrupolstruktur der Edelgase entbehrlich. Die rein theoretische Bestimmung der Molekularkrfte, welche als Strungs-effekte zweiter Ordnung zu behandeln sind, ist naturgem kaum auszufhren; dagegen lassen sich die Krfte auf Grund ihres theoretischen Zusammenhanges mit denf-Werten der Dispersionsformel aus optischen Messungen abschtzen. Die so abgeschtzten Krfte ergeben mit derjenigen Genauigkeit, mit der sie selbst sichergestellt sind, das van der Waalssche Attraktionsglied der empirischen Zustandsgieichung.
Article
Es wird ein systematisches Strungsverfahren entwickelt, welches die homopolaren und die van der Waalsschen Kraftwirkungen in einheitlichem Zusammenhang umfat und die frheren Untersuchungen zur chemischen Bindung als erste Nherung enthlt. Das Verfahren wird zur Bestimmung der van der Waalsschen Konstanten des atomaren Wasserstoffs angewandt.
Article
We recently reported a successful correlation of the normal boiling points of 298 organic compounds containing O, N, Cl, and Br with two molecular descriptors.1 In the present study the applicability of these two descriptors for the prediction of boiling points for various other classes of organic compounds was investigated further by employing a diverse data set of 612 organic compounds containing C, H, N, O, S, F, Cl, Br, and I. The data set was divided into 9 subsets, and additional descriptors were sought for each subset, which, together with the gravitation index and the charged surface area of hydrogen-donor atoms, would model the boiling points. The additional descriptors were then each tested for global relevance and retained only if this was found. A final eight-parameter correlation model was thus deduced which had R2 ) 0.965 and a standard error of 15.5 K approaching the estimated average experimental error for the data set. The model appears to be general for a wide variety of organic compounds.