Article

Non-Covalent Interactions in Triglycerides: Vaporisation

Thermodynamics for Quantification of Dispersion Forces
Sergey P. Verevkin 1,2, * and Ruslan N. Nagrimanov 2

1 Department of Physical Chemistry and Faculty of Interdisciplinary Research, Competence Centre CALOR,
University of Rostock, 18059 Rostock, Germany
2 Department of Physical Chemistry, Kazan Federal University, 420008 Kazan, Russia
* Correspondence: [email protected]

Abstract: Qualitatively, the non-covalent interactions are well-known and help to explain many
phenomena in chemistry and biochemistry. Quantitatively, determination of strength this force is
a challenging task. The vaporization enthalpy is a reliable measure not only for the intermolecular
interactions in the liquid phase, but also as the measure of intermolecular non-covalent interactions
in the gas phase for the specific group of compounds, e.g., for the triglycerides. The vaporisation
thermodynamics of four triglycerides were studied by using transpiration method, quartz crystal
microbalance, and thermogravimetric analysis. Vapour pressure–temperature dependences were
used to derive the enthalpies of vaporisation of these very low volatile liquids. Vaporisation enthalpies
of the triglycerides available in the literature were collected and uniformly adjusted to the reference
temperature 298.15 K and validated using structure–property relationships (chain-length dependence,
correlation with retention indices, and correlation with normal boiling points). The consistent sets of
evaluated vaporisation enthalpies for the linear and branched triglycerides were used to develop the
“centerpiece” based group-additivity method for predicting enthalpies of vaporisation of triglycerides.
It has turned out that the family of triglycerides do not obey the group-additivity rules. The reason
for that is that the evaporated in the gas phase triglycerides exhibit intensive non-covalent attractive
Citation: Verevkin, S.P.; Nagrimanov, dispersion interactions strongly dependent on the alkyl-chain length. For the first time the intensity
R.N. Non-Covalent Interactions of the dispersion interactions was quantified for the family of aliphatic linear triglycerides with the
in Triglycerides: Vaporisation chain length from 3 to 18 carbon atoms. The influence of the branching and unsaturation of the alkyl
Thermodynamics for Quantification chains to the strength of the non-covalent interactions was also discussed.
of Dispersion Forces. Thermo 2022, 2,
250–266. https://doi.org/10.3390/ Keywords: non-covalent interaction; vapour pressure measurements; enthalpy of vaporisation;
thermo2030018
structure–property relationships; group-additivity
Academic Editor: Jean-Noël Jaubert

Received: 31 July 2022

Accepted: 22 August 2022
1. Introduction
Published: 30 August 2022
Although non-covalent interactions play a key role in material science, chemistry and
Publisher’s Note: MDPI stays neutral
biochemistry, their interpretation and quantification are still far from being satisfactory. Dis-
with regard to jurisdictional claims in
persion forces are much more difficult to handle and therefore less is known about them, in
published maps and institutional affil-
particularly quantitatively. The dispersion forces are usually related to the attractive part of
iations.
the van der Waals potential [1]. The simplest example to show the importance of dispersion
forces is that they help explain why alkanes become liquid with increasing chain length.
Admittedly, the dispersion interactions are considered “weak”. Because of this, quantifying
Copyright: © 2022 by the authors.
dispersion forces is quite challenging. However, the dispersion interactions increase rapidly
Licensee MDPI, Basel, Switzerland. for larger and larger molecules [2]. In our recent work, very weak dispersion forces were
This article is an open access article quantified in methyl alkanoates with an alkyl chain length of 1 to 18 carbon atoms [3]. The
distributed under the terms and homologous series of triglycerides offers a three-fold increase in size compared to the ester
conditions of the Creative Commons series. We expect a significant increase in the role of dispersion forces in triglycerides and
Attribution (CC BY) license (https:// are looking for thermodynamic tools to quantify these non-covalent interactions.
creativecommons.org/licenses/by/ In this article, we have carefully collected experimental data on vaporisation ther-
4.0/). modynamics for triglycerides that are available in the literature. To ascertain the vapour

Thermo 2022, 2, 250–266. https://doi.org/10.3390/thermo2030018 https://www.mdpi.com/journal/thermo

 role of dispersion forces in triglycerides and are looking for thermodynamic tools to quan-
Thermo 2022, 2
tify these non-covalent interactions. 251
In this article, we have carefully collected experimental data on vaporisation thermo-
dynamics for triglycerides that are available in the literature. To ascertain the vapour pres-
sures and vaporisation enthalpies, complementary vapour pressure measurements were
pressures and vaporisation enthalpies, complementary vapour pressure measurements
carried out on a number of triglycerides (see Figure 1).
were carried out on a number of triglycerides (see Figure 1).

Figure 1. General structures (left) and abbreviation (right) of triglycerides evaluated in this work. The
Figure 1. General structures (left) and abbreviation (right) of triglycerides evaluated in this work.
Thetrivial
trivialnamesnamesof the
of triglycerides such assuch
the triglycerides triacetin, tricaprylin,
as triacetin, tricaprin, tricaprin,
tricaprylin, and tripalmitin
and are somewhat
tripalmitin are
awkwardawkward
somewhat for systematic data analysis.
for systematic dataInanalysis.
our opinion, the opinion,
In our numericalthedescription
numerical of the chain lengths
description of the
attached
chain lengths to attached
the glycerol moiety
to the is more
glycerol convenient.
moiety is moreToconvenient.
keep consistency
To keepwith our previous
consistency withwork
our[4],
pre-
the work
vious designation
[4], theofdesignation
triglyceridesofhas been ascribed
triglycerides hasasbeen
follows triacetin
ascribed as (TG 20 20 triacetin
follows 20 ), tricaprylin
(TG 20(TG
2020),
80 80 80 ), tricaprin
tricaprylin (TG 80808(TG 100 100 10(TG
0), tricaprin 0 ), and
100tripalmitin
100100), and(TG 160 160 160(TG
tripalmitin ). Such a designation
16016 0160). Such aisdesignation
particularly is
useful when the triglyceride has double bonds in the alkyl chains (see
particularly useful when the triglyceride has double bonds in the alkyl chains (see Figure Figure S1). To help readers, theTo
S1).
help readers,
IUPAC andthetheIUPAC and the trivialare
trivial nomenclatures nomenclatures
given for eachare given forineach
triglyceride triglyceride in the
the supplementary supple-
materials.
mentary materials.
g
We have modified the Chickos’s method [5] for calculations of ∆l Cp,m o -values, required

forWe
thehave
temperature
modified adjustments of experimental
the Chickos’s method [5]vaporisation enthalpies.
for calculations of ∆ 𝐶The available
, -values, re-
vaporisation enthalpies of triglycerides were collected, uniformly adjusted
quired for the temperature adjustments of experimental vaporisation enthalpies. to the reference
The
temperature T = 298.15 K and evaluated using the structure–property correlations based on
available vaporisation enthalpies of triglycerides were collected, uniformly adjusted to
the vaporisation enthalpy chain length dependency, retention indices, boiling temperatures,
the reference temperature T = 298.15 K and evaluated using the structure–property corre-
and group additivity. The reliable data sets for the triglycerides with the linear and branched
lations based on the vaporisation enthalpy chain length dependency, retention indices,
chains have been recommended for thermochemical calculations. The recommended
boiling temperatures,
experimental data inand group additivity.
combination The reliable data
with a group-additivity sets for
based the triglycerides
“centerpiece” with
approach
thewere
linear and branched chains have been recommended for thermochemical
used to reveal and quantify dispersion forces in triglycerides. calculations.
The recommended experimental data in combination with a group-additivity based “cen-
2. Materials
terpiece” and Methods
approach were used to reveal and quantify dispersion forces in triglycerides.
The samples of triglycerides triacetin, tricaprylin, tricaprin, and tripalmitin of com-
2. Materials
mercial originand (Sigma-Aldrich,
Methods 99%) were used as received. The degree of purity was
determined using a Hewlett Packard
The samples of triglycerides triacetin, gas chromatograph 5890 Series
tricaprylin, tricaprin, and IItripalmitin
equipped with a
of com-
flame ionization detector. A capillary column SE-30 was used with a column
mercial origin (Sigma-Aldrich, 99%) were used as received. The degree of purity was de- length of 10 m,
an insideusing
termined diameter of 0.32 mm,
a Hewlett and agas
Packard filmchromatograph
thickness of 0.255890µm. The standard
Series temperature
II equipped with a
program of the GC was T = 373 K followed by a heating rate of 0.167 K·s−1 to T = 523 K.
flame ionization detector. A capillary column SE-30 was used with a column length of 10
No impurities (greater than mass fraction 0.001) could be detected in the sample used for
m, an inside diameter of 0.32 mm, and a film thickness of 0.25 μm. The standard temper-
the vapour pressure measurements. Before starting the vapour pressure measurements,
ature program of the GC was T = 373 K followed by a heating rate of 0.167 K·s−1 to T = 523
the sample was preconditioned inside of the set-up to remove traces of water and possible
K. volatile
No impurities (greater than mass fraction 0.001) could be detected in the sample used
impurities.
for the The
vapour
standard molarmeasurements.
pressure Before starting
enthalpies of vaporisation,
g othe vapour pressure measure-
∆l Hm , of triglycerides were derived
ments,
from the
the sample was preconditioned
temperature inside of
dependences of vapour the set-up
pressures to remove
measured usingtraces of water and
the transpiration
possible
method volatile impurities.
[6,7] and the quartz-crystal microbalance (QCM) method [8]. The temperature de-
The standard
pendences molarloss
of the mass enthalpies of vaporisation,
rates measured ∆ 𝐻m o
, of triglycerides
using termogravimetric were[9]derived
analysis (TGA) were
g o
from
used derive ∆l Hmdependences
thetotemperature of vapour
-values of triacetin, pressures
tricaprylin, measuredAusing
and tricaprin. concisethedescription
transpiration
of
the experimental
method [6,7] and the methods and data treatment
quartz-crystal microbalance is given in themethod
(QCM) Supporting
[8]. Information.
The temperature
dependences of the mass loss rates measured using termogravimetric analysis (TGA) [9]
Thermo 2022, 2 252

3. Results and Discussion

3.1. Experimental Vaporisation Thermodynamics of Triglycerides
The original experimental vapour pressures of triglycerides at different temperatures
measured using transpiration method are collected in Table S1. The original experimental
vapour pressures of triglycerides at different temperatures measured using QCM method
are collected in Table S2. The mass-loss rates of triglycerides at different temperatures
measured using TGA method are collected in Table S3. These results were used to derive the
g o
standard molar enthalpies of vaporisation ∆l Hm (Tav ) which are referenced to the average
temperatures Tav . These results are shown in Table 1, column 4. For thermochemical
calculations, the vaporisation enthalpies are used to adjust to the reference temperature
T = 298.15 K according to the Kirchhoff’s equation:
g o og o
∆l Hm (298.15 K) = ∆l Hm ( Tav ) +∆Cp,m × ( Tav − 298.15 K) (1)
g
where the value ∆l Cp,mo o (g) − Co (liq) is the difference between the molar heat
= Cp,m p,m
o o (liq), respectively. The required
capacities of the gaseous Cp,m (g) and the liquid phase Cp,m
g o
∆l Cp,m -values are evaluated in Section 3.2.

g
Table 1. Compilation of available enthalpies of vaporisation ∆l Hm
o of linear triglycerides.

g g
Compound Method a T-Range ∆l Hom (Tav ) ∆l Hom (298.15 K) Ref.
CAS K kJ·mol−1 kJ·mol−1
TG 20 20 20 S 284.2–318.2 82.0 ± 0.5 82.3 ± 0.6 [10]
102-76-1 C 298.15 83.4 ± 2.0 [11]
triacetin C 298.15 (85.7 ± 0.6) [12]
E 439.5–590.2 59.7 ± 0.9 78.7 ± 3.9 [13]
T 320.1–360.9 77.1 ± 0.4 80.8 ± 0.5 [14]
T 300.2–328.2 82.3 ± 0.8 83.6 ± 0.9 [15]
T 318.1–362.9 76.4 ± 0.5 80.2 ± 0.9 Table S1
81.5 ± 0.3 b average
nc 79.3 ± 3.0 Table 4
Jx 80.3 ± 3.0 Table 5
Tb 79.4 ± 3.0 Table 6
TG 30 30 30 T 304.2–337.2 88.1 ± 0.4 90.3 ± 0.5 [15]
139-45-7 BP 403–545 77.1 ± 1.8 94.7 ± 3.9 Table S4
tripropionin 90.4 ± 0.5 b average
nc 88.9 ± 3.0 Table 4
Jx 89.5 ± 3.0 Table 5
Tb 90.2 ± 3.0 Table 6
TG 40 40 40 C 298.15 107.1 ± 5.0 [12]
60-01-5 S 318–364 81.2 ± 3.0 (86.2 ± 3.2) [16,17]
tributyrin ITGA 349.3 99.9 ± 3.5 105.6 ± 3.7 [18]
TGA 323–593 78.4 ± 5.0 94.1 ± 5.9 [19]
NTGA 476.8–584.4 78.4 ± 0.7 105.5 ± 5.5 [20]
NTGA 308.6 83.5 ± 0.4 (84.8 ± 2.5) [21]
ITGA 373 84.9 ± 3.5 93.8 ± 3.9 [22]
T 324.2–354.2 92.2 ± 0.5 97.1 ± 0.6 [15]
BP 394–583 81.4 ± 0.7 102.7 ± 4.3 Table S4
97.5 ± 0.6 b average
nc 98.4 ± 3.0 Table 4
Jx 98.1 ± 3.0 Table 5
Tb 97.4 ± 3.0 Table 6
Thermo 2022, 2 253

Table 1. Cont.

g g
Compound Method a T-Range ∆l Hom (Tav ) ∆l Hom (298.15 K) Ref.
CAS K kJ·mol−1 kJ·mol−1
TG 50 50 50 T 340.0–370.0 97.5 ± 0.5 104.8 ± 0.6 [15]
620-68-8 nc 107.9 ± 3.0 Table 4
tripentanoin Jx 107.5 ± 3.0 Table 5
TG 60 60 60 S 359–410 94.0 ± 3.0 (106.0 ± 3.8) [16,17]
621-70-5 ITGA 386.1 118.7 ± 4.2 (131.1 ± 4.9) [18]
tricapronin TGA 353–653 92.8 ± 5.0 117.4 ± 7.0 [19]
NTGA 511.0–641.3 78.0 ± 2.6 116.8 ± 8.2 [23]
NTGA 349.4 99.9 ± 2.2 (107.1 ± 2.6) [21]
NTGA 519.2–646.4 70.9 ± 5.0 110.4 ± 9.3 [24]
ITGA 373 96.4 ± 4.2 (107.0 ± 4.7) [22]
BP 428–633 84.2 ± 3.3 114.2 ± 5.6 Table S4
114.9 ± 3.6 b average
nc 117.5 ± 3.0 Table 4
Jx 116.9 ± 3.0 Table 5
Tb 115.3 ± 3.0 Table 6
TG 70 70 70 S 401.7–452.2 80.8 ± 1.6 (100.9 ± 4.3) [25]
620-67-7 nc 127.0 ± 3.0 Table 4
triheptanoin Jx 127.2 ± 3.0 Table 5
Tb 126.8 ± 3.0 Table 6
127.0 ± 1.7 b average
TG 80 80 80 S 396–453 115.8 ± 3.0 138.4 ± 5.4 [16,17]
538-23-8 ITGA 411.4 130.0 ± 4.6 150.3 ± 6.1 [18]
tricaprylin TGA 398–623 117.2 ± 5.3 152.6 ± 8.8 [19]
NTGA 551.6–658.5 103.8 ± 9.2 158 ± 14 [23]
NTGA 386.2 118.7 ± 4.7 134.5 ± 5.7 [21]
NTGA 562.2–657.8 104.0 ± 5.0 159 ± 12 [24]
ITGA 373 119.4 ± 4.6 132.8 ± 5.3 [22]
ITGA 398.3–461.6 111.7 ± 0.5 135.0 ± 4.7 Table S3
T 403.2–448.2 109.2 ± 0.5 131.9 ± 1.4 Table S1
134.1 ± 1.2 b average
nc 136.6 ± 3.0 Table 4
Jx 137.4 ± 3.0 Table 5
Tb 136.1 ± 3.0 Table 6
TG 100 100 100 S 437–485 124.5 ± 3.0 157.9 ± 7.3 [16,17]
621-71-6 ITGA 437.9 147.1 ± 5.1 (175.7 ± 7.7) [18]
tridecanoin TGA 443–673 138.6 ± 5.0 (189 ± 11) [19]
NTGA 597.6–692.5 130.3 ± 1.5 (201 ± 14) [20]
NTGA 411.5 130.5 ± 7.0 153.8 ± 8.4 [21]
I-TGA 427.9–491.6 136.2 ± 0.5 169.6 ± 4.7 Table S3
QCM 348.5–367.8 154.5 ± 0.9 166.7 ± 2.6 Table S2
T 418.2–468.2 129.5 ± 0.8 159.0 ± 1.0 Table S1
160.1 ± 0.9 b average
nc 155.6 ± 3.0 Table 4
Jx 156.1 ± 3.0 Table 5
TG 120 120 120 S 458–520 137.4 ± 3.0 (187 ± 10) [16,17]
538-24-9 ITGA 468.7 155.8 ± 5.5 (200 ± 10) [18]
trilaurin NTGA 438.0 147 ± 11 (183 ± 13) [21]
NTGA 615–667 221 ± 10 (310 ± 20) [26]
nc 174.7 ± 3.0 Table 4
Jx 174.9 ± 3.0 Table 5
174.9 ± 2.1 b average
Thermo 2022, 2 254

Table 1. Cont.

g g
Compound Method a T-Range ∆l Hom (Tav ) ∆l Hom (298.15 K) Ref.
CAS K kJ·mol−1 kJ·mol−1
TG 140 140 140 S 458–520 137.0 ± 3.0 196.4 ± 12.1 [16,17]
555-45-3 ITGA 483.1 166.3 ± 5.8 (224 ± 13) [18]
trimyristin NTGA 468.9 155.8 ± 15.9 (209 ± 19) [21]
NTGA 615–660 198 ± 10 (304 ± 23) [26]
nc 193.8 ± 3.0 Table 4
Jx 194.0 ± 3.0 Table 5
194.0 ± 2.1 b average
TG 160 160 160 S 506–572 160.6 ± 3.0 (249 ± 18) [16,17]
555-44-2 ITGA 505.8 174.9 ± 6.1 (252 ± 17) [18]
tripalmitin NTGA 483.2 166 ± 20 (235 ± 25) [21]
NTGA 615–667 474±19 (601 ± 27) [26]
QCM 384.9–433.2 169.5 ± 4.2 210.2 ± 9.2 Table S2
nc 212.9 ± 3.0 Table 4
Jx 213.0 ± 3.0 Table 5
Tb 209.0 ± 3.0 Table 6
211.6 ± 1.7 b average
TG 180 180 180 S 521–588 167.4 ± 3.0 (182.9 ± 4.4) [16,17]
555-43-1 NTGA 505.9 175 ± 23 (267 ± 30) [18]
tristearin NTGA 610–660 221 ± 10 (367 ± 31) [26]
nc 232.0 ± 3.0 Table 4
Jx 232.1 ± 3.0 Table 5
232.1 ± 2.1 b average
a Methods: T = transpiration; S = static method; C = calorimetry; J —from correlation of experimental va-
x
porisation enthalpies with Kovats indices (see text); BP—from experimental boiling temperatures reported at
different pressures compiled from the literature (see Table S4); Tb = from correlation of vaporisation enthalpies
with the normal boiling points; E = ebulliometry; ITGA = isothermal TGA; NTGA = non-isothermal TGA;
QCM = quartz-crystal microbalance. Uncertainties in the temperature adjustment of vaporisation enthalpies, are
estimates and amount to 20% of the total adjustment. b Weighted mean value (the uncertainty was taken as the
weighing factor). Uncertainty of the vaporisation enthalpy is expressed as the expanded uncertainty (0.95 level of
confidence, k = 2). Values in parentheses were not considered. Values highlighted in bold were recommended for
thermochemical calculations.

In this study, we carefully collected and evaluated the available experimental literature
data on vapour pressures of triglycerides with linear alkyl chains (see Table 1) and with
branched alkyl chains (see Table 2). Since in most studies the enthalpies of vaporisation
were not adjusted to the reference temperature or the adjustment was performed in some
g o
other way, we treated the literature results with Equation (1) and calculated ∆l Hm -values
for comparison and evaluation (see Tables 1 and 2, column 5).
g
Table 2. Compilation of available enthalpies of vaporisation ∆l Hm
o of branched triglycerides.

g g
Compound Ma T- Range ∆l Hom (Tav ) ∆l Hom (298.15 K) Ref.
CAS K kJ·mol−1 kJ·mol−1
glycerol triformate T 307.2–333.2 76.6 ± 1.1 78.2 ± 1.2 [15]
32765-69-8 Jx 73.9 ± 3.0 Table 5
GA 74.5 b this work
glycerol tri(2-methylpropanoate) T 329.1–371.2 87.8 ± 0.6 93.6 ± 0.7 [15]
14295-64-8 Jx 93.8 ± 3.0 Table 5
GA 96.0 b this work
glycerol tri(3-methylbutanoate) T 341.0–369.0 96.2 ± 1.2 104.0 ± 1.3 [15]
620-63-3 Jx 102.8 ± 3.0 Table 5
Tb 105.5 ± 3.0 Table 6
GA 104.1 b this work
Thermo 2022, 2 255

Table 2. Cont.

g g
Compound Ma T- Range ∆l Hom (Tav ) ∆l Hom (298.15 K) Ref.
CAS K kJ·mol−1 kJ·mol−1
glycerol tri(2,2-dimethylpropanoate) T 313.4–358.1 85.3 ± 0.8 90.2 ± 0.9 [15]
58006-18-1 Jx 96.1 ± 3.0 Table 5
GA 97.0 b this work
glycerol tribenzoate S 423–476 123.5 ± 3.0 141.4 ± 4.7 [16,17]
614-33-5 GA 150.7 b this work
glycerol trierucate (TG 221 ,221 ,221 ) QCM 433.7–483.7 215.0 ± 4.0 310 ± 19 [2]
2752-99-0 GA 354.4 b this work
a Methods: T = transpiration; S = static method; J —from correlation of experimental vaporisation enthalpies
x
with Kovats’s indices (see text); Tb = from correlation of vaporisation enthalpies with the normal boiling points;
GA = estimated using group-additivity (see text); QCM = quartz-crystal microbalance. Uncertainties in the
temperature adjustment of vaporisation enthalpies, are estimates and amount to 20% of the total adjustment.
b Calculated using increments listed in Table 7.

g
3.2. Adjustment of ∆l Hm
o (T)-Values to the Reference Temperature 298.15 K

In general, the adjustment of the thermodynamic properties to the reference tempera-

ture T = 298.15 K is important for the comparison and the development of the structure–
property relationships. Admittedly [3,27,28], the vaporisation enthalpies have mostly been
reported by authors as referenced to the Tav , and they have not often been adjusted to
g o
a different temperature apparently, due to the ambiguities with the ∆l Cp,m —values re-
quired in Equation (1). This ambiguity was resolved in systematic studies by Chickos
and Acree [5,29] who proposed estimating heat capacity differences using the following
empirical correlation:
g o o
− ∆l Cp,m = 10.58 + 0.26 × Cp,m (liq) (2)
which has been parameterized in general with the available data on the organic compounds
of different classes. From our experience, the parameters of Equation (2) apply successfully
to many classes of organic compounds successfully [3,27,28]. However, in our recent
g o
study on linear aliphatic esters (where the ∆l Cp,m -values were derived from temperature
dependences vapour pressures, see Table S5), we have found that the original coefficients
g o
of Equation (2) provide significantly overestimated ∆l Cp,m -values [3]. In this work we
g
correlated experimental the Cp,m (liq) and the ∆l Cp,m -values for linear aliphatic esters (see
o o

Table S5) and obtained the following empirical equation:

og o
− ∆l Cp,m = 16.4 + 0.1833 × Cp,m (liq) (with R2 = 0.979) (3)

Both empirical coefficients are significantly lower than the original values from Chickos
and Acree [5,29], but the high correlation coefficient R2 is evidence for the robustness of the
correlation according to Equation (3). Perhaps, the reason for the deviation of the empirical
coefficients from those of the original values is that not too many long-chain species were
included in the evaluation of Chickos and Acree [5,29]. It seems that for molecules with the
monotonically growing alkyl chain, there are some peculiarities that should be taken into
account. This observation should be validated with classes of organic compounds other
than esters. However, since the triglycerides are most closely related to the long-chain
g o
esters, we decided to apply Equation (3) to estimate the ∆l Cp,m -values for this class as well.
o (liq) of triglycerides are required to apply Equation (3)
Now, the molar heat capacities Cp,m
g o
and calculate the desired ∆l Cp,m -values for the temperature adjustment of vaporisa-
tion enthalpies. The compilation of the Cp,m o (liq)-values available in the literature is

given in Table 3.
Thermo 2022, 2 256

g
o
Table 3. Compilation of data on molar heat capacities Cp,m (liq) and heat capacity differences ∆l Cp,m
o
− 1 − 1
for triglycerides at T = 298.15 K (in J.K .mol ).

g
Compounds NC a Cop,m (liq) −∆l Cop,m c
glycerol triformate 1 284 b 69
TG 20 20 20 2 389.0 [12] 88
TG 30 30 30 3 481.3 [12] 105
TG 40 40 40 4 555.3 [12] 118
glycerol tri(2-methylpropanoate) 4 561 b 119
TG 50 50 50 5 610 b 128
glycerol tri(3-methylbutanoate) 5 657 b 137
glycerol tri(2,2-dimethylpropanoate) 5 640 b 134
TG 60 60 60 6 682 b 141
tribenzoin 7 555 b 118
TG 70 70 70 7 765 b 157
TG 80 80 80 8 886 [30] 179
TG 100 100 100 10 1028 [30] 205
TG 120 120 120 12 1322 [30] 259
TG 140 140 140 14 1608 [30] 311
TG 160 160 160 16 1926 b 369
TG 180 180 180 18 2288 b 436
a The NC is the number of the carbon atoms in the single side chain. b Calculated with help of equa-
o (liq, 298.15 K) = 4.7021 × N 2 + 21.0 × N + 387.4 with R2 = 0.997, which was derived by ap-
tion Cp,m C C
proximation of experimental heat capacities given in this table in bold. c Calculated with help of equation
g o
−∆l Cp,m = 16.4 + 0.1833 × Cp,m
o (liq, 298.15 K) with R2 = 0.979, which was derived by approximation of experi-
g
mental ∆l Cp,m
o -values for aliphatic long-chained esters [3].

As can be seen from Table 3, the data available are very limited, so it makes sense
to approximate the available data as a function of chain length and use interpolation and
extrapolation to estimate the heat capacities required. For homologous series, a linear
o (liq)-values with chain length is usually expected. For example, a
correlation of the Cp,m
good quality correlation was found for the linear aliphatic esters (see Table S6). To our
surprise, the dependence of the heat capacity on the chain length for triglycerides is not
linear and was approximated with the following polynomial:
o
Cp,m (liq, 298.15 K) = 4.7021 × NC2 + 21.0 × NC + 387.4 (with R2 = 0.9972) (4)

This correlation was used to estimate the missing heat capacities of triglycerides
g o
(see Table 3) and finally the heat capacity differences, ∆l Cp,m , for each triglyceride were
calculated using Equation (3). The latter values and Equation (1) have enabled the uniform
adjustment of our own and the literature data to the reference temperature T = 298.15 K
g o
and these ∆l Hm (298.15 K) results are now available for comparison and evaluation (see
Tables 1 and 2, column 5).
g
3.3. Evaluation of ∆l Hm
o (298.15 K)-Values of Triglycerides
g
A comparison of the ∆l Hm o (298.15 K)-values for the relatively short chained triglyc-

erides TG 20 20 20 , TG 30 30 30 , and for TG 40 40 40 demonstrates generally good agreement

for each molecule. Unfortunately, only single experimental values are available for TG
50 50 50 and TG 70 70 70 , which makes these results questionable without further validation.
For TG 60 60 60 the range of available experimental vaporization enthalpies from 106 to
131 kJ mol−1 makes it difficult to select a reliable value. The same ambiguity is for TG
80 80 80 , where the spread of the available experimental vaporization enthalpies ranges from
132 to 159 kJ mol−1 .
As can be seen from Table 2, the literature results of both TGA modifications (isother-
mal and non-isothermal) provide the higher and the lower values from this range of
vaporisation enthalpies. In contrast to this, the results of the conventional static and tran-
Thermo 2022, 2 257

spiration method, as well as from the ITGA method carried out in this work are definitely
close to the lower level of the values collected for TG 80 80 80 . The same trend is also ob-
served for TG 100 100 100 where the static method, transpiration, QCM, and our ITGA show
fairly similar results. At the same time, the literature modifications of the TGA provide
significantly higher values (see Table 1). The wide spread of the literature TGA results
can most likely be explained by the fact that this work [18–23] was published more than
20 years ago, when the development of this method for determining evaporation was still
in its infancy. The process and the limits of the TGA method were not sufficiently known at
the time. This statement is based on our extended investigation of the I-TGA method in
relation to measurements with heavy volatile compounds [9]. In this work, we develop
structure-property correlations to determine the general level of experimental enthalpies
of vaporization. These correlations were very helpful in establishing consistency in the
enthalpy of vaporization data for the entire TG set. For this reason, we have chosen to
avoid averaging the ‘experimental’ and ‘estimated’ results. For example, for TG 70 70 70 , TG
120 120 120 and TG 180 180 180 the available experimental data did not agree with the general
trend developed for the TG set. For this reason, averaging the “estimated” data was the
only option to get the reasonable result for the particular TG.
The volatility of the triglycerides decreased dramatically with the lengthening of the
alkyl chains, that is why all TGA results for TG 120 120 120 provide unexpectedly high
vaporisation enthalpies. The same conclusion applies for TG 140 100 100 TG 160 100 100 and
TG 180 100 100 (see Table 1). Therefore, for the latter series of the long-chained glycerides,
only results from QCM and static methods (with the exception for TG 180 180 180 ) could be
considered as reliable.
For the triglycerides with the non-linear alkyl chains (branched or phenyl substituted)
collected in Table 2, only single experimental values are available for each compound.
Therefore, the quality of these results is questionable although the data were measured
using conventional methods (transpiration, QCM, and static method).

3.4. Validation of Vaporisation Enthalpies

g
The significant disagreement among the available ∆l Hm o (298.15 K)-datasets for most

of the triglycerides compiled in Tables 1 and 2 has prompted the extended validation
using structure–property correlations, e.g., with the chain length dependence or using the
correlation between vaporisation enthalpy and retention indices, or boiling temperatures
of triglycerides. Results of these validations are given below.

3.4.1. Structure–Property Correlations: Chain-Length Dependence

g
The linear correlation of ∆l Hm o (298.15 K)-values with the number of carbon atoms

in the alkyl chain within the homologue series of organic compounds is well-established
phenomenon, e.g., for the series of aliphatic linear esters (see Table S6). We also correlated
g o
the ∆l Hm (298.15 K)-values for triglycerides (evaluated in Table 1) with the total number of
carbon atoms, nc , in the triglyceride. The following correlation was obtained (see Table 4):
g o
∆l Hm (298.15 K)/(kJ·mol−1 ) = 50.7 + 3.18 × nc with ( R2 = 0.9981) (5)

The relatively high correlation coefficient R2 was evidence of a good consistency of

the experimental data evaluated in Table 1 and approximated by Equation (5).
As it apparent from Table 4 the differences between the experimental and “theoretical”
vaporisation enthalpies are mostly below 3 kJ·mol−1 . The uncertainties of ±3.0 kJ·mol−1
(0.95 level of confidence, k = 2) were assigned to the enthalpies of vaporisation, which
g o
were estimated from the correlation of ∆l Hm (298.15 K) with the number of C-atoms in
the triglyceride. The “theoretical” results derived from Equation (5) are given in Table 1
labelled as nc .
Thermo 2022, 2 258

g
Table 4. Chain-length dependence of the experimental enthalpies of vaporisation, ∆l Hm
o (298.15), for
− 1
triglycerides with the linear saturated alkyl chains (in kJ mol ) .a

g g
Compound nc a
∆l Hom (298.15 K)exp b ∆l Hom (298.15 K)calc c ∆d
TG 20 20 20 9 81.5 ± 0.3 79.3 2.2
TG 30 30 30 12 90.4 ± 0.5 88.8 1.6
TG 40 40 40 15 97.5 ± 0.6 98.3 −0.8
TG 50 50 50 18 104.8 ± 0.6 107.9 −2.9
TG 60 60 60 21 114.9 ± 3.6 117.4 −2.5
TG 70 70 70 24 127.0 ± 1.7 126.9 0.1
TG 80 80 80 27 134.1 ± 1.2 136.5 −2.4
TG 100 100 100 33 160.1 ± 0.9 155.5 4.6
TG 120 120 120 39 174.9 ± 2.1 174.6 0.3
TG 140 140 140 45 194.0 ± 2.1 193.6 0.6
TG 160 160 160 51 211.6 ± 1.7 212.7 −1.7
TG 180 180 180 57 232.1 ± 2.1 231.7 0.4
a Total number of carbon atoms in the triglyceride. b Values evaluated and recommended (given in bold) in Table 1.
Uncertainty of the vaporisation enthalpy is expressed as the expanded uncertainty (0.95 level of confidence, k = 2).
c Difference between column 3 and 4 in this table. d Calculated using Equation (5) with the assessed expanded

uncertainty of ±3.0 kJ·mol−1 .

3.4.2. Structure–Property Correlations: Correlation with the Retention Indices Jx

g
The correlation of ∆l Hmo (298.15 K) with the retention indices is also well-known tool

to establish internal consistency within a set of structurally parent compounds, particularly

the homologous series. The linear correlations are typical for different classes of organic
compounds, e.g., alkyl-imidazoles [31], alkylbenzenes [32], and nitriles [33]. We have
correlated the vaporisation enthalpies evaluated in Table 1 for the linear triglycerides and in
Table 2 for the branched triglycerides with the data on Kovats indices, Jx , available from the
literature [15,34]. The compilation of the data used for this correlation is given in Table 5.
g
Table 5. Correlation of vaporisation enthalpies, ∆l Hm
o (298.15 K), of triglycerides with their Kovats

indices (Jx ).

g g
nc a
Jx b ∆l Hom (298.15 K)exp c ∆l Hom (298.15 K)calc d ∆e
Compound kJ·mol−1 kJ·mol−1 kJ·mol−1
glycerol triformate 6 1112 73.9
TG 20 20 20 9 1299 81.5 ± 0.3 80.3 1.2
TG 30 30 30 12 1567 90.4 ± 0.5 89.5 0.9
TG 40 40 40 15 1816 97.5 ± 0.6 98.1 −0.6
glycerol tri(2-methylpropanoate) 15 1692 93.6 ± 0.7 93.8 −0.2
TG 50 50 50 18 2089 104.8 ± 0.6 107.5 −2.7
glycerol tri(3-methylbutanoate) 18 1953 104.0 ± 1.3 102.8 1.2
glycerol tri(2,2-dimethylpropanoate) 18 1759 96.1
TG 60 60 60 21 2363 114.9 ± 3.6 116.9 −2.0
TG 70 70 70 24 2664 127.2
TG 80 80 80 27 2958 134.1 ± 1.2 137.4 −3.3
TG 100 100 100 33 3504 160.1 ± 0.9 156.1 4.0
TG 120 120 120 39 4050 174.9
TG 140 140 140 45 4604 196.4 ± 12.1 [16] 194.0 2.4
TG 160 160 160 51 5158 210.2 ± 9.2 [Table S2] 213.0 −2.8
TG 180 180 180 57 5713 232.1
aTotal number of carbon atoms in the triglyceride. b Kovats indices, Jx , on the non-polar column OV-101 [15,34].
cExperimental data evaluated in Tables 1 and 2. Uncertainty of the vaporisation enthalpy is expressed as the
expanded uncertainty (0.95 level of confidence, k = 2). d Calculated using Equation (6) with the assessed expanded
uncertainty of ±3.0 kJ·mol− 1 . Values given in italic were used for comparison with the expeimenetal results in
Table 1. e Difference between column 4 and 5 in this table.
Thermo 2022, 2 259

g
The following linear correlation was found between the ∆l Hm
o (298.15 K)-values of

triglycerides and the Jx -values:

g
o
∆l Hm (298.15 K)/(kJ·mol−1 ) = 35.6 + 0.0344 × Jx with ( R2 = 0.9972) (6)

The high correlation coefficient R2 of Equation (6) indicated the reliability of the
vaporisation enthalpies evaluated in Tables 1 and 2. We used Equation (6) to predict
vaporisation enthalpies of five triglycerides (given in Table 5 in italic), where retention
indices were available, but the experimental data were of questionable quality. As is obvious
from Table 5, the differences between the experimental and “theoretical” vaporisation
enthalpies are mostly below 3 kJ·mol− 1 . Therefore, the uncertainties of ±3.0 kJ·mol− 1
(0.95 level of confidence, k = 2) were assigned to the enthalpies of vaporisation, which are
g o
estimated from the correlation of ∆l Hm (298.15 K) with Kovats indices. The “theoretical”
results derived from Equation (6) are given in Table 4 and labelled as Jx .

3.4.3. Structure–Property Correlations: Correlation with Normal Boiling Temperatures Tb

g
The correlation of ∆l Hm o (298.15 K) with normal boiling temperatures, T , was ad-
b
ditionally examined to validate vaporisation enthalpies of triglycerides evaluated in
Tables 1 and 2. Such a correlation is usually expected to be linear, particularly within
the homologous series. The triglycerides are generally thermally stable compounds; how-
ever, due to very high boiling points, data at standard pressure are limited. The normal
boiling temperatures, Tb , of triglycerides have been compiled from the literature [35,36].
The data taken into correlation and results are given in Table 6.
g
Table 6. Correlation of vaporisation enthalpies, ∆l Hm
o (298.15 K), of triglycerides with their normal

boiling temperatures (Tb ).

g g
Tb a ∆l Hom (298.15 K)exp b ∆l Hom (298.15 K)calc c ∆d
Compound K kJ·mol−1 kJ·mol−1 kJ·mol−1
TG 20 20 20 533 [35] 81.5 ± 0.3 79.3 2.2
TG 30 30 30 563 [35] 90.4 ± 0.5 90.1 0.3
TG 40 40 40 583 [36] 97.5 ± 0.6 97.3 0.2
glycerol tri(3-methylbutanoate) 606 [36] 104.0 ± 1.3 105.5 −1.5
TG 60 60 60 633 [36] 114.9 ± 3.6 115.2 −0.3
TG 70 70 70 665 [36] 126.7
TG 80 80 80 691 [36] 134.1 ± 1.2 136.0 −2.1
TG 160 160 160 894 [36] 210.2 ± 9.2 [Table S2] 208.9 1.3
aNormal boiling temperature. b Experimental data evaluated in Tables 1 and 2. Uncertainty of the vaporisation
enthalpy is expressed as the expanded uncertainty (0.95 level of confidence, k = 2). c Calculated using Equation (7)
with the assessed expanded uncertainty of ±3.0 kJ·mol−1 . Value given in italic were used for comparison with the
expeimenetal results in Table 1. d Difference between column 4 and 5 in this table.

g
It has been found that the ∆l Hm
o (298.15 K)-values of triglycerides are also linearly

correlated with the Tb -values:

g o
∆l Hm (298.15 K)/(kJ·mol−1 ) = −112.1 + 0.3591 × Tb with ( R2 = 0.9989) (7)

The high correlation coefficient R2 in Equation (7) supports the reliability of vaporisa-
tion enthalpies evaluated in Tables 1 and 2. Indeed, the vaporisation enthalpies derived
from the correlation with the boiling temperatures are in a good agreement with the experi-
ment. This good agreement can therefore be considered as an additional validation of the
g o
experimental data for the ∆l Hm (298.15 K) evaluated in this work (see Tables 1 and 2). The
differences between the theoretical and experimental values are at the level of 2 kJ·mol− 1 .
However, considering a very limited set of experimental data included in the correlation,
g o
the uncertainties in the enthalpies of vaporisation estimated from the ∆l Hm (298.15 K)—Tb
− 1
correlation were evaluated to be ±3.0 kJ·mol (0.95 level of confidence, k = 2). We used
Equation (7) to assess the vaporisation enthalpy of TG 70 70 70 (given in Table 5 in italic),
rmo 2022, 2, FOR PEER REVIEW

Thermo 2022, 2 260

methods, particularly for long-chain triglycerides, where the experimen
array. As can be seen in Tables 1 and 2, the results of the structure–pro
where the experimental result seems to be inconsistent with other available data. The
labelled nc, gJx oand
results of the ∆l Hm
Tb agree, within the assigned uncertainties, with
(298.15 K)—Tb correlation are labelled as Tb and given in Tables 1 and g
2
the
mental
for generaldata (highlighted
comparison in bold). Therefore, these evaluated ∆l 𝐻m
of the methods.
o
(29
Finally, three independent structure–property correlations of vaporisation enthalpies
given in bold can be recommended for thermochemical calculations.
with chain length, retention indices, and boiling temperatures have demonstrated sufficient
internal consistency of the data analysis performed in Tables 1 and 2. These results are
3.5. Canforthe
valuable GrouptheAdditivity
validating Method
level of enthalpies Predict Vaporisation
of vaporisation Enthalpies
available from other methods, of Tri
particularly for long-chain triglycerides, where the experimental data are in disarray. As
can beGroup additivity
seen in Tables (GA)
1 and 2, the resultsmethods are also a correlations
of the structure–property type of structure–prop
labelled nc ,
Jx and Tb agree, within the assigned uncertainties, with the evaluated experimental data
[37,38]. The enthalpies of vaporisation g oof a set of molecules with reliab
(highlighted in bold). Therefore, these evaluated ∆l Hm (298.15 K)-values and given in bold
split
can beup into the smallest
recommended possible
for thermochemical groups, like “LEGO®” building block
calculations.
calculations, each group obtains a well-defined numerical contribution
3.5. Can the Group Additivity Method Predict Vaporisation Enthalpies of Triglycerides?
the vaporisation enthalpy
Group additivity (GA) methods are is also
then a construction
a type of the
of structure–property molecule
relationships [37,38].from th
collecting
The enthalpiesthe
of energetics
vaporisation ofof a molecule
a set of moleculesfrom the appropriate
with reliable data are usually number
split an
up into the smallest possible groups, like “LEGO® ” building blocks. Using the matrix
general, using this method for large molecules is impractical due to t
calculations, each group obtains a well-defined numerical contribution. The prediction
blocks. To overcome
of the vaporisation enthalpy this disadvantage,
is then a construction of we thedevelop a general
molecule from approach
the building
blocks, collecting the energetics of a molecule from the appropriate
isation enthalpies based on a so-called “centerpiece” molecule [39,40 number and type
of bricks. In general, using this method for large molecules is impractical due to too
“centerpiece” approach
many building blocks. To overcomeis tothisstart the prediction
disadvantage, with
we develop a potentially
a general approach larg
that can generally
to estimate mimic the
vaporisation enthalpies basedstructure
on a so-calledof “centerpiece”
the molecule of interest,
molecule [39,40]. bu
The idea of the “centerpiece” approach is to start the prediction with a potentially large
must have a that
“core” molecule reliable vaporisation
can generally enthalpy.
mimic the structure of theThe triglycerides
molecule of interest, butare
at prede
approach.
the same time The visualisation
must have of the “centerpiece”
a reliable vaporisation approach
enthalpy. The triglycerides for triglyceri
are predestined
for such an approach. The visualisation of the “centerpiece” approach for triglycerides is
ure 2.
given in Figure 2.

Figure 2. The visualisation of the “centerpiece” approach for triglycerides.

Figure 2. The visualisation of the “centerpiece” approach for triglycerides.
Indeed, the TG 20 20 20 as the “centerpiece” model already bears the main energetic
contributions
Indeed, to the
thevaporisation
TG 20202enthalpy specific for triglycerides. In order to obtain
0 as the “centerpiece” model already bears t
the “centerpiece” suitable for GA calculations we need only to cut three methyl groups
contributions
(C-(C)(H)3 ) from thetoTG
the
20 2vaporisation enthalpy specific for triglycerides. In
0 20 (see Figure 3, left).
“centerpiece” suitable for GA calculations we need only to cut three
(C)(H)3) from the TG 202020 (see Figure 3, left).
 Thermo 2022, 2, FOR PEER REVIEW 12
Thermo 2022, 2 261

Thermo 2022, 2, FOR PEER REVIEW 12

FTG = ∆ 𝐻 (298.15 K, glycerol triacetate) ∆ 𝐻 (298.15 K, glycerol trihexanoate) = FTG + 3 × С-(C)(H2)(CO2)

− 3 × C-(C)(H)3 + 9 × С(С)2(H)2 + 3 × C-(C)(H)3 + 3 × (C-CO)1–4 + 6 × (C-C)1-4.
FTG = ∆ 𝐻 (298.15 K, glycerol triacetate)
Figure 3.∆Development
Figure 3. 𝐻 (298.15 K, glycerol
Development ofofthe “centerpiece”
trihexanoate) = fragment
Ffragment FFTG from
TG + 3 × С-(C)(H glycerol
2)(CO 2) triacetate (left) and a group-
gthe “centerpiece” TG from glycerol triacetate (left) and a group-
− 3 × C-(C)(H)3 additivity calculation of2 +∆
+ 9 × С(С)2(H) g 3𝐻× oC-(C)(H)3 + 3 × (C-CO)1–4 + 6 × (C-C)1-4.
l om (298.15 K) for glycerol trihexanoate (TG 606060) as an example (right).
additivity calculation of ∆l Hm (298.15 K) for glycerol trihexanoate (TG 60 60 60 ) as an example (right).
Figure 3. Development of the “centerpiece” fragment FTG from glycerol triacetate (left) and a group-
g o
Such
additivity calculation ∆l bulk
of a 𝐻m (298.15
fragment (TriGlycerides
K) for glycerol or60F60TG
trihexanoate (TG 60)) as
and its energetic
an example (right). contribution are specific
Such a bulk fragment (TriGlycerides or FTG ) and its energetic contribution are specific
for triglycerides and
and hardly can be
beitscaptured by
by any
any other method.
method. This This special
special feature
feature of
Such for
a bulktriglycerides
fragment (TriGlycerides hardly can
or FTG ) and captured
energetic contribution other
are specific of
the “centerpiece”
the and hardly can approach
for triglycerides approach significantly
be captured significantly
by any other method. increases
increases the reliability
the reliability
This special of the property predic-
feature of of the property prediction
tion
the “centerpiece” forapproach
for similarly similarly
shaped shaped molecules,
molecules,
significantly increasese.g.,
theTGe.g.,60TG
60 660of
reliability 06
(see60 (see
0the FigureFigure
property 3, right),
3,predic-
right), wherewhere substituents
substituents with
tion for similarly
with
the known shaped
the known molecules,
contributions e.g., TG
contributions to60the
60to
60 (see
theFigure 3, right),
vaporisation
vaporisation where substituents
enthalpy
enthalpy are aresimply simply attached
attached to the
to the FTGFTG as
with the known contributions to the vaporisation enthalpy are simply attached to the FTG
as
the the “centerpiece”.
“centerpiece”. The The
group group contribution
contribution values, values,
which which
are are specific
specific for for alkanes
alkanes C-(C)(H C-
3 ),
as the “centerpiece”. The group contribution values, which are specific for alkanes C-
(C)(H
C-(C
(C)(H3), С-(С2)(H 3 ),
)(H С-(С
), as
2)(H
well
2), as
as well
the as the contribution
contribution C-(C)(H)
2 2), as2well as the contribution C-(C)(Н)2(CО2) specific for C-(C)(Н)
(CO )2 (CО )
specific
2 the2methylene- 2 specific
for thefor the methylene-
methylene-group
group
attached
group attached to attached
the to the to the carbonyl-group
carbonyl-group
carbonyl-group are well
are well established are well established
established
[14]. [14].1-4contributions
[14]. (C-C)
The contributions The The contributions
(C-C)(C-C)
1-4 and1-4
and (C-CO) and
(C-CO) are additional
(C-CO) 1–4 arecorrection terms
additional for the
correction 1–4 “gauche”
terms interactions
forthethe1–4 of carbon
1–4“gauche”
“gauche” interactions
interactions of carbon
1–4 are additional correction terms for carbon
1–4

atoms along the alkyl chain. The details on these 1–4 C-C interactions are described else-
atoms along
atoms alongthe thealkyl
alkylchain.
chain. TheThe details
details on these
on these 1–4 C-C
1–4 C-C interactions
interactions are described
are described else-
where [38,41]. The schematic representation of the group-contributions involved in this
elsewhere
where [38,41].
[38,41]. The The schematic
schematic representation
representation
study is shown in Table 7. The numerical values for these contributions were developed of of
thethe group-contributions
group-contributions involved
involved in
in this
this
studywork
study
in our previous [14] andin
is shown areTable
Table 7.7.The
also given The numerical
numerical
in Table 7. values
valuesfor forthese
thesecontributions
contributionswere weredeveloped
developed in
g o
Using our
intheourprevious
group
previous work
additivity [14] and
contributions
work [14] andare
givenalso
are in given
Table
also in in
7, the
given Table
predicted
Table ∆l 𝐻m
7. 7. (298.15
K)-values for triglycerides under study were calculated and the results are given in Tables g o
Using the group additivity contributions given in Table 7, the predicted ∆l 𝐻m (298.15
g o
Table 7. Group additivity contributions for calculation of the enthalpy of vaporisation, ∆l Hm
2, 8 and S7).
K)-values for triglycerides under study were calculated and the−results are given in Tables
(298.15 K), for triglycerides with the saturated alkyl chains (ingkJ mol 1 ).
and S7).contributions for calculation of the enthalpy of vaporisation, ∆l 𝐻mo (298.15
2, 8 additivity
Table 7. Group
K), for triglycerides with the saturated alkyl chains (in kJ mol−1).
g
Increment a ∆l Hom [14] g o
Table 7. Group additivity contributions for calculation
Increment a
g o
∆l 𝑯
of the enthalpy
m [14] of vaporisation, ∆l 𝐻m (298.15
K), for triglycerides with the saturated alkyl FTG (in kJ mol62.5
FTGchains −1). 62.5
C-(C)(H)
C-(C)(Н) 2(CО 22(CO
) 2) 3.2 3.2
g
C-(C)
C-(C) 2(Н)(CО
2 (H)(CO2 ) 2 ) Increment
−1.5 a − 1.5∆l 𝑯om [14]
C-(C)
C-(C) 3(CО 32(CO
) 2) −7.5
FTG −7.5 62.5
(C-CO)
(C-CO) 1-4 1-4 −1.0 −1.0
C-(C)(Н)2(CО2) 3.2
C-(C)
С-(С) 2(H)2 (H)2 4.52 4.52
C-(C)(H)
С-(С)(H) 3 3
C-(C) 2(Н)(CО
6.33 2 ) 6.33 −1.5
(C-C)
(С-C) 1-4 1-4 C-(C)
1.803(CО2) 1.80 −7.5
Сd(H)Cd (H) 3.8
(C-CO) 1-4 3.8 −1.0
Ph(CO
Ph(CО 2) 2) 29.4 29.4
a For calculation of unsaturated alkyl-chains an additional increment Сd(H) = 3.8 kJ mol−1 was
С-(С)2(H)2 4.52
a For calculation of unsaturated alkyl-chains an additional increment C (H) = 3.8 kJ mol−1 was developed from
developed from vaporisation enthalpies of olefines [38]. For calculation of phenyl С-(С)(H)
d 3
substituted 6.33
vaporisation enthalpies ofPh(CО
triglycerides an additional increment
olefines [38]. For calculation of phenyl substituted triglycerides an additional increment
2) = 29.4 kJ mol−1 was developed from vaporization
− 1 (С-C) 1-4
Ph(CO ) = 29.4 kJ mol was developed from vaporization enthalpy of benzyl acetate [16]. 1.80
enthalpy of benzyl2acetate [16].
Сd(H) 3.8
g o
Using the group additivity contributions given in Table Ph(CО 2) predicted ∆ Hm
7, the (298.15 K)-
29.4
l
avalues for triglycerides
For calculation under study
of unsaturated were calculated
alkyl-chains and the results
an additional are given
increment in Tables
Сd(H) 2, 8mol
= 3.8 kJ and
−1 S7).
was
Even from
developed a quick look at results
vaporisation given
enthalpies of in Table[38].
olefines 8 can
Forreveal that the
calculation “centerpiece”
of phenyl ap-
substituted
triglycerides an additional
proach systematically increment Ph(CО
overestimates ) = 29.4 kJ molenthalpies.
the 2vaporisation −1 was developed from vaporization
It is noticeable that the
enthalpy of benzylisacetate
overestimation [16]. with the growing chain-length of the triglycerides. It is ob-
increasing
vious that the GA method has completely failed to predict the vaporisation enthalpies of
Thermo 2022, 2 262

triglycerides. Is there an explanation for this phenomenon? The answer is discussed in the
following section.
g
Table 8. Comparison of experimental and additive enthalpies of vaporisation, ∆l Hm o (298.15 K), for

triglycerides with the linear saturated alkyl chains and for methyl alkanoates (in kJ mol− 1 ).

Methyl
Triglycerides
Alkanoates
g g g g
Nc a
∆l Hom (exp) b ∆l Hom (add) c Edisp (TG) d ∆l Hom (exp) e ∆l Hom (add) f Edisp (MA) g
3 90.4 91.1 −0.7 36.0 35.8 0.2
4 97.5 101.6 −4.1 39.6 40.3 −0.7
5 105.0 115.9 −10.9 43.6 45.1 −1.5
6 114.9 130.2 −15.3 48.5 49.8 −1.3
7 127.0 144.4 −17.4 53.4 54.6 −1.2
8 134.1 158.7 −24.6 57.2 59.4 −2.2
10 160.1 187.3 −27.2 66.5 69.0 −2.5
12 174.9 215.8 −40.9 75.5 78.5 −3.0
14 194.0 244.4 −50.4 85.2 88.1 −2.9
16 211.6 273.0 −61.4 93.6 97.6 −4.0
18 232.1 301.5 −69.4 103.7 107.2 −3.5
a The number of C-atoms in a single alkyl chain in the triglyceride or in methyl alkanoates. b Evaluated values
from Table 1. c Additive values calculated using increments in Table 7. d Difference between column 2 and 3,
interpreted as amount of dispersion forces in triglyceride. e Experimental values evaluated in our previous study
[3]. f Additive values calculated using increments in Table 7. g Difference between column 5 and 6, interpreted as
amount of dispersion forces in methyl alkanoates (MA).

3.6. Non-Covalent Dispersion Interactions in Triglycerides

As a matter of fact, the GA methods are not only a suitable tool to predict molecular
energetics, but also a tool to detect unusual energetic effects. When the experimental and
additive results show significant discrepancies, it is best to look for specific interactions
causing the deviation from additivity (assuming the experimental result is reliable). In the
case of triglycerides, we validated the experimental vaporisation enthalpies with different
structure–property correlations. In what follows, the profound deviation from additivity
observed in Section 3.5 should only be caused by overlooked interactions specific to these
g o
long-chained molecules. Basically, the standard molar vaporisation enthalpy, ∆l Hm , is
the portion of energy (enthalpy) required to transfer 1 mole of the liquid compound to a
gaseous state. Thus, the vaporisation enthalpy can be taken as a measure of the overall
attractive forces between the molecules in the liquid state. If these attractive interactions
are responsible for significant interlinking of alkyl chains in the liquid phase, the energy
required to take out the triglyceride with the interlocked chains from the liquid to the
gas phase should be higher and the corresponding enthalpy of vaporisation greater in
comparison to the additive value. Therefore, the relative decrease in the experimental
vaporisation enthalpy can only be explained by assuming that the attractive forces are
partially entrained into the gas phase. In this case, the attractive dispersion interactions
between chains in the gas phase could be a plausible explanation for the deviation from
additivity, since these specific non-covalent dispersion interactions are not considered in
the GA parameterization. The existence of such dispersion-stabilized conformers in the gas
phase has been theoretically supported by quantum chemical calculations [42]. Structural
optimization of the long-chain triglycerides with MOPAC-PM7 showed that the most stable
forms are folded conformers with three parallel chains that interact [43]. Similar to linear
alkanes, folded configurations are favoured over extended star conformers [44,45]. Two
possible structures of dispersion-stabilized conformers are shown in Figure 4, where the
attraction of alkyl chains in the gas phase due to dispersion forces is evident.
Thermo 2022, 2, FOR PEER REVIEW 14
Thermo 2022, 2, FOR PEER REVIEW

Thermo 2022, 2
possible structures of dispersion-stabilized conformers are shown in Figure 4, where the
263
possible
attraction of alkyl chainsstructures
in the gasofphase
dispersion-stabilized
due to dispersionconformers
forces isare shown in Figure
evident. 4, where
attraction of alkyl chains in the gas phase due to dispersion forces is evident.

Figure
Examples 4. Examples
of possible
possible of possible dispersion-stabilized
dispersion-stabilized conformations conformations
8080088008.0.for TG 808080.
Figure4.4.Examples
Figure of dispersion-stabilized conformationsfor forTG
TG
Afterdemonstrating
Afterqualitatively
qualitatively qualitatively demonstrating
existencethe
theexistence existence
ofdispersion
dispersion of dispersion
forces,we forces,
weused
usedthe we used the
the
After demonstrating the of forces, re-
results in Table 8 to quantitatively assess the strength of this interaction. To quantify the quantify
sults in Table 8 to quantitatively assess the strength of this interaction. To
sults in Table 8 dispersion
to quantitatively assess
interactions in the strength of this interaction. To quantify the
dispersion interactions in triglycerides, wetriglycerides,
assume that wethe assume thatbetween
differences the differences between exp
experimen-
dispersion interactions
mental in triglycerides,
enthalpies of we assume
vaporisation and that thevalues
additive differences
reflect between
the amountexperi-
of non-addi
tal enthalpies of vaporisation and additive values reflect the amount of non-additive forces
mental enthalpies of vaporisation
forces (denoted as Eand additive
disp, see Table values
8 and reflect
Figure 5).the amount of non-additive
(denoted as Edisp , see Table 8 and Figure 5).
forces (denoted as Edisp, see Table 8 and Figure 5).

Figure 5. Comparison of amount of dispersion interactions (in kJ mol-1 ) in triglycerides (◦) and in
Figure 5. Comparison of amount of dispersion interactions (in kJ mol-1) in triglycerides (◦) an
methyl alkanoates (•).
methyl alkanoates (●).
It is quite obvious that the differences, E , do not represent the total energy of the
disp
Figure 5. Comparison It of isamount
quite obvious that the differences,
(inEkJ , do-1)not represent the total energy of
alkylofchains
dispersion interactions dispmol in triglycerides
dispersion forces between in triglycerides. Nonetheless, Edisp -values (◦)
canand
be in
methyl alkanoatesdispersion
(●). forces between alkyl chains in triglycerides. Nonetheless, Edisp-values can
considered as an energy differences originating from dispersion forces between the three
considered as an energy differences originating from dispersion forces between the th
arms of the triglyceride, and the vaporisation enthalpies provide a reliable measure of
It is quite arms of that
obvious the triglyceride,
the differences,and theEdispvaporisation enthalpies
, do not represent the provide a reliable
total energy of themeasur
dispersion forces in triglycerides. Therefore, the dramatic increase (from −0.7 kJ mol−1
dispersion dispersion forces in triglycerides. Therefore, the dramatic increase (from −0.7be
kJ mol−
for TG 30 30forces
30 to −between
69.4 kJ mol alkyl
−1 in chains
TG 18 in18triglycerides.
0 0
Nonetheless,
180 ) in dispersion interactionsEdispwith
-values can
growing
anTG 303030 to −69.4 kJ mol in TG 18from0180180) in dispersion interactions with growing ch
−1
considered as can
chains length energy
now be differences
conceptuallyoriginating
explained and dispersionIndeed,
understood. forces inbetween the three
a homologous
length can now be conceptually explained and understood. Indeed, in a homolog
arms
seriesofofthe triglyceride,
methyl alkanoates, and the vaporisation
although the enthalpiesenthalpies provide
of vaporisation a reliable
logically measure
increase with of
dispersion forces
growing chain in triglycerides.
length Therefore,
(see Table 8 and thethe
Figure 5), dramatic increase
differences (from
between the−0.7 kJ mol−1 for
experimental
TG
and30additive
3030 to −69.4
valueskJ that
molaccount
−1 in TG for
18018 0180) in dispersion
dispersion interactionsinteractions withwith
hardly increase growing chains
increasing
length can now
chain length. bethe
Thus, conceptually explained appear
dispersion interactions and understood. Indeed,
to be negligible in a homologous
for methyl alkanoates
but not in triglycerides.
Thermo 2022, 2 264

How does alkyl chain branching affect dispersion interactions between alkyl chains in
the triglycerides? To answer this question, we calculated the additive enthalpies of vapori-
g o
sation, ∆l Hm (298.15 K), for all entries in Table 2. A comparison of the recommended (given
g o
in bold) experimental and additive enthalpies of vaporisation, ∆l Hm (298.15 K), reveals that
the values for glycerol tri(2-methylpropanoate) and glycerol tri(3-methylbutanoate) are very
close (within their uncertainties, hence the dispersion forces are small, since the branching
of the chains precludes close approximation of chains. The difference between additive and
experimental values of −6.8 kJ mol−1 for glycerol tri(2,2-dimethylpropanoate) indicates
noticeable stabilization despite the steric interactions of bulky substituents. However,
such an interesting phenomenon has already been considered specific, as shown for tert-
butyl-substituted alkanes [2]. For the glycerol tribenzoate the stabilization of −9.3 kJ mol−1
could be explained by the π–π attractive interaction of the benzene rings attached to the
TG moiety. The profound stabilization of −44 kJ mol−1 observed for glycerol trierucate
(TG 221 ,221 ,221 ) appears due to dispersive interactions of the long alkyl chains. However,
this stabilization is less intensive than expected from the Edisp trend obvious from Table 8.
It appears that the double bonds present in the alkyl chains screw them and reduce the
spatial possibilities for the attractive dispersion forces.
To draw a practical conclusion from this study, we failed to develop the group con-
tribution method for predicting the vaporisation enthalpies of triglycerides because the
non-covalent dispersion interactions are unique to each triglyceride and increase with
increasing chain length. Nonetheless, the critical evaluation and validation of the va-
porisation thermodynamics of triglycerides has enabled a qualitative and quantitative
understanding of the reasons for these dispersion forces.

Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/thermo2030018/s1, Figure S1: Schematic explanation of the TG
abbreviations; Table S1: Results of transpiration method for triglycerides measured in this work:
absolute vapour pressures p, standard molar vaporisation enthalpies and standard molar vaporisation
entropies; Figure S2: The scheme of the QCM experimental setup from; Table S2: The experimental
mass loss rates and vapour pressures determined for TG100 100 100 and TG160 160 160 with QCM
technique; Table S3: The experimental mass loss rates and vapour pressure determined for TG80 80 80
and TG100 100 100 with the I-TGA technique; Table S4: Vapour pressures p, standard (po = 0.1 MPa)
g o
molar vaporisation enthalpies, ∆l Hm , and standard (po = 0.1 MPa) molar vaporisation entropies,
g o
∆l Sm obtained by the approximation of data collected from SciFinder; Table S5: Compilation of data
o (liq) and heat capacity differences ∆g Co
on molar heat capacities Cp,m l p,m for the linear aliphatic esters
at T = 298.15 K (in J.K−1 .mol−1 ); Table S6: The chain length dependence of the molar heat capacities
o (liq) for the linear aliphatic esters at T = 298.15 K (in J.K−1 .mol−1 ); Table S7: Comparison of
Cp,m
g
experimental and additive enthalpies of vaporisation, ∆l Hm o (298.15 K), for triglycerides with the
− 1
linear saturated alkyl chains (in kJ.mol ) [3,4,6,8,9,46–51].
Author Contributions: Conceptualization, S.P.V. and R.N.N.; methodology, S.P.V. and R.N.N.; val-
idation, R.N.N.; formal analysis, R.N.N.; writing—original draft preparation, S.P.V. and R.N.N.;
writing—review and editing, S.P.V. and R.N.N.; funding acquisition, S.P.V. All authors have read and
agreed to the published version of the manuscript.
Funding: EU Project “Metrology of biofuels”, German Science Foundation (DFG) in the frame of SPP
1807 “Control of London Dispersion Interactions in Molecular Chemistry” (grant VE 265-9/2), and
Kazan Federal University Strategic Academic Leadership Program (“PRIORITY-2030”).
Data Availability Statement: Data supporting reported results can be found in the supplementary
materials to this paper.
Acknowledgments: We gratefully acknowledge the contributions of V.N. Emel´yanenko and D.H.
Zaitsau for assistance in the thermochemical experiments.
Conflicts of Interest: The authors declare no conflict of interest.
Thermo 2022, 2 265

References
1. Wagner, J.P.; Schreiner, P.R. London Dispersion in Molecular Chemistry-Reconsidering Steric Effects. Angew. Chem. Int. Ed. 2015,
54, 12274–12296.
2. Verevkin, S.P.; Kondratev, S.O.; Zaitsau, D.H.; Zherikova, K.V.; Ludwig, R. Quantification and understanding of non-covalent
interactions in molecular and ionic systems: Dispersion interactions and hydrogen bonding analysed by thermodynamic methods.
J. Mol. Liq. 2021, 343, 117547.
3. Zaitsau, D.H.; Pimerzin, A.A.; Verevkin, S.P. Fatty acids methyl esters: Complementary measurements and comprehensive
analysis of vaporization thermodynamics. J. Chem. Thermodyn. 2019, 132, 322–340.
4. Brinkmann, J.; Luebbert, C.; Zaitsau, D.H.; Verevkin, S.P.; Sadowski, G. Thermodynamic Modeling of Triglycerides using PC-SAFT.
J. Chem. Eng. Data 2019, 64, 1446–1453.
5. Acree, W., Jr.; Chickos, J.S. Phase Transition Enthalpy Measurements of Organic and Organometallic Compounds. Sublimation,
Vaporization and Fusion Enthalpies from 1880 to 2015. Part 1. C1–C10. J. Phys. Chem. Ref. Data 2016, 45, 033101.
6. Verevkin, S.P.; Emel’yanenko, V.N. Transpiration method: Vapor pressures and enthalpies of vaporization of some low-boiling
esters. Fluid Phase Equilibr. 2008, 266, 64–75.
7. Kulikov, D.; Verevkin, S.P.; Heintz, A. Enthalpies of vaporization of a series of aliphatic alcohols: Experimental results and values
predicted by the ERAS-model. Fluid Phase Equilibr. 2001, 192, 187–207.
8. Verevkin, S.P.; Zaitsau, D.H.; Emel’yanenko, V.N.; Heintz, A. A New Method for the Determination of Vaporization Enthalpies of
Ionic Liquids at Low Temperatures. J. Phys. Chem. B 2011, 115, 12889–12895.
9. Verevkin, S.P.; Ralys, R.V.; Zaitsau, D.H.; Emel’yanenko, V.N.; Schick, C. Express thermo-gravimetric method for the vaporization
enthalpies appraisal for very low volatile molecular and ionic compounds. Thermochim. Acta 2012, 538, 55–62.
10. Woodman, A.L.; Adicoff, A. Vapor Pressure of Tiracetin, Triethylene Glycol Dinitrate, and Metriol Trinitrate. J. Chem. Eng. Data
1963, 8, 241–242.
11. Fuchs, R.; Peacock, L.A. Heats of vaporization of esters by the gas chromatography–calorimetry method. Can. J. Chem. 1980, 58,
2796–2799. [CrossRef]
12. Nilsson, S.O.; Wadsö, I. Thermodynamic properties of some mono-, di-, and tri-esters enthalpies of solution in water at 288.15 to
318.15 K and enthalpies of vaporization and heat capacities at 298.15 K. J. Chem. Thermodyn. 1986, 18, 673–681. [CrossRef]
13. Daubert, T.E.; Hutchison, G. Vapor pressure of 18 pure industrial chemicals. AIChE 1990, 86, 93–114.
14. Verevkin, S.P.; Emel’yanenko, V.N.; Toktonov, A.V.; Leolko, A.S.; Duwensee, J.; Kragl, U.; Sarge, S.M. Thermochemical and Ab
Initio Studies of Biodiesel Fuel Surrogates: 1,2,3-Propanetriol Triacetate, 1,2-Ethanediol Diacetate, and 1,2-Ethanediol Monoacetate.
Ind. Eng. Chem. Res. 2009, 48, 7388–7399. [CrossRef]
15. Maslakova, A.S.; Krasnykh, E.L.; Levanova, S.V. Saturated vapor pressures and enthalpies of evaporation of esters of glycerol and
lower carboxylic acids. Rus. J. Phys. Chem. A 2010, 84, 163–168. [CrossRef]
16. Stephenson, R.M.; Malanowski, S. Handbook of the Thermodynamics of Organic Compounds; Springer: Dordrecht, The Netherlands,
1987; p. 552.
17. Perry, E.S.; Weber, W.H.; Daubert, B.F. Vapor Pressures of Phlegmatic Liquids. I. Simple and Mixed Triglycerides. J. Am. Chem.
Soc. 1949, 71, 3720–3726. [CrossRef]
18. Kishore, K.; Shobha, H.K.; Mattamal, G.J. Structural effects on the vaporization of high molecular weight esters. J. Phys. Chem.
1990, 94, 1642–1648. [CrossRef]
19. Goodrum, J.W.; Eiteman, M.A. Physical properties of low molecular weight triglycerides for the development of bio-diesel fuel
models. Bioresour. Technol. 1996, 56, 55–60. [CrossRef]
20. Goodrum, J.W.; Siesel, E.M. Thermogravimetric analysis for boiling points and vapour pressure. J. Therm. Anal. 1996, 46,
1251–1258. [CrossRef]
21. Phillips, J.C. Enthalpies of vaporization of oligomers of poly(hexamethylene sebacate) and esters of alkylcarboxylic acids. J. Appl.
Polym. Sci. 1998, 70, 731–746. [CrossRef]
22. Karis, T.E.; Nagaraj, H.S. Evaporation and Flow Properties of Several Hydrocarbon Oils. Tribol. Trans. 2000, 43, 758–766.
23. Krop, H.B.; Velzen, M.J.M.V.; Parsons, J.R.; Govers, H.A.J. Determination of environmentally relevant physical-chemical properties
of some fatty acid esters. J. Amer. Oil Chem. Soc. 1997, 74, 309–315. [CrossRef]
24. Goodrum, J.W.; Geller, D.P.; Lee, S.A. Rapid measurement of boiling points and vapor pressure of binary mixtures of short-chain
triglycerides by TGA method. Thermochim. Acta 1998, 311, 71–79. [CrossRef]
25. Bureau, N.; Jose, J.; Mokbel, I.; de Hemptinne, J.C. Vapour pressure measurements and prediction for heavy esters. J. Chem.
Thermodyn. 2001, 33, 1485–1498. [CrossRef]
26. Goodrum, J.W.; Geller, D.P. Rapid thermogravimetric measurements of boiling points and vapor pressure of saturated medium-
and long-chain triglycerides. Bioresour. Technol. 2002, 84, 75–80. [CrossRef]
27. Zherikova, K.V.; Verevkin, S.P. Ferrocene: Temperature adjustments of sublimation and vaporization enthalpies. Fluid Phase
Equilibr. 2018, 472, 196–203. [CrossRef]
28. Verevkin, S.P.; Zaitsau, D.H.; Emel’yanenko, V.N.; Yermalayeu, A.V.; Schick, C.; Liu, H.; Maginn, E.J.; Bulut, S.; Krossing, I.; Kalb,
R. Making Sense of Enthalpy of Vaporization Trends for Ionic Liquids: New Experimental and Simulation Data Show a Simple
Linear Relationship and Help Reconcile Previous Data. J. Phys. Chem. B 2013, 117, 6473–6486. [CrossRef]
Thermo 2022, 2 266

29. Chickos, J.S.; Acree, W.E., Jr. Enthalpies of Vaporization of Organic and Organometallic Compounds, 1880–2002. J. Phys. Chem.
Ref. Data 2003, 32, 519–878. [CrossRef]
30. Bogatishcheva, N.S.; Faizullin, M.Z.; Popov, A.P.; Nikitin, E.D. Critical properties, heat capacities, and thermal diffusivities of four
saturated triglycerides. J. Chem. Thermodyn. 2017, 113, 308–314. [CrossRef]
31. Emel’yanenko, V.N.; Portnova, S.V.; Verevkin, S.P.; Skrzypczak, A.; Schubert, T. Building blocks for ionic liquids: Vapor pressures
and vaporization enthalpies of 1-(n-alkyl)-imidazoles. J. Chem. Thermodyn. 2011, 43, 1500–1505. [CrossRef]
32. Verevkin, S.P. Vapour pressures and enthalpies of vaporization of a series of the linear n-alkyl-benzenes. J. Chem. Thermodyn.
2006, 38, 1111–1123. [CrossRef]
33. Emel’yanenko, V.N.; Verevkin, S.P.; Koutek, B.; Doubsky, J. Vapour pressures and enthalpies of vapourization of a series of the
linear aliphatic nitriles. J. Chem. Thermodyn. 2005, 37, 73–81. [CrossRef]
34. Andriamaharavo, N.R. Retention Data; NIST Mass Spectrometry Data Center: Gaithersburg, USA, 2014.
35. Cunico, L.P.; Hukkerikar, A.S.; Ceriani, R.; Sarup, B.; Gani, R. Molecular structure-based methods of property prediction in
application to lipids: A review and refinement. Fluid Phase Equilibr. 2013, 357, 2–18. [CrossRef]
36. Wallek, T.; Rarey, J.; Metzger, J.O.; Gmehling, J. Estimation of Pure-Component Properties of Biodiesel-Related Components:
Fatty Acid Methyl Esters, Fatty Acids, and Triglycerides. Ind. Eng. Chem. Res. 2013, 52, 16966–16978. [CrossRef]
37. Benson, S.W. Thermochemical Kinetics: Methods for the Estimation of Thermochemical Data and Rate Parameters; Wiley: New York, NY,
USA, 1976.
38. Verevkin, S.P.; Emel’yanenko, V.N.; Diky, V.; Muzny, C.D.; Chirico, R.D.; Frenkel, M. New Group-Contribution Approach to
Thermochemical Properties of Organic Compounds: Hydrocarbons and Oxygen-Containing Compounds. J. Phys. Chem. Ref.
Data 2013, 42, 033102. [CrossRef]
39. Verevkin, S.P.; Andreeva, I.V.; Emel’yanenko, V.N. Thermodynamic framework for the predicting the properties of amino-benzoic
acids. J. Chem. Thermodyn. 2022, 166, 106689.
40. Andreeva, I.V.; Verevkin, S.P. Thermochemistry of substituted benzenes: Acetophenones with methyl, ethyl, cyano and acetoxy
substituents. J. Therm. Anal. Calorim. 2022. [CrossRef]
41. Roganov, G.N.; Pisarev, P.N.; Emel’yanenko, V.N.; Verevkin, S.P. Measurement and Prediction of Thermochemical Properties.
Improved Benson-Type Increments for the Estimation of Enthalpies of Vaporization and Standard Enthalpies of Formation of
Aliphatic Alcohols. J. Chem. Eng. Data 2005, 50, 1114–1124. [CrossRef]
42. Dimakopoulos, K.D.; Papageorgiou, D.G.; Demetropoulos, I.N. Triglyceride reference database: Large scale storage of 3D
triglyceride conformers and web-based analysis tools. Mol. Simul. 2007, 33, 1057–1059. [CrossRef]
43. Energies of Stable Conformers in Heavy Alkanes and Triglycerides Using MedeA. Mater. Des. Appl. Note 2022, 13851, 1–4. Available
online: https://www.materialsdesign.com/all-application-notes/Alkane-Triglyceride-conformers (accessed on 31 July 2022).
44. Brasiello, A.; Crescitelli, S.; Milano, G. Development of a coarse-grained model for simulations of tridecanoin liquid–solid phase
transitions. Phys. Chem. Chem. Phys. 2011, 13, 16618–16628. [CrossRef] [PubMed]
45. Sum, A.K.; Biddy, M.J.; De Pablo, J.J.; Tupy, M.J. Predictive Molecular Model for the Thermodynamic and Transport Properties of
Triacylglycerols. J. Phys. Chem. B 2003, 107, 14443–14451. [CrossRef]
46. Kulikov, D.; Verevkin, S.P.; Heintz, A. Determination of Vapor Pressures and Vaporization Enthalpies of the Aliphatic Branched
C5 and C6 Alcohols. J. Chem. Eng. Data 2001, 46, 1593–1600.
47. Verevkin, S.P.; Sazonova, A.Y.; Emel’yanenko, V.N.; Zaitsau, D.H.; Varfolomeev, M.A.; Solomonov, B.N.; Zherikova, K.V.
Thermochemistry of Halogen-Substituted Methylbenzenes. J. Chem. Eng. Data 2015, 60, 89–103. [CrossRef]
48. Emel’yanenko, V.N.; Verevkin, S.P. Benchmark thermodynamic properties of 1,3-propanediol: Comprehensive experimental and
theoretical study. J. Chem. Thermodyn. 2015, 85, 111–119.
49. Sauerbrey, G. Verwendung von Schwingquarzen zur Wägung dünner Schichten und zur Mikrowägung. Z. Physik. 1959, 155,
206–222. [CrossRef]
50. Zaitsau, D.H.; Yermalayeu, A.V.; Emel’yanenko, V.N.; Butler, S.; Schubert, T.; Verevkin, S.P. Thermodynamics of Imidazolium-
Based Ionic Liquids Containing PF6 Anions. J. Phys. Chem. B 2016, 120, 7949–7957.
51. Available online: https://scifinder.cas.org (accessed on 31 July 2022).