8100
B. Barton et al. / Tetrahedron 72 (2016) 8099e8105
dedicated to their separation involving techniques such as the
aforementioned distillation, but also crystallization and absorption
onto zeolites and into metal-organic crystals.5 We therefore also
investigated the possibility of using compound 3 as a host for the
separation of these compounds from one another, a result that may
have industrial application, and report our findings here. Although
3 is not a novel host compound, it has been described to have very
low inclusion ability, forming complexes with only a few hydro-
carbon guests, cyclohexane and rac-limonene being the only two
examples provided for compound 3 in the literature.6 Our research
team, however, has shown otherwise, that 3 has a significant af-
finity for hydrocarbon guests. Further to this, Toda et al.7 have
described the selective inclusion of para-xylene from a mixture of
meta- and para-xylene using 1,1,2,2-tetraphenylethane-1,2-diol as
the host compound. Though some structural similarities do exist
between their host and compound 3, significant differences, such as
the inherent chirality and longer butyl backbone of 3, suggest that
an investigation of the host properties of 3 is warranted.
determine whether inclusion had occurred or not and, if so, the
host:guest (H:G) ratios. Table 1 summarizes the results obtained.
Each of the alkyl benzenes were included in this way, and each
with the same H:G ratio (2:1).
2.2. Competition experiments
In order to determine whether there is a preference of host 3 for
any of these guests and whether, therefore, host/guest chemistry
may have potential industrial application to effect separation of the
C8 aromatic fraction of crude oil, a number of competition exper-
iments were conducted. Here, 0.1 g (0.2 mmol) of 3 was dissolved in
various equimolar combinations of the isomeric xylenes and eth-
ylbenzene, where each of the guests weighed 0.7 g (6.6 mmol). The
vials were treated similarly to the single solvent experiments
mentioned before. The solid that crystallized out was analysed
using a gas chromatograph coupled to a mass spectrometer (GC-
MS). Table 2 displays these results where the preferred guest and its
percentage in the various combinations are shown in bold italic
print for ease of viewing. Percentage standard deviations are given
in brackets (experiments were done in triplicate).
In addition, any complexes that formed successfully were
studied using single crystal X-ray data and thermal analyses and
the data are presented in this report.
It was surprising that, in the mixed clathrates (Table 2), the
overall H:G ratio remained 2:1 throughout, irrespective of the
structure of the included guests or how many different guests were
present. This is the same ratio that was preferred in the single
solvent recrystallizations.
From experiments involving equimolar binary combinations of
the three isomeric xylenes, it is clear that 3 has a strong preference
for para-xylene [30:70 (o:p) and 25.1:74.9 (m:p)]. In its absence, the
host discriminates against the meta-isomer [60:40 (o:m)]. The
result for the ternary equimolar mixture of these three guests is in
accordance with the preferences displayed in the binary experi-
ments. It is therefore possible to state with confidence that the
2. Results and discussion
TETROL 1 was prepared according to a published procedure,2
and treated with sodium hydride to effect deprotonation of the
secondary hydroxyl groups which were subsequently methylated
with methyl iodide in the next step. This afforded the title host
compound 3 in reasonable yield.
2.1. Formation of complexes
The affinity of host 3 was investigated for guests ortho-xylene,
meta-xylene, para-xylene and ethylbenzene. Approximately 0.1 g
(0.2 mmol) of 3 was dissolved in an excess of each of these four
alkyl aromatics in glass vials. The vials were very gently heated so
as to dissolve the host, and thereafter left to stand at ambient
temperature and pressure over a period of 12 h, whereupon a solid
crystallized out. These were filtered, rinsed with petroleum ether,
sucked dry in air, and subjected to 1H NMR spectroscopy to
selectivity of host
3 for the xylenes is in the order para-
xylene [ ortho-xylene > meta-xylene.
Considering the binary mixtures involving one of the xylenes
and ethylbenzene, once again, para-xylene is preferred when eth-
ylbenzene (EB) is present [60.7:39.3 (p:EB)]. However, in the
absence of para-xylene, ethylbenzene is always favoured [46.6:53.6
(o:EB) and 38.0:62.0 (m:EB)]. A competition experiment where 3
was recrystallized from an equimolar mixture of all four of these
guests reflects the same preferences as observed from all of the
previous experiments, and host 3 may be summed up to have a
selectivity for the components of the quaternary mixture in the
order para-xylene > ethylbenzene > ortho-xylene > meta-xylene.
Table 1
Host 3:guest ratios of complexes formed during individual
recrystallization experiments.a
Guest
Host 3:guest
o-Xylene
m-Xylene
p-Xylene
2:1
2:1
2:1
2:1
2.2.1. Thermal analyses
The relative thermal stabilities of the four complexes were
assessed by subjecting them to thermogravimetric analysis (TGA,
blue curve, Fig. 1) and differential scanning calorimetery (DSC,
green curve, Fig. 1) experiments in which each was heated at 10 ꢁC/
Ethylbenzene
a
Determined using 1H NMR spectroscopy.
Table 2
Competition experiments and H:G ratios obtained.a
o-Xylene
m-Xylene
p-Xylene
Ethylbenzene
Guest ratios (% Standard deviation)b
Overall H:G ratio
X
X
X
60.0:40.0 (1.5)
30.0:70.0 (6.4)
25.1:74.9 (2.1)
26.6:19.4:54.0 (0.5) (1.8) (1.9)
46.6:53.6 (2.1)
38.0:62.0 (0.8)
60.7:39.3 (1.5)
21.2:14.1:40.6:24.2 (3.0) (0.6) (4.5) (1.0)
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
a
Determined using GC-MS.
Experiments were carried out in triplicate.
b
Barton, Benita
