A comparison of the behaviour of two closely related xanthenyl-derived host compounds in the presence of vaporous dihaloalkanes

Article abstract of DOI:10.1007/s10847-018-0833-x

The behaviour of two closely related xanthone-derived host compounds, N,N’-bis(9-phenyl-9-xanthenyl)ethylenediamine and N,N′-bis(9-phenyl-9-thioxanthenyl)ethylenediamine, which formed complexes with CH₂Cl₂, CH₂Br₂ and CH₂I₂ after recrystallization from each of these solvents, was compared when subjected to these guest and guest mixtures in the vapour phase. Surprisingly, these hosts displayed entirely different behaviours under these conditions, with only the thioxanthenyl derivative possessing the ability to clathrate these guests (or guest mixtures) from the gas phase; this ability was entirely absent in the xanthenyl host. All novel complexes were subjected to single crystal diffraction analyses in order to investigate the interactions present, as well as thermal and Hirshfeld surface experiments. The host selectivity and host–guest interactions were correlated with the differences observed in the recrystallization and vapour experiments. Furthermore, data obtained for the novel complexes by employing various analytical techniques were related back to the observed selectivity order.

Full text of DOI:10.1007/s10847-018-0833-x

Journal of Inclusion Phenomena and Macrocyclic Chemistry
https://doi.org/10.1007/s10847-018-0833-x
ORIGINAL ARTICLE
A comparison of the behaviour of two closely related xanthenyl-
derived host compounds in the presence of vaporous dihaloalkanes
Lize de Jager1 · Benita Barton · Eric C. Hosten
1
1
Received: 21 June 2018 / Accepted: 11 August 2018
©
Springer Nature B.V. 2018
Abstract
The behaviour of two closely related xanthone-derived host compounds, N,N’-bis(9-phenyl-9-xanthenyl)ethylenediamine
and N,N′-bis(9-phenyl-9-thioxanthenyl)ethylenediamine, which formed complexes with CH Cl , CH Br and CH I after
2
2
2
2
2 2
recrystallization from each of these solvents, was compared when subjected to these guest and guest mixtures in the vapour
phase. Surprisingly, these hosts displayed entirely different behaviours under these conditions, with only the thioxanthenyl
derivative possessing the ability to clathrate these guests (or guest mixtures) from the gas phase; this ability was entirely
absent in the xanthenyl host. All novel complexes were subjected to single crystal diffraction analyses in order to investigate
the interactions present, as well as thermal and Hirshfeld surface experiments. The host selectivity and host–guest interactions
were correlated with the differences observed in the recrystallization and vapour experiments. Furthermore, data obtained
for the novel complexes by employing various analytical techniques were related back to the observed selectivity order.
Keywords Host–guest chemistry · Inclusion compounds · Alkyl halides · X-Ray crystallography · Thermal stability · Gas
absorption
Introduction
techniques such as nuclear magnetic resonance (NMR) spec-
troscopy, gas chromatography (GC) and single crystal X-ray
diffraction (SCXRD).
Host–guest chemistry is a field of chemistry devoted to
investigating the synthesis, properties and applications of
new and successful host molecules for the formation of
host–guest inclusion compounds. Inclusion compounds are
There exist numerous applications of host–guest chemis-
try: in the pharmaceutical industry, host compounds are used
to clathrate drug molecules to improve transport, activity,
resistance and solubility of such drugs, and to effect chi-
ral resolution of important pharmaceutical products. Other
applications include the removal of hazardous materials
from the environment, improving taste and stability of food
products, chromatography and asymmetric synthesis [1].
The storage of organic vapours by interstitial van der
Waals confinement is an important application of host
materials. The process of absorption of guests within these
is fundamental in many industrial applications. In heterog-
enous catalysis, for example, the catalysts are solids whilst
the great majority of reactants are gases or liquids [2, 3].
Absorption of guest molecules facilitates the trapping of
highly volatile guests by forming thermally stable inclu-
sion compounds, and the separation of mixtures of such
compounds is contingent on the selectivity of the host
material [4]. Absorption of guests is controlled by molec-
ular recognition and relies on non-covalent interactions
between host and guest under defined temperatures and
“
complexes” in which one chemical compound, the host,
forms a cage, cavity or channel in which another compound,
the guest, is located. The two compounds interact to form
one new inclusion compound which does not obey the
known definition of a complex, as there are no coordinate
covalent/dative bonds between the guest and the host [1].
Hosts, usually solids, can form inclusion compounds with
guests by simple recrystallization processes. The resultant
solid inclusion compounds are analyzed by means of various
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s10847-018-0833-x) contains
supplementary material, which is available to authorized users.
*
Lize de Jager
s213337665@mandela.ac.za
1
Department of Chemistry, Nelson Mandela University, PO
Box 77000, Port Elizabeth 6031, South Africa
Vol.:(0
1
123456789)
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
pressures, the time of exposure and the porosity of the host
material [2]. The most basic principle of organic solid-
state chemistry is that molecules will arrange in a manner
to balance the effects between optimization of space and
intermolecular interactions [4]. Amombo et al. [2] investi-
gated the guest exchange of tetrakis(4-halophenyl)ethylene
hosts and noted significant halogen bonding (d > 3 Å) as
the possible reason for complex stability, based on previ-
ous work that considered the size of the halogen atoms
Novel complexes were subjected to single crystal X-ray
diffraction (SCXRD), Hirshfeld and thermal analyses in
order to better understand the mode of guest inclusion,
the nature of host–guest interactions and relative complex
stabilities, and these observations were correlated with the
selectivity and inclusion ability of the two hosts. We report
these findings here.
[
5]. A calix[4]arene-derived host was explored by Atwood
Results and discussion
et al. [4] for its ability to capture small, highly volatile
molecules that formed disordered complexes due to the
guest’s loose “fit” and the presence of high crystallo-
graphic site symmetry; they noted that the more symmet-
rical guests were not disordered [4]. Ziganshin et al. [6]
studied the thermal stability, sorption capacity and change
of morphology of amino acid derivatives with arenes, lin-
ear alcohols, alkanes and nitriles. The inclusion ability
decreased with each added methylene group and, for the
alcohols, the difference in host selectivity was correlated
to the shape of the molecules, as linear n-propanol was
highly preferred. Moreover, hosts were observed to form
more stable clathrates at lower guest content.
Synthesis of hosts 1 and 2
A simple Grignard reaction using phenylmagnesium bro-
mide was carried out on xanthone and thioxanthone, inde-
pendently, and the resultant alcohol reacted with perchloric
acid to afford the perchlorate salt. Two of these xanthenyl
units were then bridged with ethylenediamine to form the
final product 1 and 2 (yield approx. 60–65%) [7].
Analysis of inclusion compounds
Complexation by recrystallization
In this present work, we report on the inclusion ability
and behaviour of host N,N′-bis(9-phenyl-9-xanthenyl)eth-
ylenediamine 1 in the presence of dihaloalkanes CH Cl
Compound 1 was dissolved in each of the respective dihalo
guest solvents. The resultant mixtures were left at ambient
temperature and pressure to facilitate crystallization. The
solid that formed was collected by vacuum filtration and
washed efficiently with petroleum ether (40–60 °C). The
crystals were analysed using proton NMR spectroscopy to
determine whether inclusion had occurred and, if so, the
host:guest (H:G) ratios were obtained by integration of rel-
evant host and guest resonances. Table 1 is a summary of
the results so-obtained.
2
2
(
DCM), CH Br (DBM) and CH I (DIM). We also syn-
2 2 2 2
thesised a closely related compound, N,N′-bis(9-phenyl-9-
thioxanthenyl)ethylenediamine 2, and compared the poten-
tial of both hosts for gaseous guest uptake, with surprising
observations despite the similarities in each of these host
materials. These hosts have never been assessed for their
vapour inclusion ability.
Host 1 successfully formed inclusion compounds with all
three dihaloalkanes, with DCM and its dibromo analogue
preferring 1:2 H:G ratios, while diiodomethane was enclath-
rated with a 1:1 H:G ratio. Interestingly, the structurally-
similar sulfur analogue 2 showed the host to prefer the 1:1
H:G ratio consistently with the three guests [7].
Competition experiments with equimolar mixtures
of guests by recrystallization
These experiments were carried out by dissolving host 1 in
binary and ternary combinations of the guests (equimolar
amounts). The crystallization vessels were closed and stored
at 0 °C to maintain the equimolar condition. The resultant
solids were processed and analysed in the same manner as
the individual inclusions. Table 2 shows a summary of the
results obtained, where italic bold font face was employed
for preferred guests for better clarity.
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Table 1 Host:guest (H:G) ratios
provided a host selectivity order of DBM (51.72%)>DCM
35.68%)>DIM (12.61%). Interestingly, a different selectiv-
ity order was observed for the sulfur host analogue [DBM
46%)>DIM (38%)>DCM (16%)] [7].
Guest (G)
H:G
1:2a
1:2
of complexes formed by host 1
(
DCM
DBM
DIM
a
(
1:1
1
Determined using
H-NMR
Competition experiments with varied molar ratios of guests
by recrystallization
spectroscopy with CDCl as the
3
solvent
a
NMR spectra were deposited
in the Supplementary Informa-
tion (S1a–c). Thermal analy-
ses showed that these inclu-
sion compounds were unstable,
which affected the integration
values of the NMR resonance
signals. The H:G ratios were,
however, confirmed by means
of SCXRD
These experiments involved preparing binary mixtures of
two guests but varying their molar ratios beyond equimolar
in order to determine whether the host selectivity changes
with guest concentration change. The series of competition
experiments in which pairs of guest molecules A and B co-
crystallize with host H may be characterized by:
H(ꢀ, s) + nA(l or g) + mB(l or g)∕H ⋅ A (s, β) + mB(l or g)
n
where H represents the apohost in its non-porous α-phase
which, when placed in contact with a mixture of the guests
A and B, selects A and forms a solid inclusion compound
Table 2 Results of competition experiments using host 1 and various
equimolar mixtures of the guests
H·A , the β-phase, and excludes B [8]. In practice, however,
n
DCM
DBM
DIM
Average guest ratios (%)
Overall
H:G
usually both guests are found in the host crystal.
ratio
A selectivity coefficient can be defined as
x
x
x
38.46:61.54
77.79:22.21
84.16:15.84
35.68:51.72:12.61
1:2
1:1
1:1
1:1
KA:B = Z ∕Z × X ∕X , where X + X = 1
A
B
B
A
A
B
x
x
x
where X is the mole fraction of guest A in the liquid
A
x
x
mixture and Z that of guest A enclathrated in the crystal
A
x
(
Fig. 1). Here, plot ‘i’ represents no selectivity and K
=
A:B
1
1, curve ‘ii’ results when A is preferentially enclathrated
with respect to B over the entire concentration range, while
curve ‘iii’ is the case where the host selectivity is guest-
concentration dependent [9].
Ratios determined using H-NMR spectroscopy with CDCl as the
3
solvent
Experiments were conducted in duplicate; % e.s.d.’s are provided in
the Supplementary Information (S2)
We therefore prepared binary mixtures with host 1
and guest 1(G1):guest 2(G2) ratios approximating 100:0,
Host 1 displayed selectivity for DBM when mixed in
equimolar proportions with DCM and DIM in the binary
experiments (61.54 and 84.16%, respectively, Table 2). In
the presence of both DCM and DIM, the former was pre-
ferred (77.79%), while an equimolar ternary experiment
8
0:20, 60:40, 50:50, 40:60, 20:80, 0:100, and the results
of these experiments are represented as the selectivity
profiles in Fig. 2a–c. The x- and y-axes are given as mole
fractions (X
and Z
, respectively), and the
dihaloalkane
dihaloalkane
Fig. 1 General selectivity
curves obtainable for binary
non-equimolar competition
experiments [6]
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Fig. 2 Selectivity profiles of
host 1 when recrystallized from
a DBM/DCM, b DBM/DIM
and c DCM/DIM
DBM VS DCM
(
a)
1
.9
.8
.7
.6
.5
.4
.3
.2
.1
0
0
0
0
0
0
0
0
0
0
0
0
0
0.2
0. 2
0. 2
0.4
0.6
0. 6
0. 6
0.8
0. 8
0. 8
1
1
1
X OF DBM
DBM VS DIM
(
b)
1
.9
.8
.7
.6
.5
.4
.3
.2
.1
0
0
0
0
0
0
0
0
0
0
0. 4
X OF DBM
DCM VS DIM
(
c)
1
.9
.8
.7
.6
.5
.4
.3
.2
.1
0
0
0
0
0
0
0
0
0
0
0. 4
X OF DCM
straight-line plot represents the line of no selectivity (K=1)
and is hypothetical.
the data points lie closer to the straight-line plot, which con-
firms the preference of dichloromethane over diiodometh-
ane when the second solvent is DBM. This correlates with
the observed host selectivity order in the equimolar solvent
experiments. Finally, and as expected, DCM was preferred
consistently relative to DIM (Fig. 2c).
Figure 2a, b clearly show that host 1 is selective for DBM
over the entire concentration range assessed, even at low
molar fractions of this guest in the recrystallizing liquid mix-
ture (<50%). Furthermore, Fig. 2b shows a significant devi-
ation of the experimental data points relative to the hypo-
thetical line of no selectivity compared with Fig. 2a, where
In comparison, host 2 was selective for DBM over the
entire concentration range, even at low concentrations of
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
this guest. In the absence of DBM, however, the host initially
showed selectivity for DIM at low concentrations, up until
effects using the numerical method implemented in SAD-
ABS [10] All non-hydrogen atoms were refined aniso-
tropically. Carbon–bound hydrogen atoms were added in
idealized geometrical positions in a riding model. For both
1·2DCM and 1·2DBM, the nitrogen–bound hydrogen atoms
are disordered over two positions which were located on
the difference Fourier map but refined riding. The solvent
molecules have positional disorder and were refined with the
carbon–halide bonds restrained to be the same length. These
crystallographic data were deposited at the Cambridge Crys-
tallographic Data Centre [1·2DCM (1824152) and 1·2DBM
(1824153)]. The crystal structure of the inclusion compound
1·DIM could not be determined due to poor crystal quality.
The relevant crystallographic data for these experiments
are summarized in Table 3.
6
6%, and beyond this point the host preferred DCM [7].
In order to understand the observed selectivity order, we
analysed suitable crystals by means of SCXRD in order to
establish the nature and type of intermolecular interactions
present between host and guest.
Single crystal X-ray diffraction analyses
X-Ray diffraction studies were performed at 200 K using a
Bruker Kappa Apex II diffractometer with graphite mono-
chromated Mo Kα radiation (λ=0.71073 Å). APEXII [10]
was used for data collection and SAINT [10] for cell refine-
ment and data reduction. The structure was solved using
SHELXT-2014 [11] and refined by least-squares proce-
dures using SHELXL-2017/1 [11] with SHELXLE [12] as
a graphical interface. Data were corrected for absorption
Both inclusion compounds display isostructural host
packing, crystallizing in the triclinic P-1 crystal system with
very similar unit cell dimensions.
Table 3 Crystallographic data
for 1·2DCM and 1·2DBM
1
·2DCM
1·2DBM
Chemical formula
Formula weight
Crystal system
Space group
µ (Mo-Kα)/mm−
a/Å
C H N O ·2CH Cl
C H N O ·2CH Br
2
4
0
32
2
2
2
2
40 32
2
2
2
906.87
Triclinic
P-1
0.366
8.8213 (4)
8.8657 (4)
920.37
Triclinic
P-1
4.415
8.8555 (4)
8.8981 (4)
1
b/Å
c/Å
Alpha/°
Beta/°
Gamma/°
V/Å
Z
13.5055 (6)
73.237 (2)
72.035 (2)
66.861 (2)
906.87 (7)
1
13.5996 (6)
72.633 (2)
72.998 (2)
66.578 (2)
919.83 (7)
1
3
F (000)
Temp./K
Restraints
Nref
386
200
6
4501
458
200
6
4561
Npar
251
251
R
wR
S
0.0467
0.1369
1.03
0.0387
0.1013
1.05
2
θ min–max/°
Tot. data
1.6, 28.3
24,198
4501
1.6, 28.4
28,553
4561
Unique data
Observed data
3477
3360
[
I>2.0 sigma (I)]
Rint
0.023
0.999
−0.38
0.34
0.028
1
−0.60
0.37
Dffrn measured fraction θ full
Min. resd. dens. (e/Å )
Max. resd. dens. (e/Å )
3
3
The 1·DIM complex was characterized by poor crystal quality and could not be analysed by SCXRD
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Figure 3 shows unit cells of the complexes, where the host
is represented by ball-and-stick and the guests with space-fill
representation. The crystal packing of the two unit cells are
clearly isostructural.
present in 1·2DBM. This observation may be the reason for
the observed selectivity of host 1 for DBM (Table 2; Fig. 2a).
In addition, the 1·2DCM complex is held in the crystal by
means of only four short contacts (Table 4), whereas the
1·2DBM experiences a further two of these. The complex
of host 2 with DBM (the preferred guest) also experienced
a greater number of host–guest interactions compared with
the other two guests [7]. Furthermore, its diiodomethane
complex was stabilized by guest–guest interactions owing to
these molecules being in close proximity to one another as a
consequence of their increased molecular size.
The experimental powder pattern obtained when employ-
ing powder X-ray diffraction (PXRD) on the DIM complex
was compared to the calculated powder patterns of the other
two complexes (using the Mercury software). As expected,
the host packing in the DIM complex was not isostructural
with the other dihalo-analogue complexes since clear differ-
ences between these patterns were observed (Supplementary
Information S3).
We subsequently carried out Hirshfeld surface analyses
to elucidate, quantitatively, the intermolecular interactions
in the crystal.
From the SCXRD data, we obtained the significant host
̶
guest interactions present in the complexes, and these are
summarized in Table 4.
The two inclusion compounds experience no π–π, CH–π
or classical hydrogen bonding interactions (Table 4). While
the complex 1·2DCM does not display non-classical hydro-
gen bonding interactions, two of these interaction types are
Hirshfeld surface analysis
Hirshfeld surfaces describe the immediate environs of mol-
ecules. Figure 4a–d are depictions of the two-dimensional
fingerprint plots which were derived from the three-dimen-
sional Hirshfeld surfaces generated using Crystal Explorer
1
7 [13]. These surfaces were obtained using data files (.cif
files) from the crystal structure analyses. (Note that all sur-
faces were generated around the guest molecules.) Here,
de and di correspond to the distances to the nearest atom
outside and inside the surface, respectively. Figure 5 illus-
trates the comparison of the percentage of the appropriate
intermolecular interactions (G⋯H/H⋯G). Since the guests
in both 1·2DCM and 1·2DBM showed disorder, care had to
be taken and Hirshfeld surfaces were mapped for both major
and minor components.
Unfortunately, these Hirshfeld surface analyses did not
provide any information which could be used to explain the
selectivity order of the host. X–H and H–H interactions were
most predominant, where X represents the halogen atoms, as
expected since hydrogen atoms are on the periphery of the
molecules and are therefore more likely to experience these
interaction types.
We subsequently investigated the nature of the guest
accommodation.
Guest accommodation
The nature of the guest accommodation is shown in Fig. 6,
and these voids were calculated using the Mercury software
after the guests had been removed from the packing calcula-
tion [14]. Since 1·2DCM and 1·2DBM were isostructural,
only the voids for the chloro derivative were calculated, as
representative example.
In both complexes, the guest is accommodated in infinite
multidirectional channels. Interestingly and in direct contrast
to this host, the thio derivative 2 accommodated its guests
as pairs in discrete cavities [7]. Therefore, significant host
Fig. 3 Unit cells for a 1·2DCM and b 1·2DBM; isostructural host
packing is evident
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Table 4 Significant host–guest interactions for 1·2DCM and 1·2DBM
Non-covalent interaction
1·2DCM
1·2DBM
Symmetry operator
π–π
None present
None present
None present
None present
None present
Non-classical
CH⋯π (host–guest)
H-bonding (host–guest)
3
.50 Å, 144°
C–H⋯Br)
.77 Å, 145°
C–H⋯Br)
x, 1+y, z
(
3
−x, 1−y, 1−z
(
Short contacts (host/guest and
a,b
guest/guest)
2
.77 Å, 1113° (<)
(host Ar)C–C⋯H–C(guest 2)]
.77 Å, 107° (<)
(host Ar)C–C⋯H–C(guest 1)]
.82 Å, 156° (<)
(host NH)N–H⋯Cl–C(guest 2)]
.80 Å, 145° (<)
(guest 2)N–H⋯Cl–C(host Ar)]
x, y, z
[
2
1−x, 1−y, 1−z
1+x, y, z
[
2
[
2
1−x, 1−y, 1−z
x, y, z
[
2
.70 Å, 116° (<<)
(host Ar)C–C⋯H–C(guest 2)]
.90 Å, 154° (<)
(host NH)N–H⋯Br–C(guest 1)]
.91 Å, 145° (<)
(host CH )C–H⋯Br–C(guest 2)]
[
2
1−x, y, z
[
2
−x, 1−y, 1−z
1−x, 1−y, 1−z
−1+x, y, z
x,−1+y, z
[
2
2
.83 Å, 143° (<)
(guest 2)C–H⋯C–C(host Ar)]
.87 Å, 137° (<)
(guest 1)C–Br⋯H–N(host NH)]
.91 Å, 160° (<)
(guest 1)C–Br⋯H–C(host CH )]
[
2
[
2
[
2
a
Distances denoted by <are contacts that measure less than the sum of the van der Waals radii of the atoms involved while those denoted by
<
<is this sum minus 0.2 Å
b
Two-fold disorder is evident, and guest 1 is the major component and guest 2 the minor component, respectively
behaviour differences are clearly evident even though the
two hosts are so similar in structure.
temperature for the guest release process and is estimated
from the DTG traces in Fig. 7a–c.
It has been reported that higher thermal stabilities are
associated with complexes in which guests are accommo-
dated in discrete cavities compared with instances where
guests are found in channels [15]. We therefore conducted
thermal analyses of the three complexes with 1 in order to
determine whether this was indeed the case here.
Figure 7a, b show that the DBM and DCM complexes
are unstable at room temperature and pressure since these
complexes experienced mass loss from the outset of these
experiments as is observed in the two relevant TG traces.
However, the diiodomethane complex is significantly more
stable than the latter two: when 1·DIM was heated, mass
loss was only experienced at approximately 81.5 °C.
Thermal analysis
The T is the temperature at which the first of the guest
on
is released from the host crystal and T the boiling point of
b
Thermal analyses were carried out on each of the three com-
plexes of host 1, and thermogravimetric (TG), the deriva-
tive of the TG (DTG) and differential scanning calorimetry
pure guest solvent. The value of T is estimated from the
on
derivative of the TG curve. Taking into account these T
on
values for the three guest release processes, the DIM com-
plex is the most stable, followed by complexes contain-
ing DBM and DCM. These data do not correlate with the
selectivity order observed for host 1, which was also the
case for the thioxanthenyl-derived host compound 2 [7].
(
DSC) (overlaid) traces are provided in Fig. 7a–c. The tem-
perature range used was 20–250 °C, and the heating rate
was 10 °C/min.
The relevant thermal data from these traces have
been summarized in Table 5. The term T is the onset
on
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Fig. 4 Two-dimensional fingerprint plots for a 1·2DCM major component, b 1·2DCM minor component, c 1·2DBM major component, and d
1
·2DBM minor component
What is, however, evident is that complexes with 1:1
Vapour inclusion
H:G ratios had higher relative thermal stabilities than
1
·2DCM and 1·2DBM. Furthermore, all three complexes
Vapour inclusion has, to date, not been experimentally inves-
tigated for hosts 1 and 2, and we therefore carried out appli-
cable experiments in order to assess the behaviour of these
hosts in the presence of gaseous guests.
with 2, with H:G ratios of 1:1, were also thermally stable
at room temperature [7] These results correlate exactly
with the findings of Zinganshin et al. [6] who reported that
hosts form more stable inclusion compounds when their
guest content is lower (e.g., 1:1 H:G ratios) relative to
complexes with higher guest content (e.g., 1:2 H:G ratios),
as in the present study.
In closed glass vials, the crystalline host compound 1 and
2 (0.5 mmol) were separately suspended on a plastic tray
well above each of liquid phase guests (30 mmol) (Fig. 8). In
this way, the solid host was effectively subjected to vaporous
guest. The vials were fitted with lids and parafilmed to allow
the gaseous guest to saturate the vessel. In addition, a similar
experiment was conducted but using an equimolar mixture
of the three guests (10 mmol each). The crystals were treated
Vapour inclusions experiments were subsequently con-
ducted with both hosts to determine whether the host behav-
iour is dependent on the method employed for complex
formation.
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Fig. 5 Summary of the percent-
age and type of interactions
experienced by the guests in
HIRSHFELD SURFACE ANALYSIS SUMMARY
1
·2DCM and 1·2DBM
DCM major
DCM minor
DBM major
DBM minor
X - C
X - H
C - H
X - X
H - H
INTERACTION
Fig. 6 Channel occupation for
guests in 1·2DCM, as repre-
sentative example
and analysed as before after monitoring the progress via pro-
ton NMR over several days (1–31 days).
accordance with the volatility of the three guests as DCM is
the most volatile, followed by DBM and DIM.
Host 1 did not include any of the three guests over the
allocated time (1–31 days) period. Additionally, this host
also displayed no inclusion ability in the presence of the
mixture of gaseous guests. Surprisingly, and in direct con-
trast, host 2 possessed the ability to absorb guests from the
gaseous phase, and Fig. 9 illustrates the results obtained
when this host was subjected to these gaseous guests. The
y-axis indicates the percentage of guest inclusion that was
calculated using the integration of applicable resonances
In the mixed guest experiment (Fig. 9b), 2 initially
selected for the guests in the order DCM > DBM > DIM,
according to guest volatility once more. After 3 days, how-
ever, a guest exchange was observed to occur, and while
the DBM percentage remained relatively constant (since the
3-day analysis), DCM was exchanged by DIM molecules.
This observation correlated with the recrystallization experi-
ments (DBM>DIM>DCM) [7] where 2 showed increased
selectivity for DIM relative to DCM as the amount of DCM
decreased from 45 to 35%, while that of DIM increased from
10 to 23%. The overall H:G ratio remained 1:1 throughout
the entire experiment.
1
from the H-NMR spectra, and the x-axis displays the
amount of time the host was subjected to these gases.
From Fig. 9a, it is clear that DCM was included with
a H:G ratio of 1:1 after only 6 h, and this ratio remained
relatively consistent until 54 h had lapsed. DBM uptake
was much slower, with a 1:1 ratio being observed at 24 h,
while DIM only reached 67% after 54 h. These results are in
The fact that host 1 did not include any guest from the
gas phase was thought-provoking. Host 2 is generally less
selective than host 1, and guests taken up by 2 experi-
ence fewer host–guest interactions compared with host 1
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Fig. 7 TG, DTG and DSC traces for a 1·2DCM, b 1·2DBM and c 1·DIM
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Table 5 Thermal properties of the complexes of host 1 with the three
Conclusion
dihalomethanes
Guest (G)
T /°C
Mass loss
Actual mass
loss meas-
ured/%
In this investigation, the inclusion ability of N,N′-bis(9-
phenyl-9-thioxanthenyl)ethylenediamine 2, from a previous
report, was compared with that of the oxygen derivative,
N,N′-bis(9-phenyl-9-xanthenyl)ethylenediamine 1. Both
compounds included the dihalomethanes CH Cl , CH Br
on
expected/%
a
DCM
DBM
DIM
21.0
24.2
81.5
22.9
37.8
32.6
12.8
32.3
a
2
2
2
2
and CH I , but while 2 preferred a 1:1 H:G ratio consist-
28.8
2 2
ently, 1 complexed with CH Cl and CH Br with 1:2 ratios,
2
2
2
2
a
The inclusion compounds with DCM and DBM were unstable at
room temperature; therefore, the observed mass loss is significantly
lower than that expected due to loss of some guest during sample
preparation
and only CH I experienced a 1:1 ratio. Equimolar binary
2
2
and ternary competition experiments were correlated to
results from single solvent experiments. For both hosts, the
bromine derivative was the preferred guest, and SCXRD
showed that this guest experienced a significantly larger
number of host–guest interactions, in accordance with the
selectivity order. Whilst 2 accommodated its guests consist-
ently in discrete cavities, with two guests found in each, the
CH Cl and CH Br guests in complexes with 1 were always
[
7]. Therefore it is plausible that guests are readily taken
up from the gas phase by the less-discerning 2 compared
with 1 (which did not take up guest from the gas phase).
Furthermore, we observed that host 2 recrystallized from
all three dihaloalkanes to form stable inclusion complexes
with a H:G ratio of 1:1 and, in contrast, host 1 only formed
one stable complex, that with DIM. It is thus conceivable
that host 2 successfully included these guests from the
vapour phase since the resulting complexes were stable.
The experiment of this host with a mixture of guests was
initially affected by volatility and, after a period of time,
the selectivity of the host became more prominent, and
guest exchange was observed to occur, with the more pre-
ferred guest (DIM) being absorbed in favour of the less
favoured one (DCM)—this process was possibly facili-
tated by the fact that the host packing in all the complexes
was isostructural. Host 1, on the other hand, did not form
complexes with the preferred guests DBM and DCM when
these were in vapour phase and this may be as a result
of the fact that the resultant complexes would have been
unstable, and hence inclusion was not favoured. However,
this host formed a stable inclusion compound with DIM in
the recrystallization process but was possibly unsuccess-
fully absorbed owing to the low preference of this host for
DIM (DBM > DCM > DIM).
2
2
2
2
found in multidirectional and infinite channels. For both host
materials, Hirshfeld surface analyses and relative complex
stabilities (based on onset temperatures) did not correlate
with the selectivity orders. Vapour inclusion experiments
indicated that 1 did not clathrate any of the three guests
from this phase, whilst 2 clathrated all three, with the host
selectivity correlating exactly with guest volatility, initially,
and then, later, the preference mimicked the selectivity order
from the recrystallization experiments. Complex stabilities
explained these observations.
Experimental
General methods
Melting points were recorded on an Electrothermal IA9000
Series digital melting point apparatus and are uncorrected.
Infrared spectra were recorded on a Bruker Tensor 27 Infrared
1
spectrometer (analysing with OPUS software), and H-NMR
13
and C-NMR spectra on a Bruker Ultrashield Plus 400 MHz
Fig. 8 Vapour inclusion experimental setup
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
Fig. 9 Graphical representation
(
a)
TIME INCLUSION STUDY FOR 2 WITH INDIVIDUAL GUESTS
of the inclusion behaviour of
%DBM
%DCM
%DIM
2
with a independent gaseous
guests, and b an equimolar
1
00
mixture of the three guests
90
80
70
60
50
40
30
20
10
0
0
10
20
30
HOURS
40
50
TIME INCLUSION STUDY OF 2 WITH MIXTURE OF GUESTS
DCM %DBM %DIM
(
b)
%
5
4
4
3
3
2
2
1
1
0
5
0
5
0
5
0
5
0
5
0
0
5
10
15
20
25
30
35
DAYS
spectrometer. Thermal experiments were conducted using a
TA SDT Q600 Module system and analysed using TA Univer-
sal Analysis 2000 data analysis software. Samples were placed
in open platinum pans with an empty platinum pan functioning
as a reference. High purity nitrogen gas was used as purge gas
in both cases. PXRD experiments were carried out using a
Bruker D2 PHASER X-ray diffractometer. The resultant trace
has been placed in the Supplementary Information. All start-
ing materials and guest solvents were purchased from Sigma
Aldrich, South Africa.
Synthesis of N,N′‑bis(9‑phenyl‑9‑xanthenyl)
ethylenediamine 1
9-Phenylxanthen-9-ol (Y=O) (Supplementary Information,
S4) [16]
Magnesium turnings (0.78 g, 32 mmol), bromobenzene
(5.40 g, 34 mmol) and xanthone (5.39 g, 27.5 mmol)
afforded 9-hydroxy-9-phenylxanthene (5.12 g, 74.7%) as a
cream solid, m.p. 160–162 °C (lit., [16] 159 °C); ν(film)/
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
−
1
−1
cm 3294 (OH) and 1582 (Ar); δ (CDCl , 400 MHz)/ppm
°C (lit., [16] 239 °C); ν(film)/cm 1448 (Ar); δ (CDCl ,
H
3
H
3
2
4
1
.67 (s, 1H, OH) and 7.07–7.45 (m, 13H, ArH); δ (CDCl
400 MHz)/ppm 7.3–8.8 (m, 13H, ArH).
C
3,
00 MHz)/ppm 70.5 (COH), 116.4 (ArC), 123.6 (ArC),
26.2 (ArC), 126.8 (ArC), 127.2 (ArC), 128.0 (ArC), 129.0
N,N′-Bis(9-phenyl-9-thioxanthenyl)ethylenediamine
(Supplementary Information, S9) [16]
(
ArC), 129.1 (ArC), 148.0 (quaternary ArC) and 149.7 (qua-
ternary ArC).
Perchlorate salt (3.58 g, 9.6 mmol) and ethylenediamine
(0.322 g, 5.36 mmol) afforded N,N′-bis(9-phenyl-9-
xanthenyl)ethylenediamine (1.9009 g, 58.72%) as a white
9
(
-Phenylxanthen-9-ylium perchlorate (Y=O),
Supplementary Information, S5) [16]
−1
solid, m.p. 174–176 °C (lit., [16] 174–175 °C); ν(film)/cm
Perchloric acid (6.60 mL) and xanthenol (5.00 g, 8.2 mmol)
afforded 9-phenylxanth-9-ylium perchlorate (6.08 g,
3056 (CH) and 1432(Ar); δ (CDCl , 400 MHz)/ppm 2.48
H 3
(2H, broad s, NH), 2.48 (4H, s, CH ) and 7.19–7.47 (m,
2
9
3.7%) as a bright yellow solid, m.p. 284–286 °C (lit., [16]
26H, ArH); δ (CDCl , 400 MHz)/ppm 44.4 (CH ), 66.2
C
3
2
−1
2
80–281 °C); ν(film)/cm 1595 (Ar); δ (CDCl , 400 MHz)/
(Ph–C–NH), 125.8 (ArC), 126.0 (ArC), 126.8 (ArC), 126.9
(ArC), 128.0 (ArC), 129.8 (ArC), 131.6 (ArC), 137.9 (qua-
ternary ArC) and 146.5 (quaternary ArC).
H
3
ppm 7.8–8.5 (m, 13H, ArH).
N,N′-Bis(9-phenyl-9-xanthenyl)ethylenediamine
Supplementary Information, S6) [16]
Acknowledgements Financial support is acknowledged from Nelson
Mandela University and the National Research Foundation (NRF). Mr
L. Bolo and Mr K. Ngobo are acknowledged for thermal analyses and
Prof M. Caira for his input in the crystal packing discussion.
(
Perchlorate salt (3.00 g, 8.4 mmol) and ethylenedi-
amine (0.30 g, 5.0 mmol) afforded N,N′-bis(9-phenyl-
9
-xanthenyl)ethylenediamine (1.54 g, 58.7%) as a white
solid, m.p. 202–203 °C (lit., [16] 204–206 °C); ν(film)/
References
−
1
cm 3019 (CH) and 1573 (Ar); δ (CDCl , 400 MHz)/ppm
H
3
2
.32 (2H, broad s, NH), 2.25 (4H, s, CH ) and 7.02–7.43
1. Atwood, J.L., Steed, J.W., Encyclopedia of supramolecular chem-
istry. CRC Press, New York (2004)
2
(
(
(
m, 26H, ArH); δ (CDCl , 400 MHz)/ppm 43.4 (CH ), 59.9
C 3 2
2
.
Amombo Noa, F.M., Bourne, S.A., Su, H., Nassimbeni, L.R.:
Guest exchange in halogenated host–guest compounds: structures
and kinetics. Cryst. Growth Des. 16(3), 1636–1642 (2016)
Ph–C–NH), 116.2 (ArC), 123.3 (ArC), 125.7 (ArC), 126.5
ArC), 127.2 (ArC), 128.0 (ArC), 128.2 (ArC), 129.0 (ArC),
1
49.9 (quaternary ArC) and 151.3 (quaternary ArC).
3. Bönnemann, H.: 279 pages;(hardback). ISBN-978-3 527-31824-
7
. Gadi Rothenberg. Catalysis—concepts and green Applications
Wiley-VCH 2008. Appl. Organomet. Chem. 22(7), 412–412
Synthesis of N,N′‑bis(9‑phenyl‑9‑thioxanthenyl)
ethylenediamine 2
(
2008)
4
5
.
.
Atwood, J.L., Barbour, L.J., Jerga, A.: Storage of methane
and freon by interstitial van der Waals confinement. Science
2
96(5577), 2367–2369 (2002)
9
-Phenylthioxanthen-9-ol (Y=S) (Supplementary
Information, S7) [16]
Amombo Noa, F.M., Bourne, S.A., Nassimbeni, L.R.: Halogen
bonding in host–guest compounds: structures and kinetics of
enclathration and desolvation. Cryst. Growth Des. 15(7), 3271–
3279 (2015)
Magnesium turnings (1.82 g, 75 mmol), bromobenzene
6
7
.
.
Ziganshin, M.A., Gubina, N.S., Gerasimov, A.V., Gorbatchuk,
V.V., Ziganshina, S.A., Chuklanov, A.P., Bukharaev, A.A.: Inter-
action of L-alanyl-L-valine and L-valyl-L-alanine with organic
vapors: thermal stability of clathrates, sorption capacity and the
change in the morphology of dipeptide films. Phys. Chem. Chem.
Phys. 17(31), 20168–20177 (2015)
(
11.1 g, 70.7 mmol) and thioxanthone (15.00 g, 75 mmol)
afforded 9-hydroxy-9-phenylxanthene (5.12 g, 74.74%) as a
white solid, m.p. 105–107 °C (lit., [16] 105–106 °C); ν(film)/
−
1
cm 3294 (OH) and 1443 (Ar); δ (CDCl , 400 MHz)/ppm
H
3
2
4
1
.89 (s, 1H, OH) and 7.05−8.07 (m, 13H, ArH); δ (CDCl
C 3,
Barton, B., de Jager, L., Hosten, E.C.: An investigation of the
complexation of host N,N′-bis (9-phenyl-9-thioxanthenyl) ethyl-
enediamine with dihaloalkane guests. J. Incl. Phenom. Macrocycl.
Chem. 89, 1–12 (2017)
00 MHz)/ppm 77.1 (COH), 126.2 (ArC), 126.6 (ArC),
26.7 (ArC), 127.0 (ArC), 127.4 (ArC), 127.8 (ArC), 128.1
(
ArC), 131.6 (ArC), 140.0 (quaternary ArC) and 143.4 (qua-
8
9
.
.
Nassimbeni, L.R., Marivel, S., Su, H., Weber, E.: Inclusion of
picolines by a substituted binaphthyl diol host: selectivity and
structure. RSC Adv. 3(48), 25758–25764 (2013)
ternary ArC).
9
(
-Phenylthioxanthen-9-ylium perchlorate (Y=S),
Supplementary Information, S8) [16]
Pivovar, A.M., Holman, K.T., Ward, M.D.: Shape-selective sepa-
ration of molecular isomers with tunable hydrogen-bonded host
frameworks. Chem. Mater. 13(9), 3018–3031 (2001)
1
1
0. Bruker, A.: APEX2, SAINT and SADABS. Bruker AXS Inc.,
Madison (2007)
Perchloric acid (4 ml, 66.5 mmol) and thioxanthenol (3 g,
0.3 mmol) afforded 9-phenylxanth-9-ylium perchlorate
3.58 g, 93%) as a bright yellow solid, m.p. 230–235 °C
1
1. Sheldrick, G.M.: SHELXT–Integrated space-group and crystal-
structure determination. Acta Crystallogr. A 71(1), 3–8 (2015)
(
1
3

Journal of Inclusion Phenomena and Macrocyclic Chemistry
1
1
1
2. Hübschle, C.B., Sheldrick, G.M., Dittrich, B.: ShelXle: a Qt
the visualization and investigation of crystal structures. J. Appl.
Crystallogr. 41(2), 466–470 (2008)
graphical user interface for SHELXL. J. Appl. Crystallogr. 44(6),
1
281–1284 (2011)
15. Tanaka, K., Hori, K., Tsuyuhara, S., Motoki, S., Shide, S.,
Arakawa, R., Caira, M.R.: Role of halogen bonding in clathrate
formation of tetra-and hexa-salicylides derived from halogenated
salicylic acids. Tetrahedron 69(3), 1120–1127 (2013)
3. Wolff, S., Grimwood, D., McKinnon, J., Turner, M., Jayatilaka,
D., Spackman, M.: Crystal explorer ver. 3.1. University of West-
ern Australia, Perth, Australia (2013)
4. Macrae, C.F., Bruno, I.J., Chisholm, J.A., Edgington, P.R.,
McCabe, P., Pidcock, E., Rodriguez-Monge, L., Taylor, R.,
Streek, J.v., Wood, P.A.: Mercury: CSD 2.0—new features for
16. Barton, B.: Host-guest chemistry: the synthesis and assessment of
host compounds based on the 9-arylxanthenyl and related systems.
PhD, University of Port Elizabeth, Port Elizabeth (1997)
1
3