Cyclohexanone-Driven Discriminatory Behavior Change of Host

Nov 9, 2017 - Cyclohexanone-Driven Discriminatory Behavior Change of Host Compound (+)-(2R,3R)-TETROL for Isomeric Methylcyclohexanone Guests ... *E-m...
1 downloads 14 Views 1MB Size
Subscriber access provided by READING UNIV

Article

Cyclohexanone-driven discriminatory behaviour change of host compound (+)-(2R,3R)-TETROL for the isomeric methylcyclohexanone guests Benita Barton, Sasha-Lee Dorfling, and Eric C. Hosten Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.7b01334 • Publication Date (Web): 09 Nov 2017 Downloaded from http://pubs.acs.org on November 10, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Crystal Growth & Design is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

1 Cyclohexanone-driven discriminatory behaviour change of host compound (+)-(2R,3R)TETROL for the isomeric methylcyclohexanone guests

Benita Barton,* Sasha-Lee Dorfling and Eric C. Hosten

Department of Chemistry, PO Box 77000, Nelson Mandela University, Port Elizabeth, 6031, South Africa. E-mail: [email protected]

ABSTRACT

Using appropriate competition experiments, host compound TETROL [(+)-(2R,3R)-1,1-4,4tetraphenylbutane-1,2,3,4-tetraol] was revealed to discriminate between the isomeric methylcyclohexanone guests; the host selectivity was in the order 2- >> 3- > 4methylcyclohexanone.

Surprisingly, addition of unsubstituted cyclohexanone to these

competitions instigated a complete reversal in the host’s preference for the alkylcyclohexanones: a selectivity order of cyclohexanone > 4- > 3- > 2- methylcyclohexanone was attained. It is proposed that it is the significant preference of TETROL for cyclohexanone that is responsible for the host behaviour change in the presence of this guest. As evidence, the

crystal

packing

of

a

mixed

complex,

TETROL∙(72%)cyclohexanone∙(28%)4-

methylcyclohexanone, was shown to be isostructural with pure TETROL∙cyclohexanone (which differs from TETROL∙4-methylcyclohexanone).

The addition of cyclohexanone to the

competition experiments therefore steers crystallization towards crystals with similar packing to that of the preferred guest complex accounting for TETROL's change in selectivity in the presence of this guest.

Hirshfeld surface analyses were used to demonstrate that

TETROL∙cyclohexanone experiences a significantly larger number of stabilizing O···H/H···O interactions (15.2%) compared with the other complexes (11.5−14.4%), accounting for the host's overwhelming preference for this guest. Thermogravimetric analysis further confirmed that cyclohexanone was more strongly bound in the crystal than the other guests.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

2 KEYWORDS

Host-guest

Chemistry;

Inclusion;

Complexes;

Selectivity;

Cyclohexanone;

Methylcyclohexanones; X-Ray Crystallography; Supramolecular Chemistry.

1. INTRODUCTION

The separation of isomeric compounds in a mixture is usually an arduous task owing to the similarity in the physical properties, such as boiling and freezing points, of the component isomers. As a result, fractional distillations or recrystallizations of such mixtures are tedious, time-consuming and costly, and hence alternative methods to achieve these separations become attractive. It is well known that host compounds are able to display selectivity in the presence of guests with similar structural features.1−3 As an example, Barbour et al4 have demonstrated that the octahedral nickel complex, Ni(NCS)2(para-phenylpyridine)4, may be used to separate each of the xylene isomers by means of solid-vapour sorption of these guests, which follows the work of Alaerts et al5 who used a microporous vanadium(IV) terephthalate material in a chromatographic separation of these same guests. Nassimbeni and his research team very recently explored the selective enclathration by hosts comprising the fluorenyl moiety for methyl- and dimethyl- piperidine guests,6 while Mitra and coworkers7 focused their research efforts on the synthesis of molecular organic cages for the separation of mesitylene from its structural isomer 4-ethyltoluene simply as a result of size and shape differences between these two molecules. The cyclodextrins are another class of highly discriminatory hosts, finding applications in widely varied fields of analytical chemistry, and are efficient at forming reversible inclusion complexes and are therefore able to recognize guests selectively.8

During the course of our investigations into the inclusion ability of TETROL [(+)-(2R,3R)-1,14,4-tetraphenylbutane-1,2,3,4-tetraol] 1, we discovered that cyclohexanone and the three isomeric methylcyclohexanones were all enclathrated when crystals of this host compound were grown from each of these guest solvents; the host:guest ratio was consistently 1:1 (Scheme 1). An analysis of the single crystal X-ray diffraction data for each complex showed that, surprisingly, the 3- and 4- methylcyclohexanones were included exclusively in their

ACS Paragon Plus Environment

Page 3 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

3 energetically unfavourable axial conformations, while the 2-methyl analogue was enclathrated as the more conventional equatorial conformer, and we subsequently reported these results.9,10

In that investigation, we did not explore the selectivity of TETROL for the four cyclohexanones when this host was recrystallized from various mixtures of these guests, an aspect of TETROL that has not been considered before. We therefore recently pursued this particular avenue and were surprised by the results obtained, where the presence of cyclohexanone in the competition experiments completely reversed the host's selectivity for the three alkylcyclohexanones compared with competitions where cyclohexanone was absent. We explored the reasons for these observations by considering the results of single crystal X-ray diffraction (SCXRD) and thermogravimetric analyses and report our findings here. We thus obtained the crystal structure for a mixed complex containing cyclohexanone and 4methylcyclohexanone and discuss some interesting characteristics pertaining to this inclusion complex and relate these findings back to the unusual host behaviour change.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 22

4 Scheme 1. Host and Guest Compoundsa

HO

OH

O

O

Cy 154 156 C

2-MeCy 162 163 C

O

O

3-MeCy 169 170 C

4-MeCy 169 171 C

OH

HO

1

a1

= (+)-(2R,3R)-1,1-4,4-tetraphenylbutane-1,2,3,4-tetraol (TETROL). Cy = cyclohexanone.

2MeCy = 2-methylcyclohexanone. 3MeCy = 3-methylcyclohexanone. 4MeCy = 4methylcyclohexanone. Boiling points of the cyclohexanones are also provided below each structure.

2. EXPERIMENTAL SECTION

General. The host melting point was recorded on an Electrothermal IA9000 Series digital melting point apparatus and is not corrected. The infrared spectrum was recorded on a Bruker Tensor 27 Platinum ATR system and analysed using Opus version software. 1H- and 13C-

NMR spectra were obtained using a Bruker Ultrashield Plus 400 MHz spectrometer and

examined using TopSpin 3.2 software. The optical rotation was measured using an A. Krüss Optronic polarimeter (Germany) furnished with a sodium lamp. GC-MS experiments were carried out on an Agilent 7890A gas chromatograph fitted with an Agilent 5975C VL mass spectrometer and a DB-WAX column (30 m). The method involved an initial 1 min hold time at 55°C, followed by a ramp of 1 °C/min until 150 °C was reached, and then a hold time at this

ACS Paragon Plus Environment

Page 5 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

5 temperature for 10 min. This was followed by a ramp of 1 °C/min until 160 °C was reached, and then a hold time at this temperature for 10 min, and lastly a ramp of 20 °C/min until 220 °C was reached, and then a final hold time at this temperature for 10 min. The split ratio was 2:1 and inlet temperature 250 °C.

Materials. The host compound, (+)-TETROL 1, was synthesized using a standard Grignard addition reaction of PhMgBr (≥ 6 molar equiv.) to optically pure diethyl tartrate, according to a published procedure.11 After an aqueous acidic workup, an orange gum was obtained which was crystallized and recrystallized from CH2Cl2/hexane/MeOH to afford (+)-(2R,3R)-1,1,4,4tetraphenylbutane-1,2,3,4-tetrol 1 as a white solid (45 %), mp 147-149 °C (lit.,12 mp 150−151 °C); [α]23D +166 (c = 9.32, CH2Cl2) {lit.,12 [α]25D +154 (c = 1.2, CHCl3)}; νmax(solid)/cm-1 3440 (br, OH), 3294 (br, OH), 3057 (Ar), 3033 (Ar), 1598 (Ar) and 1494 (Ar); δH(CDCl3) 3.86 (2H, d, 2COH), 4.44 (2H, d, 2HCOH), 4.72 (2H, s, 2CPh2OH) and 7.2-7.4 (20H, m, Ar); δC(CDCl3) 72.11 (HCOH), 81.71 (CPh2OH), 124.97 (Ar), 126.05 (Ar), 127.15 (Ar), 127.27 (Ar), 128.10 (Ar), 128.37 (Ar), 128.55 (Ar), 130.08 (Ar), 143.85 (quaternary Ar) and 144.16 (quaternary Ar). The (+)diethyl tartrate and bromobenzene were purchased from Sigma-Aldrich and used as received.

Formation of inclusion complexes of host 1 with each of the cyclohexanones. TETROL 1 (0.3 mmol) was dissolved independently in an excess (10−15 mmol) of each of the four cyclohexanone guests, which were also obtained from Sigma-Aldrich and used without further purification. Dissolution was ensured by heating the mixtures at 75 °C using a hot water bath. These experiments were conducted in glass vials which were left open to the ambient atmosphere to facilitate evaporation of some of the solvent, after which crystallization ensued and whereupon the vials were closed and left overnight to enforce further crystallization. The crystals were recovered by means of vacuum filtration and, in order to remove superficial guest solvent, carefully washed with small quantities of petroleum ether (40−60 °C) followed by ethanol. The compounds were analysed using 1HNMR spectroscopy with CDCl3 as the NMR solvent. The recovery of host from the solutions in this way ranged between 60 and 72%.

Competition experiments.

TETROL (0.3 mmol) was dissolved and recrystallized from

equimolar amounts (5 mmol each) of binary, ternary and quaternary mixtures of the four

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 22

6 cyclohexanones. In order to maintain these equimolar conditions, the vials were closed and stored at 0 °C. After crystallization had occurred (1−5 days), the crystals were recovered and treated in an identical manner to the single solvent experiments mentioned previously. However, 1H-NMR spectroscopy was not a suitable technique to analyse these mixed complexes due to resonance overlap of the various guest signals in the crystals [see Supporting Information, Figure a(i) and a(ii)].

GC-MS was therefore selected as the

appropriate method for analysis, with dichloromethane being used to dissolve each sample after crystal recovery.

[A GC-MS trace showing its suitability for the separation and

quantification of these guests has been deposited in the Supporting Information (Figure b).]

Molar ratios of binary mixtures of the guests were also varied beyond equimolar [the mol% ratios that were used were approximately 10:90, 20:80, 30:70, 40:60, 50:50, 60:40, 70:30, 80:20 and 90:10, for guest 1:guest 2 (G1:G2), respectively] and the host (0.3 mmol) recrystallized from each of these mixtures. Vials were treated identically to the binary experiments just mentioned, and both the mother liquor from which crystallization had occurred and the crystals were analysed by GC-MS (dichloromethane served once more as the solvent). Data obtained in this manner were used to construct selectivity curves.

X-Ray crystallography. Each of the complexes of host 1 with the four cyclohexanones were subjected to single crystal X-ray diffraction experiments, and the experimental conditions for these have been published in a prior report,10 and are therefore not furnished here for the sake of brevity. These structures were deposited at the Cambridge Crystallographic Data Centre [CCDC 989251 (1∙Cy), 989081 (1∙2MeCy), 989004 (1∙3MeCy) and 1007403 (1∙4MeCy)]. In addition, crystals that resulted from the recrystallization of TETROL from an equimolar binary mixture of Cy and 4MeCy were suitable for SCXRD analysis and the crystal structure is reported here. This experiment was conducted at 200 K using a Bruker Kappa Apex II diffractometer with graphite-monochromated Mo Kα radiation (λ = 0.71073 Å). APEXII was used for data collection, and SAINT for cell refinement and data reduction, respectively.13 SHELXT-201414 was used to solve the structures, and these were refined by least-squares procedures using SHELXL-2016;14 here, SHELXLE15 served as a graphical interface. All non-hydrogen atoms were refined anisotropically, and the hydrogen atoms were added in idealized geometrical positions in a riding model. Data were corrected for

ACS Paragon Plus Environment

Page 7 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

7 absorption effects using the numerical method implemented in SADABS.13 The H atoms of the hydroxyl groups were allowed to rotate with a fixed angle around the C−O bonds to best fit the experimental electron density (HFIX 147 in the SHELX program suite 14). The structure has also been deposited at the Cambridge Crystallographic Data Centre (CCDC-1541279).

Thermogravimetric analyses. These experiments were carried out according to a published procedure.9,10

3. RESULTS AND DISCUSSION

The host compound, (+)-TETROL 1, was readily prepared by reacting optically pure diethyl tartrate with an excess of phenylmagnesium bromide in anhydrous THF. A moderate yield of 45% was obtained.11 Recrystallizing this material from each of Cy, 2MeCy, 3MeCy and 4MeCy resulted in 1:1 host:guest (H:G) inclusion complexes in each instance, as observed from the relative integrals of relevant host and guest resonances of their respective 1H-NMR spectra [see Supporting Information, Figures c(i)−c(iv)].

Competition experiments. Since each guest is enclathrated individually, the selectivity of 1 for these guests was consequently investigated in order to establish if the host would discriminate between them.

Here we considered mixtures of the three isomeric

methylcyclohexanones (Table 1) independently from those having cyclohexanone present (Table 2).

Therefore, recrystallizing TETROL 1 from the various equimolar binary and ternary combinations of 2MeCy, 3MeCy and 4MeCy (see Experimental Section), and analysing the soformed crystals using GC-MS with dichloromethane as the solvent, allowed for the population of Table 1, which shows the relative amounts of each guest in the crystals from each experiment. [The experiments were carried out in triplicate, and an average ratio is given here, with estimated percentage standard deviations (e.s.d.'s) provided in parentheses. The raw data has been deposited in the Supporting Information, Table a.] The preferred guest species is given in bold red italic font face.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

8 Table 1. Competition experiments of TETROL in the presence of equimolar mixtures of the isomeric methylcyclohexanonesa 2-MeCy

3-MeCy

X

X

X

X aExperiments

4-MeCy

X X

X

X

X

Guest ratios (%e.s.d.) 89 : 11 (1.37) 95 : 5 (2.03) 26 : 74 (0.74) 79 : 14 : 7 (4.54) (3.87) (0.95)

were carried out in triplicate, and an average ratio is provided here with %

e.s.d.'s in parentheses. From Table 1, it is clear that 2-MeCy was considerably preferred in equimolar binary competition studies whenever it was present (89 and 95% were found in the host crystals when TETROL was recrystallized from equimolar mixtures of 2-/3-MeCy and 2-/4-MeCy, respectively).

In the absence of 2-methylcyclohexanone, the 3-methyl derivative was

discriminated against in favour of 4-MeCy (26 versus 74%).

The equimolar ternary

experiment involving the three methylcyclohexanones yielded unsurprising results in that 2MeCy remained the preferred isomer (79%), but its presence prompted a selectivity switch where 1 now displayed a higher selectivity for 3-MeCy (14%) than the 4-methyl analogue (7%). The host's selectivity for these isomers may therefore be written as in the order 2MeCy >> 3-MeCy > 4-MeCy.

ACS Paragon Plus Environment

Page 9 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

9 Table 2. Competition experiments of TETROL in the presence of equimolar mixtures of cyclohexanone and the isomeric methylcyclohexanonesa Cy

2-MeCy

X

X

X

3-MeCy

X

X

X

X

X

X

X

X X aAn

Guest ratios (%e.s.d.) 84 : 16 (1.39) 75 : 25 (1.15) 68 : 32 (0.87) 67 : 13 : 21 (0.58) (0.52) (0.06) 61 : 10 : 29 (1.26) (2.30) (1.05) 60 : 15 : 25 (1.11) (0.33) (1.40) 52 : 5 : 13 : 30 (0.67) (0.53) (1.14) (2.31)

4-MeCy

X

X X X

X

X

X

average of three experiments are provided here; e.s.d.'s are in parentheses; the raw data

is also obtainable in the Supporting Information, Table a.

Table 2 displays the results obtained when similar equimolar binary, ternary and quaternary competitions were carried out but with unsubstituted cyclohexanone present. instance, this guest was now the preferred one.

In each

From a consideration of the binary

experiments comprising Cy and one of each of the alkylcyclohexanones, it is clear that the preference for the alkylcyclohexanones decreases in the order 4-MeCy > 3-MeCy > 2-MeCy (32, 25 and 16%, respectively), while an equimolar quaternary experiment furnished a host selectivity order of Cy (52%) > 4-MeCy (30%) > 3-MeCy (13%) > 2-MeCy (5%). The ternary experiments not yet discussed correlate exactly with these orders (Table 2).

We also considered binary guest (G1 and G2) competition experiments, where the guests were present in unequal molar amounts, by sequentially decreasing the molar concentration of one guest relative to the other in the mixture from which 1 was recrystallized (see Experimental section). Selectivity curves could thus be constructed and these are provided in Figures 1a−f, six in all to account for all permutations. Here, the mole fraction (Z) of G1 present in the crystals recovered from the experiment was plotted against the mole fraction of G1 (X) in the liquid mixture of G1 and G2. Analyses were carried out using the GC-MS method as before.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Page 10 of 22

10 Figure 1. Selectivity Curves for a) Cy/2-MeCy, b) Cy/3-MeCy, c) Cy/4-MeCy, d) 2-MeCy/3-MeCy, e) 2-MeCy/4-MeCy and f) 3-MeCy/4-MeCya

(a)

(b)

(c)

ACS Paragon Plus Environment

Page 11 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Crystal Growth & Design

11 (d)

(e)

(f)

aIt

was not possible to carry out these experiments in duplicate, let alone triplicate, owing to the impracticality of preparing solutions with

identical molar amounts of each guest; it is also not standard practise to do so.16

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

12 In each of these figures, the straight line plot (orange data points) is a theoretical one, inserted for ease of comparison with the experimentally-determined curves (blue data points), and represents the case where the host is completely unselective towards both guests (i.e., the host selects an amount of G1 and G2 in direct accordance with the amount of G1 and G2 present in the liquid mixture).

The curve obtained for Cy/2-MeCy (Figure 1a) shows that the selectivity of the host is guestconcentration dependent: at low concentrations of Cy (up to approximately 10%), the host does not display significant selectivity for either guest but, after this point, 2MeCy is favoured until a Cy concentration of 30% is reached (where the selectivity constant16 K = 1). Thereafter, Cy remains preferred. Figures 1b and c also suggest a preference for Cy over 3MeCy and 4MeCy, respectively, but this is consistent for the entire concentration range, while similar observations can be made for the 2MeCy/3MeCy selectivity study (Figure 1d) but where 3MeCy is always discriminated against in favour of 2MeCy. In the presence of both 2MeCy and 4MeCy (Figure 1e), the host selects more of 4MeCy at high concentrations of this isomer while at low concentrations, 2MeCy is more prevalent in the crystals; for this mixture, K = 1 at approximately 44% 2MeCy. Finally, in an experiment using 3MeCy and 4MeCy (Figure 1f), the host consistently discriminates against 3-MeCy.

It was rather surprising to observe that the addition of cyclohexanone (clearly a highly favoured guest species) to the competition experiments comprising the isomeric methylcyclohexanones completely reversed the selectivity of the host for these guests relative to those competitions where cyclohexanone was absent. This feature of the present work is unprecedented in our laboratories. One example is the results of experiments we conducted on a derivative of TETROL with guests xylenes and ethylbenzene, where the host selectivity order remained constant, irrespective of whether the favoured guest, para-xylene, was present or not.17 We were therefore determined to uncover the reasons for these observations, and consequently analysed the data obtained from SCXRD experiments.

Single crystal X-ray diffraction. The crystal structures for the four complexes have been published by our team on a prior occasion [CCDC 989251 (1∙Cy), 989081 (1∙2MeCy), 989004 (1∙3MeCy) and 1007403 (1∙4MeCy)].10 The 1∙Cy and 1∙3MeCy complexes are isostructural

ACS Paragon Plus Environment

Page 13 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

13 and crystallize in the monoclinic crystal system and P21 space group while the 1∙2MeCy (orthorhombic, P212121) and 1∙4MeCy (triclinic, P1) differ from these and one another. Furthermore, both of the chiral 2- and 3-methyl analogues are disordered over two positions [s.o.f. 0.645(6) and 0.778(3), respectively] owing to the presence of the two enantiomers, with the (R)-isomer being favoured in both instances.

A pair of 1,3-intramolecular hydrogen bonds maintains the geometry of each host molecule, and guests are held in the crystal by means of (guest)C=O∙∙∙H−O(host) hydrogen bonds involving only secondary host hydroxyl groups, as illustrated in the stereoview provided in Figure 2 (left) for the 1∙Cy complex as a representative example. (Host)O∙∙∙O(guest) distances range between 2.705(19)−3.230(4) Å with associated angles 141−167°. Furthermore, all guests are accommodated in discrete cavities in each crystal (Figure 2, right).

Figure 2. Stereoview (left) showing hydrogen bonding and discrete cavity guest packing mode (right) of the 1∙Cy complex as a representative example

Since it can be difficult to visualise the many contacts in these kinds of structures, we considered Hirshfeld surface analyses using the Crystal Explorer software package18 to allow at least a quantitative comparison between them. This software is useful for exploring modes of packing as well as host−guest intermolecular interactions and does so via Hirshfeld surfaces, which are surfaces that display the immediate environment of a molecule.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 22

14 Figure 3. Two dimensional fingerprint plots obtained from Hirshfeld surfaces for a) 1∙Cy, b) 1∙(R)-2MeCy, c) 1∙(S)-2MeCy, d) 1∙(R)-3MeCy, e) 1∙(S)-3MeCy and f) 1∙4MeCy

(a)

(d)

(b)

(e)

(c)

(f)

After generating the Hirshfeld surfaces, the three dimensional topographies were translated into two dimensional fingerprint plots, given in Figure 3, where de and di are distances to the nearest atom outside and inside the surface, respectively. The 'spikes' (1) and 'wings' (2,3) represent O∙∙∙H, H∙∙∙H and H∙∙∙C interactions. Figure 4 compares the percentage of the intermolecular interactions (G∙∙∙H/H∙∙∙G) present in each complex graphically while Table 3 lists these values numerically.

ACS Paragon Plus Environment

Page 15 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

15 Figure 4. Graphical display showing the percentage intermolecular forces of each type for a) 1∙Cy, b) 1∙(R)-2MeCy, c) 1∙(S)-2MeCy, d) 1∙(R)-3MeCy, e) 1∙(S)-3MeCy and f) 1∙4MeCy

Table 3. Percentage intermolecular interactions in each inclusion complex (G···H/H···G)

aOnly

Complex

O···H/H···O

H···H

H···C/C···H

O···Ca

C···C

O···O

1·Cy

15.2

57.2

26.5

0.5

0.6

0

1·(R)-2MeCy

14.4

62.4

22.1

0.9

0

0.1

1·(S)-2MeCy

13.8

62.2

22.8

1.1

0

0.1

1·(R)-3MeCy

11.5

62.1

25.8

0.2

0.4

0

1·(S)-3MeCy

12.3

61.7

25.6

0.5

0

0

1·4MeCy

14.1

58.5

26.7

0.2

0.4

0

G∙∙∙H interactions were observed.

The predominant interactions are H∙∙∙H in nature (57.2−62.4%), while all molecules also experience a number of H∙∙∙C interactions (22.1−26.7%, Figure 4, Table 3). However, what is more striking upon analysis of these data is that the 1·Cy complex experiences significantly more O···H/H···O interaction types (15.2% compared with the second most prevalent, 1·(R)2MeCy, with 14.4%) which is in direct accordance with the observation that the host consistently favours Cy.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

16 We subsequently grew host crystals from an equimolar mixture of Cy and 4MeCy. The crystals were of suitable quality for SCXRD, and the selected crystal was determined to have a s.o.f. for the major component (Cy) of 0.723 and the minor component 0.277 (4MeCy) which is in agreement with the GC-MS experiment discussed previously (Table 2). [In order to further confirm that this 0.72Cy:0.28(4MeCy) ratio is a bulk property rather than that of an individual crystal, an 1H-NMR spectrum was also obtained (in addition to the GC-MS) for the mixed complex, which showed a Cy:4MeCy ratio of 71:29, which correlates with the SCXRD result; this aspect of the investigation was deposited in the Supporting Information, Appendix a. Furthermore, a powder pattern was obtained for this mixed clathrate and compared with the pattern generated from the single crystal .cif file; the two patterns also correlate and have been deposited in the Supporting Information, Appendix b.] The structure was subsequently deposited at the Cambridge Crystallographic Data Centre (CCDC-1541279). Figure 5 shows a stereoview of the host and guests' orientations in the unit cell.

Figure 5. Stereoview of the host and guests' orientations in the unit cell of the Cy:4MeCy (72%:28%) mixed complex

It is clear that both of the guests occupy the same site in the crystal, Cy 72% and the 4MeCy 28% of the time, according to their s.o.f.'s. In fact, both of their orientations are identical, with each cyclohexyl ring carbon atom exactly overlapping, so much so that it appears as though only 4MeCy is present (Figure 5). The reasons provided for the axial orientation of the methyl-substituted guest reported previously was that this conformation was stabilized by means of two (guest)C−H∙∙∙π(host) and four (guest)H∙∙∙CAr interactions. More specifically, the methyl group experiences two interactions of the (guest methyl)H∙∙∙CAr type and, in the current investigation, this is also the case, despite the different packing systems (2.86 and

ACS Paragon Plus Environment

Page 17 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

17 2.77 Å, 152 and 153°, respectively). This explains why the axial orientation of the alkyl group is maintained in the mixed complex as well. The hydrogen bonds assisting in retaining these guests in the crystal measure 2.711(3) Å [(host)O∙∙∙O(guest)] with (host)O−H∙∙∙O(guest) angles of 150°. Crystal data and refinement parameters are listed in Table 4. Here we also include data from our previous work for the individual enclathrations of cyclohexanone and 4methylcyclohexanone for ease of comparison.10

Table 4. Crystal structure data for the complexes formed by host 1 with Cy,10 4MeCy10 and a 72%:28% mixed Cy:4MeCy complex Chemical formula Formula weight Crystal system Space group µ (Mo Kα)/mm–1 a/Å b/Å c/Å α/° β/° γ/° V/Å3 Z F (000) Temp (K) Restraints Nref Npar R1 wR2 S ϴ min–max/° Tot. data Unique data Observed data [I>2.0σ(I)] Rint Diffrn measured fraction ϴ full Min. resd. dens. (e/Å3) Max. resd. dens. (e/Å3)

1˖Cy C28H26O4·C6H10O 524.63 monoclinic P21 0.084 12.5944(4) 8.1531(2) 13.4570(5) 90 94.025(2) 90 1378.40(8) 2 560 200 1 6501 356 0.0388 0.1045 1.04 1.5, 28.3 13 345 6501 5556

1˖4MeCy C28H26O4·C7H12O 538.65 Triclinic P1 0.084 8.181(2) 9.952(3) 10.163(3) 79.296(6) 68.813(5) 65.825(5) 703.2(3) 1 288 173 3 9403 366 0.0515 0.1338 0.94 2.2, 27.1 9415 9403 5729

1˖(0.72)Cy˖(0.28)4MeCy C28H26O4·C6.28H10.55O 528.52 monoclinic P21 0.083 12.6617(6) 8.1716(4) 13.4898(7) 90 93.883(2) 90 1392.54(12) 2 565 200 1 6643 366 0.0440 0.1232 1.03 1.5, 28.3 37634 6643 5452

0.017 1.000

0.000 0.952

0.022 1.000

-0.21 0.22

-0.25 0.26

-0.23 0.37

A significant detail arising from a close analysis of the data in Table 4 is that the 1∙Cy and 1∙(0.72)Cy∙(0.28)4MeCy mixed complex are isostructural, crystallizing in the monoclinic crystal system and P21 space group. The complex containing only 4MeCy, on the other hand, has a different packing (triclinic, P1). As we know, Cy is the overwhelmingly favoured guest,

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 22

18 as observed from the competition experiments, and it appears as though, because of this fact, the host compound is encouraged to crystallize in the crystal system of the preferred Cy guest despite the presence of 4MeCy. The presence of Cy therefore appears to steer the host packing towards the monoclinic crystal system (P21) and it is plausible that this packing type is favoured for the mixed complex because the preferred guest is then able to experience a larger number of stabilizing O···H/H···O interactions (Figure 4, Table 3). This is quite possibly what occurs in all of the competitions where Cy is present, and explains the change in the host selectivity for the alkylcyclohexanones in the presence of this guest.

We subsequently removed only the Cy guest in the 1∙(0.72)Cy∙(0.28)4MeCy mixed complex from the Hirshfeld surface determination and recalculated the 2D fingerprint plot; the procedure was then repeated by removing only the 4MeCy guest. The two plots are provided in Figures 6a and b.

Figure 6.

Hirshfeld fingerprint plots for the a) 1∙(0.72)Cy∙(0.28)4MeCy complex after

removal of Cy guest, and the b) 1∙(0.72)Cy∙(0.28)4MeCy complex after removal of 4MeCy guest from the surface calculations

(a)

(b)

Figure 6a is very similar indeed to Figure 3a, which is not surprising: the Cy guest molecules in the mixed complex are expected to experience similar host−guest intermolecular interaction types compared with those in the pure 1∙Cy complex owing to the identical packing mode in each of these complexes. On the other hand, one would expect the fingerprint plot for 4MeCy (Figure 6b) in the mixed complex to differ from that for the 1∙4MeCy (Figure 3f)

ACS Paragon Plus Environment

Page 19 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

19 complex because the packing modes in the two complexes are not the same, and this is indeed the case, especially in the shapes of the wings (2) in these plots.

Thermogravimetric analyses. Thermal experiments were conducted on the four complexes and the results of these reported in previous manuscripts.9,10 From these plots, it was clear that Cy was held more tightly in the crystal relative to the other three guests, as witnessed by the higher temperatures associated with its release, and this is due to the increased number of stabilizing O···H/H···O interactions experienced by this host-guest complex (Hirshfeld surface analyses). These data further explain the high affinity for Cy demonstrated by TETROL, forcing the host behaviour change witnessed earlier in the presence of this guest species. (In Supporting Information, Appendix c, an overlay of the four TGA traces has been provided to illustrate this fact.)

4. CONCLUSIONS

In this work, we have demonstrated that the selectivity order of TETROL for the three isomeric methylcyclohexanones in competition experiments was reversed in the presence of cyclohexanone compared with when this guest was absent. The reason proposed for this unusual host behaviour was that cyclohexanone, as a guest species, was overwhelmingly favoured over any of the alkylcyclohexanones, thus altering the host behaviour whenever this guest was present. As evidence, single crystal X-ray data showed that in a mixed complex of 1∙(0.72)Cy∙(0.28)4MeCy, the mode of packing was isostructural with the pure 1∙Cy complex (whose packing differed from pure 1∙4MeCy), and it was suggested that the presence of cyclohexanone encourages the host to pack in the same mode as for the favoured guest species, cyclohexanone. This was energetically favourable since cyclohexanone experiences a higher number of stabilizing O···H/H···O interactions in the crystal, as demonstrated by Hirshfeld 2D fingerprint plots.

Thermogravimetric analyses from earlier reports further

confirmed the observation that cyclohexanone is much more tightly bound in the crystal than the other guests, and hence the host's consistent preference for it.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 22

20 SUPPORTING INFORMATION

The Supporting Information contains 1H-NMR spectra, GC-MS traces, the raw data for Tables 1 and 2, as well as further NMR data that provides evidence that the mixed complex (078)Cy:(0.22)4MeCy ratio is a bulk property and not just that of a single crystal. Powder patterns and TG plots may also be found here. The crystallographic data has been deposited at the Cambridge Crystallographic Data Centre, and these data may be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif; CCDC-989251 (1∙Cy), 989081 (1∙2MeCy), 989004 (1∙3MeCy), 1007403 (1∙4MeCy) and 1541279 (1∙ (0.78)Cy∙(0.22)4MeCy).

Acknowledgements

Financial support is acknowledged from the Nelson Mandela University. L. Bolo is thanked for thermogravimetric analyses, and Professor M. Caira (University of Cape Town) for informative informal discussions.

ACS Paragon Plus Environment

Page 21 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

21 REFERENCES

1.

Toda, F.; Bishop, R., Eds. Separations and Reactions in Organic Supramolecular Chemistry; Wiley: Chichester, 2004.

2.

Gokel, G. Crown Ethers & Cryptands. Monographs in Supramolecular Chemistry; Royal Society of Chemistry: Cambridge, 1991.

3.

Seebach, D.; Beck, A. K.; Heckel, A. Angew. Chem. Int. Ed. 2001, 40, 92.

4.

Lusi, M.; Barbour, L. J. Angew. Chem. Int. Ed. 2012, 51, 3928.

5.

Alaerts, L.; Kirschhock, C. E. A.; Maes, M.; van der Veen, M. A.; Finsy, V.; Depla, A.; Martens, J. A.; Baron, G. V.; Jacobs, P. A.; Denayer, J. F. M.; de Vos, D. E.; Angew. Chem. Int. Ed. 2007, 46, 4293.

6.

Sykes, N. M.; Su, H.; Weber, E.; Bourne, S. A.; Nassimbeni, L. R. Cryst. Growth Des. 2017, 17 (2), 819.

7.

Mitra, T.; Jelfs, K. E.; Schmidtmann, M.; Ahmed, A.; Chong, S. Y.; Adams, D. J.; Cooper, I. A. Nat. Chem., 2013, 5, 276.

8.

Szente, L.; Szemán, J. Anal. Chem., 2013, 85 (17), 8024.

9.

Barton, B.; Caira, M. R.; Hosten, E. C.; McCleland, C. W.; Weitz, S. Chem. Commun. 2014, 50, 13353.

10.

Barton, B.; Caira, M. R.; Hosten, E. C.; McCleland, C. W.; Weitz, S. J. Org. Chem. 2015, 80, 7184.

11.

Barton, B.; Caira, M. R.; Hosten, E. C.; McCleland, C. W. Tetrahedron 2013, 69, 8713.

12.

Shan, Z.; Hu, X.; Zhou, Y.; Peng, X.; Li, Z. Helv. Chim. Acta, 2010, 93, 497.

13.

APEX2, SADABS and SAINT; Bruker AXS: Madison, Wisconsin, USA, 2010.

14.

Sheldrick, G. M. Acta Crystallogr. 2015, C71, 3–8.

15.

Hübschle, C. B.; Sheldrick, G. M.; Dittrich, B. J. Appl. Crystallogr. 2011, 44, 1281

16.

Nassimbeni, L. R.; Marivel, S.; Su, H.; Weber, E. RSC. Adv., 2013, 3, 25758.

17.

Barton, B.; Hosten, E. C.; Pohl, P. L. Tetrahedron, 2016, 72, 8099.

18.

Wolff, S. K.; Grimwood, D. J.; McKinnon, J. J.; Jayatilaka, D.; Spackman, M. A. Crystalexplorer

17.5,

University

of

Western

(hirshfeldsurface.net).

ACS Paragon Plus Environment

Australia,

Perth,

2007.

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 22

22 FOR TABLE OF CONTENTS USE ONLY

Cyclohexanone-driven discriminatory behaviour change of host compound (+)-(2R,3R)TETROL for the isomeric methylcyclohexanone guests

Benita Barton,* Sasha-Lee Dorfling and Eric C. Hosten

Host compound TETROL [(+)-(2R,3R)-1,1-4,4-tetraphenylbutane-1,2,3,4-tetraol] discriminates between the isomeric methylcyclohexanone guests, with a host selectivity order of 2- >> 3- > 4- methylcyclohexanone. However, upon addition of unsubstituted cyclohexanone to these competitions, a complete reversal in the host’s preference for the alkylcyclohexanones is observed.

The significant preference of TETROL for cyclohexanone is proposed to be

responsible for this host behaviour change.

ACS Paragon Plus Environment