WO2020053589A1

WO2020053589A1 - Organic cage microparticle dispersions for gas absorption

Info

Publication number: WO2020053589A1
Application number: PCT/GB2019/052557
Authority: WO
Inventors: Andrew Cooper; Rebecca GREENAWAY; Aiting KAI; Benjamin EGLESTON; Stuart James
Original assignee: University of Liverpool; Queens University of Belfast
Current assignee: University of Liverpool; Queens University of Belfast
Priority date: 2018-09-13
Filing date: 2019-09-13
Publication date: 2020-03-19
Anticipated expiration: 2021-03-13
Also published as: GB201814881D0

Abstract

This invention relates to adispersion ofporous particles comprisingorganic cage molecules, the porous particles beingdispersed in a liquid phase, wherein the liquid is selected from the group consisting of liquid oligomers, size-excludedliquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof. The invention also relates to amethod of adsorbing a gas into a liquid, comprising at least the step of bringing the gas into contact with thedispersion. In addition, the invention includes a method for preparing thedispersion comprising at least the step of:mixing (i) porous particles comprisinganorganic cagemolecule,and (ii) a liquid selected from the group consisting of liquid oligomers, size-excludedliquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof. The invention also relates to an assemblage of the dispersion, the organic cage molecule comprising a cavity and a gas contained within the cavity.

Description

ORGANIC CAGE MICROPARTICLE DISPERSIONS FOR GAS ABSORPTION

[001] This invention relates to dispersions of porous particles comprising organic cage molecules, the porous particles being dispersed in a liquid phase, as well as to methods of preparing such dispersions. The invention also relates to a method of adsorbing a gas into a liquid, comprising at least the step of bringing the gas into contact with the dispersions. In addition, the invention relates to an assemblage of such a dispersion, the organic cage molecule comprising a cavity and a gas contained within the cavity.

[002] Background

[003] Porous solids such as zeolites are useful in molecular separation due to their permanent porosity. Porous solid adsorbents have significant advantages, for instance in terms of lower energy penalties in adsorption-desorption cycles when compared with their liquid counterparts, but they are difficult to incorporate into conventional flow processes.

[004] Liquid phases for the dissolution of gases are known. Solutions of various amines in water, or other solvents, are known to dissolve CO2 and are applied industrially in natural gas“sweetening”. However, these methodologies comprise the use of toxic materials; are corrosive towards steel, which limits their uses industrially; and require large amounts of energy to regenerate. They are also non-specific and therefore cannot be used for specific or targeted gas separation.

[005] Porous liquids (liquids with permanent porosity) for use in molecular separation have subsequently been developed. These porous liquids have been categorised into three different types ( Chem . Eur. J., 2007, 13, 3020) as follows:

• Type 1 - neat liquid hosts comprising molecules having an internal cavity, and that the molecules cannot collapse or interpenetrate,

• Type 2 - host molecules having an internal cavity, the molecules being dissolved in a solvent that cannot occupy the host’s cavities, and • Type 3 - particles of a host molecular framework dispersed in a liquid that cannot occupy the host’s cavities.

Thus, each type of porous liquid comprises a“host” having a cavity into which, for example, gas molecules could be absorbed.

[006] Porous liquids having improved properties, particularly high gas uptake, are sought.

[007] Statement of invention

[008] According to one aspect of the invention, there is provided a dispersion of porous particles comprising organic cage molecules, the particles being dispersed in a liquid phase, wherein the liquid is selected from the group consisting of liquid oligomers, size-excluded liquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof.

[009] Generally, a dispersion is a system in which particles are dispersed in a continuous phase of a different composition. The term “dispersion” is used in relation to the invention to refer to a system in which particles of a porous solid are dispersed in a liquid phase or medium. The dispersion may optionally comprise additives, such as surfactants, in order to increase the stability of the dispersion. Such additives are known to those skilled in the art.

[0010] The term“organic cage molecule” is used in relation to the invention to refer to cage molecules which do not include a metal atom.

[0011] In particular, the porous particles may be microparticles and/or nanoparticles. More particularly, the porous particles may be microparticles. Microparticles are generally defined as particles having a mean diameter in the range 0.1-100 pm (ie 100-100,000 nm). More particularly, the porous particles may have a mean diameter in the range 0.1-2 pm (ie 100-2000 nm). Nanoparticles are generally defined as particles having a mean diameter in the range 1-100 nm. [0012] More particularly, the porous particles may be microporous (ie that they have a mean pore diameter of less than 2 nm).

[0013] In particular, the porous particles comprising the organic cage molecules may comprise co-crystals of two chiral forms, or enantiomers, of the same organic cage molecule. In alternative embodiments, the porous particles comprising the organic cage molecules may comprise at least 90 wt%, more particularly at least 95 wt%, of a single chiral form, or enantiomer, of the organic cage molecule. In other embodiments, the porous particles comprising the organic cage molecules may comprise co-crystals of more than one type of organic cage molecules, for example, two or three different types.

[0014] In particular, the porous particles comprising the organic cage molecules may comprise co-crystals of:

(i) the same organic cage molecule, but in different homochiral forms - for example CC3-S/CC3-R as included, another example is FT-RCC3- R/FT-RCC3-S ( CrystEngComm , 2017, 19, 4933),

(ii) two different organic cage molecules (i.e. binary systems) - these can be selected to have complimentary chirality but differ based on their functionalisation, derivative, or topology - examples include CC1/CC3- and CC1/CC4- (Angew. Chem. Int. Ed. 2012, 51 , 7154); FT-RCC3-R/CC1 , FT-RCC3-R/CC2-S, and FT-RCC3-R/CC4-S 0 CrystEngComm , 2017, 19, 4933); CC3- /CC2-S and CC13/CC3- (, Nanoscale , 2017, 9, 6783); CC3-S/CC15- (ACS Cent. Sci. 2017, 3, 734); TCC2-R/CC3-S and TCC1-R/CC3-S (Nat. Chem., 2017, 9, 17); CC3- /CC19-S (Angew. Chem. Int. Ed., 2018, 57, 11228); and

(iii) three different organic cage molecules (i.e. ternary systems) - examples include CC1/CC3- /CC4- (Angew. Chem. Int. Ed. 2012, 51 , 7154) and CC3-S/CC4-S/CC13 (Nanoscale, 2017, 9, 6783).

[0015] In this way, the properties of the cage can be controlled, for example allowing porosity to be tuned for certain guest molecules. More particularly, the two chiral forms of the same organic cage molecule may be CC3-R and CC3-S, which can be formed using ( , )-1 ,2-diaminocyclohexane and (S,S)-1 ,2-diaminocyclohexane respectively. [0016] In particular, the porous particles comprising a single chiral form, or enantiomer, of the organic cage molecule may comprise CC3-S, CC15 -R, CC19 -R or TCC2-

[0017] More particularly, the liquid may be an ionic liquid. In particular, the ionic liquid may be 1 -butyl- or 1 -ethyl- 3-methylimidazolium bis(trifluoromethylsulfonyl) imide, 1-butylpyridinium bis(trifluoromethanesulfonyl)imide, trihexyltetradecylphosphonium bis(trifluoromethanesulfonyl)imide or benzyl(ethyl)dimethylammonium bis(trifluoromethanesulfonyl)imide.

[0018] In particular, the liquid may be a halogenated oil, which may be a chlorinated, brominated or fluorinated oil. Suitable halogenated oils include, for example a perfluoropolyether oil such as Fomblin Y (RTM) oil, a fluorinated silicone such as Fluorosil J15 (RTM), a fluorinated polydimethylsiloxane such as Fluorosil D2 (RTM), a fluorinated alkyl polydimethylsiloxane copolymer such as Fluorosil H418 (RTM), a fluorocarbon ether such as Krytox oil, chlorinated silicone oil and halocarbon oil 27 (RTM).

[0019] More particularly, the liquid may be a triglyceride oil. Suitable triglyceride oils include, for example olive oil, castor oil, sesame oil, sunflower oil, safflower oil, rapeseed oil, walnut oil, peanut oil, almond oil, clove oil, soybean oil, corn oil, cottonseed oil, anise oil and linseed oil.

[0020] In particular, the liquid may be a liquid polymer.

[0021] More particularly, the liquid may be selected from the group consisting of silicone oils, halogenated oils, triglyceride oils, paraffin oils, ionic liquids and size- excluded liquids.

[0022] In particular, the liquid may be a silicone oil (e.g. polydimethylsiloxane). Suitable silicone oils include, for example, pure silicone oil (more particularly silicone oil 5 cSt, silicone oil 20 cSt, and/or silicone oil 50 cSt) and silicone oil AR 20. Silicone oils are cheap, readily available and biocompatible making them advantageous for use in the dispersions of the present invention. In addition, silicone oil has high thermal stability which facilitates material regeneration.

[0023] In particular, the liquid may be an ionic liquid. Ionic liquids are generally defined as salts in a liquid state. More particularly, the ionic liquid may be 1 -butyl-3- methylimidazolium bis(trifluoromethylsulfonyl)imide (ie BMIM.NTf2), 1-butylpyridinium bis(trifluoromethanesulfonyl)imide (ie BPy.NTf2), trihexyltetradecylphosphonium bis(trifluoromethanesulfonyl)imide (ie P666i4.NTf2) and benzyl(ethyl)dimethylammonium bis(trifluoromethanesulfonyl)imide (ie BEMA.NTf2).

[0024] More particularly, the liquid may be a size-excluded liquid. A size-excluded liquid may be defined in the context of the invention as a liquid which is excluded from the pores (ie the cavities) within the organic cage molecule. This can be because the size-excluded liquid has a molecular size which is too large to enter the pores of the porous particles. Alternatively, entry into the pores of the porous particles may be thermodynamically or kinetically unfavourable. In particular, the size-excluded liquid may be 1 -‘butyl-3, 5-dimethylbenzene, 15-crown-5, hexachloropropene, methyl salicylate, 2,4-dichlorotoluene or 4- (trifluoromethoxy)benzyl alcohol. More particularly, the size-excluded liquid may be 1 -‘butyl-3, 5-dimethylbenzene or 15-crown-5.

[0025] In particular, the dispersion may comprise 0.1-50 wt% of the porous particles, more particularly 0.5-30 wt%. Even more particularly, the dispersion may comprise

I-20 wt% of the porous particles. More particularly, the dispersion may comprise 10- 15 wt% of the porous particles. In some embodiments, the dispersion may comprise

I I-14 wt% of the porous particles, more particularly 12-13 wt% of the porous particles.

[0026] In the dispersion according to the present invention, the pores of the porous particles are accessible to a gas. Optionally, the gas is CO₂, ChU, N₂, C₂H₄, C₂H₆, Xe, SFe, C₃H₈ or H₂, or a mixture thereof.

[0027] Particular embodiments of the dispersion are as follows: (i) the porous particles comprising the organic cage molecules comprise co-crystals of CC3-R and CC3-S and the liquid is silicone oil, more particularly silicone oil 5 cSt or silicone oil AR 20;

(ii) the porous particles comprising the organic cage molecules comprise co-crystals of CC3-R and CC3-S and the liquid is an ionic liquid, more particularly P666i4.NTf2 or BPy.NTf2, even more particularly BPy.NTf2;

(iii) the porous particles comprising the organic cage molecules comprise CC3-S, CC15- , CC19- or TCC2- and the liquid is silicone oil, more particularly silicone oil 5 cSt.

[0028] According to a further aspect of the present invention, there is provided a method of adsorbing a gas into a liquid, comprising at least the step of bringing the gas into contact with a dispersion of porous particles comprising an organic cage molecule, the porous particles being dispersed in a liquid phase, wherein the liquid is selected from the group consisting of liquid oligomers, size-excluded liquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof. More particularly, the dispersion may be as defined above.

[0029] In particular, the gas may be CO₂, ChU, N₂, C₂H₄, C₂H₆, Xe, SF₆, C₃H₈ or H₂, or a mixture thereof. In particular, the gas may be ChU or CO₂.

[0030] According to a further aspect of the present invention, there is provided a method for preparing a dispersion comprising at least the step of: mixing (i) porous particles comprising an organic cage molecule, and (ii) a liquid selected from the group consisting of liquid oligomers, size-excluded liquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof. The porous particles may be as defined above. More particularly, the dispersion formed by the method may be as defined above.

[0031] Optionally, the mixing includes agitating, stirring, sonication or grinding or a combination thereof. More particularly, the method comprises sonicating the mixture. [0032] According to a further aspect of the present invention, there is provided an assemblage of a dispersion of porous particles comprising an organic cage molecule, the porous particles being dispersed in a liquid phase, wherein the liquid is selected from the group consisting of liquid oligomers, size-excluded liquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof, wherein the organic cage molecule comprises a cavity and a gas contained within the cavity. More particularly, the gas may be CO₂, CFU, N₂, C₂H₄, C₂H₆, Xe, SF₆, C₃H₈ or H₂, or a mixture thereof. In particular, the gas may be CO₂ or CFU.

[0033] Brief description of the drawings

[0034] This invention will be further described by reference to the following Figures which are not intended to limit the scope of the invention claimed, in which:

Figure 1 shows a reaction scheme for the synthesis of porous organic cages CC3-R and CC3-S, the dashed bond being a cyclohexane ring,

Figure 2 illustrates the formation of microporous organic particles by chiral recognition between CC3-R and CC3-S upon mixing in solution, with a schematic of the packing in the CC3 -R/S particles

Figure 3 shows a series of PXRD traces of the individual cages CC3-R and CC3-S, as well as for the CC3 -R/S microparticles formed by Examples 1-6 and 7, Figure 4 shows SEM images of the CC3 -R/S microparticles formed by Examples 1-6 and 7,

Figure 5a-5d show DLS histograms comparing the particle sizes and distributions of CC3 -R/S microparticles formed by Examples 1-6 and 7,

Figure 5e shows a series of PXRD traces of the CC3 -R/S microparticles formed by Examples 7, 8, 9, 10 and 11 of the scale-up procedure,

Figure 5f shows DLS histograms comparing the reproducibility of the CC3 -R/S microparticle sizes and distributions formed by Examples 7, 8 and 9 of the scale- up procedure,

Figure 5g shows DLS histograms comparing the reproducibility of the CC3 -R/S microparticle sizes and distributions formed by Examples 10 and 11 of the scale- up procedure,

Figure 6 shows the structures of liquid dispersants which were investigated for the formation of cage microparticle dispersions, Figure 7 shows the structure of the control molecule, homochiral cage CC3 -R, homochiral CC15 -R, homochiral CC19 -R, homochiral TCC2- and a schematic of the packing of the CC3 -R/S particles,

Figure 8 shows the experimental setup for the measurement of volumetric uptake of gases by porous liquids on a Quantachrome Nova 4200e, with stirrer plates in place to ensure agitation of the liquid samples during degas and adsorption,

Figure 9 shows FTIR traces for the CC3 -R/S microparticles of Example 7, a 12.5 wt% dispersion of CC3 -R/S microparticles in silicone oil (5 cSt), and of silicone oil (5 cSt),

Figure 10 shows PXRD traces for the CC3 -R/S microparticles of Example 7 in air (lower trace) and a 12.5 wt% dispersion of CC3 -R/S microparticles in silicone oil (5 cSt) (upper trace),

Figure 11 shows DLS histograms comparing the particle sizes and distributions of CC3 -R/S microparticles dispersed in DCM (left most peak), a 12.5 wt% dispersion of CC3 -R/S microparticles in silicone oil (5 cSt) after 1 day (middle peak), and a 12.5 wt% dispersion of CC3 -R/S microparticles in silicone oil (5 cSt) after 24 days (right most peak),

Figure 12a shows TGA analysis of the CC3 -R/S solid microparticles, silicone oil (5 cSt), and the 12.5 wt% dispersion of CC3 -R/S microparticles in silicone oil (5 cSt) under air,

Figure 12b shows TGA analysis of the CC3 -R/S solid microparticles, BPy.NTf2, and the 5 wt%, 12.5 wt% and 20 wt% dispersions of CC3 -R/S microparticles in BPy.NTf2 under nitrogen,

Figure 13 shows CO2 uptake (298-303 K) in an ionic liquid (BMIM.NTf2, left) and a size-excluded solvent (15-crown-5, right),

Figure 14 shows CO2 (left) and ChU (right) isotherms on the solid CC3 -R/S microparticles of Examples 7 and 9,

Figure 15a-p shows CO2 uptake isotherms (298-303 K) for a range of dispersions and dispersants investigated,

Figure 16 shows a comparison of CO2 uptake isotherms (298-303 K) for the 12.5 wt% CC3 -R/S microparticle dispersions in silicone oil (5 cSt),

Figure 17a shows repeated CO2 uptake isotherms for the 12.5 wt% CC3 -R/S microparticle dispersions in silicone oil (5 cSt), Figure 17b shows repeated C0₂ uptake isotherms for the 12.5 wt% CC3 -R/S microparticle dispersions in BPy.NTf2,

Figure 18a shows repeated CO2 uptakes at 1 bar for the 12.5 wt% CC3 -R/S microparticle dispersions in silicone oil (5 cSt),

Figure 18b shows repeated CO2 uptakes at 1 bar for the 12.5 wt% CC3 -R/S microparticle dispersions in BPy.NTf2,

Figure 19a shows repeated CO2 uptake isotherms for fresh and aged 12.5 wt% CC3 -R/S microparticle dispersions in silicone oil (5 cSt),

Figure 19b shows repeated CO2 uptake isotherms for fresh and aged 12.5 wt% CC3 -R/S microparticle dispersions in BPy.NTf2,

Figure 20a shows a comparison of ChU uptake isotherms (298-303 K) for the 12.5 wt% and 20 wt% CC3 -R/S microparticle dispersions in silicone oil (5 cSt), with neat silicone oil (5 cSt),

Figure 20b shows a comparison of ChU uptake isotherms (298-303 K) for the 5 wt%, 12.5 wt% and 20 wt% CC3 -R/S microparticle dispersions in BPy.NTf2, with neat BPy.NTf2,

Figure 21a shows a Lumisizer image for a longer term stability test on 12.5 wt% CC3 -R/S (9) microparticles in silicone oil (5cSt),

Figure 21 b shows a Lumisizer image for a longer term stability test on 12.5 wt% CC3 -R in silicone oil (5cSt),

Figure 21c shows a Lumisizer image for a longer term stability test on 12.5 wt% CC3 -R/S (9) microparticles in silicone oil AR 20,

Figure 21 d shows a Lumisizer image for a longer term stability test on 12.5 wt% CC3 -R/S (9) microparticles in olive oil,

Figure 21e shows a Lumisizer image for a longer term stability test on 12.5 wt% CC3 -R/S (9) microparticles in Fomblin Y,

Figure 21f shows a Lumisizer image for a longer term stability test on 12.5 wt% CC3 -R/S (9) microparticles in Halocarbon 27,

Figure 21 g shows a Lumisizer image for a longer term stability test on 5 wt% CC3 -R/S (10) microparticles in BPy.NTf2.

Figure 21 h shows a Lumisizer image for a longer term stability test on 12.5 wt% CC3 -R/S (10) microparticles in BPy.NTf2.

[0035] Examples [0036] Porous Organic Cage Microparticle Screen

[0037] CC3-R and CC3-S were synthesised according to the procedure shown in JACS, 2012, 134, 588. The reaction scheme is shown in Figure 1. CC15- was synthesised according to the procedure shown in ACS Cent. Sci., 2017, 3, 7, 734. CC19- was synthesised according to the procedure shown in Angew. Chem. Int. Ed., 2018, 57, 35, 11228. TCC2- was synthesised according to the procedure shown in Nat. Chem., 2017, 9, 17.

[0038] A solution (1.5 mg/ml_) of organic cage CC3-S (150 g) in dichloromethane (100 ml_) was added to a solution (1.5 mg/ml_) of CC3-R (150 mg) in dichloromethane (100 ml_) using various addition rates, stirring types and temperatures (see Examples 1-7 in Table 1 below). Example 7 was carried out on a larger scale, i.e. CC3-S (1.35 g) in dichloromethane (900 ml_) was added to CC3-R (1.35 g) in dichloromethane (900 ml_).

[0039] The resulting colourless precipitate (CC3 -R/S microparticles) for each example was collected by vacuum filtration through a 0.2 pm nylon membrane filter paper and dried under vacuum overnight (maximum mass recovery = 300 mg).

dichloromethane (900 ml_) + CC3-S (1.35g) in dichloromethane (900 ml_)) to allow dispersion studies to be carried out.

Table 1 [0040] The resulting collected microparticles from Examples 1-7 were analysed by powder X-ray diffraction (PXRD) (see Figure 3) to confirm formation of the CC3 -R/S co-crystal. The PXRD patterns of the CC3 -R/S microparticles suggest phase-pure material, and closely resemble that observed for homochiral CC3-R and CC3-S. This suggests a similar crystal packing, but the peak angles are displaced slightly due to changes in the unit cell volume. This has been previously reported where the unit cell size of CC3 -R/S is smaller than CC3-R and CC3-S due to the ability of homochiral cages to pack more closely together than their chiral counterparts ( JACS , 2012, 134, 588).

[0041] The samples were also analysed by scanning electron microscopy (SEM) (see Figure 4) and dynamic light scattering (DLS) (see Figure 5) to determine particle morphology, size and distribution. SEM was conducted on a Hitachi (RTM) S4800 scanning electron microscope, and powdered samples of the particles were deposited on to adhesive graphite tabs mounted on 15 mm aluminium stages. DLS was carried out on a Zetasizer (RTM) Nano Particle Sizer by re-dispersing the powdered samples into dichloromethane in a quartz cuvette. Laboratory powder X- ray diffraction data (PXRD) were collected in transmission mode on samples held on thin Mylar (RTM) film in aluminium well plates on a Panalytical (RTM) X'Pert PRO MPD equipped with a high throughput screening (HTS) XYZ stage, X-ray focusing mirror and PIXcel detector, using Ni-filtered Cu Ka radiation. Data were measured over the range 4-40° in -0.013° steps over 60 minutes.

[0042] The SEM images show that spherical particles are obtained when magnetic stirring with a stirrer bar and hotplate are used (Examples 1-5), or on rapid mixing of the two solutions even when overhead stirring is employed (Example 7), whilst octahedral particles are formed on slow addition with overhead stirring (Example 6). It is worth noting that the particle sizes are similar on rapid mixing at room temperature whether the stirring be magnetic or overhead (Example 2 vs Example 7), and that whilst the morphology of the particles can be changed, the PXRDs confirm the same co-crystal I ised species are present.

[0043] The DLS histograms in Figure 5 are used to compare the CC3 -R/S microparticle sizes and distributions formed by a range of methods. Figure 5(a) is a comparison of the effect of rapid solution mixing with either prolonged magnetic stirring (Example 1), magnetic stirring and immediate filtration (Example 2), or overhead stirring and immediate filtration (Example 7) at the same temperature. Figure 5(b) is a comparison of the effect of rapid solution mixing (Example 2) and slow addition (Example 3) with the same stirring method at the same temperature. Figure 5(c) is a comparison of the effect of different temperatures for solution mixing - ambient or room temperature (RT) (Example 3), 0 °C (Example 4) and -78 °C (Example 5) with the same addition rate and stirring method. Figure 5(d) is a comparison of the effect of magnetic stirring (Example 3) and overhead stirring (Example 6) at the same temperature and addition rate.

[0044] A summary of the results from both the SEM and DLS measurements is shown in Table 2 below:

*Reported size is the z-average over 3 runs

Table 2

[0045] Due to the ease of scaling the procedure up, Example 7 was used to investigate dispersion formation in oils, ionic liquids and size-excluded solvents.

[0046] Scale-up of Porous Organic Cage Microparticles

[0047] A solution (1.5 mg/ml_) of organic cage CC3-S (1.35 g) in dichloromethane (900 ml_) was rapidly added in a single addition to a solution (1.5 mg/ml_) of CC3-R (1.35 g) in dichloromethane (900 ml_) with overhead stirring at room temperature (ie the Example 7 procedure). The resulting colourless precipitate (CC3 -R/S microparticles) was collected by vacuum filtration through a 0.2 pm nylon membrane filter paper and dried under vacuum overnight (52% mass recovery, maximum mass recovery = 2.7 g).

[0048] This preparation was repeated a further four times to ensure reproducibility in particle size/distribution as determined by DLS - for a summary of the five runs see Table 2a below. PXRD and DLS data for Examples 7, 8 and 9 is shown in Figures 5e and 5f, and Examples 10 and 11 are shown in Figures 5e and 5g. In Figure 5f, the particle size can be seen to vary slightly between the three Examples, but all particle sizes are within the range 200-2000 nm.

*Reported size is the z-average over 3 runs

Table 2a

[0049] Cage Microparticle Dispersions - Formation, Characterisation and Gas Uptake

[0050] Dispersions of the scaled-up microparticles (Examples 7-11) were investigated in silicone oil (PDMS, 5 cSt and 50 cSt), size-excluded ionic liquids (1- butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (BMIM.NTf2), 1- butylpyridinium bis(trifluoromethanesulfonyl)imide (BPy.NTf2) and a size-excluded solvent (15-Crown-5) that is known to be size-excluded from the cage cavity. Other oils were also tested, including triglycerides, hydrocarbon oils, halocarbon oils, and polyethyleneglycol (PEG) derivatives (see Figure 6) using the scaled-up microparticles of Examples 7-11.

[0051] The solid microparticles and the potential liquid dispersants were stirred or sonicated and the resulting suspension visually inspected. In silicone oil (5 cSt), the ionic liquid and the size-excluded solvent, a 12.5 wt% flowable dispersion, that visually remains dispersed, was formed. The microparticles in 5 cSt silicone oil and BPy.NTf2 were also made up to a 20 wt% dispersion.

[0052] A series of other oils were also investigated as dispersants at a 12.5 wt% loading of CC3 -R/S microparticles (Examples 7-9) including: silicone oil AR 20, olive oil and sunflower oil (triglycerides), paraffin oil (hydrocarbon) and Genesorb (RTM) 1753. For both Halocarbon 27 (RTM) and Fomblin Y (RTM), 5 wt% dispersions were made.

[0053] Representative Procedure for formation of 5 wt% microparticle dispersions

[0054] The CC3 -R/S microparticles (150 mg) were desolvated at 90 °C overnight in a vacuum oven, before the direct addition of the liquid dispersant (2.85 g), and the resulting suspension sonicated for 10 minutes to afford a 5 wt% dispersion.

[0055] Representative Procedure for formation of 12.5 wt% microparticle dispersions

[0056] The CC3 -R/S microparticles (200 mg) were desolvated at 90 °C overnight in a vacuum oven, before the direct addition of the liquid dispersant (1.4 g), and the resulting suspension sonicated for 10 minutes to afford a 12.5 wt% dispersion.

[0057] Representative Procedure for formation of 20 wt% microparticle dispersions

[0058] The CC3 -R/S microparticles (100 mg) were desolvated at 90 °C overnight in a vacuum oven, before the direct addition of the liquid dispersant (0.4 g), and the resulting suspension sonicated for 10 minutes to afford a 20 wt% dispersion.

[0059] Alongside this, to confirm that the gas uptake was due to the cavities in the cages, a control liquid using a molecule that contains all the functionality of the porous organic cages used (CC3 -R and CC3-S), but no cavity, was investigated as a 12.5 wt% dispersion in silicone oil (5 cSt). To determine if the homochiral cage CC3- R has the same uptake capacity as the formed CC3 -R/S microparticles, the solid was ground in a mortar and pestle and the powder investigated as a 12.5 wt% dispersion in silicone oil (5 cSt) (see Figure 7 for the structure of the control molecule (left), resulting homochiral cage (middle), and a schematic of the packing of the formed microparticles). Alternative microparticles of a different size (Example 5), and different morphology (Example 6), were also investigated as 12.5 wt% dispersions in silicone oil (5 cSt). These were formed using the same representative particle synthesis and dispersion formation procedures described above. [0060] Representative Procedure for formation of 10 wt% dispersions using alternative organic cage species

[0061] The alternative organic cages (CC15- , CC19- and TCC2- ) were ground in a mortar and pestle. The resulting solid (100 g) was then desolvated at 90 °C overnight in a vacuum oven, before the direct addition of the liquid dispersant (0.9 g silicone oil, 5 cSt), and the resulting suspension sonicated for 10 minutes to afford a 10 wt% dispersion.

[0062] These materials were prepared in order to test whether alternative cage molecules could modify gas uptake in the resulting dispersions.

[0063] CO2 uptake of microparticle dispersions

[0064] The uptake of CO2 was measured for the liquid dispersants and dispersions using a Quantachrome (RTM) Nova 4200e to confirm the formation of Type 3 porous liquids. The equipment is shown in Figure 8.

[0065] To ensure reliable gas uptake measurements, the sorption method on the Quantachrome (RTM) Nova 4200e was optimised to benchmark the system against liquids with known Henry’s constants for CO2 in the literature. The sample size, effect of stirring, stirrer bar type, number of pressure points, pressure tolerance and equilibration time were investigated and optimised, and the CO2 uptake measured on an ionic liquid (BMIM.NTf2, literature - Accounts of Chemical Research, 2007, 40, 11 1208-1216) and a size-excluded solvent (15-crown-5, literature - J. Phys. Chem. B, 2017, 121 8367), and compared to those calculated from the literature (see Table 2b below).

[0066] The standardised setup and method was as follows. Adsorption isotherms were measured using an in-house modification of a Quantachrome (RTM) Nova 4200e for liquid samples at ambient temperature (298-303 K). A 9 mm sample cell with large bulb (P/N:74064), filler rod (P/N: 74105-LP), and small glass coated magnetic stirrer bars were used for all measurements. The sample size used was approximately 0.5 ml_ injected directly into the bulb using a long syringe needle. Liquid samples were degassed under vacuum at ambient temperature for a minimum of 18 hours, whilst stirring (500 rpm) - no filler rod, He backfill selected to prevent refill with gas being measured. Adsorption of CO2 was carried out using the following settings: backfill = He, He Pump Down = 60 min, pressure points P/P_o = 0.05 to 1.0, in 0.05 increments; pressure tolerance = 0.05 mmHg; equilibration time = 1800 seconds; equilibration timeout = 5400 seconds, and was stirred (500 rpm) throughout the measurement, with a filler rod in place. The software automatically measures the free-space and takes this into account when calculating the uptakes. Separate samples of each dispersion were measured a minimum of 2 times to calculate an average uptake and standard deviation.

Table 2b

[0067] The C0₂ uptake measurements of both an ionic liquid and size-excluded solvent were comparable to those calculated from the reported Henry’s constants in the literature, and the uptakes reproducible between samples (see Figure 13). Therefore, this method was used to measure the uptakes with the dispersants and microparticle dispersions. The results are set out in Table 3 below.

[0068] To enable comparison to a calculated uptake for the amount of porous solid included in the dispersions, CO2 and CH₄ isotherms at 298 K were measured on the solid CC3 -R/S microparticles (Examples 7 and 9) using a Micromeritics (RTM) ASAP 2020 volumetric adsorption analyser (see Figure 14). The uptakes from the two measurements at 1 bar were used to calculate an expected uptake for the solid present in the dispersions. Powder samples were degassed offline at 90 °C for 15 h under dynamic vacuum (10⁵ bar) before analysis, followed by degassing on the analysis port under vacuum, also at 90 °C. Whilst the uptakes are in the same region, slight variation is seen between batches so the average uptake at 1 bar was calculated and used to compare to the uptakes in the dispersions - CO2 = 1895 mGTΐoI/gs (7) and 1690 pmol/gs (9), therefore average = 1792 ± 72 pmol/gs; ChU = 1 119 pmol/gs (7) and 1009 pmol/gs (9), therefore average = 1064 ± 77 pmol/gs.

*Average uptake measurements for CC3 -R/S solid microparticles carried out on a Micromeritics (RTM) ASAP 2020 (C0₂, 298 K, 1 bar)

Table 3 [0069] For the C0₂ uptake measurements, the isotherms and uptakes of duplicate samples for each example dispersion were comparable and reproducible, and all showed an increase in uptake over the neat dispersant with the exception of Genesorb 1753 (see Figure 15). [0070] On comparison of the CO2 uptake in dispersions consisting of different CC3-

R/S microparticles (varying the sizes and morphologies), and using homochiral CC3- R not formed into microparticles, in silicone oil (5 cSt), the uptake was comparable. This suggests that particle size and morphology does not have a direct effect on the porosity of the dispersion (see Figure 15c). Furthermore, the control molecule dispersed in silicone oil (5 cSt) at 12.5 wt% showed no increase in CO2 uptake over neat silicone oil (5 cSt), confirming that the observed uptake is reliant on the presence of the cage cavities (see Figure 15a). There was, however, variation in the uptake depending on which liquid dispersant was used - see Table 4 below for more information. Compared to the CO2 uptake in a previously reported‘scrambled’ cage porous liquid of ~55 pmol/gpi_. ( Chem . Sci., 2017, 8, 2640), the dispersions of microparticles in silicone oil (5 cSt) showed an approximate 4-6 fold increase in gas uptake.

[0071] In Table 4 below, the average CO2 volumetric uptake measurement (1 bar, 298-303 K) on the different CC3 -R/S microparticle dispersions was compared to the calculated theoretical maximum uptake for the dispersion based on the ratio of solid and liquid present in that dispersion. For example, for a 12.5 wt% dispersion, the calculated uptake (pmol/g) would be 0.125 times the uptake in the solid (pmol/gs), plus 0.875 times the uptake in the neat liquid (pmol/g .

Calculated using the average CO2 uptake in the solid CC3 -R/S microparticles - 1792 pmol/gs

Table 4

[0072] Overall, two thirds (66%) of the calculated theoretical porosity from the solid microparticles and liquid dispersant is maintained in the dispersions in silicone oils (Silicone oil 5 cSt and Silicone AR20) and some ionic liquids (BPy.NTf2), even at different weight loadings. On changing to halocarbon oils (Halocarbon 27 (RTM) and Fomblin Y (RTM)) the percentage porosity is slightly reduced compared to the theoretical calculated uptake for the dispersion, with -57-60% of the calculated uptake being achieved. This reduction in porosity is also observed with triglyceride and hydrocarbon oils, although to differing extents depending on the oil used - paraffin oil and olive oil had 25-28% of the calculated uptake being achieved, and sunflower oil achieved 46% of the calculated uptake. The use of an alternative ionic liquid (BMIM.NTf2), or a size-excluded solvent (15-crown-5), also showed a reduction in porosity compared to the calculated uptake, but still exhibited -25-32% of the calculated theoretical porosity. The porosity of the microparticles is lost when dispersed in Genesorb (RTM) 1753, with a lower uptake than the neat liquid itself, likely due to interpenetration of the polyethylene glycol dimethyl ethers into the cage cavities.

[0073] The use of fewer pressure points and the effect on the measured CO2 uptake was investigated using the 12.5 wt% CC3 -R/S microparticle dispersion in silicone oil (5 cSt). As shown in Figure 16, the uptakes measured using 4 pressure points (0.25, 0.5, 0.75, and 1.0 bar, the squares in Figure 16) are comparable to those measured using 20 pressure points (0.05 to 1.0, in 0.05 bar increments, the triangles in Figure 16) showing that the gas does not need long equilibration times in order for gas uptake to occur in the porous liquid.

[0074] CH4 uptake of microparticle dispersions

[0075] An alternative gas (CFU) was investigated for uptake in the porous dispersions using the 12.5 wt% and 20 wt% CC3 -R/S microparticle dispersions in silicone oil (5 cSt) ), and the 5 wt%, 12.5 wt% and 20 wt% CC3 -R/S microparticle dispersions in BPy.NTf2 - see Figure 20a and 20b. Both dispersions showed an increased uptake compared to the dispersant alone. These results are shown in Table 5 below.

*Average update measurements for CC3 -R/S solid microparticles carried out on a Micromeritics (RTM) ASAP 2020 (CH₄, 298 K, 1 bar)

Table 5

[0076] In Table 6 below, the average ChU volumetric uptake measurement (1 bar, 298-303 K) on the different CC3 -R/S microparticle dispersions was compared to the calculated theoretical maximum uptake for the dispersion based on the ratio of solid and liquid present in that dispersion, as described above.

Calculated using the average ChU uptake in the solid CC3 -R/S microparticles - 1064 pmol/gs

Table 6

[0077] Overall, -47-60% of the calculated theoretical porosity (ChU) from the solid microparticles and liquid dispersant is maintained in the dispersions in silicone oil (5 cSt) and an ionic liquid (BPy.NTf2). As observed for the CO2 uptake, on increasing the microparticle dispersion from 12.5 wt% to 20 wt%, the percentage porosity maintained in the dispersion is comparable.

[0078] Recyclability of CO2 uptake

[0079] The 12.5 wt% CC3 -R/S microparticle (Example 9) dispersion in silicone oil (5 cSt) and the 12.5 wt% CC3 -R/S microparticle in BPy.NTf2 were also investigated for recyclability of CO2 uptake (see Figures 17a, 18a, 17b and 18b), and the effect of aging on CO2 uptake (Figures 19a and 19b). For the 12.5 wt% CC3 -R/S microparticle (Example 9) dispersion in silicone oil (5 cSt), two separate samples had sorption measurements repeated over 2 days with a short vacuum degas (He pump down = 60 min) between each measurement. Both samples, and their repeats, showed similar isotherms (Figure 17a) and uptakes at 1 bar (Figure 18a), demonstrating both the ease of removal of gas from the porous liquid, and the recyclability of gas uptake. For the 12.5 wt% CC3 -R/S microparticle (Example 11) dispersion in BPy.NTf2, sorption measurements were only performed on a single sample but repeated 7 times, with a thorough overnight degas between each measurement. Corresponding isotherms and uptakes at 1 bar are shown in Figure 17b and Figure 18b respectively. [0080] Figures 19a and 19b then show CO2 uptake isotherms for aged samples (24 days or 30 days) of 12.5 wt% CC3 -R/S microparticles (Example 9) in silicone oil (5 cSt) and 12.5 wt% CC3 -R/S microparticles (Example 11) in BPy.NTf2 espectively. These show that the dispersions are still porous after ageing at ambient temperature under normal atmospheric conditions. For the dispersion in silicone oil (5 cSt), the uptake on day 1 = 178 pmol/gi_., compared to uptake on day 24 = 207 pmol/gi_. (298- 303 K, 1 bar). For the dispersion in BPy.NTf2, the uptake on day 1 = 170 pmol/gi_., compared to day 30 = 172 pmol/gi_.

[0081] Visual stability of dispersions

[0082] Table 7 below shows data on the visual stability of a selection of dispersions after being left to stand at ambient (or room) temperature for one day. A tick indicates that the dispersion was visibly stable, a cross that it wasn’t, and a dash if the sample was not tested. Apart from the 12.5 wt% control dispersion in silicone oil (5 cSt), which rapidly settled, the dispersions were visually stable on standing for 1 day. All samples also appeared to be dispersed when stirring was maintained, for example, during the degassing and adsorption studies.

[0083] Representative characterisation of 12.5 wt% CC3 -R/S microparticle dispersion in silicone oil (5 cSt, PDMS)

[0084] This dispersion was characterized using FTIR (see Figure 9), PXRD (see Figure 10) and DLS (see Figure 11).

[0085] Figure 9 shows, from top to bottom, stacked FTIR spectra of solid CC3 -R/S microparticles (Example 7), the 12.5 wt% CC3 -R/S microparticle dispersion in silicone oil (5 cSt) and silicone oil (5 cSt).

[0086] In relation to Figure 10, the powder X-ray diffraction data was collected using a Panalytical (RTM) Empyrean diffractometer producing Cu-Kcr radiation and equipped with an X-ray focusing mirror, PIXcel 3D detector, and capillary spinner. X- ray diffraction data was collected in transmission geometry over the 2Q range, 2-50 (°). Samples were contained in 0.7 mm borosilicate glass capillaries that were spun to improve powder averaging. To improve peak shape, 0.02 mm Soller slits, and 1/2° anti scatter slits were used during data collections.

[0087] In Figure 10, the bottom graph shows powder X-ray diffraction data for solid CC3 -R/S microparticles (Example 7) prepared according to the scale-up procedure. The top graph shows powder X-ray diffraction data for CC3 -R/S microparticles dispersed in silicone oil (PDMS, 5 cSt) at 12.5 wt%.

[0088] Figure 11 is a comparison of particle size and distribution of CC3 -R/S microparticles (Example 9) dispersed in dichloromethane, in silicone oil taken from the 12.5 wt% dispersion at T = 1 day, and in silicone oil taken from the 12.5 wt% dispersion at T = 24 days. Sizes in silicone oil (5 cSt) were based on the measured refractive index - 1.397, with the latter samples prepared by taking a small aliquot from the dispersion and diluting in silicone oil (5 cSt).

[0089] The DLS measurements in Figure 11 show an increase in particle size after the dispersion is left for 24 days, from a z-average of 1206 nm on day 1 (comparable to when the microparticles are dispersed in dichloromethane) to 1573 nm on day 24, which could be due to either aggregation or swelling. However, the aged dispersion still demonstrates the same porosity as a newly formed sample (see Figure 19). [0090] Thermal stability

[0091] Thermogravimetric analysis (TGA) of the CC3 -R/S solid microparticles, silicone oil (5 cSt) or BPy.NTf2, and the 12.5 wt% dispersions of CC3 -R/S microparticles in silicone oil (5 cSt) or BPy.NTf2 were carried out. This data is shown in Figures 12a and 12b. TGA was performed using a Q5000IR analyser (TA instruments) with an automated vertical overhead thermobalance. The samples were heated in platinum pans under nitrogen at a rate of 20 °C/min from ambient (room) temperature to 500 °C or 600 °C, either under air or nitrogen.

[0092] The cage microparticles are stable to a temperature of 310 °C under air, and to 400 °C under nitrogen, before thermal degradation occurs. Whilst the literature boiling point of silicone oil (5 cSt) is reported as >140 °C at 0.002 mmHg (supplier - Sigma Aldrich), under a flow of air the silicone oil is lost from 60 °C, with no sample remaining at 310 °C. In the dispersion once the silicone oil has been lost, approximately 12.5 wt% mass is remaining due to the presence of the cage. The ionic liquid, BPy.NTf2, was thermally stable until 325 °C under a nitrogen flow. The 5 wt%, 12.5 wt%, 20 wt% dispersions of CC3 -R/S microparticles in BPy.NTf2 also remained unaffected at 325 °C. There were small differences in the TG curve patterns between pure BPy.NTf2 and the corresponding dispersions. The curve showed a gradual decrease around the decomposing temperature for BPy.NTf2, with a steeper drop-off around the decomposition temperature for the 12.5 wt% and 20 wt% dispersions, and increasing mass remained in accordance with the increasing loadings of CC3 -R/S microparticles in the 5 wt%, 12.5 wt%, 20 wt% dispersions.

[0093] Long-term stability

[0094] A series of samples were then studied for long term dispersion stability using a LUMiSizer (RTM) 6512-48 at 25 °C with a wavelength of 865 nm, in disposable sample cells (2 mm, rectangular synthetic cell 110-131 mm), with 700 x 30 (or 1000 x 21) second intervals measured and a total run time of 350 min - see Figures 21a-h. This allows accelerated testing to determine the stability of dispersions (e.g. does sedimentation or creaming occur), and what the sedimentation velocity (V, pm/s) of the particles is at a relative centrifugal force (RCF - calculated using the size of the centrifuge and rpm), which can be then be converted to a sedimentation velocity at gravity, assuming Newtonian sample behaviour - see Table 8 below. RCF is calculated as follows:

where RCF = relative centrifugal force, RPM = revolutions per minute, and r = centrifugal radius in .

able 8

[0095] By using the same batch of microparticles (9) with different dispersants, and running the stability tests at the same RPM, it is possible to directly compare the stability of the dispersions and the effect of changing the liquid component. Depending on the dispersant used, the stability varied in terms of the sedimentation velocity (V), and whether the dispersion creams (particles go to the top of the sample) or sediments (particles go to the bottom of the sample). If long-term stability is of interest for a particular application, it is possible to design a more stable dispersion of microparticles whilst maintaining a similar level of porosity. For example, for 12.5 wt% CC3 -R/S (Example 9) dispersions in silicone oils a similar uptake is achieved in silicone oil (5 cSt) and Silicone Oil AR 20 (CO2 = 198.0 ± 3.8, and 181.8 ± 3.8, respectively, see Table 3), but the dispersion is more stable using the latter (V (mm/day) = 0.24 and 0.06 respectively, see Table 8). Further, the oil can be substituted for an ionic liquid (BPy.NTf2), which has a comparable stability but a lower vapour pressure.

[0096] The Lumisizer (RTM) images obtained for the samples in Table 8 are shown in Figures 21a-h. In Figures 21a, 21c, 21e, 21 f, 21g and 21 h, propagation from right to left indicates creaming. In Figures 21 b and 21d, propagation from left to right indicates sedimentation.

Claims

1. A dispersion of porous particles comprising organic cage molecules, the porous particles being dispersed in a liquid phase, wherein the liquid is selected from the group consisting of liquid oligomers, size-excluded liquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof.

2. A dispersion as claimed in claim 1 , wherein the porous particles have a mean diameter in the range 0.1-100 pm.

3. A dispersion as claimed in either claim 1 or claim 2, wherein the porous particles comprising the organic cage molecules comprise co-crystals of either the same organic cage molecule in two chiral forms, or two or more different organic cage molecules.

4. A dispersion as claimed in claim 3, wherein the porous particles comprising the organic cage molecules comprise co-crystals of CC3-R and CC3-S.

5. A dispersion as claimed in either claim 1 or claim 2, wherein the porous particles comprising the organic cage molecules comprise at least 90 wt% of a single chiral form, or enantiomer, of the organic cage molecule.

6. A dispersion as claimed in claim 5, wherein the porous particles comprising the organic cage molecule comprise CC3-S, CC15 -R, CC19 -R or TCC2-

7. A dispersion as claimed in any one of the preceding claims wherein the liquid is selected from the group consisting of silicone oils, halogenated oils, triglyceride oils, paraffin, oils, ionic liquids and size-excluded liquids.

8. A dispersion as claimed in claim 7, wherein the silicone oil is selected from the group consisting of silicone oil 5 cSt, silicone oil 50 cSt and silicone oil AR 20.

9. A dispersion as claimed in claim 7, wherein the ionic liquid is 1 -butyl-3- methylimidazolium bis(trifluoromethylsulfonyl)imide), 1-butylpyridinium bis(trifluoromethanesulfonyl)imide, trihexyltetradecylphosphonium bis(trifluoromethanesulfonyl)imide or benzyl(ethyl)dimethylammonium bis(trifluoromethanesulfonyl)imide.

10. A dispersion as claimed in claim 7, wherein the size-excluded liquid is 1- ‘butyl-3,5-dimethylbenzene or 15-crown-5.

11. A dispersion as claimed in any one of the preceding claims comprising 0.1-50 wt% of the porous particles.

12. A dispersion as claimed in claim 11 comprising 1-20 wt% of the porous particles.

13. A method of adsorbing a gas into a liquid, comprising at least the step of bringing the gas into contact with a dispersion as claimed in any one of the preceding claims.

14. A method of adsorbing a gas into a liquid as claimed in claim 13, wherein the gas is selected from the group consisting of CO₂, CFU, N₂, C₂H₄, C₂H₆, Xe, SF₆, C₃H₈ or H2.

15. A method for preparing a dispersion as claimed in any one of claims 1-12, comprising at least the step of:

mixing (i) porous particles comprising an organic cage molecule, and (ii) a liquid selected from the group consisting of liquid oligomers, size-excluded liquids, ionic liquids, liquid polymers, silicone oils, halogenated oils, paraffin oils, petroleum oils, triglyceride oils and combinations thereof.

16. An assemblage of a dispersion as claimed in any one of claims 1-12, the organic cage molecule comprising a cavity and a gas contained within the cavity.

17. An assemblage as claimed in claim 16, wherein the gas is selected from the group consisting of CO2, CH₄, N₂, C2H4, C2H6, Xe, SF₆, C3H8 or H₂.