WO2019028519A1

WO2019028519A1 - Method of forming a carbon dioxide adsorbent for a rebreather or other breathing apparatus

Info

Publication number: WO2019028519A1
Application number: PCT/AU2018/050840
Authority: WO
Inventors: Timothy Paul LYONS; Matthew Ronald HILL; Xavier MULET; Cher Hon LAU; Michael Batten
Original assignee: Commonwealth Scientific And Industrial Research Organisation; One Atmosphere Pty Ltd
Priority date: 2017-08-09
Filing date: 2018-08-09
Publication date: 2019-02-14

Abstract

A method of forming a carbon dioxide adsorbent of a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas, the method comprising: preparing a SIFSIX-3-Ni paste comprising a mixture of SIFSIX-3-Ni material and a solvent, the SIFSIX-3-Ni material substantially comprising 2- dimensional SIFSIX-3-Ni structure; forming the SIFSIX-3-Ni paste into a shaped body having at least one mean dimension of greater than 0.5 mm; and heat treating the shaped body in a reduced pressure environment comprising a pressure of less than 500 mbar at a temperature of at most 160 °C to substantially remove the solvent from the shaped body and form 3-dimensional SIFSIX-3-Ni crystal structure in the shaped body, thereby producing a shaped adsorbent body for use in a rebreather apparatus.

Description

METHOD OF FORMING A CARBON DIOXIDE ADSORBENT

FOR A REBREATHER OR OTHER BREATHING APPARATUS

CROSS-REFERENCE

[001 ] The present application claims priority from Australian provisional patent application No. 2017903176 filed on 9 August 2017, the contents of which should be understood to be incorporated into this specification by this reference.

TECHNICAL FIELD

[002] The present invention generally relates to a method of forming a carbon dioxide adsorbent for rebreathers and other breathing apparatus, and a carbon dioxide filter for rebreathers and other breathing apparatus for the removal of carbon dioxide from breathing gasses such as a user's exhaled breath. The invention is particularly applicable to rebreather apparatus for an individual user's use in underwater environments and it will be convenient to hereinafter disclose the invention in relation to that exemplary application. However, it is to be appreciated that the invention is not limited to that application and could be used in any applications involving the scrubbing of a carbon dioxide content of an expelled breath in other breathing systems including anaesthetics.

BACKGROUND OF THE INVENTION

[003] The following discussion of the background to the invention is intended to facilitate an understanding of the invention. However, it should be appreciated that the discussion is not an acknowledgement or admission that any of the material referred to was published, known or part of the common general knowledge as at the priority date of the application.

[004] Rebreather apparatus are used in closed or partially closed systems to recycle breathing gas, both the oxygen and inert gas components of breathing gas, exhaled from a user. A carbon dioxide removal agent (typically an absorbent or adsorbent) is used to remove carbon dioxide from exhaled or breathing gas allowing the remaining oxygen and other gasses in the breathing gas to be recycled for breathing. The removal agent, in many cases is contained in a canister or other suitable housing connected to the airway of the rebreather apparatus.

[005] A number of carbon dioxide removal agents are known to be suitable for removing carbon dioxide. The most prevalent absorbents are soda lime, as for example taught in British patent publication No. GB1438757, and calcium hydroxide, as for example taught in International Patent Publication No. WO200183294. These absorbents use chemisorption mechanism where CO₂ is chemically converted, irreversibly to the absorbent. In underwater applications, soda lime could be hazardous to the user should sea water enter the breathing loop while in use as the chemical reaction between the soda lime and sea water results in caustic gases which are potentially harmful. The general use of soda lime can also be harmful if particles and/or dust of soda lime are inhaled whilst filling and handling CO₂ scrubbing canisters.

[006] Alternate removal agents such as physisorbents can also be used in a rebreathing apparatus. Physisorbents rely on adsorption to internal surfaces of a porous material, the strength of such an interaction being primarily governed by attraction between CO₂ and the adsorbent surface. One example of a physisorbent material being used in a rebreather apparatus is taught in United States Patent Publication No. 20090032023A1 ("the 023 patent"). Here metal organic framework (MOF) adsorbents are utilised for adsorbing carbon dioxide in a rebreather apparatus. CO₂ is physisorbed to the surfaces within the pores of the MOF adsorbent. This process is reversible, allowing the adsorbent material to be regenerated for example by using either pressure swing or temperature swing techniques to release the adsorbed CO₂ from the pores of the MOF. The regenerated MOF material can then be reused.

[007] However, whilst the 023 patent teaches a vast array of possible MOFs that could be used in the rebreather apparatus, the 023 patent does not exemplify any MOF that is optimally suited for the dynamic and fast kinetics required for the dynamic adsorption of low partial pressure concentrations of CO₂ used in rebreathers and other breathing apparatus. Additionally, the 023 patent does not teach any way of enhancing or optimising the adsorbent properties of the taught MOFs when used in a filter or other adsorbent arrangement fitted in an actual rebreather or other breathing apparatus.

[008] International patent publication WO2016036814 ("the 814 patent") teaches techniques for removing trace and low concentration CO₂ from fluids using SIFSIX-n-M MOFs, wherein n is at least two and M is a metal such as zinc or copper. A general adsorption device is taught in relation to Figure 5 having a housing that contains one or more SIFSIX-n-M MOF compositions. The 814 patent does not teach any way of enhancing or optimising the adsorbent properties of the taught MOFs when used in a filter or other adsorbent arrangement fitted in an actual rebreather apparatus. Furthermore, the 814 patent does not teach any way of optimising the configuration or properties of the taught MOFs for use in an actual rebreather apparatus. A person skilled in the art could not build a workable rebreather or other breathing apparatus whereby adequate CO₂ is removed from the breathing loop based solely on the disclosure of the 814 patent.

[009] Thus, there is still opportunity to refine selection and further optimise MOF adsorbents to enhance, and importantly provide a viable method for the removal of carbon dioxide from breathing gas in a rebreather using MOF technology. It is therefore desirable to provide an improved or alternate rebreather apparatus which utilises metal organic frameworks for adsorbing carbon dioxide.

SUMMARY OF THE INVENTION

[010] A first aspect of the present invention provides a method of forming a carbon dioxide adsorbent of a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas, the method comprising:

preparing a SIFSIX-3-Ni paste comprising a mixture of SIFSIX-3-Ni material and a solvent, the SIFSIX-3-Ni material substantially comprising a 2- dimensional SIFSIX-3-Ni structure;

forming the SIFSIX-3-Ni paste into a shaped body having at least one mean dimension of greater than 0.5 mm; and heat treating the shaped body in a reduced pressure environment comprising a pressure of less than 500 mbar at a temperature of at most 160 °C to substantially remove the solvent from the shaped body and form 3- dimensional SIFSIX-3-Ni crystal structure in the shaped body,

thereby producing a shaped adsorbent body for use in a rebreather apparatus.

[01 1 ] It should be appreciated that "mean dimension" refers to the mean (average) dimension of at least one of the width, depth or height of the shaped body. Accordingly, at least one of the mean width, mean depth or mean height must be greater than the specified dimensional value.

[012] The method of this first aspect of the present invention produces a shaped SIFSIX-3-Ni MOF body which is suitable to adsorb and strip CO₂ from gas mixtures having gas content of exhaled breath (3 to 4% CO₂), and thus be used in a rebreather apparatus.

[013] It is to be understood that a rebreather apparatus is a breathing apparatus that removes the carbon dioxide of an individual user's exhaled breath to permit the rebreathing (recycling) of the substantially unused oxygen content, and unused inert content when present, of each breath. In some versions, oxygen can be added to replenish the amount metabolised by the user. This differs from an open-circuit breathing apparatus, where the exhaled gas is discharged directly into the environment.

[014] It should also be appreciated that "rebreather" indicates that the rebreather is designed to be used by an individual user for removing carbon dioxide from that individual user's breathing gas, and in particular that user's exhaled breath. In contrast, rebreather does not simultaneously process breathing gas or exhaled breath from a plurality or group of individuals.

[015] It should be understood that breathing gas comprises a gaseous mixture comprising oxygen. The gaseous mixture preferably comprises oxygen mixed with a diluent such as nitrogen, air, or helium. In some forms, the breathing gas may comprise an atmospheric or environmental gas, or may be supplied to a user via a breathing apparatus such as a mask or other device designed to fit around or over a user's head to supply gas thereto. It should be appreciated that in some embodiments, the breathing gas may include an oxygen content or air content with one or more additive gasses. For example, the breathing gas may include an anaesthetic gas as an additive gas, for example nitrous oxide, halothane, enflurane, isoflurane, desfiurane or sevoflurane. In a rebreather device, breathing gas preferably comprises oxygen mixed with a diluent like nitrogen, air or helium (for example used in deep diving). In an anaesthetic application, the breathing gas comprises oxygen, and a diluent such as nitrogen and an additive gas or gasses used to or to assist in anaesthetising the patent. In some embodiments, breathing gas comprises breathing air.

[016] Metal organic frameworks (MOFs) consist of metal atoms or clusters linked periodically by organic molecules to establish an array where each atom forms part of an internal surface. MOFs are capable of scrubbing CO₂ while remaining safe in the event of sea water ingress into the miniature rebreather's breathing loop. MOFs as a physisorbent achieve strong adsorption characteristics through the internal surfaces of the MOF porous structure. The strength of this interaction depends on the makeup of the adsorbent surface of the MOF to capture CO₂ molecules. Advantageously, the surface chemistry and structure of MOFs are able to be tuned for a specific application, where performance criteria such as adsorption/desorption rate, capacity as a function of pressure, and operating temperature may be of particular importance.

[017] For rebreather applications, the inventors have surprisingly and unexpectedly found that three-dimensional shaped bodies of SIFSIX-3-Ni have excellent CO₂ adsorption properties at low CO₂ partial pressures (3 to 4% CO₂) and suitable adsorption kinetics required for rebreather apparatus. SIFSIX-3-Ni also has useful breakthrough test properties for CO₂ capture, and has been found to have suitable stability when consolidated, shaped and heat treated.

[018] SIFSIX-3-Ni is from the SIFSIX-3-M family of hybrid ultramicroporous materials which can be classified as a metal organic framework (MOF), where M = Zn (Zn²⁺), Ni (Ni²⁺), Cu (Cu²⁺) or Co (Co²⁺). The kinetics of adsorption of SIFSIX-3-Ni has been found to be comparable to the kinetics of adsorption of a calcium hydroxide material.

[019] The Inventors have found that the vast difference in performance of MOFs at the required pressure of CO₂ is linked to the capacity of the MOF at low partial pressures.

[020] Nevertheless, as formed SIFSIX-3-Ni powder is not ideally used as produced in a packed bed filter of a rebreather apparatus. The inventors have found that the CO₂ adsorption characteristics of this material can be enhanced for rebreather application through shaping and specific heat treatment operations.

[021 ] Shaping is important to produce a shaped body suitable for packed bed adsorption unit processes used in a filter arrangement of a rebreather apparatus. Shaped SIFSIX-3-Ni bodies (for example noodles) preferably show an increased capacity for CO₂ uptake than SIFSIX-3-Ni-powder alone.

[022] Furthermore, specific heat treatment conditions are required to properly treat the formed SIFSIX-3-Ni material for use in the packed bed. The heat treatment step performs two important functions: Firstly, the selected temperature regime ensures that the particles are not heated to a temperature that would detrimentally affect the structure of the material, for example char the MOF material. Secondly, the heat treatment step converts the MOF particles consolidated in the shaped body from a 2D structure to a 3D linked MOF structure. It is therefore essential that the heat treatment step is carried out correctly to assist the initial SIFSIX-3-Ni material to undergo a structural or phase transformation from a 2D structure to the 3D structure, sometimes known as activation of the SIFSIX-3-Ni. The 2-dimensional structure comprises the metal, pyridine and hexafluorosilicate (SIFSIX) pillars being aligned in an aligned pillar structure in the material. These pillars are not laterally linked in this 2D structure. In the phase transformation, these 2D pillars are cross-linked to form the 3-dimensional crystallised tetragonal structure with P4/mmm symmetry forming the porous MOF structure required for CO₂ adsorption. [023] It is desirable to achieve as complete a conversion of 2D to 3D structure of the SIFSIX-3-Ni material as possible to form as a complete 3D porous crystalline MOF structure as possible. In embodiments, at least 60%, preferably at least 70%, more preferably at least 90%, yet more preferably at least 95%, yet even more preferably at least 99% of the 2-dimensional SIFSIX-3-Ni structure of the shaped body is transformed into 3-dimensional SIFSIX-3-Ni crystal structure in the heat treatment step. Moreover, it is preferred that the adsorbent bodies have a CO₂ breakthrough for 3.8% CO₂ gas stream of greater than 1 hour following the breakthrough testing detailed in section 5 of the detailed description.

[024] The heating step is also used to remove the solvent from the SIFSIX material. Importantly, the removal of solvent from the shaped body leaves the pores and surfaces of the SIFSIX-3-Ni shaped body in an activated state, free of material and therefore primed to adsorb any carbon dioxide passed over the material. Furthermore, in some embodiments, an initial heat treatment step is used to further dry, i.e. remove the solvent from the shaped body. The heat treatment step can therefore include an initial heating step of:

heat treating the shaped body at least 80 °C, preferably about 80 °C at a pressure of less than 500 mbar, preferably less than 100 mbar for at least 12 hours, preferably at least 24 hours.

[025] Of course, in other embodiments, this step can be integrated into the overall heat treatment regime, with solvent removal and 2D to 3D phase transformation being achieved in a single or multiple heating steps.

[026] The heat treatment step is conducted at a reduced pressure to assist removal of the solvent and any other off gases produced in the duration of the step. In embodiments, the heat treatment step is conducted at a pressure of less than 100 mbar, preferably less than 50 mbar, more preferably less than 35 mbar.

[027] The conditions of the heat treatment step can be varied to suit the amount of material, shape and configuration and other variables of the shaped material. For example, the duration of the heat treatment step can be varied to reach a desired phase conversion (2D to 3D structure). In embodiments, the heat treatment step is conducted for at least 5 hours, preferably at least 8 hours, more preferably at least 10 hours. Similarly, the temperature regime can be varied to suit desired conversion and shape body parameters. In embodiments, the heat treatment step is conducted at a temperature of between 1 10 to 160 °C. In some embodiments, the heat treatment step includes a heat treatment regime having more than one heating step. The various temperatures can be selected to assist the 2D to 3D phase transformation, and to avoid any damage to the material that could be caused by prolonged exposure to high temperatures. For example, the heat treatment step can in some embodiments, comprises a temperature regime of:

a first heating step in which the temperature is kept from 140 °C to 160 °C, preferably 150 °C for less than 5 hours, preferably 2 to 5 hours; and

a second heating step in which the temperature is lowered to at most 130 °C for at least 5 hours, preferably at least 10 hours, more preferably at least 12 hours, and yet more preferably from 8 to 12 hours.

[028] Heating to between 1 10 and 160 °C transforms the structure of the SIFSIX-3-Ni from a 2D to a 3D structure. Conversion of the shaped SIFSIX-3- Ni at the temperatures specified forms a key inventive aspect of the invention as it provides a means to activate MOFs in industrially required quantities. It should be appreciated that these heat treatment steps are conducted at the specified reduced pressure of less than 500 mbar, preferably less than 100 mbar, more preferably less than 50 mbar. In some embodiments the pressure is less than 35 mbar.

[029] The 2D structured SIFSIX-3-Ni material is typically formed as a powder (light violet powder) that is consolidated in the forming step into the shaped body. In embodiments, this 2-dimensionally structured SIFSIX-3-Ni material is synthesised from a mixture of a SiF₆ precursor compound, a Ni precursor compound and pyrazine, in which the SiF₆ precursor compound is selected from a SiF₆ salt, for example (NH₄)₂SiF₆ and the Ni precursor compound is selected from a salt comprising Ni²⁺,for example Ni(NO₃)₂. These precursor compounds provide NiSiF₆ which is co-ordinated with organic ligand pyrazine to form SIFSIX-3-Ni.

[030] It should be appreciated that the SIFSIX-3-Ni paste comprises a thick, soft, moist mixture. The paste preferably has sufficient viscosity to retain a form when shaped into a desired configuration in the forming/ shaping step. Preferably, the amount of solvent and SIFSIX-3-Ni material (preferably powder or particulates) is mixed to provide a suitable paste consistency for shaping processes such as extrusion or pelletising.

[031 ] It is also important to appreciate that the shaped body comprises a two component mixture, the SIFSIX-3-Ni MOF and a solvent. The solvent is evaporated or otherwise removed from the SIFSIX-3-Ni material during the heat treatment step, leaving a substantially pure MOF having solvent free pores. In some embodiments, the two component mixture, (SIFSIX-3-Ni MOF and a solvent) does not use binders, lubricants and/or other additives (for example pasting agent, adsorbents) or the like. The two component mixture consists of SIFSIX-3-Ni MOF and a solvent. The resulting shaped body is therefore a substantially pure MOF body. Where binders or fillers are used, at least a portion of the filler, binders or adhesives may end up fouling pores in the formed MOF, reducing the adsorption capacity of that material. Therefore where additional components are not used, other than the SIFSIX-3-Ni MOF and solvent, after heating, the shaped body comprises a substantially pure MOF material with pores that are substantially free of gas or solvent. The inventors consider that such a MOF is in an activated state because the surface area of the MOF is substantially free of gas or solvents and available for the adsorption of the target gas - carbon dioxide.

[032] The SIFSIX-3-Ni paste can be formed into the shaped body using a variety of processes. In embodiments, forming the SIFSIX-3-Ni paste into a shaped body comprises at least one of extruding, pelletising or moulding the SIFSIX-3-Ni paste into a desired 3-dimensional configuration. Preferred methods include rod extrusion or tableting. Where the shaped body is formed by an extrusion or similar process, such that the SIFSIX-3-Ni paste is extruded into an elongate body, that elongate body is preferably subsequently longitudinally divided, typically to a length suitable used in a packed bed.

[033] The shaped body is preferably formed having dimensions that are suitable for use in a packed bed adsorption type device, in which a plurality of the shaped bodies are packed at a high packing density 0.10 to 1 .0 g/cm, preferably 0.25 to 0.5 g/cm between two support surfaces. The dimensions of the shaped body can be optimised to suit this application. In some embodiments, the shaped body has at least one mean dimension of greater than 0.8 mm, preferably at least 1 mm, preferably at least 1 .2 mm, and yet more preferably at least 1 .5 mm. Preferably, each of the mean width, mean depth and mean height of the shaped body are greater than 0.5 mm, and preferably greater than 1 mm.

[034] The shaped body can have any suitable geometry. For example, the shaped body could comprise pellets, for example, disk-shaped pellets, pills, spheres, granules, extrudates, for example rod extrudates, honeycombs, meshes or hollow bodies. In embodiments, the shaped body is three dimensional, preferably three dimensionally shaped. In particular embodiments, the shaped body comprises an elongate body having a circular, or regular polygonal cross-sectional shape. For example, the shaped body may have a square or triangular cross-sectional shape. In one form, the shaped body has equilateral triangle cross-section, preferably the sides of the equilateral triangle are at least 1 mm in length, preferably between 1 .0 and 1 .5 mm in length. The elongate shaped body is preferably from 1 to 5 mm in length (longitudinal length), more preferably 1 to 4 mm in length.

[035] The solvent used to form the shaped body can be any suitable solvent that has good interaction with SIFSIX-3-Ni. Suitable solvents are preferably selected from a non-basic polar solvent and/or a non-self ionising polar solvent. The solvent preferably comprises an alcohol, such as methanol, C2-C9 alcohols including their branched isomers, supercritical water, supercritical carbon dioxide, n-methylpyrrolidone, tetrahydrofuran or combinations thereof. [036] In some embodiments, the shaped body undergoes a drying process prior to being subjected to the heat treatment step. In embodiments, the method of the first aspect of the present invention can further comprise:

drying the shaped body at a temperature of at least 20 °C for at least 1 hour prior to the heat treatment step.

[037] The drying temperature can be selected to meet drying requirements (time, location, equipment etc). In embodiments, the drying temperature is at least 30 °C, preferably at least 50 °C. Similarly, the duration of the drying step can be selected to meeting drying requirements. Whilst at least 1 hour is typical, the drying time may be at least 2 hours in some embodiments. The shaped body is preferably dried to a final density of at least 0.5 g/ cm³, preferably at least 0.60 g/cm³.

[038] The heat treatment step can produce small quantities of hydrofluoric acid (HF). As HF is an extremely toxic and corrosive material, this off-gas is preferably extracted during the heat treatment process to avoid damage to the shaped body and other materials. Accordingly, the heat treatment step is preferably conducted in a fluid tight housing. The method of the first aspect of the present invention therefore may further comprise the step of:

flushing an inert gas through the fluid tight housing.

The inert gas preferably comprises nitrogen, a Noble gas for example helium or argon.

[039] The shaped body typically undergoes a cooling process after the heat treatment step to cool the body to a temperature in which the body can be more readily handled. After the heat treatment step, the method may further comprise the step of:

cooling the shaped body to at most 80 °C, preferably at most 60 °C under reduced pressure of at most 500 mbar, preferably at most 100 mbar.

[040] The cooling step preferably has a duration of at least 2 hours, preferably at least 3 hours, more preferably at least 4 hours. [041 ] After CO₂ uptake inside SIFSIX-3-Ni has occurred, the adsorbed CO₂ can be removed from the MOF allowing it to be reused in removing CO₂ from breathing gasses. To remove the CO₂ in readying the MOF for re-use the MOF is subjected to heat less than 120 °C and a vacuum. Thus SIFSIX-3-Ni, following a period of reactivation, can be reused. SIFSIX-3-Ni can therefore be recycled and reused in CO₂ capture at exhaled breath CO₂ concentration.

[042] In this regenerative process, the method of the first aspect of the present invention can further comprise:

adsorbing carbon dioxide onto the surfaces of the SIFSIX-3-Ni shaped body; and

heating at a temperature of at least 80 °C, preferably at least 100 °C, more preferably at least 1 10 °C, and yet more preferably at least 120 °C for at least 1 hour, preferably at least 2 hours, at less than 100 mbar, preferably less than 50 mbar, thereby desorbing the carbon dioxide from the surfaces of the SIFSIX-3-Ni shaped body to produce a regenerated SIFSIX-3-Ni shaped body.

[043] The heating step is preferably conducted for sufficient time to remove CO₂ from the shaped body. The heating step is conducted for at least 1 hour, preferably at least 2 hours, more preferably at least 5 hours, yet more preferably at least 8 hours, and yet more preferably at least 10 hours. Similarly, the pressure is selected to assist CO₂ removal. In embodiments, the pressure is less than 100 mbar, preferable less than 50 mbar, more preferable less than 35 mbar. In other embodiments, the pressure is less than 500 mbar.

[044] The heating step is preferably conducted under inert gas flushing. This provides an inert gas atmosphere surrounding the shaped body, ensuring that once the CO₂ is removed, that only the inert gas remains in the pores of the constituent SIFSIX-3-Ni MOF. Once all CO₂ is removed, this leaves the SIFSIX-3-Ni material in an activated state, i.e. with pores substantially free of gas or solvent, ready for CO₂ adsorption. The inert gas preferably comprises nitrogen or a Noble gas for example helium or argon. [045] After heating, the material is preferably cooled down to at most 80 °C, preferably at most 60 °C under reduced pressure of at most 500 mbar, preferably at most 100 mbar. In embodiments, the regenerated material then cooled down for 4 hours under reduced pressure (35 mbar) and continuous inert gas flushing.

[046] Regeneration of a MOF adsorbent is one of the key advantages of the use of this material in a rebreather apparatus as compared to traditional chemisorption materials. A second aspect of the present invention provides a method for regenerating a SIFSIX-3-Ni adsorbent material from a closed or partially closed system as described above comprising the steps of:

optionally removing the SIFSIX-3-Ni material from a housing containing the SIFSIX-3-Ni material; and

heating the SIFSIX-3-Ni material at a temperature of at least 80 °C, preferably at least 100 °C, more preferably at least 1 10 °C, and yet more preferably at least 120 °C in an inert gas atmosphere.

[047] It is preferred that the SIFSIX-3-Ni material is kept insitu, within the housing containing that material for the regeneration step. The SIFSIX-3-Ni material (in the form of shaped adsorption bodies) can therefore remain undisturbed. Where that material is compressed in a packed bed within the container, that packing remains undisturbed, allow reuse of the adsorption material without the difficulties and additional time expended in unpacking and repacking that adsorption material.

[048] Regeneration of the SIFSIX-3-Ni material is conducted at a sufficient temperature to remove (desorb or other process which breaks the physisorption bond) CO₂ from the pores and adsorption surfaces of the SIFSIX-3-Ni material. The heating step is also preferably conducted at a reduced pressure, preferably less than 500 mbar bar, preferably less than 100 mbar, preferably less than 50 mbar (typically less than 35 mbar). Pressure reduction assists the removal process. Again, the inert gas is preferably a Noble gas, or nitrogen, or a mixture thereof. In some embodiments, the inert gas comprises one of nitrogen, helium, or argon. The heating step is conducted at a sufficient time period to remove CO₂ from the SIFSIX-3-Ni material. In some embodiments, the heating step is conducted for at least 1 hour, preferably at least 2 hours, more preferably at least 5 hours and in some embodiments at least 8 hours.

[049] A third aspect of the present invention provides a method of forming a carbon dioxide filter of a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas, the method comprising:

producing a plurality of shaped adsorbent bodies according to any one of the preceding claims; and

sealing a plurality of said shaped adsorbent bodies into housing under an inert gas atmosphere.

[050] This third aspect of the present invention relates to the formation of a housing to house the produced SIFSIX-3-Ni shaped bodies produced according to the first aspect of the present invention. In this aspect, the shaped bodies are enclosed in a housing, preferably a fluid tight housing, under an inert gas atmosphere to ensure that the SIFSIX-3-Ni material remains in an activated state once sealed in that housing. The inert gas atmosphere is also sealed within that housing, ensuring that the pores of the SIFSIX-3-Ni material are not contaminated with other gases before use in a rebreather apparatus. The housing is preferably only unsealed in the event the material is needed for use to scrub CO₂ from breathing gas when the rebreather is in use.

[051 ] The carbon dioxide filter is preferably used for removing carbon dioxide from breathing gas. Again, the inert gas atmosphere preferably comprises at least one of nitrogen or a Noble gas (for example helium, argon).

[052] The housing has a fluid inlet and a fluid outlet through which a fluid, preferably breathing gas is configured to flow. The housing can have any suitable configuration. In some embodiments, the housing comprises a container or canister for example a substantially cylindrical container or canister. The housing is preferably fluid tight, with only fluid access and egress through the inlet and outlet of that housing. In other embodiments, the housing comprises a flat, high surface area container. A high surface area container can be used to enable the exchange of heat generated by CO₂ adsorption in the MOF to reduce the adverse effect of heat on performance. It should be appreciated that a variety of container and canister shapes and configurations could be used. The inlet and outlet of the housing are sealed to retain the inert has atmosphere within the fluid prior to use. Those seals are broken in operation to allow fluid, again typically breathing gas, to flow from the inlet, through the plurality of said shaped adsorbent bodies, and to the outlet. The housing may be exchangeable or is installed fixed in the system.

[053] The plurality of said shaped adsorbent bodies is preferably arranged in the housing in a packed bed arrangement. In some embodiments, the housing includes two spaced apart support membranes configured to allow gas flow therethrough each membrane, the plurality of said shaped adsorbent bodies forming a packed bed therebetween and preferably being compressed therebetween. As with any packed bed type adsorber, it is important that the adsorbent is packed tightly and substantially uniformly throughout the packed bed volume to avoid short circuiting of any adsorbent in that packed bed. Any flow that is able to avoid or follow a shorter/ short circuit route through the packed bed will avoid having CO₂ removed from that stream. Short circuit flow can be detrimental or possibly fatal to a user when the rebreather apparatus is in operation.

[054] In some embodiments, the housing includes an inner container housing the support membranes and packed bed, the inner container being seated within the housing using a fitted insert which at least friction fits the inner container within the inner walls of the housing. The fitted insert is preferably formed from a resilient material such as a rubber, resilient polymer, foam or the like. The inner container may be removably fitted within the housing. This can allow the inner container to be removed and then subjected to regeneration steps (heating etc as discussed above) and then replaced in the housing for further use. [055] A fourth aspect of the present invention provides a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas, the rebreather apparatus including a carbon dioxide filter comprising:

a housing containing therein a packed bed of shaped adsorbent bodies; wherein the shaped adsorbent bodies comprise SIFSIX-3-Ni having at least 60% 3-dimensional SIFSIX-3-Ni crystal structure, the balance being 2- dimensional SIFSIX-3-Ni structure, the shaped adsorbent bodies having at least one mean dimension of greater than 0.5 mm.

[056] This fourth aspect can also relate to a carbon dioxide filter of a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas, comprising

[057] In some embodiments, the shaped adsorbent bodies have pores that are substantially free of foreign material or fluid. Foreign material should be understood to refer to any material other than SIFSIX-3-Ni or the inert gas. The adsorbent bodies are preferably sealed within the housing in an inert gas atmosphere prior to use. This prevents the adsorbent bodies adsorbing gases such as carbon dioxide from atmospheric air or the like prior to use, leaving the pores activated for use. Again, the inert gas atmosphere preferably comprises at least one of nitrogen or a Noble gas.

[058] This fourth aspect of the present invention relates to a carbon dioxide filter container or canister containing SIFSIX-3-Ni material that can be produced using the method of the first aspect of the present invention. This carbon dioxide filter is configured for fitment to a rebreather apparatus. The product has all the advantages described above in relation to the first aspect of the present invention. The filter may be exchangeable or is installed fixed in the system. [059] Again, it is desirable that the shaped adsorbent bodies comprise a high wt% of 3D structure of the SIFSIX-3-Ni material, and therefore comprise a high proportion of 3D porous crystalline MOF structure. In embodiments, at least 70%, preferably at least 90%, more preferably at least 95%, even more preferably at least 99% of the SIFSIX-3-Ni structure comprises 3-dimensional SIFSIX-3-Ni crystal structure. Moreover, it is preferred that the adsorbent bodies have a CO₂ breakthrough for 3.8% CO₂ gas stream of greater than 1 hour following the breakthrough testing detailed in section 5 of the detailed description. Similarly, it is preferred that the housing includes an outlet that, in use, has a gas flow having about 0% CO₂ gas stream for at least 1 hour.

[060] As discussed above for the third aspect of the present invention, the housing has a fluid inlet and a fluid outlet through which a fluid, preferably breathing gas is configured to flow. The housing can have any suitable configuration. In some embodiments, the housing comprises a container or canister, preferably a substantially cylindrical container or canister. The inlet and outlet of the housing are sealed to retain the inert has atmosphere within the fluid prior to use. Those seals are broken in operation to allow fluid, again typically breathing gas, to flow from the inlet, through the plurality of said shaped adsorbent bodies, and to the outlet.

[061 ] The plurality of said shaped adsorbent bodies is preferably arranged in the housing in a packed bed arrangement. In some embodiments, the housing includes two spaced apart support membranes configured to allow gas flow therethrough each membrane, the plurality of said shaped adsorbent bodies forming a packed bed therebetween and preferably being compressed therebetween. As with any packed bed type adsorber, it is important that the adsorbent is packed tightly and substantially uniformly throughout the packed bed volume to avoid short circuiting of any adsorbent in that packed bed. Any flow that is able to avoid or follow a shorter/ short circuit route through the packed bed will avoid having CO₂ removed from that stream. Short circuit flow can be detrimental or possibly fatal to a user when the rebreather apparatus is in operation. [062] In some embodiments, the housing includes an inner container housing the support membranes and packed bed, the inner container being seated within the housing using a fitted insert which at least friction fits the inner container within the inner walls of the housing. The fitted insert is preferably formed from a resilient material such as a rubber, resilient polymer, foam or the like. The inner container may be removably fitted within the housing. This can allow the inner container to be removed and then subjected to regeneration steps (heating etc as discussed above) and then replaced in the housing for further use.

[063] In embodiments, a first liquid free gas flow space is in fluid communication with an inlet for the user's exhaled breathing gas, said exhaled breathing gas including carbon dioxide. A second liquid free gas flow space is also in fluid communication with an outlet for fluid flow of gas that is substantially free of carbon dioxide. In such embodiments, the first liquid free gas flow space is in fluid communication with the second liquid free gas flow space through the packed bed such that, in use, an exhaled gas stream can flow through the packed bed to facilitate adsorption of carbon dioxide from the user's exhaled breathing gas onto the adsorbent bodies. It should be appreciated that the term liquid free gas flow space refer to the spaces being substantially free of liquid in use, apart from liquid entrained in a user's exhaled breathing gas. It is noted that the first and second liquid free gas flow spaces are much smaller than the equivalent space in a soda lime filter canister as this space must function as a liquid drainage area in a soda lime filter to contain liquids such as water that accumulates from the absorption reaction that occurs when using that material. This space is at least half the volume that must be provided in an equivalent soda lime filter canister.

[064] The shaped adsorbent bodies have dimensions that are suitable for use in a packed bed adsorption type device, in which a plurality of the shaped adsorbent bodies are packed at a high packing density, 0.10 to 1 .0 g/cm, preferably 0.25 to 0.8 g/cm, more preferably 0.25 to 0.60 g/cm between two support surfaces. The dimensions of the shaped adsorbent bodies can be optimised to suit this application. In some embodiments, the shaped adsorbent bodies have at least one mean dimension of greater than 0.8 mm, preferably at least 1 mm, preferably at least 1 .2 mm, and yet more preferably at least 1 .5 mm. Preferably, each of the mean width, mean depth and mean height of the shaped adsorbent bodies are greater than 0.5 mm, and preferably greater than 1 mm.

[065] The shaped adsorbent bodies can have any desired shape considered suitable for use in a packed bed adsorber type configuration. In embodiments the shaped adsorbent bodies comprises elongate bodies having a circular, or regular polygonal cross-sectional shape. In preferred embodiments, the shaped adsorbent bodies have a square or triangular cross-sectional shape. In one form, the shaped adsorbent bodies have equilateral triangle cross-section, preferably the sides of the equilateral triangle are at least 1 mm in length, preferably between 1 .0 and 1 .5 mm in length. The elongate shaped adsorbent bodies are preferably from 1 to 5 mm in length (longitudinal length), more preferably 1 to 4 mm in length.

[066] One advantage of using a MOF adsorbent is that the MOF adsorbent can be regenerated for reuse of the filter within the rebreather apparatus. As described in relation to the first aspect of the present invention, the shaped adsorbent bodies are configured to be regenerated insitu in the carbon dioxide filter by heat treatment. In such embodiments, the adsorbent bodies are configured to adsorb carbon dioxide from an individual user's exhaled breathing gas, and be regenerated to a substantially carbon dioxide free state in situ, within the filter, by heat treatment. That heat treatment is conducted with the adsorbent bodies kept insitu within the filter housing, thereby avoiding time consuming task of unpacking and repacking the adsorbent bodies from the filter. In such a form, the filter can be reused with ease once carbon dioxide is desorbed from the shaped adsorbent bodies in the filter.

[067] The heat treatment regime to regenerate the shaped adsorbent bodies can be any suitable heat treatment regime known in the art. For example, in some embodiments heat treatment comprises heating the shaped adsorbent bodies at a temperature of at least 80 °C for at least 1 hour. [068] The rebreather apparatus is preferably a closed or partially closed system which comprises at least one breathing apparatus and also a breathing mask, a breathing suit, or other life support systems in fluid connection to the carbon dioxide filter. The breathing mask can be a diving mask, respiratory protection mask, helmet or the like. The mask or helmet may be integrated as part of a suit, for example a diving suit, hazardous environment suit, space suit or the like.

[069] A fifth aspect of the present invention provides a method for removing carbon dioxide from breathing gas in closed or partially closed systems comprising:

contacting the breathing gas with a SIFSIX-3-Ni shaped body having at least one mean dimension of greater than 0.5 mm, wherein the closed or partially closed system comprises at least one breathing apparatus and also a breathing mask, a breathing suit or other life support systems.

[070] Again, it is desirable that the shaped bodies comprise a high wt% of 3D structure of the SIFSIX-3-Ni material, and therefore comprise a high proportion of 3D porous crystalline MOF structure. Accordingly, the shaped bodies preferably comprise SIFSIX-3-Ni having at least 60% 3-dimensional SIFSIX-3- Ni crystal structure, the balance being 2-dimensional SIFSIX-3-Ni structure. In some embodiments, the SIFSIX-3-Ni material has pores substantially free of foreign material or fluid other than the inert gas.

[071 ] The SIFSIX-3-Ni shaped body is preferably at least part of an adsorber bed in a filter. In embodiments, the SIFSIX-3-Ni shaped body is part of a carbon dioxide filter of a rebreather apparatus according to the fourth embodiment of the present invention.

[072] A sixth aspect of the present invention provides a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas which comprises:

at least one breathing apparatus, a breathing mask, a breathing suit or other life support system; and housing containing a plurality of SIFSIX-3-Ni shaped bodies having at least one mean dimension of greater than 0.5 mm, the housing being fluidly connected to the breathing apparatus.

[073] Once again, it is desirable that the shaped bodies comprise a high wt% of 3D structure of the SIFSIX-3-Ni material, and therefore comprise a high proportion of 3D porous crystalline MOF structure. Accordingly, the shaped bodies preferably comprise SIFSIX-3-Ni having at least 60% 3-dimensional SIFSIX-3-Ni crystal structure, the balance being 2-dimensional SIFSIX-3-Ni structure. In some embodiments, the SIFSIX-3-Ni material has pores substantially free of foreign material or fluid other than the inert gas.

[074] The SIFSIX-3-Ni shaped bodies are preferably at least part of an adsorber bed in a filter. The rebreather apparatus can therefore further comprise a filter, in which the SIFSIX-3-Ni shaped bodies are present at least as part of an adsorber bed. In embodiments, the filter comprises a carbon dioxide filter of a rebreather apparatus according to the fourth embodiment of the present invention. The filter may be exchangeable or is installed fixed in the system.

[075] The adsorbent bodies are preferably sealed within the housing in an inert gas atmosphere prior to use. This prevents the adsorbent bodies adsorbing gases such as carbon dioxide from atmospheric air or the like prior to use, leaving the pores activated for use.

[076] The rebreather apparatus is preferably a closed or partially closed system which comprises at least one breathing apparatus and also a breathing mask, a breathing suit, or other life support systems. Closed systems comprise systems which have no connections (in or out) with the surroundings. For example, a closed circuit rebreather apparatus recycles breathing gas through the breathing circuit without expelling any breathing gas to the atmosphere, such that no bubbles are therefore produced when underwater. Partially closed systems or semi closed-circuit systems are those in which fluid is recycled through the system, but small quantities of gas is exhaled to the surroundings, i.e. bubbles will be produced underwater. In both cases, atmospheric oxygen does not enter the system. The only breathing gas is available through the rebreather recycle and/or any gas supply fluidly linked to the system, for example by a gas or oxygen tank/ cylinder. The closed or partially closed systems of the present invention are typically used in oxygen depleted or oxygen absent environments, such as underwater, in a toxic gas environment, in space or the like. An oxygen depleted environment is an environment in which the atmosphere (such as inhaled breathing gas) has an oxygen fraction or partial pressure too low for life sustaining breathing and/or which has other harmful constituents.

[077] The breathing mask described above can be a diving mask, respiratory protection mask, anaesthetic mask, helmet or the like. The mask or helmet may be integrated as part of a life support system or suit, for example a diving suit, hazardous environment suit, space suit or the like.

[078] The rebreather apparatus of the various aspects of the present invention may include one or more additional filter for the removal of other constituents of breathing gas, for example particulates, water or other gases. The filter can be exchangeable or be installed fixed in the system.

[079] The various aspects of the present invention are applicable to rebreather apparatus suitable for use for underwater applications such as diving and underwater rescue and emergency use. However, it is to be appreciated that the invention is not limited to that application and could be used in any applications involving the scrubbing of a carbon dioxide content of an expelled breath in other breathing systems including anaesthetics.

BRIEF DESCRIPTION OF THE DRAWINGS

[080] The present invention will now be described with reference to the Figures of the accompanying drawings, which illustrate particular preferred embodiments of the present invention, wherein: [081 ] Figure 1 is a schematic diagram illustrating the structural phase transformation of the 2D SIFSIX-3-Ni structure to the 3D SIFSIX-3-Ni structure.

[082] Figure 2 provides three representative optical microscopy images of a pellet from a sample of 20 soda lime pellets, showing (a) side view of pellet; (b) front view of pellet showing triangular cross-section; and (c) top view of pellet.

[083] Figure 2A provides a photograph of a carbon dioxide scrubbing canister used for testing direct exhaled CO₂ capture.

[084] Figure 2B illustrates one form of rebreather apparatus that can include a carbon dioxide filter or scrubber according to an embodiment of the present invention.

[085] Figure 3 provides a plot of Adsorption rate of CO₂ in a Soda Lime system. ^* highlights CO₂ injections (1 1 .49 ml at 0.05 bar) used to compare rates of CO₂ capture, specifically the 1 st, 2nd, 3rd 4th and 6th injections.

[086] Figure 4 provides a photograph showing the corrosion of a brass gas diffuser situated below the adsorbent column showing clear signs of corrosion; compared to the brass piece placed ahead of the material adsorbent column.

[087] Figure 5 provides a plot of CO₂ adsorption isotherms for different MOFs selected. The region of interest for relevant exhaled breath concentration is highlighted.

[088] Figure 6 provides photograph of (a) a photograph of experimental extrusion device (b) a jar of SIFSIX-3-Ni 1 .5mm equilateral triangle extrudates (or "noodles"); (c) a single SIFSIX-3-Ni 1 .5mm equilateral triangle extrudates.

[089] Figure 7 provides a plot of Isotherms of CO₂ uptake of SIFSIX-3-Ni MOF in different morphologies, extrudates or powder. For ease of reference, elongated shaped bodies with an equilateral triangle cross-section are referred to as noodles.

[090] Figure 8 provides plots of the adsorption of CO₂ over different Metal Organic Frameworks over time, each CO₂ injection (1 1 .49 ml at 0.05 bar) sampled for rate calculation for (a) UiO66-NH2; (b) CAU-1 ; (c) SIFSIX-3-Ni; and (d) CuBTC.

[091 ] Figure 9 provides plots of the adsorption of CO₂ over different porous, hypercrosslinked polymers over time, each CO₂ injection (1 1 .49 ml at 0.05 bar) sampled for rate calculation for (a) PDCS; and (b) PDCX-NH2.

[092] Figure 10 provides a comparison of the Rate of Adsorption of CO₂ over different porous, materials over time at different CO₂ injection (1 1 .49 ml at 0.05 bar).

[093] Figure 1 1 provides photographs of small-scale (gram scale) breakthrough testing system for adsorbent testing showing (a) pressure differential measurement; (b) small sample housing and adsorbent sample; (c) sample housing fluidly connected within the testing rig.

[094] Figure 12A provides (a) a schematic of the CO₂ breakthrough apparatus; and (b) a schematic of the adsorbent bed used in the break-through testing apparatus.

[095] Figure 12B provides (a) a schematic of a minimised design of the CO₂ breakthrough apparatus; and (b) a schematic of Apparatus with provision for in- situ adsorbent activation, inert fluid sweeping and inlet stream calibration.

[096] Figure 12C provides a schematic of the testing apparatus with provision for in-situ adsorbent activation, inert fluid sweeping and inlet stream calibration.

[097] Figure 12D provides schematics of adsorbent bed configurations, (a) In its simplest form the bed may simply hold the packed sample in place with frits or filter media, heating/cooling systems may not be required; and (b) Larger- scale or more complex measurements or purifications may warrant fluid flow distribution systems, the addition of internal heating/cooling systems, data acquisition systems or sample points.

[098] Figure 13 provides a plot of breakthrough testing for Soda Lime at a rate of flow of 40 ml/min per 2.45 gram of material. [099] Figure 14 provides a plot of breakthrough testing for Soda Lime at a rate of flow of 8 ml/min per gram of material.

[100] Figure 15 provides plots of breakthrough testing for MOFs and hypercrosslinked polymers at a rate of flow of 8 ml/min per gram of material, for (a) Al-fum; (b) pDCX; (c) MIL53-NH2; (d) results summary; (e) UiO-66; (f) UiO66-NH2; (g) pDCX-NH2; and (h) results summary.

[101 ] Figure 16 provides plots of breakthrough testing for MOFs and hypercrosslinked polymers at a rate of flow of 8 ml/min per gram of material, for (a) CuBTC shaped bodies (noodles); (b) SIF-SIX-3-Ni; (c) CALM ; and (d) results summary.

[102] Figure 17 provides a comparison of the results of breakthrough testing for MOFs and hypercrosslinked polymers at a rate of flow of 8 ml/min per gram of material.

[103] Figure 18 provides a plot showing a comparison of CO₂ uptake and desorption of SIFSIX-3-Ni MOFs with CALM MOFs. The star highlights the actual quantity of gas taken up by SIFSIX-3-Ni during a breakthrough experiment.

[104] Figure 19 provides a plot showing a comparison of CO₂ uptake of SIFSIX-3-Ni MOFs in different forms - comparing powder and extrudates.

[105] Figure 20 is a plot providing a CO₂ isotherm uptake of SIFSIX-3-Ni MOFs following non-optimal activation.

[106] Figure 21 provides a plot of breakthrough testing for Soda Lime, scaled to mass (19.2 litres/min of air and 0.8 litres/min of CO₂ for 1 .23 kg of material).

[107] Figure 22 provides a plot of breakthrough testing for Soda Lime, scaled to mass (40 litres/min of air and approx. 2 litres/min of CO₂ for 0.96 kg of material). [108] Figure 23 provides a plot of breakthrough testing for SIFSIX-3-Ni, full gas flow (40 litres/min of air and approx. 2 litres/min of CO₂ for 0.96 kg of material).

[109] Figure 24 provides a plot of breakthrough testing for SIFSIX-3-Ni, full gas flow (13.5 litres/min of air and approx. 0.5 litres/min of CO₂ for 0.96 kg of material).

[1 10] Figure 25 provides a schematic diagram of a testing canister designed for full flow test.

[1 1 1 ] Figure 26 a) provides comparison CO₂ isotherms for poorly activated SIFSIX-3-Ni shaped bodies and optimally activated SIFSIX-3-Ni shaped bodies b) provides comparison CO₂ breakthrough curves for poorly activated SIFSIX-3- Ni shaped bodies and optimally activated SIFSIX-3-Ni shaped bodies.

DETAILED DESCRIPTION

[1 12] The present invention provides a Metal Organic Frameworks (MOFs) replacement of a chemisorption system such as soda lime or other absorbent, for example a mixture of Ca(OH)₂, NaOH, KOH and H₂O, for the removal of carbon dioxide from breathing gases. MOFs provide a reversible CO₂ adsorption through a regenerative step that desorbs the adsorbed quantity of CO₂ in the MOF. The selected MOF of the present invention is used in a rebreather apparatus, preferably a rebreather apparatus suitable for use for underwater applications such as diving and underwater rescue and emergency use. However, it is to be appreciated that the invention is not limited to that application and could be used in any applications involving the scrubbing of a carbon dioxide content of an expelled breath in other breathing systems including anaesthetics

[1 13] The inventors note that MOFs have been intensively investigated for intermediate and high CO₂ concentration removal applications such as post- combustion, pre-combustion capture, natural gas and biogas upgrading. However, the potential of MOFs to remove traces and low CO₂ concentration from gas streams has not been extensively considered as most reported MOFs exhibit relatively low CO₂ selectivity and uptake particularly at relatively low CO₂ partial pressure. Amine grafting chemistry has been considered a prospective pathway to enhance the CO₂ adsorption energetics and uptake in MOFs. However, such amine grafted MOFs can be difficult and complex to manufacture.

[1 14] The inventors have found that MOFs can adsorb and strip CO₂ from a gas mixture that mimics the gas content of exhaled breath (3 to 4% CO₂), that this performance is reproducible and that the lead MOF studied can be reused following a period of activation. However, in this study, the Inventors have surprisingly and unexpectedly found that SIFSIX-3-Ni has excellent CO₂ adsorption properties at low CO₂ partial pressures (3 to 4% CO₂) and suitable adsorption kinetics required for rebreather apparatus. Inventors surprising found that the kinetics of adsorption of the lead MOF SIFSIX-3-Ni is comparable to the kinetics of adsorption of a calcium hydroxide material, and therefore can be favourably compared to the existing rebreather adsorption materials. SIFSIX-3-Ni also has a significant capacity for CO₂ uptake which can be assessed through breakthrough testing. Breakthrough testing tests the time taken for an adsorbent material to saturate with target gas. Once the adsorbent is saturated with the target gas, the gas breaks through the adsorbent column. This technique permits an assessment of CO₂ performance mimicking real gas flow rates. These experiments test both adsorption capacity and rate of gas adsorption. It is thus useful to test breakthrough properties for CO₂ capture, and SIFSIX-3-Ni has been found to have suitable stability when consolidated, shaped and heat treated.

[1 15] SIFSIX-3-Ni is from a family of MOFs (SIFSIX-3-M) comprising SIFSIX-3- Zn, SIFSIX-3-Ni, SIFSIX-3-Cu and SIFSIX-3-Co. All of these compounds are crystallised in tetragonal structure with P4/mmm symmetry. In these "SIFSIX" materials, two-dimensional (2D) nets of organic ligand (in this case pyrazine) and metal node are pillared with hexafluorosilicate SiF₆ ^" (SIFSIX) anions in the third dimension to form 3D coordination networks that exhibit primitive cubic topology and, importantly, pore walls lined by inorganic anions. The different metal nodes in these structures provide different unit cell and pore sizes. SIFSIX-3-Zn MOFs comprising pyrazine ligands can have average pore sizes of about 3.84 A and BET apparent surface areas of about 250 m²/g (determined from the CO₂ adsorption isotherm at 298K). SIFSIX-3-Cu MOFs comprising pyrazine ligands can have average pore sizes of about 3.50 A (NLDFT) and BET and Langmuir apparent surface areas of ca. 300 m²/g (determined from the CO₂ adsorption isotherm at 298K).

[1 16] The properties, in particular the adsorption properties of SIFSIX-3-M MOFs has been extensively studied. For example, Ziaee, A (2016) "Theoretical optimisation of pore size and pore chemistry of SIFSIX-3-M hybrid ultramicroporous materials (HUMs)", PhD Thesis, University of Limerick - the contents of which should be understood to be incorporated into this specification by this reference - conducted an extensive study of the adsorption properties of the SIFSIX-3-M family. Ziaee showed that SIFSIX-3-M family have exceptional affinity toward small polarizable molecules such as CO₂. Isosteric heats of adsorption (Q_st) were found to decrease as M varies from Cu to Ni to Zn in SIFSIX-3-M. Furthermore, the interaction energy between a CO₂ molecule and HUMs with varying pore-size was calculated using a DFT-D2 level of theory and showed that the strongest interaction energy was calculated for SIFSIX-3-Cu (56.89 kJ mol^-1) where the pore-size is the smallest. The interaction energy decreased in SIFSIX-3-Ni (52.21 kJ mol^-1) and SIFSIX-3-Zn (48.46 kJ mol^-1), each of which exhibited larger pore dimensions. Ziaee attributed the increase in the strength of the interaction in SIFSIX-3-Cu to the shorter distance between the negatively charged equatorial fluorine atoms of the SiF₆ ^2_ pillar and the positively charged carbon atom of CO₂.

[1 17] However, a suitable material must also be stable when handled and heat treated. Significant stability issues were found with SIFSIX-3-Cu and although literature values claim good CO₂ adsorption performance, the poor stability of this MOF makes it incompatible with use in exhaled breathing gas CO₂ capture. SIFSIX-3-Cu was found to degrade to its constituents rapidly on purification and processing, being relatively unstable to ambient conditions. SIFSIX-3-Cu was difficult to process, degraded quickly (liberating HF) and therefore not a realistic proposition to use at any large scale. [1 18] In comparison, SIFSIX-3-Ni was found to have suitable stability on purification and processing. SIFSIX-3-Ni is cheap to make, easy to handle and process. SIFSIX-3-Ni was selected as a MOF of choice for rebreather CO₂ capture due to several key factors:

1 ) Ease of manufacture - this MOF can be synthesised in water and washed using a small alcohol (typically methanol or ethanol). Following synthesis, processing the MOF is simple. As outlined in the Examples, a key advantage that was found was that following three water washes and three ethanol washes a MOF paste was obtained that was ideal for processing using centrifugation.

2) It is robust to handling in ambient conditions and can withstand multiple temperature cycles without degradation.

3) SIFSIX-3-Ni also has kinetics of adsorption that are similar to those of Soda Lime chemisorbents and a high capacity at low partial pressures of CO₂ in a mixed gas stream.

[1 19] SIFSIX-3-Ni material can comprise two different structures, a 2- dimensional SIFSIX-3-Ni structure (A in Figure 1 ) and a 3-dimensional SIFSIX- 3-Ni crystal structure (B in Figure 1 ). The 2-dimensional structure comprises the metal, pyridine and hexafluorosilicate (SIFSIX) pillars being aligned in an aligned pillar structure in the material. These pillars are not laterally linked in this 2D structure. This material has a pale magenta colour. The transformation from the 2D and 3D structure involves a structure phase transformation in which the 2D pillars of the 2D SIFSIX-3-Ni structure are cross-linked to form a crystallised tetragonal structure with P4/mmm symmetry of the 3D SIFSIX-3-Ni structure. The 3D SIFSIX-3-Ni structure provides the requisite 3- dimensional porous MOF structure requisite for providing surface area for physisorption of CO₂ in this particular MOF. This material has a pale blue colour.

[120] It is desirable that the SIFSIX-3-Ni materials comprise a high wt% of 3D structure of the SIFSIX-3-Ni material and therefore consist of a high proportion of 3D porous crystalline MOF structure. That transformation is conducted using a heating step, at a reduced pressure as will be described below in the Examples.

EXAMPLES

[121 ] The inventors conducted a study to determine the most suitable metal organic framework for use in a rebreather apparatus, for example adsorbent systems for CO₂ scrubbing in emergency rebreathers used in aircraft water crash scenarios. The study was conducted in a number of stages. The first stage would start by determination of the most suitable porous materials for the removal of CO₂ from breathing gases. The second stage concentrated on synthesis and characterisation of selected porous materials to ensure appropriate quality. In-depth characterisation is essential to determine that the materials are fit-for-purpose and to ensure on going reproducibility and material quality. The third stage then concentrated on production, characterisation of the lead material identified in the first two stages. That material was processed and tested in a trial rebreather canister. Sufficient material was produced to fill one canister of material, estimated to be 1 .5 kg. The performance of the material was benchmarked against a commercial soda lime (Soda Lime)_canister used in rebreather apparatus.

[122] There are two key elements to the requirements for a novel material here. The first is that the rate of adsorption of CO₂ into the porous materials is comparable to that of the current chemisorbents. This was tested through a rate of adsorption of CO₂ at different pressures. The second is the capacity of the MOF under dynamic testing. Porous adsorbents are typically tested in regimes that are at equilibrium. The breakthrough test performed herein permits the quantification of the material performance under dynamic conditions over time. These experiments (or can be scaled to) mimic the rates of gas flow during breathing. This allows a comprehensive assessment of material performance.

1. Soda Lime baseline

[123] The first requirement of assessing the performance of MOFs or porous polymers as potential alternatives to soda lime materials is characterise the performance of Soda Lime. Despite the significant differences in adsorption mechanisms, the performance of an adsorbent MOF species must be similar to that of the soda lime to have potential as a replacement technology.

[124] Soda Lime comprises a scuba diving grade soda lime product, available in different grades. Soda Lime Grade N1025 non-indicating soda lime was used, which comprises a pure (>99.9 wt%) soda lime product having granule sized smaller than 8-12 mesh (particles 1 .0 to 2.5 mm with triangular cross- section). Soda Lime absorbs 150 L/Kg of CO₂.

1.1 Soda Lime Shape.

[125] Reducing the number of variables within the system is key to this project. The shape and packing of Soda Lime has been optimised to correlate between the kinetics of adsorption and the work of breathing (WOB). The shape and size of the soda lime baseline material was established and can be observed in Figure 2.

[126] A sample of more than 20 soda lime pellets was taken to determine shape and morphology. Typical shape distribution consists of equilateral triangles with 1 .5mm +/- 0.05mm (scale bar 0.5 mm) and varying lengths between 2 and 5 mm. This morphology offers an initial starting point for any MOF extrudates used as an adsorbent.

1.2 Soda Lime Kinetics

[127] The kinetics of Soda Lime CO2 absorption were assessed on a PCTPro 2000 (Sievert instrument for measuring gas sorption properties of materials) to determine the rate at which the CO2 was taken up at varying relative pressures of CO₂. The results of the kinetics of adsorption over time for a Soda Lime sample are shown in Figure 3, which clearly show uptake of CO₂ at different pressure over time. In Figure 3, "^*" denotes CO2 injection (1 1 .49 ml at 0.05 bar) sampled for rate calculation.

1.3 Soda Lime Breakthrough [128] The first step of performing the breakthrough experiments was to design, develop and construct a set-up able to test the breakthrough of CO₂. The parameters for this are outlined below.

1.4 Requirements for an adsorbent testing rig- Direct Exhaled Carbon Capture (DECC):

[129] Equipment was designed to quantify exhaled CO₂ adsorption on MOFs. The rig was used for both small scale and larger scale experiments. The rig was adaptable to small scale samples and to a commercial soda lime scrubbing canister 100, shown in Figure 2A. As shown in that Figure, a commercial soda lime scrubbing canister 100 comprises a cylindrical canister having a base inlet 1 10, removable lid 1 15 containing an outlet (not illustrated) and an adsorbent cartridge 120 (best shown in Figure 2A(b), which is removable and is filled with Soda Lime when required to be used. The cartridge includes porous filter base and lid which retains the adsorbent and allows breathing gas to flow therethrough. The cartridge 120 holds approximately 2 litres of adsorbent. Small samples of adsorbent to be tested were approximately 5 ml in size. The scrubbing canister 100 can be used in a rebreathing apparatus 150 as for example is illustrated in Figure 2B (described in more detail below).

[130] The equipment aims to mimic exhaled gas flow (41 .6 L/min) to the commercial soda lime scrubbing canister and quantify the amount of CO₂ gas adsorbed by measuring CO₂ levels in the exit gas. If the assumption of correlating sample volume to gas flow rate holds true, the flow rates required for the small volume sample holder will therefore be approximately 10.4 ml/min. The exhaust gas can then be fed to the Gas Chromatograph and CO₂ levels quantified.

[131 ] The required technical specifications of the designed system is as follows:

(A) Gas Supply:

- Gas Cylinder doped with 3.8% CO₂

- Can prepare own mix up with up to 800ml/min of CO₂ flow (B) Pressure required:

- Between 1 and 2 bar.

(C) Flow rate required:

- Upper range is 41 .6L/min for testing filled canister.

- Lower range is 10.4ml/min for testing small scale sample holder.

(D) Sample holder required:

- Commercial soda lime canister (as shown in Figure 2A) - large scale testing;

- Designed & built sample holder as shown in Figure 1 1 .

[132] It is important to note that rapid evidence of Soda Lime induced corrosion was observed after the first flow-through test in the diffuser of the small scale sample holder. The diffuser comprises brass discs placed above and below the sample to support and distribute gas through the sample. The downstream brass piece showed clear signs of corrosion after a single run of CO₂ capture, as shown in Figure 4. The disc on the left is post-exhaled gas capture experiment, the right side is a pristine original disc.

2. MOFs chosen

[133] 4 MOFs and 2 hyper-crosslinked porous polymers were chosen to study. Specifically these were CuBTC (also known as HKUST-1 ), UiO-66, UiO-66- NH2, SIFSIX-3-Ni, CAU-1 , PDCX, PDCX-NH2. These materials were chosen based on the variables (i) Safety; (ii) Performance; (iii) Stability; (iv)Cost. Several materials such as Mg-MOF74, were ruled out due to stability and cost of manufacture issues. At this stage, performance was a crucial element to demonstrate that MOFs can be used as an adsorbent in rebreather CO₂ adsorption applications.

[134] These materials were synthesised, purified and characterised: Synthesis of p-DCX

[135] To a solution of α,α'-dichloro-p-xylene (DCX, 1 ,4, dichloroxylene) monomer (0.171 mol, 30 g) in anhydrous DCE (388 mL), a DCE solution (388 imL) of FeCI3 (0.173 mol, 28 g) was added. The resulting mixture was stirred in an open vessel at room temperature. The precipitated p-DCX was washed once with water, three times with methanol (until the filtrate was clear), and with diethyl ether followed by drying for 24 h at 60 °C.

Synthesis of NH2- p-DCX

[136] p-DCX (2 g) was swollen in 65ml_ of DCE in the presence of aniline (0.54 g, 5.8 mmol) and formaldehyde methyl acetal (FDA, 0.028 mol). After swelling for 60 mins iron(lll) chloride (0.028 mol) in 20 imL of DCE was added and the solution was heated to 80 °C for 18 hours. The NH2-p-DCX was washed once with water, three times with methanol (until the filtrate 2 was clear), and with diethyl ether followed by drying for 24 h at 60 °C. The surface area of the resulting polymer was 750 m² /g.

CAU-1 Synthesis

[137] A mixture of AICI₃-6H₂O (1461 .0 mg, 0.61 mmol) and H₂BDC-NH₂ (365.0 mg, 0.20 mmol) were charged and suspended in methanol (20 imL). This mixture was sonicated for 20 minutes prior to solvothermal reaction. The reactions were carried out under stirring in a sealed glass pressure vessel at 120 °C.

Ui066 synthesis

[138] Equimo!ar quantities (43 mmol) of zirconium tetrachloride and 1 ,4- benzenedicarboxyiic acid were reacted in the presence of a large excess (684 mmol) of benzoic acid in a DMF:H₂0 (1650:83 mL) solvent. The resulting product was washed sequentially with D F and eOH before being dried under vacuum at 1 00°C.

UiO-66-NH2

[139] 1 .4914 g of ZrCI4 (6.4 mmol) and 1 .1464 g of H2N- H2BDC were charged in a Schott bottle and dissolved in 150 ml in DMF The bottle was heated to 100 ^°C overnight. Following cooling, the resultant yellow powder was filtered. After washing three times with ethanol, the product was dried at 70 °C in a drying oven to yield the metal-organic framework UiO-66-NH2.

HKUST-1 [140] In a typical reaction, solutions of 0.1 M Cu(N0₃)₂- 3H₂O and 0.24 M benzene-1 ,3,5-tricarboxyIic acid (BTC) also in ethanol were mixed under continuous flow conditions and heated in a tubular reactor. The synthesis was conducted at 140°C using a total flow rate of 90 mL-mirf , giving a total residence time of 1 .2 min. The material was washed twice with ethanol and dried under vacuum for 8 hours at 40°C. Yield: 1 00%.

[141 ] After synthesis, purification and characterisation, CO₂ uptake isotherms at low pressure (between 0 and 1 bar) were performed at 25°C (298K), as described below in section 5. This enabled the inventors to assess the performance of each of these materials at pressures which were relevant to the target 3.8% of C0₂ in breathing gas (for example air). Whilst this test is performed in pure CO₂ rather than a mixed gas system, it provides information on the optimal performance of the materials at such low partial pressure (0.04 atm). The region of interest is highlighted in Figure 5.

[142] Key to the performance of the MOF materials is the level of CO₂ adsorption at low partial pressure. Whilst these materials have comparatively similar adsorptions at 1 atm, at the low partial pressures required SIFSIX-3-Ni clearly stands out as having the highest performance.

3. Material Shaping

[143] For material shaping, a hand-extrusion device was used to produce the MOF into tubular or triangular extrudates. This device comprised a modified domestic pasta extruder, having a steel die retrofitted with equilateral triangle dies of 1 .5 mm triangular cross-sectional area retrofitted thereon to extrude through. A photograph of the device is shown in Figure 6(a).

[144] An example of SIFSIX-3-Ni triangular extrudates of the final product is shown in Figure 6 (b) and (c).

[145] Shaping the material into noodles did not affect the performance of the materials tested. This is clearly evidenced by the CO₂ isotherm at 25°C of SIFSIX-3-Ni noodles compared to the powder in Figure 7 (measured using test equipment described below in section 5). 4. Kinetics of C0₂ adsorption

[146] An assessment of the potential use of MOFs or adsorbents as a replacement to a soda lime chemisorbent requires a comparison of the rate of adsorption of the materials compared to that of Soda Lime. Kinetics of Soda Lime CO₂ absorption were assessed on a PCTPro 2000 (gas sorption analysis apparatus) to determine the rate at which the CO₂ was taken up at varying relative pressures of CO₂ as described below. To perform this experiment, multiple injections of CO₂ were done on each material sample and the pressure monitored over time as described below in section 5. For comparison purposes, the rate of CO₂ adsorption/absorption was compared at five injection intervals for the six candidates. All of these experiments were performed at 298K (25°C).

4.1 Soda Lime kinetics

[147] To determine whether the rate of the CO₂ adsorption was comparable for the porous adsorbents tested and the soda lime system tested, rates were calculated at several injection points. This rate compared the amounts of CO₂ adsorbed per gram of adsorbent over time. The results of individual trials are shown in Figures 3,7 to 9 and are summarised in Figure 10.

[148] The results clearly show that the rate of CO₂ adsorption for the MOFs and porous materials is comparable to that of soda lime. It is noted that SIFSIX-3-Ni is superior to the other tested materials, for example as shown in Figure 8 and 10 due to a balance of the capacity with the uptake velocity.

5. C0₂ breakthrough testing

5.1 Testing conditions

[149] To facilitate the experimental outlook of testing the adsorption of CO₂ on an adsorbent system a gas break-through testing methodology based on a simple break-through column was developed to quantify the presence of CO₂ (Figure 12A(a)). The apparatus was designed so that the number of fittings, number of joints and bends, and hence dead volume, between the adsorbent test bed, the rotameters and the other components such as the gas analyser is minimised. The resulting low back-pressure and dead-volume increases the precision of the apparatus, for example when testing small samples at high flow rates. The inherent simplicity of the apparatus provides a very precise, accurate, robust device that is both cheap to construct and simple to use and calibrate. The system may be designed such that the adsorbent bed (Figure 12A(b)) can be loaded into a glovebox to allow reactive or sensitive materials to be loaded into it and transferred back to the apparatus without exposure to the atmosphere or other things that might damage the adsorbent capacity of the adsorbent under test. The adsorbent bed holds the adsorbent sample in place with frits and/or filter media. External heating and/or cooling may be applied if required, for example for in situ activation of the sample - where the active form of the adsorbent is formed inside the test device.

[150] To operate the apparatus, contaminated gas is fed at a constant pressure via the inlet gas regulator to the inlet rotameter, which provides a constant flow-rate of the gas across the bed. A pre-mixed cylinder of compressed gas mixed with approximately 3.8% of CO₂ as the contaminant of interest was used. Using a pre-mix gas cylinder enabled rapid reproduction of results. The rotameter also allows the flow of gas into the adsorbent bed to be observed and measured.

[151 ] The pressure was measured upstream (Adsorbent Bed Inlet Pressure) and downstream (Adsorbent Bed Outlet Pressure) of the adsorbent bed. The upstream pressure was reported as the assumed bed pressure and the flow was taken as the flow leaving the set-up for the GC. This typically was the closest point to atmospheric pressure. The adsorbent bed inlet and outlet pressure measurement devices allow pressure correction of the inlet rotameter reading and the calculation of pressure drop across the adsorbent bed.

[152] The start of the break-through run is timed from when the flow of contaminated gas is introduced to the adsorbent bed. The adsorbent bed removes contaminants in the fluid stream and the outlet rotameter and pressure measurement device allow the control and measurement of the total flow-rate through the bed as well as measurement of the pressure-drop over the adsorbent bed. The purified effluent from the adsorbent bed is delivered to a downstream consumer process or to vent. A gas chromatograph (GC) was included to observe the saturation of the adsorbent bed with the contaminant (CO₂). A means of restricting the flow to downstream processes and controlling the flow to the gas analysis system is provided using a combination of outlet needle valve and Analysis rotameter provide a stable, known flow of material to the GC. The GC used was a Perkin Elmer Clarus 500 gas chromatograph fitted with a gas sampling valve, using a helium carrier, a ShinCarbon™ carbon molecular sieve packed bed column held at 220 degrees Celsius (isothermal) and a thermal conductivity detector. Typical time resolution achieved was 3.5 minutes. The method was typically quantitative down to 200 ppm (0.02 %) with qualitative detection of CO₂ to below 100 ppm. The time at which break-through occurs is taken as the first time at which the CO₂ reading from the GC exceeds 200 ppm, saturation of the adsorbent is then measured by taking further GC measurements as required. A person skilled in the art will realise that there are various variations of the flow control systems described here may be implemented to achieve the same result and that the apparatus as a whole could be fitted with things such as automation systems to increase the functionality of the apparatus.

[153] In some cases it is necessary to remove traces of gas or other fluids from the apparatus, for example when the gas interferes with an analysis, where the gas can damage the adsorbent, or where the adsorbent needs to be activated or formed in-place inside the adsorbent bed. Additionally the gas being fed to the adsorbent bed may need to be tested without being passed through the adsorbent bed in order to collect 'baseline' data before, after, or during the run. In such cases selector valves 1 and 2 and the inlet isolation valve (Figure 12A(a)) can be set to either direct the feed gas directly to the GC for analysis or to place the adsorbent bed under vacuum or an inert or pre-treatment atmosphere such as, but not limited to nitrogen, argon, or helium. For example the adsorbent bed, if fitted with a heater may be placed under vacuum and heated to activate the adsorbent bed in situ.

[154] The pressure was measured upstream and downstream of the adsorbent bed. The upstream pressure was reported as the assumed bed pressure and the flow was taken as the flow leaving the set-up for the GC. This typically was the closest point to atmospheric pressure. The values are typically not pressure corrected. The typical set-up is shown in Figure 1 1 .

5.2 Test Equipment

[155] The designed testing system comprises a monitored adsorbent bed filter. The system may be configured so that it accurately compares the performance fluid purification adsorbent systems. The system described here is specifically designed to purify gaseous fluids, and in particular remove CO₂ from a gas stream. However a person skilled in the art will recognise that such a system may equally be applied to other fluids such as liquids, slurries, emulsions or supercritical fluids. The designed system has a specific and unique benefit in that it can quickly, cheaply, and simply compare different adsorbents (e.g. a commercial adsorbent and a newly designed adsorbent) with a high degree of accuracy. The system is accurate enough to enable most implementations to be used to measure the specific amount of fluid contaminant removed from the contaminated fluid and allow the loading of the contaminant on the adsorbent to be calculated to a degree of precision and accuracy suitable for research or optimisation testing purposes. The system may also be designed so that the precision and accuracy achieved is suitable for analytical measurements.

[156] The device is deliberately designed using common interchangeable fittings such as flared, threaded, VCO, VCR or compression type fittings, with a minimal number of welded joints or parts made from bespoke fittings. In fact in the three implementations described below no welded or bespoke fittings are needed. This design feature allows the device to be re-configured to suit the adsorbent bed under test. For example the same system might be mounted on a benchtop frame for testing small samples in a laboratory setting, mounted to a test-gas delivery trolley for portable testing or be mounted directly onto a large or fixed-in-place adsorbent bed. In fact the interchangeability of the various components allows the same apparatus to be used over several orders of magnitude of fluid flow rate by, for example hot-swapping the flow control components for larger or smaller ones.

[157] One implementation of the device is given in Figure 1 2B(a). This version is designed so that the number of fittings, and hence dead volume, between the test bed, the rotameters and the other components such as the gas analyser is minimised. This increases the precision of the apparatus, for example when testing small samples at high flow rates, such as might be demanded for testing rebreather applications or environmental contamination control applications. The inherent simplicity of this implementation of the apparatus provides a very precise, accurate, robust device that is both cheap to construct and simple to use and calibrate. By using compact components, for example rotameters and pressure measurement devices, the system may be designed such that the section of the apparatus from the inlet rotameter to the outlet rotameter can be small enough to be loaded into a glovebox to allow reactive or sensitive materials to be loaded into it. A feature of all implementations of this apparatus is that the length and diameter of tubing and components on lines 2 to 4 are minimised. The tubing/components inner diameter is ideally only large enough to allow unimpeded flow of the contaminated or analyte fluid under the conditions of operation specified for the apparatus. More importantly the length and number of joints and bends in lines 2 to 4 are rigorously minimised. This is essential to the precision, accuracy and efficiency of the apparatus.

[158] To operate the apparatus, contaminated gas (i.e. CO₂ loaded gas) is fed at a constant pressure via the inlet gas regulator to the inlet rotameter, which controls the flow-rate of the gas across the bed. The rotameter also allows the flow of gas into the adsorbent bed to be observed and measured. The adsorbent bed inlet pressure measurement device allows pressure correction of the inlet rotameter reading and the calculation of pressure drop across the adsorbent bed. The adsorbent bed removes contaminants (in this case CO₂) in the fluid stream and the outlet rotameter and pressure measurement device allow the control and measurement of the total flow-rate through the bed as well as measurement of the pressure-drop over the adsorbent bed. The effluent from the adsorbent bed is delivered to a downstream consumer process or to vent. A gas analysis system may be included where desired, for example where the purification system is used to protect human health or where the system is being used to compare the performance of two different adsorbents during sequential adsorption runs, or for some other analysis. Where gas analysis is required a means of restricting the flow to downstream processes and controlling the flow to the gas analysis system is provided. The combination of outlet needle valve and Analysis rotameter shown in the system described in Figure 12B(a) allows a known flow of material to the Gas Analysis System allowing both stable operation of the analysis system and calculation of the distribution of fluid flow to the analysis system and the consumer processes. A person skilled in the art will realise that there are various variations of this flow control systems may be implemented to achieve the same result.

[159] When in some cases it is necessary to remove traces of gas or other fluids from the apparatus, for example when the gas interferes with an analysis, where the gas can damage the adsorbent, or where the adsorbent needs to be activated or formed in-place inside the adsorbent bed. Additionally some implementation such as a laboratory adsorbent testing station may require that the gas being fed to the adsorbent bed is tested to collect 'baseline' data before, after, or during the run. In such cases additional gas distribution functions can be built into the apparatus as shown in Figure 12B(b). The use of interchangeable components means that such an apparatus can be easily stripped back to the configuration described in Figure 1 2(a), for example if it needs to be removed to an adsorbent bed in a remote or confined location.

[160] The added functions described in Figure 12B(b) allow the following additional operations compared to the implementation described in Figure 12B(a). Selector Valves 1 and 2 can select between either one of two flow paths or a closed state. When the Inlet Isolation Valve is closed and Selector Valve 1 is set to direct the flow of fluid to the adsorbent bed Selector Valve 2 may be opened to either a vacuum system or a secondary inert or pre-treatment gas supply. For example an adsorbent such as a Zeolite or MOF adsorbent may be loaded into the adsorbent bed and a vacuum applied via Selector Valve 2 in order to remove adsorbed solvent or other species thereby providing the adsorbent in an active form. Inert gas or pre-treatment gas may then be introduced via Selector Valve 2 in order to bring the system back up to the required pressure and/or introduce pre-treatment species (e.g. tracer gas) into the adsorbent bed. With the adsorbent bed ready and Selector Valve 2 closed the inlet isolation valve may be opened and fluid directed to the adsorbent bed for an adsorption process operation as described for Figure 1 2B(a). Additionally, at any point before, during or after the adsorption process Selector Valve 1 may be set to deliver a sample of fluid directly to the analysis equipment in order to verify the 'baseline' composition of the inlet fluid. In some implementations Selector Valve 1 may be set to deliver gas both to the adsorbent bed and the analysis equipment, for example through a flow restrictor or flow control device in the path to the analysis equipment to allow continuous analysis of the inlet fluid stream. Such an implementation may include a valve to select the input source for the analysis system, if required.

[161 ] In some cases a degree of automation may be desirable, for example where the breakthrough test occurs over an extended period, where many tests are to be performed simultaneously by a coordinated network or adsorption apparatus or where the apparatus must work in an autonomous fashion. Figure 12C describes an alternate implementation of the device where the various control and measurement elements are automated by connection to a control system such as a computer, PLC or a distributed control system. The implementation of the system described in Figure 12C performs similar processes to the implementations described in Figure 1 2B(a) or Figure 12B(b), but also includes modifications so that an automated flow controller (Flow Control-1 ) can be set to flow continuously by-passing the adsorbent bed. This allows the flow-controller to pre-equilibrate and thereby provide the correct and steady flow-rate of gas to the adsorbent bed from the time that Selector valve 1 directs the flow of gas towards the adsorbent bed and, as a result, the breakthrough test begins. This bypass system, while convenient to have in the implementation described in Figure 12B(b), greatly improves the stability, precision and accuracy of implementations using automated flow control.

[162] A person skilled in the art will understand that different configurations of flow and pressure control may be used to the same or similar effect. Furthermore a person skilled in the art will recognise that various iterations of complexity varying from that of Figure 1 2B(a) to that of Figure 12C may be arranged using any combination of manual or automated control and measurement components as dictated by the scale and complexity of the system under test.

[163] The adsorbent bed itself may be configured in a number of ways (Figure 12D). The simplest form simply holds the adsorbent sample in place with filter media - external heating and/or cooling may be applied if required. In the most complex example the adsorption bed may include internal temperature transducers linked to a data-logging system, heating elements or heat exchangers to provide heating and/or cooling, sample points for monitoring fluid composition along the bed and gas distribution systems consisting of a plethora of channels the size of which may be tuned to achieve and optimal distribution of fluids within the adsorbent bed. The degree of complexity of the implementation determines the degree of complexity of the bed and the bed design may be at any level of complexity within, but not limited to the range of complexity exemplified by the range from Figure 12B(A) to Figure 12B(B). External heating/cooling may include but is not limited to solid-state heater/coolers, induction coils or fluid baths

[164] Initial experiments tested were performed based on real-world flows (40 litres/minute to 2.45 kg of adsorbent, 40:2.45) and scaled down to match the quantity of adsorbent present in the small test chamber (typically between 1 and 2.45 grams). A test using 2.45 grams of adsorbent would therefore be performed at flows of 40 ml of gas per minute. Unfortunately, these tests gave rapid breakthroughs, below the resolution of the GC runs, which did not permit efficient rankings of the MOFs and did not allow the target adsorbents to be assessed adequately. This rapid breakthrough meant that uncertainty in the point of breakthrough was unacceptably high. For this reason lower flowrates of 8:1 were tested to get a more robust comparison of the adsorbent.

5.2 Soda Lime

[165] The soda lime adsorbent was first benchmarked to establish whether the experimental set-up was able to reproduce the expected results. As can be seen below, with the breakthrough occurring at the expected 3:15:00 mark, the system on which it was tested is robust (Figure 1 3). [166] Table 1 : Results of breakthrough testing for Soda Lime at a rate of flow of 40 ml/min per 2.45 gram of material.

[167] A greater time resolution was required to be able to compare adsorbents with lower performance than Soda Lime. Therefore a low gas flow rate was also tested (Figure 14). The performance of Soda Lime was found to be scalable to the flow rate. At a lower flow rate of 8:1 , breakthrough was found to occur at 7:45:00. This reduction in flow of about 50% results in a doubling of the breakthrough time as expected.

[168] Table 2: Results of breakthrough testing for Soda Lime at a rate of flow of 8 ml/min per gram of material.

5.3 MOFs/Porous Polymers with rapid breakthrough.

[169] The performance of several of the candidates which showed good performance from a kinetic aspect was established. As evidenced in the Figure 15, the breakthroughs were rapid.

[170] The three best candidates, CAU-1 , CuBTC and SIFSIX-3-Ni showed longer (improved) breakthrough time. The best candidate of all, SIFSIX-3-Ni achieved more than 1 hour in breakthrough as shown in Figure 1 6. The breakthrough testing times are summarised in Figure 1 7.

5.4 The shape of the C0₂ adsorption Isotherm is key to achieving C0₂ uptake. [171 ] The vast difference in performance of MOFs at the required pressure of CO₂ is linked to the capacity of the MOF at low partial pressures. Figure 18 highlights two MOFs, SIFSIX-3-Ni and CAU-1 , that have similar CO₂ uptakes at 1 atmosphere. At the lower, more relevant, partial pressure of pure CO₂ (0.04 atm) a significant difference in uptake is clearly visible.

[172] The potential for improvement of the MOF has also been highlighted in Figure 18. CO₂ uptake in an ideal system is shown by the red curve. The star highlights the actual calculated CO₂ uptake of a SIFSIX-3-Ni extrudate adsorbent system from a mixed exhaled gas mimic system, showing significant potential for improvement. There are clearly differences in the conditions of the two experiments. Most notably, the red isotherm curve measures the CO₂ uptake in vacuum using pure CO₂ whilst the adsorbent breakthrough test measures the uptake in a mixed gas system.

5.5 Increasing bed pressure can increase C0₂ uptake.

[173] Improvements in quantities of CO₂ taken up by the SIFSIX-3-Ni system during breakthrough testing can be driven through an increase in the pressure across the adsorbent bed. Two experiments were performed on scaled flow (equivalent to 42 litres/min over 2.45kg Soda Lime). These experiments demonstrate comparable breakthrough times, as expected, although the increased pressure across the bead means that the actual quantity of CO₂ captured is increased. Increasing the pressure across the adsorbent bed therefore permits greater uptake of CO₂ per gram of adsorbent as illustrated in Figure 19 and shown in Table 3 below.

[174] Table 3: Comparison of Gas uptake for the extrudate and the powder of SIFSIX-3-Ni.

[175] As shown in Figure 7, using the SIFSIX-3-Ni powder results in an approximately 4-fold increase in CO₂ uptake during a breakthrough experiment compared to SIFSIX-3-Ni noodles. This technique offers a clear way to increase the performance of the MOF, although this will have to be balanced against other breathing requirements.

5.6 SIFSIX-3-Cu - not a stable alternative to SIFSIX-3-Ni

[176] A potential replacement of Nickel in the best performing MOF, SIFSIX-3-

Ni was trialled by replacing the Ni salt by a Cu salt species in the MOF and then determining the performance of this material. Several synthetic attempts were made for the SIFSIX-3-Cu derivative both in water and methanol solvents. Significant stability issues were found with the Cu derivative and although literature values claim good performance, the poor stability of this MOF makes it incompatible with use in exhaled breathing gas CO₂ capture. SIFSIX-3-Cu was found to degrade to its constituents rapidly on purification and processing.

5.7 Intermediate size experiments

[177] The initial experiments were performed on a small scale of approximately 2 grams with a squat aspect ratio of adsorbent bed. This increased the potential for packing issues, and the potential for short-paths which would yield non- optimal results during breakthrough testing. Note that no problems were observed with Soda Lime despite retaining the same particle size as used in the full-scale units. In order to assess the performance in a larger scale system a different adsorbent bed was used.

[178] The initial tests using real-world flows (40:2.45) scaled down to 8:1 g of adsorbent allowed the target adsorbents to be ranked. From these materials, a verification of performance was performed at a higher, three fold scale, using triangular noodles of CuBTC and SIFSIX-3-Ni to establish whether performance was scalable. For this reason tests were run using the 8:1 flow ratio but using an adsorbent bed three times larger but with the same diameter (i.e. more adsorbent over a longer path). These breakthrough tests were performed using triangular shaped extrudates to mimic the Soda Lime packing as closely as possible. [179] Initial experiments were performed with CuBTC to validate the 3x scale set-up. Follow-on experiments were performed with CAU-1 and SIFSIX-3-Ni, the other two top MOF based materials.

6. Preparation, testing and assessment of Large SIFSIX-3-Ni Batch.

6.1 Large scale synthesis

[180] A batch synthesis technique was employed to synthesise the SIFSIX-3- Ni batch. The synthesis protocol used, which can be scaled depending on the quantity required, was as follows.

[181 ] A 1 litre centrifuge bottle was charged with 12.01 1 g (67.42 mmol) of (NH₄)2SiF₆ ^*, followed by 19.01 1 g (67.42 mmol) of Ni(N0₃)₂, and 10.800 g (134.8 mmol) of pyrazine in 1 10 mL of water for 4 days. Prior to the addition of water, the powders were mixed by shaking until a purple moiety was formed. The final suspension was centrifuged to separate the SIFSIX-3-Ni product from the supernatant. If the supernatant had a strong blue colour the reaction mixture was allowed to stir for an additional day. The product was washed with water (3 x 100 mL) (ensuring the supernatant was colourless before moving on to the next washing stage). SIFSIX-3-Ni was then soaked in dry methanol for 2 days and then washed twice with dry methanol (80 mL). Centrifugation was used to separate the materials. The final paste for extrusion was prepared by centrifuging the SIFSIX-3-Ni at 4800 to 8000 RPM for 10 minutes in a 1 litre flask. The 500 ml of supernatant was discarded to yield approximately 300 ml of thick paste having the correct paste consistency for extrusion.

[182] It should be noted that (NH₄)₂SiF₆ is a toxic chemical and so it is added to a pre-weighed schott flask in the fumehood. Ni(NO₃)₂ and pyrazine amounts are then calculated according to the mass of (NH₄)2SiF₆ added. All material handling must be performed in with adequate PPE and the relevant laboratory equipment (including material handling in a fume cupboard).

6.2 Shaping

[183] The material pastes produced during the synthesis were extruded in a custom designed hand extruder having equilateral triangle dies of 1 .5 mm (see Figure 6(a)). The extrudates were dried for two hours prior to being cut to size with a scalpel (between 2 and 5 mm long). A total of 1 kilogram of MOF was manufactured from 2 kg of paste. The final density of the material is 0.60g/cm³.

6.3 Activation

[184] Activation of the shaped SIFSIX-3-Ni proved to be the complex element of the manufacturing process. The process required for activation of the material is typically heating to 120 °C under vacuum. The translation of scale from a maximum of 4 grams of material to in excess of 1 kg showed several unexpected problems. The first of these was that at the full scale the presence of small quantities of Hydrofluoric acid (HF) were detected during the activation process. This was coupled to the problem of getting sufficient heat into the material for successful activation. As HF is an extremely toxic and corrosive material, its generation had to be closely controlled and monitored.

[185] The first activation of the material (transitioning from the 2D pillars to the 3D structure) was achieved in three steps. The first step was effective solvent removal which was performed at 80 °C under reduced pressure for 48 hours. This was followed by the crucial activation step which changed the structure from a 2D to 3D structure, by heating the material to 1 50 °C at 35 mbar under continuous flushing of Nitrogen for two hours. The flushing was essential to prevent the build-up of any toxic component. This was followed by a final step of activation for 12 hours at 120 °C at 35 mbar. The material then cooled down for 4 hours under reduced pressure (35 mbar) and continuous nitrogen flushing prior to removal from the oven (60 °C removal temperature). All exhausts were performed through a fume cabinet.

[186] Subsequent activation can be performed simply through heating at 120 °C overnight at 35 mbar under nitrogen flushing. The material then cooled down for 4 hours under reduced pressure (35 mbar) and continuous nitrogen flushing prior to removal from the oven (60 °C removal temperature).

6.4 Activated material adsorption

[187] An adsorption isotherm of the activated material was performed to establish the efficacy of large scale material activation compared to established protocols at the smaller scale. Gas adsorption isotherms between the range of 0 - 780 mmHg were measured by a volumetric approach using a Micrometrics ASAP 2420 instrument. All the samples were transferred to predried analysis tubes, sealed with Transeal stoppers in a glove bag under N₂(_g). Ultra-high purity CO₂ gas was used for these experiments. CO₂ adsorption measurements were conducted at 298 K.

[188] The results are shown in Figure 20 which indicates that performance is approximately 50% of the performance obtained with optimised small-scale activation. There is therefore room for improvement of the large scale material activation protocol.

6.5 Low Flow benchmarking

[189] A Soda Lime test was performed using the testing apparatus described in section 5.2 using 1 .23kg of Soda Lime to reflect the quantity of MOF available for the full scale test. The flow of gas was scaled to reflect the quantity of material present so that the flow of gas 19.2 litres/min of air and 0.8 litres/min of CO₂.

[190] The breakthrough result was 3 hours and 1 1 minutes, with time to 3.5% 5 hours and 29 minutes. These results are completely within the expected standard and proved that the large scale testing methodology is robust. This test was performed in the commercial soda lime canister, unmodified (Figure 21 ).

6.6 High flow benchmarking

[191 ] This Soda Lime test was performed using the testing apparatus described in section 5.2 using 0.96kg of Soda Lime which filled a column inserted into the commercial soda lime canister (Figure 2A) to reflect the quantity of MOF available for the full scale test. The commercial soda lime canister was fitted to the test equipment described in section 5.2. The flow of gas was scaled to reflect the quantity of material present so that the flow of gas was 40 litres/min of air and 2 litres/min of CO₂.

[192] The breakthrough result was 50 minutes, with time to 4.5% 1 hour and 15 minutes (Figure 22). These results are completely within the expected standard and again proved that the large scale testing methodology is robust. This test was performed in a column inserted into the commercial soda lime canister to ensure the integrity of the small adsorbent bed in the presence of large gas flows.

[193] With these two base line tests performed to show that the equipment is performing as expected several breakthrough experiments were performed with SIFSIX-3-Ni. It should be noted that multiple unsuccessful attempts were made to perform these larger scale tests with no CO₂ capture observed. The tests reported here use the material tested in Section 1 .64. This material was recycled at least 8 times prior to successful tests being performed. This is recyclability was a key element to demonstrate in this project.

6.7 SIFSIX-3-Ni full flow test run

[194] A full flow test was performed with SIFSIX-3-Ni. 920 grams of the SIFSIX-3-Ni was placed in a 1 .2L column which was fitted to the inside of the canister (described below in section 6.9). This enabled us to obtain as tight a bed as possible in a geometry which closely resembled that of the canister. The bed was tested using 40 litre flow of air doped with 3.5% of CO₂ (2 litres/minute) using the testing apparatus described in section 5.2 fitted with this canister.

[195] CO₂ Breakthrough was found to happen also immediately, although there was notable CO₂ uptake observed as the point taken to reach system saturation took over 20 minutes. This experiment was performed twice with activation of the MOF between each run and comparable performance was noted, as shown in Figure 24. Heat was generated during the runs (as is typically observed with Soda Lime). Upon removal from the commercial soda lime canister, the bed was demonstrably hot in one section.

[196] This confirms that the bed is working and that CO₂ adsorption is happening, but the localisation of the heat is evidence that channelling or short paths are potentially developing which would account for some of the reduction in performance observed.

6.8 SIFSIX-3-Ni scaled flow test run [197] The same material and column insert were reloaded into the canister for further testing with a reduced gas flow (scaled to the quantity of adsorbent present in the column). The gas flow used was 13 litres per minute of air and 0.53 litres per minute of CO₂ with 960 grams of SIFSIX-3-Ni adsorbent. Figure 24 shows that the breakthrough time is extended and the time to saturation is in excess of one hour. This highlights that there are significant performance gains that remain to be made for this system with optimisation.

[198] It is evident from these results, that there is a significant performance gap between the small scale and the full scale results. There is significant room for optimisation of the physisorption MOF-based system.

6.9SIFSIX-3-Ni Test Canister

[199] A testing canister 190 was designed for full flow test as shown in Figure 25. This canister comprised a commercial soda lime canister 100 as previously described (Figure 2A) modified to include a smaller inner container containing the shaped SIFSIX-3-Ni material. The canister 100 includes inlet 1 10 and outlet 1 1 1 . As shown in Figure 25, the commercial soda lime canister 100 was modified to include an inner container 200 comprising a 1 .2L column 200 fitted with a fluid distributor disc 210 proximate the base and lid to retain the shaped SIFSIX-3-Ni material 215 between the discs 200. Each fluid distributor disc 210 comprised a metal disc with multiple holes drilled therethrough to allow fluid to flow through the packed SIFSIX-3-Ni material. The shaped SIFSIX-3-Ni material formed a compressed packed bed between the discs 200, and were manually compressed therebetween so that the adsorbent shaped bodies were tightly packed, thereby avoiding any flow short circuiting. The inner container 200 is designed to be removable to allow the SIFSIX-3-Ni material to be removed and then regenerated and activated for reuse. Regeneration processes can be conducted with the SIFSIX-3-Ni material in the inner container 200 if that container is constructed of suitable material for heating and reduced pressure, for example a metal. Alternatively, the SIFSIX-3-Ni material could be removed from the inner container 200 for regeneration.

[200] The inner container 200 is tightly fitted within the canister 100 using a resilient foam insert 230 which tightly fits the inner container 200 within the inner walls of the canister 100, and forms a fluid seal around the inner container forcing any gas flow to flow through the inner container 200 and the packed bed of SIFSIX-3-Ni material 215 inside that inner container 200.

[201 ] The fluid distributors 210 are spaced away from the respective top and base of the outer canister 100 to provide small gas circulation spaces 235 and 236 spaced apart from the packed bed of SIFSIX-3-Ni material 215. Each of these spaces comprise a liquid free gas flow space that receive and direct the breathing gas flow of a user through the packed bed of SIFSIX-3-Ni material 215 such that, in use, an exhaled gas stream can flow through the packed bed 215 to facilitate adsorption of carbon dioxide from the user's exhaled breathing gas. The spaces are substantially free of liquid in use, apart from liquid entrained in a user's exhaled breathing gas. It is noted that these spaces 235, 236 are much smaller than the equivalent space in a soda lime filter canister as this space must function as a liquid drainage area in a soda lime filter to contain liquids such as water that accumulates from the absorption reaction that occurs when using that material. The gas circulation spaces 235 and 236 are at least half the volume that must be provided in an equivalent soda lime filter canister.

[202] The testing container 190 can be fitted to a rebreather apparatus such as is illustrated in Figure 2A. The illustrated rebreather apparatus 150 includes a mouth piece 155 which is fitted to a user's mouth for breathing which is fluidly connected to the inlet and outlet of the testing container 190 via conduits 153 and 154. The rebreather apparatus may also include an additional gas cylinder 160 which includes make-up gas, for example oxygen, for use in this closed system.

6.10 Activation Isotherms

[203] Gas adsorption isotherms between the range of 0 - 780 mmHg were measured by a volumetric approach using a Micrometrics ASAP 2420 instrument. All the samples were transferred to predried analysis tubes, sealed with Transeal stoppers in a glove bag under N₂(_g). Ultra-high purity CO₂ gas was used for these experiments. CO₂ adsorption measurements were conducted at 298 K. [204] Activation protocols.

1 : For large quantities of materials (in excess of 200 grams of material per vessel) only poor activation can be achieved by placing the vessel at reduced pressure (35 mbar) at 120 ^°C for 24 hours under continuous flushing of Nitrogen. The material then cooled down for 4 hours under reduced pressure (35 mbar) and continuous nitrogen flushing prior to removal from the oven (60 °C removal temperature). All exhausts were performed through a fume cabinet.

2. For large quantities of materials (in excess of 200 grams of material per vessel) only intermediate activation can be achieved by placing the vessel at reduced pressure (35 mbar) at 150 ^°C under continuous flushing of Nitrogen for two hours a reduction of the temperature to 140 °C at 35 mbar under continuous flushing of Nitrogen for a further 12 hours. The flushing was essential to prevent the build-up of any toxic component. The material then cooled down for 4 hours under reduced pressure (35 mbar) and continuous nitrogen flushing prior to removal from the oven (60 °C removal temperature). All exhausts were performed through a fume cabinet.

3. Optimal activation of the material can only be achieved with small quantities of material (less than 5 grams) by heating at 120 °C for 12 hours under reduced pressure (35 mbar or less).

[205] Adsorption tests were run using the using the testing apparatus described in section 5.2. Gas adsorption isotherms between the range of 0 - 780 immHg were measured by a volumetric approach using a Micrometrics ASAP 2420 instrument. All the samples were transferred to predried analysis tubes, sealed with Transeal stoppers. Ultra-high purity CO₂ gas was used for these experiments. CO₂ adsorption measurements were conducted at 298 K.

[206] The results of these tests are shown in Figure 26, which illustrate the difference between poor activation of the shaped bodies (i.e. poor conversion from 2D to 3D structure), intermediate activation (partial conversion from 2D to 3D structure) and optimal activation (substantially all SIFSIX-3-Ni converted from 2D to 3D). 7. Conclusions

[207] The potential of adsorbents systems for the replacement of soda lime has clearly been demonstrated in this study. Performance at the gram-scale has been shown to be approximately 37% that of Soda Lime.

[208] The kinetics of CO₂ adsorption to the top performing system, SIFSIX-3- Ni, seem to be comparable to those of Soda Lime, as demonstrated by small scale experiments. This is a crucial finding and the preliminary results from the commercial soda lime canister scale tests seem to confirm that the rate of gas capture is not a limiting factor to the physisorption of CO₂.

[209] Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is understood that the invention includes all such variations and modifications which fall within the spirit and scope of the present invention.

[210] Where the terms "comprise", "comprises", "comprised" or "comprising" are used in this specification (including the claims) they are to be interpreted as specifying the presence of the stated features, integers, steps or components, but not precluding the presence of one or more other feature, integer, step, component or group thereof.

Claims

1 . A method of forming a carbon dioxide adsorbent of a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas, the method comprising:

preparing a SIFSIX-3-Ni paste comprising a mixture of SIFSIX-3-Ni material and a solvent, the SIFSIX-3-Ni material substantially comprising 2- dimensional SIFSIX-3-Ni structure;

forming the SIFSIX-3-Ni paste into a shaped body having at least one mean dimension of greater than 0.5 mm; and

heat treating the shaped body in a reduced pressure environment comprising a pressure of less than 500 mbar at a temperature of at most 160 °C to substantially remove the solvent from the shaped body and form 3- dimensional SIFSIX-3-Ni crystal structure in the shaped body,

thereby producing a shaped adsorbent body for use in a rebreather apparatus.

2. A method according to claim 1 , wherein at least 60%, preferably at least 70%, more preferably at least 90%, yet more preferably at least 95%, yet even more preferably at least 99% of the 2-dimensional SIFSIX-3-Ni structure of the shaped body is transformed into 3-dimensional SIFSIX-3-Ni crystal structure in the heat treatment step.

3. A method according to claim 1 or 2, wherein the heat treatment step is conducted at a pressure of less than 100 mbar, preferably less than 50 mbar, more preferably less than 35 mbar.

4. A method according to any preceding claim, wherein the heat treatment step is conducted at a temperature of between 1 10 to 160 °C.

5. A method according to any preceding claim, wherein the heat treatment step is conducted for at least 5 hours, preferably at least 8 hours, more preferably at least 10 hours.

6. A method according to any preceding claim, wherein the heat treatment step comprises an initial heating step of:

7. A method according to any preceding claim, wherein the heat treatment step comprises a temperature regime of:

8. A method according to any preceding claim, wherein the inactivated SIFSIX-3-Ni material is synthesised from a mixture of a SiF₆ precursor compound, a Ni precursor compound and pyrazine, in which the SiF₆ precursor compound is selected from a salt of SiF₆ and the Ni precursor compound is selected from at least one of a salt comprising Ni²⁺Ni(NO₃)₂.

9. A method according to any preceding claim, wherein forming the SIFSIX- 3-Ni paste into a shaped body comprises at least one of extruding, palletising or moulding the SIFSIX-3-Ni paste into a desired 3-dimensional configuration.

10. A method according to any preceding claim, wherein the shaped body has at least one mean dimension of greater than 0.8 mm, preferably at least 1 mm, preferably at least 1 .2 mm, and yet more preferably at least 1 .5 mm.

1 1 . A method according to any preceding claim, wherein each of the mean width, mean depth and mean height of the shaped body are greater than 0.5 mm, and preferably greater than 1 mm.

12. A method according to any preceding claim, wherein in the forming step, the SIFSIX-3-Ni paste is extruded into an elongate body which is subsequently longitudinally divided.

13. A method according to any preceding claim, wherein the shaped body comprises an elongate body having a circular, or regular polygonal cross- sectional shape.

14. A method according to any preceding claim, wherein the solvent comprises a non-basic polar solvent or a non-self ionising polar solvent, preferably an alcohol.

15. A method according to any preceding claim, further comprising the step of:

16. A method according to any preceding claim, wherein the heat treatment step is conducted in a fluid tight housing, and the method further comprises the step of:

flushing an inert gas through the fluid tight housing.

17. A method according to any preceding claim, wherein after the heat treatment step further comprising the step of:

18. A method according to any preceding claim, further comprising:

adsorbing carbon dioxide onto the surfaces of the SIFSIX-3-Ni shaped body; and

heating at a temperature of at least 80 °C, preferably 120 °C for at least 1 hour, preferably at least 2 hours at less than 100 mbar, preferably less than 50 mbar, thereby desorbing the carbon dioxide from the surfaces of the SIFSIX-3- Ni shaped body to produce a regenerated SIFSIX-3-Ni shaped body.

19. A method according to claim 18, wherein the heating step is conducted under inert gas flushing.

20. A method of forming a carbon dioxide filter of a rebreather apparatus for removing carbon dioxide from an individual user's breathing gas, the rebreathing apparatus, the method comprising:

producing a plurality of shaped adsorbent bodies according to any one of the preceding claims;

sealing a plurality of said shaped adsorbent bodies into a housing under an inert gas atmosphere.

21 . A method according to claim 20, wherein the inert gas atmosphere comprises at least one of nitrogen, or a Noble gas, preferably helium or argon.

22. A method according to claim 20 or 21 , wherein the housing includes two spaced apart support membranes configured to allow gas flow therethrough each membrane, the plurality of said shaped adsorbent bodies forming a packed bed therebetween and being compressed therebetween.

23. A method according to claim 22, wherein the housing includes an inner container housing the support membranes and packed bed, the inner container being seated within the housing using a fitted insert which at least friction fits the inner container within the inner walls of the housing.

24. A method according to claim 23, wherein the inner container is removably fitted within the housing.

25. A rebreathing apparatus for removing carbon dioxide from an individual user's breathing gas, the rebreathing apparatus including a carbon dioxide filter comprising:

a housing containing therein a packed bed of shaped adsorbent bodies,; wherein the shaped adsorbent bodies comprise SIFSIX-3-Ni having at least 60% 3-dimensional SIFSIX-3-Ni crystal structure, the balance being 2- dimensional SIFSIX-3-Ni structure, the shaped adsorbent bodies having at least one mean dimension of greater than 0.5 mm.

26. A rebreathing apparatus according to claim 25, wherein the housing includes an outlet that, in use, has a gas flow having about 0% CO₂ gas stream for at least 1 hour.

27. A rebreathing apparatus according to claim 25 or 26, wherein at least 70%, preferably at least 90%, more preferably at least 95%, yet more preferably at least 99% the shaped adsorbent bodies comprises the 3-dimensional SIFSIX-3-Ni crystal structure.

28. A rebreathing apparatus according to any one of claims 25 to 27, wherein the adsorbent bodies are sealed within the housing in an inert gas atmosphere prior to use.

29. A rebreathing apparatus according to any one of claims 25 to 28, wherein the housing includes two spaced apart support membranes configured to allow gas flow therethrough each membrane, the plurality of said shaped adsorbent bodies forming a packed bed therebetween and being compressed therebetween.

30. A rebreathing apparatus according to claim 29, wherein the housing includes an inner container housing the support membranes and packed bed, the inner container being seated within the housing using a fitted insert which at least friction fits the inner container within the inner walls of the housing.

31 . A rebreathing apparatus according to claim 30, wherein the inner container is removably fitted within the housing.

32. A rebreathing apparatus according to claim 29, 30, or 31 , further including:

a first liquid free gas flow space that is in fluid communication with an inlet for the user's exhaled breathing gas, said exhaled breathing gas including carbon dioxide;

a second liquid free gas flow space that is in fluid communication with an outlet for fluid flow of gas that is substantially free of carbon dioxide; wherein the first liquid free gas flow space is in fluid communication with the second liquid free gas flow space through the packed bed such that, in use, an exhaled gas stream can flow through the packed bed to facilitate adsorption of carbon dioxide from the user's exhaled breathing gas onto the adsorbent bodies.

33. A rebreathing apparatus according to any one of claims 25 to 32, wherein the shaped adsorbent bodies comprises elongate bodies having a circular, or regular polygonal cross-sectional shape.

34. A rebreathing apparatus according to any one of claims 25 to 33, wherein, the adsorbent bodies are configured to adsorb carbon dioxide from an individual user's exhaled gas, and be regenerated to a substantially carbon dioxide free state in situ, within the filter, by heat treatment.

35. A rebreathing apparatus according to claim 34, wherein heat treatment comprises heating the shaped adsorbent bodies at a temperature of at least 80 °C for at least 1 hour.

36. A rebreathing apparatus according to any one of claims 25 to 35, further including at least one breathing apparatus, a breathing mask, a breathing suit or other life support system in fluid connection to the carbon dioxide filter.