Field of the Invention

The present invention is directed to metal-organic frameworks which can be used to remove hydrocarbons from an inert gas by adsorption, as well as a process for removing

hydrocarbons from an inert gas.

Background

Inert gases are used to create an inert atmosphere in a variety of industrial processes.

Accordingly, the removal of contaminants from inert gases is of interest in a number of fields. Therefore, solid sorbent materials capable of adsorbing contaminants from inert gases are of great interest.

For example, argon is used in the preparation of silicon ingots for applications including the solar cell industry. The production of silicon for the solar cell industry requires silicon to be melted and recrystallized in large furnaces under an argon blanket. Cost is a huge driver in this industry and so recycling of the argon used is beneficial. The current solution for recycling argon uses vacuum pumps within the system to facilitate gas flow into the recycling unit. The use of these oil pumps can introduce methane into the system in unacceptable amounts, which can reduce the recycling rate of the argon.

Summary of the Invention

There remains in particular a need for efficient methods of removing hydrocarbons from inert gas systems, so that recycling rates of inert gases can be improved. There is a direct environmental benefit associated with reduction in the use of inert gases.

Accordingly, the present invention provides a method of removing one or more

hydrocarbon(s) from an inert gas, said method comprising: i) contacting said inert gas with a metal-organic framework adsorbent to adsorb the hydrocarbon(s) from the inert gas; wherein the metal-organic framework has the MOF-74 structure with a pore size in the range of and including 10 to 75 A and comprises a metal M, wherein M is Mg, Mn, Fe,

Co, Cu, Ni or Zn.

It is surprising that hydrocarbons, in particular low levels of hydrocarbons (e.g. up to and including 100 ppm), can be selectively removed from inert gases by adsorption using a metal-organic framework adsorbent wherein the metal-organic framework has the MOF-74 structure with a pore size in the range of and including 10 to 75 A and comprises a metal M, wherein M is Mg, Mn, Fe, Co, Cu, Ni or Zn. It is also beneficial that the adsorption occurs at ambient temperature.

It is also surprising that, even though hydrocarbons can be selectively adsorbed in the presence of an inert gas (in particular a vast excess of inert gas), the hydrocarbons are not bound so strongly that that they cannot be readily removed. Therefore, it is also a benefit of the invention that the metal-organic framework can be regenerated by supplying a desorbing gas flow at ambient temperature and pressure.

Due to excellent selective hydrocarbon adsorption and ease of regeneration, the inventors consider that the invention is of particular use in removing hydrocarbons such as methane from inert gases such as argon in silicon furnaces used in the solar cell industry. However, it will be understood that the benefits of the invention are not limited to this application. Indeed, any application which requires hydrocarbons to be removed from an inert gas will benefit from the present invention.

The present invention also provides a method of preparing pellets of a metal-organic framework, wherein the metal-organic framework has the MOF-74 structure with a pore size in the range of and including 10 to 75 A and comprises a metal M, wherein the method comprises the steps of:

(i) roll-compacting the metal-organic framework at a pressure in the range of and including 1000 to 100000 KPa to provide pellets of the metal-organic framework;

(ii) sieving the pellets produced in step (i) to provide pellets having a size of no more than 1500 pm; wherein

M is Mg, Mn, Fe, Co, Cu, Ni or Zn.

Additionally, the present invention provides pellets of a metal-organic framework, wherein the metal-organic framework has the MOF-74 structure with a pore size in the range of and including 10 to 75 A and comprises a metal M, wherein: the pellets have a size of no more than 1500 pm as measure by sieving; the pellets have a crush strength of at least 20 g; and

M is Mg, Mn, Fe, Co, Cu, Ni or Zn.

The inventors have surprisingly found that roll-compacting the metal-organic framework to give pellets having a size of no more than 1500 pm and a crush strength of at least 20 g further improves the ability of the metal-organic framework to selectively adsorb hydrocarbons in the presence of an inert gas. Moreover, the metal-organic framework can be regenerated over a number of cycles.

Brief Description of the Drawings

Fig. 1a shows a breakthrough curve for 100 ppm methane in Ar for a reactor blank.

Fig. 1 b shows a breakthrough curve for 100 ppm methane in Ar for a reactor blank, Fe-BTC and Cu-BTC.

Fig. 1c shows a breakthrough curve for 100 ppm methane in Ar for a reactor blank, Fe-BTC, Cu-BTC, and Ni-CPO.

Fig. 2a shows a breakthrough curve for 100 ppm methane in Ar and subsequent

regeneration at a temperature of 175°C for Ni-CPO.

Fig. 2b shows a closeup of the two adsorption events in Fig 2a.

Fig. 3 shows a breakthrough curve for Ni-CPO showing initial uptake and uptake after flow degassing with clean Ar at room temperature.

Fig. 4 shows a breakthrough curve for 100 ppm methane in Ar for a reactor blank and roll- compacted Ni-CPO.

Fig. 5a shows repeat adsorption and desorption cycles for 100 ppm methane in Ar and roll- compacted Ni-CPO.

Fig. 5b shows recorded breakthrough times per cycle of adsorption and desorption shown in Fig. 5a.

Detailed Description

Preferred and/or optional features of the invention will now be set out. Any aspect of the invention may be combined with any other aspect of the invention unless the context demands otherwise. Any of the preferred and/or optional features of any aspect may be combined, either singly or in combination, with any aspect of the invention unless the context demands otherwise.

The inert gas containing the hydrocarbon (s) may be any inert gas. Accordingly, the inert gas includes N2 as well as noble gases such as He, Ne, Ar, Kr, Xe or Rn and mixtures thereof. Suitably, the inert gas is Ar or N2 or a mixture thereof. Preferably, the inert gas is Ar.

In the invention, the inert gas is contacted with the metal-organic framework, which removes some or all of the hydrocarbon(s) from the inert gas by adsorbing the hydrocarbon(s). The inventors have found that, in particular, the metal-organic framework adsorbent can remove hydrocarbon(s) from inert gases containing low concentrations of hydrocarbon(s).

Accordingly, the inert gas may comprise no more than 500 ppm hydrocarbons, typically no more than 100 ppm hydrocarbon (s), suitably no more than 75 ppm hydrocarbon(s), preferably no more than 50 ppm, more preferably no more than 25 ppm hydrocarbon(s), for example no more than 10 ppm hydrocarbon(s). The hydrocarbon(s) may be selected from linear and branched, saturated and unsaturated, C1-C12 hydrocarbons including alkanes, alkynes and alkenes. For example, all isomers and stereoisomers (enantiomers and diastereoisomers) of methane, ethane, propane, butane, pentane, hexane, heptane, octane, nonane, decane, undecane and dodecane. Also, all isomers and stereoisomers

(enantiomers and diastereoisomers) of ethene, propene, butene, pentene, hexene, heptene, octene, nonene, decene, undecene and dodecene. Suitably, the hydrocarbon (s) is/are selected from linear and branched, saturated and unsaturated, Ci-Cs hydrocarbons, preferably linear and branched, saturated and unsaturated, C1-C5 hydrocarbons. Preferably, the hydrocarbon(s) is/are linear. Preferably, the hydrocarbon(s) is/are saturated. For example, the hydrocarbon(s) typically is/are selected from methane, ethane, propane (preferably /7-propane) butane (preferably /7-butane) and pentane (preferably /7-pentane), suitably the hydrocarbon is methane.

It is a benefit of the invention that the hydrocarbon(s) can be removed from the inert gas by contacting said inert gas with the metal-organic framework at ambient temperature.

Accordingly, the step of contacting the inert gas with the metal-organic framework may advantageously be carried out at a temperature of no more than 40 °C, typically no more than 35 °C, suitably no more than 30 °C. The minimum temperature is not particularly limited in the present invention, but may for example be at least 0 °C, at least 10 °C or at least 15 °C.

The inert gas is contacted with the metal-organic framework by subjecting the metal-organic framework to a flow of the inert gas. The flow rate of the gas is not particularly limited, but is typically no more than 50 Nm ³ /hr, suitably no more than 40 Nm ³ /hr, preferably no more than 30 Nm ³ /hr. The flow rate is typically at least 5 Nm ³ /hr, at least 10 Nm ³ /hr, or at least 15 Nm ³ /hr.

The method of the invention may also comprise a step ii) of regenerating the metal-organic framework by desorbing hydrocarbon(s) from the metal-organic framework. The

regeneration step may be carried out by supplying a desorbing gas flow to the metal-organic framework. The nature of the desorbing gas flow is not particularly limited. Conveniently, the desorbing gas flow can be a noble gas, such as He, Ne, Ar, Kr, Xe or Rn, or N2, or a mixture thereof. For example, the desorbing gas flow can be Ar or N2 or a mixture thereof.

Accordingly, the desorbing gas flow can be Ar. It may be preferable to use a desorbing gas with the same identity as the gas from which hydrocarbon (s) is/are adsorbed.

The desorbing gas flow preferably contains no trace of hydrocarbon(s) or other impurities, such as H2O. Accordingly, the desorbing gas flow is preferably moisture free. The flow rate of the desorbing gas is not particularly limited but may typically be no more than 50 Nm ³ /hr, suitably no more than 40 Nm ³ /hr, preferably no more than 30 Nm ³ /hr. The flow rate may typically be at least 5 Nm ³ /hr, at least 10 Nm ³ /hr, or at least 15 Nm ³ /hr. The desorbing gas flow will contain desorbed hydrocarbon(s) after being supplied to the metal-organic framework. After being supplied to the metal-organic framework (i.e. after flowing through the metal-organic framework) the gas will be vented.

It is advantageous that regeneration of the metal-organic framework can occur at ambient temperature e.g. at a temperature of no more than 40 °C e.g. no more than 35 °C, or no more than 30 °C, typically at least 10 °C, at least 15 °C or at least 20 °C. Suitably, regeneration occurs approximately in the range of and including 20 to 23 °C.

Notwithstanding this, if desired, regeneration may also be carried out at elevated

temperature e.g. in the range of and including 50 °C to 300 °C, suitably 100 °C to 200 °C.

It is also advantageous that regeneration of the metal-organic framework can occur at ambient pressure e.g. at a pressure of no more than 120 KPa, e.g. no more than 110 KPa, typically at least 90 KPa. Suitably regeneration occurs at approximately 100 KPa. A skilled person will be aware that ambient pressure can vary slightly. All variations are included here. Notwithstanding this, regeneration can also be performed at reduced pressure e.g. at a pressure less than ambient pressure. For example, regeneration may occur at a pressure in the range of and including 25 KPa and ambient pressure, suitably in the range of and including 50 KPa and ambient pressure.

It will be understood that in the process of the present invention the steps of contacting the inert gas with a metal-organic framework and subsequently regenerating the metal-organic framework may be repeated one or more times over the lifetime of the metal-organic framework. The number of times that the metal-organic framework is regenerated is not particularly limited, and, for example, may be repeated hundreds or thousands of times.

Prior to contacting the inert gas with the metal-organic framework, the metal-organic framework may be degassed to activate the metal-organic framework for adsorption of the hydrocarbon(s). For example, the metal-organic framework may be subjected to a vacuum for a period of time that is not particularly limited but may be, for example, at least 5 hours or at least 10 hours. For example, the time may be no more than 48, or no more than 36 hours. Typically, the metal-organic framework may be left under vacuum overnight. The temperature at which the metal-organic framework is subjected to vacuum is also not particularly limited, but can be, for example, in the range of and including 100 °C to 200 °C e.g. approximately 150 °C.

The metal-organic framework in the present invention has the MOF-74 structure. Put another way, the metal-organic framework is MOF-74, or a metal-organic framework which is isoreticular with MOF-74. MOF-74 has the formula metal ₂ (2,5-dihydroxy-1 ,4- benzenedicarboxylate) comprising a metal and the 2,5-dihydroxy-1 ,4-benzenedicarboxylate ligand (also known as a“linker”) otherwise known as H4DOBDC or DOT

(dioxidoterephthalate). The term“isoreticular” is known in the art. Two metal-organic frameworks that are isoreticular possess the same topology, otherwise known as net.

Accordingly, a metal-organic framework which is isoreticular with MOF-74 possesses the same topology or net as MOF-74, which is given the three letter identifier etb in the art (a list of net structures and their exact meaning and be found at the Reticular Chemistry Structure Resource, http://rcsr.anu.edu.au). Compounds isoreticular with MOF-74 possess a hexagonal honeycomb pore structure with metals lining the pores, in which the metals are joined along the long axes (i.e. in the hexagonal plane) by the ligand. The metals have uncoordinated sites accessible to the pores. MOF-74 and metal-organic frameworks which are isoreticular with MOF-74 can be prepared by methods known in the art, for example by reacting a salt containing the divalent cation of the metal with an appropriate ligand in a suitable solvent for example, as outlined in Deng, H.X. et al. Science, 2012, 336, 1018.

The pore size of the metal-organic framework is at least 10 A. The pore size of the metal- organic framework may be no more than 75 A, suitably no more than 45 A, typically no more than 25 A, preferably no more than 12 A. It is known in the art that the term“pore size” used in connection with metal organic frameworks means the diameter of the largest sphere that can fit within a given cavity. Pore size can be obtained from the pore size distribution (PSD) curve, i.e. the probability to find pores of a given size in the structure. PSD is calculated using the methodology of Gelb and Gubbins (L.D. Gelb, K.E. Gubbins, Pore size

distributions in porous glasses: a computer simulation study, Langmuir, 1999, 15, 305-308) using the Universal Force Field (UFF) for Lennard Jones parameters. The methodology consists of trying to position spheres of increasing diameter into the free volume of the unit cell to determine the largest sphere able to fit in the structure, using Monte Carlo

calculations. Evidently, the sphere occupies the free pore volume of the unit cell and cannot be superposed with the space occupied by atoms of the structure. Using this methodology, it is possible to determine the PSD. The pore size of the metal-organic framework is determined by the size of the ligand.

Accordingly, the metal-organic framework can be any compound M2L, wherein is a L is a di- carboxylate ligand and M is as defined herein, providing that the metal organic-framework has the MOF-74 structure with a pore size in the required range. Suitably, the di-carboxylate ligand is 2,5-dihydroxy-1 ,4-benzenedicarboxylate, 3,3'-dihydroxy-4,4'-biphenyldicarboxylate, or a linear oligophenylene ligand terminated at both ends with a 2-hydroxy-benzoate moiety. A linear oligophenylene ligand in this context is a ligand comprising three or more phenylene groups, including the phenylene groups of the terminal 2-hydroxy-benzoate moieties. The terminal 2-hydroxy-benzoate moieties are linked to adjacent phenylene groups by a sigma bond at the para- position relative to the carboxylate group, and the remaining phenylene groups are liked to adjacent phenylene groups by sigma bonds which are in the para- position relative to one another. An example of such a linear oligophenylene ligand is 4-[4- (4-carboxylate-3-hydroxyphenyl)phenyl]-2-hydroxybenzoate, which has three phenylene groups.

Suitably, the oligophenylene ligand has no more than nine phenylene groups, typically no more than five, preferably no more than three. The phenylene group(s) other than the terminal 2-hydroxy-benzoate moieties may be unsubstituted or independently substituted.

For example, each phenylene group may be independently substituted with one or more substituent(s) which may be the same or different. The oligophenylene ligand may contain a mixture of substituted and unsubstituted groups but, typically, either all the phenylene groups are substituted or none of the phenylene groups are substituted. Substituted phenylene groups may have different substitution patterns but, typically, when more than one phenylene group is substituted all the substituted phenylene groups have the same substitution pattern. The substituents on a single substituted phenylene group may be different but, typically, all the substituents on a single substituted phenylene group having more than one substituent are the same. Each substituted phenylene group may have the maximum four substituents, or three substituents but, preferably, each substituted phenylene group has no more than two substituents. Preferably, the substituent(s) is/are an alkyl group. Suitable alkyl groups include C1-C5 alkyl groups, typically C1-C3 alkyl groups. Preferably, the alkyl group(s) is/are linear. An example of such a linear substituted oligophenylene ligand is 4-[4-(4-carboxylate-3-hydroxyphenyl)-2,5-dimethylphenyl]-2-h ydroxybenzoate.

The metal M in the metal-organic framework is Mg, Mn, Fe, Co, Cu, Ni or Zn, suitably Ni, Co or Zn, preferably Ni.

Accordingly, the metal-organic framework may be M ₂ (2,5-dihydroxy-1 ,4- benzenedicarboxylate), M ₂ (3,3'-dihydroxy-4,4'-biphenyldicarboxylate), M ₂ (4-[4-(4- carboxylate-3-hydroxyphenyl)phenyl]-2-hydroxybenzoate), or M ₂ (4-[4-(4-carboxylate-3- hydroxyphenyl)-2,5-dimethylphenyl]-2-hydroxybenzoate), wherein M is Mg, Mn, Fe, Co, Cu, Ni or Zn, Ni, Co or Zn, preferably Ni. Preferably the metal-organic framework is M ₂ (2,5- dihydroxy-1 ,4-benzenedicarboxylate). Accordingly, the preferred metal-organic framework is MOF-74(Ni), also called Ni-CPO which has a pore size of 11.45 A.

The inventors have surprisingly found that the ability of the metal-organic framework to adsorb hydrocarbon(s) from the inert gas can be improved yet further when it is prepared by the pellet preparation method of the invention or is a pellet according to the invention. The term“pellet” takes its conventional meaning in the art e.g. a small compacted mass of substance. The term roll-compacting also takes its conventional meaning in the art e.g. application of pressure to a powder by compaction between two rollers. Accordingly, the MOF provided for roll-compaction is preferably in powder form. Roll-compaction takes place at a pressure in the range of and including 100 to 100000 KPa, suitably 2500 to 7500 KPa, preferably 4000 to 6000 KPa, for example approximately 5000 KPa. The pellets of the invention, including the pellets prepared by the corresponding method of the invention, have a size of no more than 1500 pm, suitably no more than 1250 pm, preferably no more than 1000 pm. The pellets may have a size of at least 250 pm, suitably at least 500 pm, preferably at least 750 pm. Accordingly, the pellets may have a size in the range of and including 250 to 1500 pm, suitably 500 to 1250 pm, preferably 750 to 1000 pm.

The term“size” in this context means the dimension of the pellets determined by sieving using calibrated sieves. It is within the capability of a skilled person to determine pellet size using sieves, and it is within the capability of a skilled person to produce pellets with the required pellet size by sieving fractions.

As well as having a size of no more than 1500 pm, the pellets of the invention and the pellets prepared by the method of the invention have a crush strength of at least 20 g, typically at least 30 g, suitably at least 40 g, preferably at least 50g. The pellets may have a crush strength of no more than 150 g, typically no more than 125 g, suitably no more than 100 g, preferably no more than 90 g. For example, the pellets may have a crush strength in the range of and including 20 to 150 g, typically 30 to 125 g, suitably 40 to 100 g, preferably 50 to 90 g. Crush strength is known in the art and is the greatest compressive stress that a brittle solid (i.e. a solid which fractures without significant plastic deformation when subjected to stress) can sustain without fracture. As a skilled person knows, crush strength is measured using a mechanical strength testing machine having a load cell. Specifically, crush strength is measure by placing a specimen between two platens on a mechanical strength testing machine, e.g. a CT6 Small Mechanical Strength Testing Machine (Engineering Systems Ltd®) with a 5 kg load cell. The cell is moved downward, applying force to the specimen until it fractures. The fracture point is detected, in this specific example

automatically, and the force at fracture value is recorded. The recorded value is the crush strength, which is expressed in grams.

It will be understood that the method of removing one or more hydrocarbon (s) from an inert gas of the invention encompasses the use of both metal-organic frameworks which are not pellets of the invention and are not prepared by the pellet preparation method of the invention, as well as those that are.

When used to remove one or more hydrocarbon(s) from an inert gas, the metal-organic framework may be present in a packed bed. Accordingly, the invention also provides a packed bed containing the pellets of the invention. However, this is not a limiting

arrangement for the metal-organic framework and a skilled person will be aware of other arrangements that can be used for effective contact of the metal-organic framework with the inert gas containing hydrocarbon(s). A packed bed (or other arrangement) of metal-organic framework will be present in a position through which the inert gas containing

hydrocarbon(s) can be flowed. Accordingly, a packed bed (or other arrangement) may be located within a system which is flowing inert gas. Alternatively, some or all of the inert gas may be removed from the system, contacted with the packed bed in an external unit, and returned to the system. External units include, for example, gas recycling units appended to industrial apparatus such as, for example, a silicon ingot furnace. The gas recycling unit may contain facilities for the removal of other impurities from the inert gas, such as carbon dioxide and/or water. Such a recycling unit may also have a monitoring system to monitor the performance of the metal-organic framework (and other impurity removing facilities) which can, for example, indicate when the metal-organic framework needs to be regenerated by desorbing hydrocarbon(s) from the metal-organic framework.

Examples

Hydrocarbon adsorption performance general method

Formed metal-organic framework (MOF) material (-0.2 g) was placed as a bed in a plug flow reactor tube with an internal dimeter of 0.8 mm. The sample was subjected to a gas stream of Ar (100 ml min ¹ ) prior to switching to methane/Ar (100 ppm, 100 ml min ¹ ). The methane in the resulting gas downstream of the MOF bed was monitored using an IR spectrometer. The breakthrough time was determined as the time required for detection of either 50 ppm or greater than 0 ppm of methane by the IR spectrometer. Degassing experiments were conducted using a clean Ar flow (100 ml min ¹ ) at either room temperature or 175 °C. Results without roll-compaction of the MOF

Fig 1a shows the breakthrough curve of a blank reactor with a recorded break through time of 1.36 min (i.e. the time lapse between the switch to methane/Ar, represented by the distance between the vertical line in the figure and the breakthrough trace curve at 50 ppm methane).

Fig. 1b shows that MOFs Cu-BTC and Fe-BTC (which have the tbo topology, not the MOF- 74 structure) display no improvement in breakthrough time over the blank reactor (i.e. the breakthrough curves for Cu-BTC and Fe-BTC lie on top of the curve for the blank reactor).

It can be seen clearly in Fig 1c that Ni-CPO (or MOF-74(Ni)) shows a marked improvement in breakthrough time when compared to the blank, Fe-BTC, and Cu-BTC (i.e. it is longer before 50 ppm methane is detected in the gas downstream of the MOF).

Furthermore, Fig. 2a shows that Ni-CPO not only improves the breakthrough, but can be regenerated using heat (i.e. degassing with a clean Ar flow (100 ml min ¹ ) at 175 °C). The initial material shows a breakthrough time of 13.4 min g ^-1 (Fig 2. a shows the breakthrough time for 0.2 g which is about 2.68 mins). After this adsorption event the material was heated to 175 °C under an Ar flow to remove methane from the pores and regenerate the material. After degassing the material was re-exposed to 100 ppm methane/Ar and the breakthrough time was 17.05 min g ^-1 . Fig. 2b shows a closeup of the two adsorption events in Fig 2a.

Fig. 3 shows how the Ni-CPO can also be regenerated using flowing Ar gas alone, without an increase in temperature. The initial breakthrough time was recorded as 13.4 min g ^-1 however after only 10 mins of the sample being subjected to a flow of clean Ar a breakthrough time of 11 min g ^-1 was achieved - representing a capacity regeneration of 82 %. Breakthrough times for this procedure are summarised in Table 1.

Table 1: A summary of breakthrough times.

Condition Breakthrough time (min g ¹ )

Initial 13.4

After degas @ 175 °C 17.05

After degas with Ar 11 Roll-compaction

Ni-CPO was roll compacted using 5000 KPa and sieved to produce pellets between 850-915 p , and having a crush strength of 75g. Crush strength was measured by placing a specimen between to platens on a CT6 Small Mechanical Strength Testing Machine (Engineering Systems Ltd®) with a 5 kg load cell. During testing the cell was moved downward, applying force to the specimen until it fractured. The fracture point was automatically detected, and the force at fracture is recorded. The recorded value was 75g.

Prior to breakthrough measurements the material was degassed at 150 °C under vacuum overnight. Breakthrough measurements were the same as above in accordance with the general method, but in these experiments breakthrough time was measured as the time taken before any methane is observed in the FTIR above instrument error (i.e. a stricter test than the detection of 50 ppm methane).

Fig. 4 shows the breakthrough curve for roll compacted Ni-CPO. Superior breakthrough times of ~ 14 min/ g ^_1 were observed. This is significantly better than the Ni-CPO used in the example above, where breakthrough times were measured at 50 ppm methane.

The same material was subjected to multiple adsorption-desorption events. Between exposure to methane the samples were degassed (at room temperature) by flowing dry N2 through the samples. Fig. 5a shows the breakthrough curves obtained while Fig. 5b shows the recorded breakthrough times per cycle. Capacity is retained between cycles and only flowing N2 (without an increase in temperature) in required to regenerate the material.