WO2008012559A1

WO2008012559A1 - Process for separating a fluoropropene from a mixture

Info

Publication number: WO2008012559A1
Application number: PCT/GB2007/002879
Authority: WO
Inventors: Robert Elliott Low; Stuart Corr
Original assignee: Ineos Fluor Holdings Limited
Priority date: 2006-07-27
Filing date: 2007-07-27
Publication date: 2008-01-31
Also published as: GB0614927D0

Abstract

A method for the separation of a fluoropropene from a mixture of the fluoropropene and at least one other fluorochemical, which method comprises the step of passing said mixture through a first distillation column whereby to separate a first azeotrope or near-azeotrope of the fluoropropene and at least one other fluorochemical from a first residue comprising the fluoropropene and passing said first residue through a second distillation column whereby to separate a second azeotrope or near-azeotrope of the fluoropropene and at least one other fluorochemical from a second residue comprising at least one other fluorochemical and collecting said second residue from the second distillation column.

Description

INTERNATIONAL SEARCH REPORT

(PCT Article 18 and Rules 43 and 44)

Applicant's or agent's file reference FOR FURTHER see Form PCT/ISA/220 ACTION as well as, where applicable, item 5 below. INECI/ P38384PC

International application No. International filing date (day/month/year) (Earliest) Priority Date (day/month/year) PCT/GB2007/ 002879 27/ 07/2007 27/07/2006

Applicant

INEOS FLUOR HOLDINGS LIMITED

This international search report has been prepared by this International Searching Authority and is transmitted to the applicant according to Article 18. A copy is being transmitted to the International Bureau.

This international search report consists of a total of 4 sheets.

[xl It is also accompanied by a copy of each prior art document cited in this report.

1. Basis of the report a. With regard to the language, the international search was carried out on the basis of: [x] the international application in the language in which it was filed

I I a translation of the international application into . . , which is the language of a translation furnished for the purposes of international search (Rules 12.3(a) and 23.1(b))

"^■ D With regard to any nucleotide and/or amino acid sequence disclosed in the international application, see Box No. I.

2. Q Certain claims were found unsearchable (See Box No. II)

3. Q Unity of invention is lacking (see Box No III)

4. With regard to the title,

Q] the text is approved as submitted by the applicant [x] the text has been established by this Authority to read as follows: PROCESS FOR SEPARATING A FLUOROPROPENE FROM A MIXTURE

5. With regard to the abstract,

[x] the text is approved as submitted by the applicant

Q the text has been established, according to Rule 38.2(b), by this Authority as it appears in Box No. IV. The applicant may, within one month from the date of mailing of this international search report, submit comments to this Authority

6. With regard to the drawings, a. the figure of the drawings to be published with the abstract is Figure No. _1

[xl as suggested by the applicant

I I as selected by this Authority, because the applicant failed to suggest a figure

I J as selected by this Authority, because this figure better characterizes the invention

"^• D none of the figures is to be published with the abstract

Form PCT/ISA/210 (first sheet) (April 2005) PROCESS FOR SEPARATING A FLUOROPROPENE FROM A MIXTURE

This invention relates to the separation of azeotropic or azeotrope-like mixtures comprising a fluorinated propene and other fluorochemicals.

Fluids containing fluorinated propenes have been suggested for use in various applications, such as heat transfer compositions, blowing agents, propellants, solvents and foaming agents. In order to provide suitable properties for use in these applications, the fluorinated propenes are often mixed with other components, such as other fluorochemicals.

For use a heat transfer compositions, such as refrigerants, it is often, desirable to use a blend of components which form an azeotrope or azeotrope-like mixture. Although such mixtures provide desirable., properties in use, they can lead to problems when handling the mixtures for disposal or other purposes. In particular, heat transfer compositions are often removed from heat transfer devices, such as air conditioning units, during maintenance or decommissioning. It would be desirable to be able to separate at least one of the components of such mixtures in order to make effective use of the various components.

According to one aspect of the invention, there is provided a method for the separation of a fluoropropene from a mixture of the fluoropropene and at least one other fluorochemical, which method comprises the step of passing said mixture through a first distillation column whereby to separate a first azeotrope or near-azeotrope of the fluoropropene and at least one other fluorochemical from a first residue comprising the fluoropropene and passing said first residue through a second distillation column whereby to separate a second azeotrope or near-azeotrope of the fluoropropene and at least one other fluorochemical from a second residue comprising at least one other fluorochemical and collecting said second residue from the second distillation column.

Preferably, the fluoropropene is selected from (E)-l,2,3,3,3- pentafluoropropene (R-1225ye), (Z)-l_J2,3,3,3-pentafluoropropene (R- 1225ye), 1,1 ,3 ,3, 3 -pentafluoropropene (R-1225zc)₅ 1,1,2,3,3- pentafluoropropene (R-1225yc)₅ cis-l,3,3,3-tetrafluoropropene (R-1234ze), trans-l,3,3,3-tetrafluoropropene (R-1234ze), 2,3,3,3-tetrafluoropropene (R- 1234yf), 3,3,3-trifluoropropene (R-1243zf)₅ and mixtures thereof.

Conveniently, the fluoropropene is 2,3,3,3-tetrafluoropropene (R-1234yf).

Advantageously, the first residue comprises at least about 90 % by weight ofR-1234yf.

Preferably, the first residue comprises at least about 95 % by weight of R- 1234yf.

Conveniently, the first residue comprises at least about 99 % by weight of R-1234yf.

Advantageously, the at least one other fluorochemical is selected from iodotrifluoromethane (CF₃I) and iodopentafluoroethane (CF₃CF₂I).

Preferably, the at least one other fluorochemical comprises iodotrifluoromethane (CF₃I).

Conveniently, the second residue comprises at least about 90 % by weight OfCF₃I. Advantageously, the second residue comprises at least about 95 % by- weight of CF₃I.

Preferably, the second residue comprises at least about 99 % by weight of CF₃I.

Conveniently, the second column operates at a lower pressure than the first column.

Advantageously, the second column operates at about 2 bar (200 kPa).

Preferably, the first column operates at about atmospheric pressure (101 kPa).

Conveniently, the second column operates at a lower temperature than the first column.

Advantageously, the method further comprises passing the second residue through a third distillation column.

Preferably, the mixture of a fluoropropene and at least one other fluorochemical comprises about 55 % by weight R~1234yf and about 45 wt % by weight CF₃I.

Conveniently, the mixture is a heat transfer composition which comprises R-1234yf and CF₃I.

Advantageously, the method further comprises the step of obtaining the mixture of R-1234yf and CF₃I from a heat transfer device. Preferably, the heat transfer device is an air-conditioning device.

The invention will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 shows a graph of the variation of the R-1234yf/CF₃I azeotrope with pressure,

Figure 2 shows a graph of the relative volatility of R-1234yf to CF₃I,

Figure 3 shows a graph of the component vapour pressure of R-1234yf and CF₃I,

Figure 4 shows a graph of the error plot for the fit of R-1234yf/CF₃I data .to: the PWRS model, and

Figure 5 is a schematic diagram of a distillation process using two columns.

2,3,3,3-Tetrafiuoropropene, formula CF₃-CF=CH₂, also described as R- 1234yf or HFC-1234yf₅ has been disclosed in mixtures comprising iodotrifluoromethane (CF₃I) by Singh, in US Patent Application Publication

No US2005/0233934A1. The mixture is taught by Singh as being azeotrope-like and hence difficult to separate. The evidence for the azeotrope is given as a depression of the boiling point of a mixture of the fluids as measured at close to atmospheric pressure in an ebulliometer.

We have found, surprisingly, that although this pair of fluids exhibits azeotropy at atmospheric pressure, the azeotropic composition is a sensitive function of system pressure, and disappears at higher pressures (above about 4 atmospheres). The azeotropes found are minimum-boiling azeotropes; that is, the boiling point is depressed relative to that of the pure components. Above about 4 atmospheres pressure the system no longer exhibits boiling point depression; instead an extremely close-boiling mixture having effectively constant boiling temperature is exhibited at high concentrations of R- 1234yf.

This feature of the system is unexpected and has allowed the invention of a separation process as will be further disclosed.

The variation of azeotropic composition (and its associated boiling point) with total pressure, based on measured data as discussed below, are shown in Figure 1.

Figure 2 shows the variation in relative volatility of R-1234yf to CF₃I as..a ranction of the mixture composition (expressed as weight fraction of R- 1234yf) at two different pressures.

Relative volatility is defined in. many standard chemical engineering textbooks, for example in Perry's Chemical Engineers' Handbook (pub. McGraw Hill).

The relative volatility of a fluid in a mixture is a measure of the ease of separation of that fluid from the mixture by distillation. A relative volatility of 1 for a component of a binary mixture signifies an azeotrope. Generally speaking a mixture having a relative volatility of greater than about 1.1 is more economical to separate by distillation and a relative volatility of less than 1.1 signifies increasing difficulty of separation. It is still possible however to design equipment that will effect separation of the mixture if the relative volatility is between 1.1 and 1.0. TMs composition variation in azeotropy and relative volatility makes it possible to design a distillation system for separation of the two fluids by exploiting the sensitivity of the azeotropic composition to pressure. Such separation may be desired: for example in reclamation of mixtures for reuse.

Exploitation of the variation of an azeotropic mixture composition with pressure to allow its separation is taught in EP0467531B1, which discloses a process for the separation of HFC-134a (1,1,1,2-tetrafluoroefhaiie) from mixtures comprising HF₅ with which it forms an azeotrope.

In the current invention the separation process consists of at least two distillation operations. Depending on the overall composition of the mixture to be separated the skilled man can choose to use additional, auxiliary distillation operations, as appropriate, to yield a process optimised for the application requirements; which may be to nώrimise total operating cost, total capital cost, or purity of one or more of the products.

A -nnn-Hmiting description of the invention is now provided with reference to Figure 5.

In the first (higher pressure) column, a feed material comprising a mixture of R-1234yf and CF₃I, having an average R-1234yf composition lower than the azeotropic composition at that column's operating pressure, is separated into a first stream having a composition equivalent to the azeotrope or at some composition intermediate between the azeotrope and the feed; and a second stream, having a composition intermediate between the feed and essentially pure CF₃I. The first stream is then passed to a second column, operating at a lower pressure than the first column. The pressure may be any convenient value which is lower than that of the first column, providing that it is sufficiently low that the azeotropic composition of R-1234yf at that pressure is lower than the composition of the first stream being fed to the column.

In this second column is separated a third stream, having a composition equivalent to the azeotrope or at some composition intermediate between the azeotrope and the feed; and a fourth stream, having a composition intermediate between the feed and essentially pure R-1234yf.

The third stream is then preferably recycled by a pump or other pressure . changing means to the first column. If it is removed as vapour it may be compressed and recycled as a gas; or it may be condensed to liquid before. recycle. It may be returned using gravity to the first column if suitable arrangement of relative elevations of the equipments are provided.

It may be admixed with the feed to the first column or may be admitted at some other point in the distillation column. Preferably, it is admitted at a point where the composition of the equivalent phase in the column matches the composition of the third stream. For example if it is recycled as liquid then it is preferably admitted at a point in the column where the the local liquid composition will be close to that of the third stream.

The component vapour pressure data, which can be used to infer the relationship between operating pressure and temperature, is presented in Figure 3

The skilled man, armed with the azeotropic information in Figure 1, the relative volatility data presented in Figure 2 and the vapour pressure data in Figure 3 can readily adapt the design to other desired combinations of product purity and operating conditions.

Example of Invention

For the sake of illustration the invention is described firstly for separation of a 50% (w/w) mixture of R-1234yf with CF₃I into two streams of 99% purity.

The process is as sketched in Figure 5. A feed mixture F comprising 50% w/w R-1234yf in CF₃I at a rate of 1000 kg/hr is admitted to column A₅ operating at a pressure of 2 bar absolute (200 kPa).

The azeotropic composition of the system at 200 kPa is found from Figure 1 to be 82% w/w R-1234yf. A practical column design will operate at a lower purity than the azeotrope; here a purity of 78% w/w is arbitrarily chosen for illustration. The top product Dl is therefore selected to be of this composition. The bottoms product Bl is specified as 99% pure CF₃I.

The top distillate Dl is then let down in pressure and passed to Column B, operating at atmospheric pressure (101.3 kPa). In this column the azeotropic composition is 71% w/w R-1234yf. Accordingly a tops product purity of 68% w/w is arbitrarily selected. This tops product D2 is pumped back to column Dl₅ where it may be admixed with the feed F or admitted to a different location, depending on the overall economics of the design. The bottoms product B2 from Column B can now be achieved as 99% pure R- 1234yf, as the feed to the column (at 78% purity) was more pure in R- 1234yf than the azeotropic composition at that operating pressure.

The material flows in the system are presented in Table 1. TABLE 1: Example of distillation system

The flows have been derived from these data by applying the principles of mass balance to the two columns. The solution to the mass balance was chosen so as to minimise the interchange of material between columns. This calculation can be done in any suitable software package, for example; a .. spreadsheetiαg tool such as Microsoft Excel, or a process simulation package such as Aspen Technology's AES.

Similarly, dependent on the cost of energy or equipment the designer may choose to vary the size of column or the target purities as appropriate to his or her needs. This may give rise to solutions where larger interchanges of material between the columns are obtained than the ones shown above for the same desired end purities. This reflects the nature of engineering design optimisation in that one may select process parameters to give a globally optimum process.

Details of experimental work

The vapour pressures of R-1234yf, CF₃I and binary mixtures thereof have been measured in a static cell apparatus of the general type described in "Modelling Fluorocarbon Vapor-Liquid Equilibria Using the Wong-Sandier Model" by MB Shiflett & SI Sandler in Fluid Phase Equilibria 147 (1998)

9 ppl45-162. The general procedure followed in this work is similar to that described in SMfLett and is described below. This procedure applies for both pure fluid and mixture measurements.

The cell is of accurately known volume and is first purged with nitrogen and evacuated to vacuum, then charged with a known mass of the material (or materials) of interest; whose quantity is accurately determined by weighing. The charge is repeatedly cycled by freezing with liquid nitrogen followed by thawing and venting of pressure, to allow degassing of light impurities. Following this operation the cell is brought to equilibrium in a thermostatically controlled oven. The contents of the cell are stirred throughout the experiment using an agitator inside the cell. The pressure and temperature in the cell are monitored electronically and recorded when they are stable within desired limits for a desired period of time. By variation of the cell temperature this generates a sequence of recorded pressures for a given charge composition.

The materials used in this measurement were analysed by GC beforehand and found to each be of better than 99% purity. Data were measured over the composition range from 100% CF₃I to 100% R-1234yf, and over the temperature range -20 to +6O⁰C.

The data found for the mixtures studied were then subjected to regression to suitable thermodynamic models capable of describing the phase equilibrium of an azeotropic mixture over this range of temperature and pressure.

The models selected were the Peng Robinson equation of state, using a temperature independent interaction constant with van der Waal's mixing rules, and the Peng Robinson equation of state, using the mixing rules of Wong and Sandler, with the NRTL equation to represent the excess Gibbs

10 energy of the liquid phase at infinite pressure. These methods are described more fully in the reference textbook "Modeling Vapor-Liquid Equilibria: Cubic Equations of State and Their Mixing Rules" by H Orbey and SI Sandler (pub. Cambridge University Press 1998) and in Shiflett.

For the simple Peng Robinson equation (using a temperature independent interaction constant with van άer Waal's mixing rules) the only adjustable parameter used was the interaction constant ku. For the more complicated model (using the mixing rules of Wong and Sandler, hereinafter referred to as PWRS) five adjustable parameters were used: the interaction constant ki₂, and four energy interaction parameters.

The regression technique used was a modification of the Barker method, in which for each datapoint the thermodynamic model selected is used to calculate the pressure in the cell from knowledge of the cell volume, temperature and mass of materials charged. This calculated pressure is then compared to the measured pressure to yield a pressure deviation (calculated pressure - measured pressure). A suitable mathematical technique is then used to vary the parameters of the phase equilibrium model to nrinimise an objective function derived from the sum of squares of the pressure deviations.

The objective function used in this work was a weighted sum of squares of pressure deviations. The weighting factors for each datapoint were calculated using the maximum likelihood principle, as described in "Vapor- Liquid Equilibrium Part V: Data Reduction by Maximum Likelihood" by HC Van Ness, F Pedersen and P Rasmussen, AIChE Journal 24(6) (1978) pp 1055-1063, to account for the effect of measurement errors on pressure and cell temperature.

11 The calculation was carried out in the MATLAB software package using an internal Levenberg-Marquardt non-linear optimisation routine to find the optimal parameter values.

The Barker method also requires a thermodynamic model of the saturated liquid and vapour densities as functions of temperature. In this work the liquid density was modelled using the relation of Hankinson, Brobst and Thomson as described in "The Properties of Gases and Liquids 4^th Edition" by RC Reid, JM Prausnitz and BE Poling (pub. McGraw Hill). The correlation parameters were optimised for prediction of the known density of saturated liquid CF₃I and R-1234yf. The vapour density was modelled using the Peng Robinson equation of state.

In addition to the measurements taken in this work the vapour pressure of CF₃I has been published in Fluid Phase Equilibria 1996 121 227-234; these data have been used to check the quality of the measured vapour pressure for the fluid, which were found to be in close agreement.

Both thermodynamic models require the critical temperature, critical pressure and acentric factor of each pure fluid. The representation of pure component vapour pressures by the Peng Robinson equation when using the PV^⁷RS model was improved by fitting the data to the Mathias Copeman function as described in F Rivollet et al, Fluid Phase Equilibria 218 (2004) ρp95-101.

The critical temperature and pressure for CF₃I were taken from the literature (Duan et al J Chem Eng Data 1999 44 501-504), The measured vapour pressure data from this work and from Duan were used to derive the acentric factor for CF₃I

12 The critical pressure and temperature of R-1234yf were measured using a critical point apparatus of the visual cell type. The acentric factor was derived from the vapour pressure measurements carried out in this work.

Both models gave good representation, of the cell pressure over the range of datapoints. Bofh models also showed similar variation of the azeotropic composition with pressure, although the calculated azeotropic composition given by the models was slightly different . The more sophisticated PWRS model was found to give a smaller average absolute error in calculated cell pressure over the temperature range and has been used to generate the data presented in Figures 1 and 2. Figure 4 is a plot of pressure deviation for each data point, as calculated using the optimised PWRS model. This shows the data set is well represented by the model.

13

Claims

1. A method for the separation of a fluoropropene from a mixture of the fhioropropene and at least one other fiuorochemical, which method comprises the step of passing said mixture through a first distillation column whereby to separate a first azeotrope or near-azeotrope of the fluoropropene and at least one other fiuorochemical from a first residue comprising the fluoropropene and passing said first residue through a second distillation column whereby to separate a second azeotrope or near- azeotrope of the fluoropropene and at least one other fiuorochemical from a second residue comprising at least one other fluorochernical and collecting said second residue from the second distillation column.

2. A method according to Claim 1 wherein the fluoropropene is selected from (Ε)-l,2,3,3,3-pentafluoroproρene (R-1225ye), (Z)- 1,2,3,3,3- pentafluoropropene (R-1225ye)₅ 1,1,3,3,3-pentafluoropropene (R-1225zc), 1,1,2,3,3-pentafluoropropene (R-i225yc), cis-l,3,3,3-tetrafluoropropene (R- 1234ze), trans~l,3,3,3-tetrafluoroρroρene (R-1234ze), 2,3,3,3- tetrafluoropropene (R-1234yf), 3,3,3-trifluoroρropene (R-1243zf), and mixtures thereof.

3. A method according to Claim 1 or 2 wherein the fluoropropene is 2,3,3,3-tetrafluoroproρene (R-1234yi).

4. A method according to Claim 3 wherein the first residue comprises at least about 90 % by weight of R-1234yf.

5. A method according to Claim 3 wherein the first residue comprises at least about 95 % by weight of R-1234yf.

14

6. A method according to Claim 3 wherein the first residue comprises at least about 99 % by weight of R-1234yf.

7. A method according to any preceding claim wherein the at least one other fluorocheπήcal is selected from iodotrifluoromethane (CF₃I) and iodopentafluoroethane (CF₃CF₂I).

8. A method according to any preceding claim wherein the at least one other fluorochemical comprises iodotrifluoromethane (CF₃I).

9. A method according to Claim 8 wherein the second residue comprises at least about 90 % by weight Of CF₃I.

10. A method according to Claim 8 wherein the second residue comprises at least about 95 % by weight of CF₃I.

11. A method according to Claim 8 wherein the second residue comprises at least about 99 % by weight of CF₃L

12. A method according to any preceding claim wherein the second column operates at a lower pressure than the first column.

13. A method according to any preceding claim wherein the second column operates at about 2 bar (200 kPa).

14. A method according to any preceding claim wherein the first column operates at about atmospheric pressure (101 fcPa).

15. A method according to any preceding claim wherein the second column operates at a lower temperature than the first column.

15

16. A method according to any preceding Claim further comprising passing the second residue through a third distillation column.

5 17. A method according to any preceding claim wherein the mixture of a fluoropropene and at least one other fluorochemical comprises about 55 % by weight R-1234yf and about 45 wt % by weight CF₃I

18. A method according to any preceding claim wherein the mixture is a l o heat transfer composition which comprises R- 1234yf and CF₃I.

19. A method according to Claim 18 further comprising the step of obtaining the mixture of R-1234yf and CF₃I from a heat transfer device. 5

20. A method according to Claim 19 wherein the heat transfer device is an air conditioning device.

21. A method for the separation of 2,3,3,3-tetrafluoropropene (R-1234yf) from a mixture of R-1234yf and iodotrifluoromethane (CF₃I) substantially0 as hereinbefore described with reference to and as shown in the accompanying drawings.

22. A method for the separation of iodotrifluoromethane (CF₃I) from a mixture of R-1234yf and CF₃I substantially as hereinbefore described with5 reference to and as shown in the accompanying drawings.

16