Field of Art

The present invention relates to a process for adsorptive separation of propene from propane. The process employs high-silica narrow-pore PCR zeolite prepared by top-down synthesis protocol from UTL precursor.

Background Art

Light olefins are one of the most important building blocks for production of various chemicals. They are produced mainly by catalytic and steam cracking, methanol-to-olefins process or oil pyrolysis. Propene, the most important light olefin after ethene, is used mainly for large-scale production of polypropylene, acrylonitrile, cumene, propylene oxide etc. The production of propene is also accompanied by the formation of a saturated analogue (propane) in a molar ratio of propene-to-propane from 4:1 to 1:1 depending on the conditions and method of cracking. Propane is a less valuable compound which is used as a working medium in refrigerators, fuel and aerosol propellant or it is converted to propene by catalytic dehydrogenation. Due to the stringent requirements for the purity of propene for further use (especially in polymer production), it is necessary to separate propane from propene. However, this separation is very difficult due to very close boiling points of both compounds (atmospheric pressure boiling points are -42.04 °C and - 47.53 °C for propane and propene, respectively) and thus very low relative volatility ranging from 1.09 to 1.15 depending on the mixture composition and pressure. At present, the mixture of propane and propene is separated by fractional distillation, also called superfractionation, which is the most energy-intensive distillation process practiced commercially. The large number of distillation trays and the high reflux ratio is the reason why the propene/propane separation consume about 80% of all energy to produce propylene. The energy intensity of this separation creates the need to find an alternative cost effective method of separation. Another motivating impetus in the search for an alternative, low-cost separation of C3 hydrocarbons is the measures taken by governments (e.g. the Clean Air Act in the USA) to suppress greenhouse gas emissions (which include, but are not limited to, hydrocarbons). Facilities for production and storage of polymers can no longer afford to dispose of waste hydrocarbons in their flare systems, hence the waste hydrocarbons must be redirected to the recovery process.

A separation process that does not require phase changes could reduce energy consumption by up to ten times.

Disclosure of the Invention

A very promising method for separation of propene from a mixture of gases containing propene and propane is adsorption on a suitable type of adsorbent. A suitable adsorbent should adsorb only one component of the mixture, be sufficiently mechanically and chemically stable, and the interaction of the adsorbate with the adsorbent should not be too strong to reduce the energy required to recover the adsorbed component and regenerate the adsorbent. The separation efficiency can be advantageously enhanced, in the case of microporous materials, by the so-called molecular sieve effect, where the components of the mixture differing in their kinetic cross-sections diffuse into the internal pore volume of the adsorbent at significantly different rates, or one of them does not enter the pores at all. Such materials may include certain types of zeolites. Zeolites are crystalline elementosilicates comprising tetrahedral connected three-dimensional frameworks, which possess orderly distributed micropores with the diameter of sub-nanometer size. Compared with other adsorbents, zeolites exhibit high surface area, pore volumes, thermal and chemical stability, they can act as a molecular sieve and are environmentally benign. Several zeolitic materials having pores with 8-rings defining their size have been investigated in the C3 hydrocarbon separation, especially zeolites with CHA, DDR and LTA structure. These zeolitic materials have entrance windows into channels with dimensions of about 0.4 nm, i.e. with dimensions very close to the kinetic diameters of propene and propane (kinetic diameter for propane and propene is 0.42 and 0.40 nm, respectively). Since propane is slightly larger, it diffuses significantly more slowly and the separation is kinetically controlled. However, these adsorbents have not yet been implemented on an industrial scale, mainly due to very slow diffusion of molecules, problematic regeneration or insufficient separation efficiency. Thus, an alternative separation process based on adsorption has not yet been satisfactorily provided and new adsorbents need to be sought which will be effective in separation of propene from propene/propane mixture. The present invention provides a process for separation of binary propane/propene mixture using a zeolite adsorbent, comprising the steps of contacting a zeolite adsorbent with a mixture of propane and propene, thereby adsorbing propene, and collecting non-adsorbed propane, and then recovering propene, wherein the zeolite adsorbent is a PCR zeolite.

The PCR zeolite (also denoted as IPC-4) is synthesized by so-called assembly-disassembly- organization-reassembly (ADOR) process, alternative synthesis route to the solvothermal synthesis. It starts by synthesis of parent UTL zeolite according to procedure reported in: Verified Synthesis of Zeolitic Materials, Third Revised Edition, p. 392; ISBN 978-0-692-68539-6. The UTL zeolite is then hydrolyzed in acidic aqueous solution at elevated temperature. The hydrolyzed UTL zeolite (also denoted as IPC-1P) is treated with neat octylamine, which intercalated between layers, followed by calcination resulting in condensation of individual silica layers through silanols. The process is described in: Roth, W., Nachtigall, P., Morris, R. et al. A family of zeolites with controlled pore size prepared using a top-down method. Nature Chem. 5, pp. 628-633 (2013); https://doi. org/10.1038/nchem.1662.

The PCR zeolite is a structural type of zeolite having channels defined by ten-membered and eight- membered rings that intersect in perpendicular directions. Entrance windows into pores have dimensions 3.4 x 5.6 A (10-MR) and 3.5 x 4.7 A (8-MR).

The PCR zeolite usually consists of Si and O with optional presence of residual Ge. Silicon-to- germanium atomic ratio is typically greater than 50. PCR is aluminium-free material, it means that aluminium is not detectable by X-ray fluorescent spectrometry in the material.

Because the zeolite is aluminum-free, the number of cationic centers that would increase the propene interaction energy and the risk of adverse and side reactions, such as propene polymerization at these centers, is strongly reduced.

Preferably, the PCR zeolite has BET surface area of at least 200 m ² /g, preferably at least 220 m ² /g. The micropore volume of the PCR zeolite is at least 0.075 mL/g, preferably at least 0.08 mL/g. The BET surface is as measured by physical adsorption of nitrogen at -196 °C, and the micropore volume is determined from nitrogen adsorption isotherm by t-plot methodology employing Harkins- Jura equation.

PCR zeolites are typically prepared by ADOR process, which comprises the steps of:

- hydrolysis of UTL zeolite in aqueous solution of HC1 or acetic acid at a temperature within the range of 25 - 90 °C,

- intercalation of octylamine at a temperature of 60 °C or 80 °C,

- calcination of the product at 550 - 650 °C for 4 - 16 hours.

The propane/propene mixture can have different ratios of propane and propene, usually in the range of molar propene- to-propane ratios from 4:1 to 1:1.

The PCR zeolite has shown surprisingly high differences in diffusivity of propane and propene, which allows a very effective separation of the mixtures. An advantage of the PCR zeolite is that while it selectively adsorbs one component of the mixture (propene), thus allowing an effective separation of the components of the mixture, it does not catalyze the polymerization of the olefin due to the absence of trivalent heteroatoms in the framework, so there is no risk of undesirable side reactions.

For example, for the propane/propene mixture, the PCR zeolite shows a propene/propane diffusion ratio greater that 100 (at 30 °C and 115 mbar), as well as an adsorption capacity for propene higher than 25 mg/1 g of adsorbent at 30 °C and 900 mbar, while showing the adsorption capacity for propane lower than 3 mg propane/1 g of adsorbent at 30 °C and 900 mbar. The combination of these characteristics then results in selectivity of the adsorptive separation estimated by Ideal Adsorbed solution Theory (IAST) higher than 10 ⁷ at 1000 mbar and 30 °C for equimolar mixture of propene and propane.

The step of adsorbing propene is preferably carried out by at least one of: multi-stage pressure swing adsorption, single pressure swing adsorption, vacuum swing adsorption, single temperature swing adsorption, multi-stage temperature swing adsorption and combinations thereof. The process can be conducted in flow (continuous) systems as well as batch (discontinuous) systems. The non-adsorbed propane flows into a buffer vessel at the discharge side of the adsorber.

The step of recovery of propene is preferably carried out by a desorption process selected from stripping with a stripping gas, by reducing the pressure, raising the temperature, or any combination of these methods.

The stripping gas should be selected from cheap and abundant gases which can be easily separated from propene or which do not interfere with further use of propylene. Stripping gases to regenerate the adsorbent and recover the adsorbed component of the separated mixture can be e.g. steam or nitrogen.

The adsorbent (PCR zeolite) used in the process of the present invention exhibits low adsorbed amounts of propane and high separation selectivities even at temperatures up to 120 °C and can therefore be operated over a wide temperature range.

The adsorption temperature should be kept below the temperature at which propene begins to react (below the cracking temperature). The lower the temperature, the higher the amount of propene that is adsorbed to the adsorbent at a given working pressure. Therefore, it is advisable to keep the temperature below 250 °C, more preferably below 120 °C, even more preferably below 100 °C and most preferably below 35 °C.

In particularly preferred embodiments, the adsorption is carried out at an adsorption temperature between 25 °C and 80 °C and total pressure for adsorption above 1 bar.

Brief Description of Drawings

FIG. 1 shows SEM images of the PCR adsorbent (Z-l).

FIG. 2 shows XRD patterns of the Z-l (a), Z-2 (b) and Z-3 (c) PCR adsorbents. To clarify the data diffractograms of Z-2 and Z-3 were moved vertically by 3000 and 6000 counts, respectively. FIG. 3 shows nitrogen adsorption isotherms of the Z-l (a), Z-2 (b) and Z-3 (c) PCR adsorbents. To clarify the data isotherms of Z-2 and Z-3 were moved vertically by 20 and 40 cm ³ /g, respectively.

FIG. 4 shows adsorption isotherms for propene (empty points) and propane (full points) on the Z- 1 (a), Z-2 (b) and Z-3 (c) PCR adsorbents at pressures of about 0 to 1000 mbar and temperature of 30 °C.

FIG.5 shows the amount of propene (empty points) and propane (full points) adsorbed on Z-3 PCR adsorbent over time at 30 °C under equilibrium pressure 115 mbar of propene.

FIG.6 shows adsorption selectivity of adsorptive separation of propene from the mixture consisting of 20 mol.% propane and 80 mol.% propene at 30°C on the Z-l (a), Z-2 (b) and Z-3 (c) PCR adsorbents estimated by IAST as a function of equilibrium pressure.

FIG. 7 shows adsorption isotherms for propene and propane on Z-3 at pressures of about 0 to 1000 mbar and temperature of 80°C.

Materials and Methods

Crystallinity of synthesized adsorbents were checked by X-ray powder diffraction (XRD) on D8 Advance Eco (Bruker AXS) applying CuKa radiation (l = 1.5406 A). The step size of 0.02° and a step time of 0.5 s were used. The patterns were collected over the 2Q range from 2° to 60° and evaluated by using the Diffrac.Eva V 4.1.1.

Morphology of the adsorbent particles were evaluated by scanning electron microscopy using a JEOL JSM-7500F instrument with a cold cathode - field emission and LYRA 3 (Tescan) microscope equipped with EDS analyzer AZtec X-Max 20 (Oxford Instruments). The EDS measurements of Ge content were performed at 5 kV and 20 kV acceleration voltage on five 400x400 pm spots for each studied sample.

All adsorption isotherms were measured using a Micromeritics ASAP 2020 volumetric apparatus. Before each measurement, the adsorbents were degassed by heating in vacuum at 250 °C for 8 h. Textural properties of adsorbents were evaluated from nitrogen adsorption isotherms measured at -196 °C (liquid nitrogen temperature). BET method was used to calculate surface area of materials using adsorption data in the range of relative pressures p/po= 0.05 - 0.2. The adsorbed amount at p/po = 0.98 reflects the total pore volume. The micropore volumes and external surface area of adsorbents were obtained using standard Micromeritics software by plotting and evaluation of t- plot. Adsorption isotherms of individual hydrocarbons (propane and propene) were measured at a stepped pressure increments from about 1 mbar to about 1000 mbar. The experimentally determined isotherms were fitted to the Sips isotherm model and the fitted parameters were then used to calculate the selectivity of adsorption by Ideal Adsorbed Solution Theory (IAST).

Rate of adsorption was measured on MK2-M5 vacuum microbalances (Cl Precision, UK) connected with vacuum dosing line. The equilibrium rate was measured at 30 ° C on a Z-3 adsorbent that was exposed to a single dose of a given hydrocarbon leading to the reaching of equilibrium pressure between 110 and 120 mbar.

Examples

A series of adsorbents with PCR crystal structure were prepared according to the general procedure described above in various amounts and from different batches of starting precursor (UTL zeolite).

EXAMPLE 1

Three batches of PCR zeolite (denoted Z-l, Z-2 and Z-3) were prepared from UTL zeolite with Si/Ge = 5.1 (Si/Ge = atomic ratio of Si and Ge) under the reaction conditions summarized in TABLE 1. Calcined UTL was hydrolyzed with acid of selected type and concentration (TABLE 1, entries “acid” and “concentration”, respectively), at appropriate <zeolite mass/acid solution volume> ratio (TABLE 1) for 16 h at 90 °C in a plastic bottle. The solid was isolated by filtration and centrifugation, washed with water and dried at 40 °C for 10 h. The obtained solid was mixed with octylamine and heated (component ratios and conditions are listed in TABLE 1). The solid was isolated by centrifugation, decantation of the supernatant, dried in an open tube in air at 90 °C for 6 h, and then calcined at 550 °C for 6 h. The structure and morphology of the particles of the prepared adsorbents were subjected to characterization by electron microscopy, XRD and measurement of nitrogen adsorption isotherms. A small amount of the sample (ca. 2 mg) was sprayed onto the sample holder and placed in the microscope chamber. After reaching the operating vacuum the micrographs of particles at various magnifications were taken. The sizes of the particles were evaluated by the tool of ImageJ software package. The adsorbent shows crystallites in the form of thin plates of approximately rectangular shape with a side length of 20 - 60 pm x 10 - 30 mih and a thickness of about 0.3 - 0.4 mih (FIG.l). Diffraction patterns of the samples were measured in the thin layer of the fine powders in air (in hydrated form). The material is crystalline, having a monoclinic Cm space group crystal lattice, characterized by typical diffraction lines at angles 7.94°, 9.63°, 12.67°, 14.66°, 15.8°, 15.95°and 19.39° as is shown in the FIG.2. Surface area and porosity of the adsorbents (summarized in TABFE 2) were derived from nitrogen adsorption isotherms displayed in FIG. 3. A fine powder sample weighing 60-95 mg (TABFE 2, entry “m”) was used to measure textural properties. All three samples were gently degassed on an automatic degassing unit (SmartVacPrep, Micromeritics) by gradually reducing the pressure at a rate of 1 Torr/s, and after reaching a vacuum (setpoint 5 mTorr) by slowly heating to 100 ° C at a rate of 0.5 °C/min. At this temperature, the sample was evacuated for 60 min and then heated at 1 °C/min to 250 ° C, where it was further evacuated for 6 h. Subsequently, the sample was freely cooled to room temperature under dynamic vacuum and then the cuvette was pressurized 760 Torr with pure dry nitrogen (99.9999%). N2 adsorption was measured in a liquid nitrogen bath over the entire range of relative pressures (p/po = 0 - 0.99) at a minimum of 60 points. Equilibrating time for each dose of adsorptive was 5 s (equilibrium is reached when the pressure change per equilibration time period (5 s) is less than 0.01% of the average pressure during the interval). Measured data were evaluated by MicroActive software package (Micromeritics). All three adsorbents exhibit microporous character with volume of micropores around 0.083 cm ³ /g and BET area around 235 m ² /g (see TABLE 2).

TABLE 1 ^a Si-to-Ge atomic ratio in final PCR adsorbent TABLE 2 ^a mass of degassed samples used for volumetric measurements in ASAP 2020 device ^b specific surface area determined from linearized BET equation using adsorption data in the range of relative pressures p/po = 0.05 - 0.2 ^c volume of micropores determined by t-plot approach using Harkins-Jura equation ^d total pore volume was determined from amount adsorbed at p/po = 0.98

EXAMPLE 2

Equilibrium adsorption isotherms of pure propane and pure propene at 30 °C on all three adsorbents are shown in FIG. 4. The isotherms were measured using a Micromeritics ASAP 2020 volumetric apparatus on the same samples of adsorbent as in EXAMPLE 1 after the same degas procedure as described in EXAMPLE 1. The temperature of adsorption was controlled (±0.1 °C) by Peltier thermostat with iso-propanol as heat transfer medium. In the case of hydrocarbons adsorption equilibrium measurements, the equilibration time interval was set to 20 s. The isotherms clearly show significant propene adsorption over the entire pressure range, while propane adsorption is very low. The maximum adsorption capacity of adsorbents Z-l to Z-3 on propene and propane at 30 °C, and of Z-3 at 80 °C is shown in TABLE 3. The maximum adsorption capacity was estimated from the Sips adsorption isotherm model using the equation:

Adsorption selectivity at equilibrium pressure of 1000 mbar was determined for equimolar mixture

_ A/» _ P ° and for 20/80 mixture of propane and propene, according to the equation: — - — 5- . x _] /y p TABLE 3 ^a adsorption of hydrocarbon at 80 °C ^b propane(C ₃ )-to-propene(C _3- ) molar ratio ^c adsorption selectivity (a _ad s) at equilibrium pressure of gas mixture 1000 mbar

EXAMPLE 3

Gravimetric determination of diffusion coefficients was performed in a volumetric vacuum apparatus equipped with three pressure gauges in the range 0-10, 0-100 and 0-1000 mbar and thermostated at 35 ° C, which was connected to very sensitive vacuum microbalances MK2-M5 (Cl Precision). A 61.87 mg of Z-3 sample was placed in the balance, and a thin stainless steel wire of the same weight was used as a counterweight. The sample was degassed by slow evacuation at room temperature overnight, then heated to 100°C with heating rate 3°C/min and kept for 30 min. Subsequently, the temperature was increased to 250°C (by heating rate of 5 °C/min) and maintained overnight. Then, the sample was cooled down to room temperature, Peltier thermostat of the same type as in the case of volumetric measurements was installed (cf. EXAMPLE 2) and temperature of 30°C was set. In subsequent step, the Z-3 sample was contacted with dose of propane (or propene) expanding from the dosing volume of the volumetric part of device until equilibrium was obtained. FIG. 5 shows the time dependence of propene and propane uptake on Z-3 adsorbent at 30 °C measured during equilibration of adsorption system. Time dependence of relative uptake (mt/mequii, where mt means adsorptive uptake at time t, m _eqUii. is equilibrium uptake) is related to D/r ² , where D is diffusion coefficient and r is the crystal radius based on mathematical description of diffusion by equation given by J. Crank in The mathematics of diffusion, page 327, econd edition 1975, Oxford University Press, ISBN 0-19-853411-6: Zeolite used in present invention is characterized by its unexpectedly high diffusional distinction between propane and propene. Resulting D/r ² coefficients are 2.5 10 ⁴ s ¹ for propene and 1.7 10 ⁶ s ¹ for propane. Ratio of the respective D/r ² coefficients is 147, thus propene diffuses into PCR adsorbent 147-times faster than propane.

EXAMPLE 4

Co-adsorption of propene and propane is simulated by Ideal Adsorbed Solution Theory (IAST), described in: A.L. Myers, J.M. Prausnitz, Thermodynamics of mixed-gas adsorption, AIChE J. 11 (1965) 121-127, based on the isotherm data presented in EXAMPLE 2. Based on the knowledge of the adsorption equilibrium of the pure components, the selectivity of propene adsorption (a _a ds) from a mixture of 80 mol. % propene and 20 mol. % propane (typical ratio for fluid catalytic cracking products) at 30 °C is predicted by IAST (FIG. 6). There are a number of software tools for solving the IAST equation, in our case the procedure programmed in the SciLab software package was used. The selectivity reaches the value between 10 ³ and 10 ¹⁹ depending on the adsorbent and equilibrium pressure, indicating very selective preferential adsorption of propene over propane. For all the adsorbents, which are the subject of the present invention, the selectivity of the adsorption increases very steeply with increasing the equilibrium pressure. Thus, performing an adsorption process utilizing the adsorbent described herein at higher pressures is advantageous and desirable. The molar ratio of propene to propane in the C3 fraction of cracking products may vary depending on the conditions and type of process. Usually, however, the propane content does not exceed 50 mol. %. With the increase of the propane content in the mixture up to 1:1, the selectivity values decrease, but by no more than two orders of magnitude (TABLE 3) , so that the selectivity value does not fall below 10 ⁷ . Thus, the adsorbent of the invention is selective in wide range of mixture compositions.

EXAMPLE 5

FIG. 7 shows the single component adsorption isotherms of propene and propane recorded at 80 °C on Z-3 adsorbent. The experimental procedure was the same as in EXAMPLE 2 using of the same sample after measurement of N2 adsorption isotherm and hydrocarbons isotherms (thus the same mas of sample was used). Prior the measurement, sample was degassed by the same procedure as described in EXAMPLE 2. The propene isotherm is less steep compared to the isotherm measured at 30 °C due to a shift in equilibrium in favor of free gas, but still the amount adsorbed is high compared to propane amount adsorbed at the same temperature. The adsorption capacity (TABLE 3) decreased to 60% of the capacity at 30 °C, while the capacity on propane by 70% compared to the capacity at 30 ° C. Resulting adsorption selectivity (aads calculated for equilibrium pressure of gas mixture 1000 mbar) estimated based on IAST theory (TABLE 3) is even higher than the adsorption process carried out at 30 °C. Thus, adsorption process of this invention is very selective in wide range of temperatures as is documented by calculated IAST selectivity reported in TABLE 3. Industrial Applicability

The separation process employing the adsorbent of the present invention can be used in variety of petrochemical and petroleum refinery processes to purify and separate propene from the mixture of propene and propane. It can optionally be combined with fractionation.