Ethanol - combined pressurized distillation and membrane filtering
PT 20161738 (344244)
Key information
Title
Ethanol - combined pressurized distillation and membrane filtering
Application number
20161738
Case type
National
Status
Filed
02.11.2016
Effective date
02.11.2016
Publicly available
03.05.2018
Next annual fee due
30.11.2026
Applicant
ABADJOM CONSULTING AS (NO)
Owner
ABADJOM CONSULTING AS (NO)
Inventor
Martin Haugstvedt (NO) et al.
Granted
21.10.2019
Patent number
344244
Expiry date
02.11.2036
Abstract and drawing
Disclaimer: This text has been machine-scanned and may contain errors – please refer to "Publications" for legally binding content.
An improved hybrid process for production of ethanol from aqueous solution is previously investigated within the project. The modular process consists of a pressurized distillation/stripper column followed by vapor permeation module using a water selective membrane under the same pressurized conditions. Thorough experimental work on the membrane separation has been performed, applying commercially available HybSi® membranes; the results have been used as input for comprehensive process modelling. Pressures in the range of ~6-10 bar(a) and temperatures in the range of ~130-150°C have been investigated. The results indicate that the energy consumption and cost for production of fuel grade ethanol (<1 wt % water) can be significantly reduced with this process as compared to the traditional set up with 2 columns and dehydration step (molecular sieves). The combination of one pressurized distillation column followed by a membrane separation step operated under the same pressurized condition, make these findings an invention.
Publications
B1A1
Checking availability of the latest published version...
Documents
ExpandDownload files (46)
Date
Date
Doc. No.
Process
Case number
In/Out
Journal description
To/from
02.10.2026
26-01
Saksbehandling
20161738(344244)
UT
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
ABADJOM CONSULTING AS
02.10.2025
25-01
Saksbehandling
20161738(344244)
UT
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
ABADJOM CONSULTING AS
02.10.2024
24-01
Saksbehandling
20161738(344244)
UT
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
ABADJOM CONSULTING AS
09.04.2024
23-01
Saksbehandling
20161738(344244)
UT
PT Påminnelse om ikke betalt årsavgift (3329) (PT20161738)
ABADJOM CONSULTING AS
04.10.2023
22-01
Saksbehandling
20161738(344244)
UT
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
ABADJOM CONSULTING AS
04.10.2022
21-01
Saksbehandling
20161738(344244)
UT
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
04.10.2021
20-01
Saksbehandling
20161738(344244)
UT
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
06.10.2020
19-01
Saksbehandling
20161738(344244)
UT
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
PT Varsel om betaling av årsavgift for år 4 + (3352) (PT20161738)
28.08.2019
16-01
Saksbehandling
20161738
UT
PT Meddelelse om patent
ABADJOM CONSULTING AS
24.05.2019
15-06
Saksbehandling
20161738
INN
Krav
ABADJOM CONSULTING AS
24.05.2019
15-05
Saksbehandling
20161738
INN
Beskrivelse
ABADJOM CONSULTING AS
24.05.2019
15-04
Saksbehandling
20161738
INN
Patenttegninger
ABADJOM CONSULTING AS
24.05.2019
15-03
Saksbehandling
20161738
INN
Sammendrag
ABADJOM CONSULTING AS
24.05.2019
15-02
Saksbehandling
20161738
INN
Krav NO
ABADJOM CONSULTING AS
24.05.2019
15-01
Saksbehandling
20161738
INN
Korrespondanse (Hovedbrev inn)
ABADJOM CONSULTING AS
25.02.2019
14-01
Saksbehandling
20161738
UT
Realitet patent
ABADJOM CONSULTING AS
19.12.2018
13-04
Saksbehandling
20161738
INN
Beskrivelse
ABADJOM CONSULTING AS
19.12.2018
13-03
Saksbehandling
20161738
INN
Patenttegninger
ABADJOM CONSULTING AS
19.12.2018
13-02
Saksbehandling
20161738
INN
Sammendrag
ABADJOM CONSULTING AS
19.12.2018
13-01
Saksbehandling
20161738
INN
Korrespondanse (Hovedbrev inn)
ABADJOM CONSULTING AS
02.10.2018
12-01
Saksbehandling
20161738
UT
PT Batch Varsel om betaling av første årsavgift (3317)
28.09.2018
11-01
Saksbehandling
20161738
UT
Realitet patent
ABADJOM CONSULTING AS
10.08.2018
10-04
Saksbehandling
20161738
INN
Patenttegninger
ABADJOM CONSULTING AS
10.08.2018
10-03
Saksbehandling
20161738
INN
Sammendrag
ABADJOM CONSULTING AS
10.08.2018
10-02
Saksbehandling
20161738
INN
Beskrivelse
ABADJOM CONSULTING AS
10.08.2018
10-01
Saksbehandling
20161738
INN
Korrespondanse (Hovedbrev inn)
ABADJOM CONSULTING AS
15.05.2018
09-01
Saksbehandling
20161738
UT
Realitet patent
ABADJOM CONSULTING AS
29.11.2017
08-04
Saksbehandling
20161738
INN
Sammendrag
ABADJOM CONSULTING AS
29.11.2017
08-03
Saksbehandling
20161738
INN
Patenttegninger
ABADJOM CONSULTING AS
29.11.2017
08-02
Saksbehandling
20161738
INN
Beskrivelse
ABADJOM CONSULTING AS
29.11.2017
08-01
Saksbehandling
20161738
INN
Korrespondanse (Hovedbrev inn)
ABADJOM CONSULTING AS
31.05.2017
07-02
Saksbehandling
20161738
UT
PT report 10:47:22
31.05.2017
07-01
Saksbehandling
20161738
UT
Realitet patent
ABADJOM CONSULTING AS
04.04.2017
06-03
Saksbehandling
20161738
INN
Avtale Bio Oil AS dets konkursbo og Abadjom AS
ABADJOM CONSULTING AS
04.04.2017
06-02
Saksbehandling
20161738
INN
Erklæring Søkers rett til oppfinnelse
ABADJOM CONSULTING AS
04.04.2017
06-01
Saksbehandling
20161738
INN
Korrespondanse (Hovedbrev inn)
ABADJOM CONSULTING AS
13.02.2017
05-01
Saksbehandling
20161738
UT
Henlagt manglede svar
ABADJOM CONSULTING AS
04.11.2016
04-01
Saksbehandling
20161738
UT
Formalia 1
ABADJOM CONSULTING AS
04.11.2016
03-01
Saksbehandling
20161738
UT
Infobrev til oppfinner
ABADJOM CONSULTING AS
04.11.2016
02-01
Saksbehandling
20161738
UT
Generelt Brev Patent
ABADJOM CONSULTING AS
02.11.2016
01-04
Saksbehandling
20161738
INN
Sammendrag
ABADJOM CONSULTING AS
02.11.2016
01-03
Saksbehandling
20161738
INN
Beskrivelse
ABADJOM CONSULTING AS
02.11.2016
01-02
Saksbehandling
20161738
INN
Tegninger
ABADJOM CONSULTING AS
02.11.2016
01-01
Saksbehandling
20161738
INN
Søknadsskjema Patent
ABADJOM CONSULTING AS
Description, claims and drawings
Disclaimer: This text has been machine-scanned and may contain errors – please refer to "Publications" for legally binding content.Description
Ethanol – combined pressurized distillation and membrane filtering -Detailed description and claimsField and background of the inventionSeveral possible separation strategies are found in the literature for separation of ethanol from fermentation broth. Distillation is currently the workhorse of this process with a well-earned reputation of being both robust, energy efficient and easy to implement. Most alternative separation strategies still rely on the difference in relative volatility of water and ethanol and include the usage of membranes or gas stripping.The heat of combustion of ethanol is approximately 30 MJ/kg. For ethanol concentration in the range 3 − 5 wt % the energy required for distillation to 94 wt % is in the order of 5 − 8 MJ/kg. For more dilute feeds the energy demand rises rapidly to 20 MJ/kg for an ethanol concentration of 1 wt % feed. These figures are for distillation schemes with highly optimized heat recovery and integration. For dilute ethanol-water systems there is therefore incentives for finding less energy intensive separation techniques lowering the total cost of operation and capital. Alternative separation strategies using membranes can include pervaporation and gas permeation coupled with distillation or fractional condensation. Some of these options and their challenges lead to this invention.A lot of attention has recently been given to ethanol selective membranes and the pervaporation process. When using pervaporation, liquid feed evaporates through the membrane and gives a vapor permeate. An enrichment of one component relative to others are achieved due to the difference in flux for the individual compounds. This phase change from liquid to vapor for the permeate requires an energy equal to the heat of evaporation. The driving force for the flux is the difference in partial pressures (e.g. component chemical potentials) over the membrane. For diluted systems with typical 3 − 5 wt% ethanol in water a significant vacuum is required to create a partial pressure on the permeate side that is lower than on the feed side, and thus have a positive driving force. For a feed of 5 wt% ethanol at 1 bara, a typical permeate side pressure used in lab scale membrane testing is less than 1 millibar. The flux is strongly coupled with the difference in partial pressures over the membrane, thus lowering the permeate side pressure will increase the flux.Cooling is required to condense the vapor permeate, and the dew point temperature of the permeate is highly dependent on pressure. For a membrane with high ethanol selectivity yielding a permeate of 80 wt% ethanol at 2 kPa absolute pressure, the permeate has a dew point temperature of 8 ⁰ C. A heat exchange, with reasonable limit on heat surface exchange area will require a cooling medium to be at least 20 ⁰ C lower than this. This cooling requires quite significant investments in equipment and adds to the cost of operation for the compressors in the cooling cycle. The cooling alone might be more expensive than the alternative base case distillation unit.Permeate pressure in the 10 kPa magnitude impose challenges to the operation of the equipment. To maintain this low pressure in an industrial setting, the equipment must
be very resilient to leaks. Minor misalignments in fittings and changes in flanges and seals can over time cause leaks making it impossible to maintain the required vacuum level.Although pervaporation requires low temperature coolant and low pressures on the permeate side, the advantage compared to normal distillation is that less energy is required for evaporation. In distillation, a significant amount of energy is used to generate the reflux needed to attain a given purity within the bounds given by the number of plates available, whereas pervaporation requires only the heat of evaporation for the mass passing through the membrane. However, the added complexity of low vacuum and refrigeration adds significantly to both operation and investment.Given this background, the process was changed from ethanol selective membranes and the pervaporation process. Instead The process of the invention involves mechanical compression of the system followed by vapor stripping, vapor permeation and condensation.There is little to be found in the literature on the pervaporation of ethanol and water at low concentrations and high pressure. This is most likely due to the ease of lab scale experiments using vacuum, and the availability of cooling traps using liquid nitrogen.Patents investigated – Comparing to prior artAlthough there are several patents available relevant for ethanol production using a combination of distillation and membrane separation, the idea of using an overall pressurized process appear to be unexplored. From an industrial viewpoint, the new invention will dramatically reduce the membrane area. Hence, it is of interest to evaluate the possibility of patenting this process. A number of patents/patent applications have been investigated in order to evaluate the patentability of the mentioned process. The most relevant are listed in the following.Leland Vane et. Al:- Leland M. Vane, Franklin Alvarez, U.S. Patent Application publication 2013/0015052 A1 “Liquid separation by membrane assisted vapour stripping process” (filed June 6, 2012).- Humbird, D., et al., Process Design and Economics for Biochemical Conversion of Lignocellulosic Biomass to Ethanol, National Renewable Energy Laboratory Report NREL. TP-5100-47764, 2011.- Leeland M. Vane, Franklin Alvarez, Yu Huang, and Richard Baker, US Patent 8,114,255: “Membrane Augmented Distillation with Compression to Separate Solvents From Water” (issued February 14, 2012)- Leeland M. Vane, Franklin Alvarez, U.S. Patent Application 11/896201, publication 2009/0057128 A1 “Liquid separation by membrane assisted vapour stripping process” (filed August 30, 2007). International publication #WO/2009/032143 (A1).
- Yu Huang, Richard Baker, Ramin Daniels, Tiem Aldajani, Jennifer Ly, Leland Vane and Franklin Alvarez, International Patent Application PCT/US2008/074488, publication WO/2009/029668 (A1), “Membrane Augmented Distillation To Separate Solvents From Water” (priority date August 27, 2008)Vane et al. has worked on the process for some time and has proposed low pressure distillation with re-compression before the membrane module, originally1-1.5 bar pressure on the feed side of the membrane and0.2 bar on the permeate side. In the latest patent they proposed a pressurized membrane system, with 3 bar feed pressure, however, not combined with pressurized distillation column.Especially the patent US 8,114,255 B2 and patent WO 2009/029668 A1 were constructed as part of the research for the present invention. Results were promising, but at the end the up-scaling would be complex and uncertain. Industrial solution needed a simpler and more robust solution. Hence a pressurized column and membrane system at same pressure, was tested with successful results.Pervaporation and vapor permeation membrane processes compete with molecular sieve adsorption processes for the dehydration of solvents, including ethanol for use as a fuel. Hybrid vapor stripping-vapor permeation processes have been proposed to efficiently recover ethanol and butanol from fermentation broths. In the hybrid process, the vapor permeation membrane is used to selectively remove water from an alcohol-water vapor containing high water concentrations, returning the permeated water vapor to the stripping column to significantly reduce energy demand in the stripping unit. This process requires a membrane which can withstand high water concentrations at temperatures sufficient to maintain the stream as a vapor – typically up to 130 °C. Commercially available organicinorganic hybrid silica membranes have been demonstrated to be stable in pervaporation applications removing low levels of water from organic solvents at temperatures as high as 160 °C.In this paper, the performance characteristics of hybrid-silica membranes removing water from an ethanol-water vapor with water concentrations of 80-85 mol% at 115 °C will be described. The membranes were operated continuously for 3 months. As has been reported in the literature, there is some falloff in water permeance and rise in water selectivity over time with this type of membrane. In our tests, water permeance decreased from 1.1E−6 to 0.17E−6 kmol/m2-s-kPa (from 3000 to 500 GPU) over the 3 month test period. Over that same time, water-ethanol molar permselectivity (α) increased from 19 to 330, indicating the ethanol permeance dropped more rapidly than did the water permeance. For a week near the end of the 3 month test period, an acetone-butanol-ethanol-water (ABE-water) vapor mixture was dehydrated with the membrane. Solvent-water permselectivities for the ABE tests were 2000, 5000, and 320 for acetone, butanol, and ethanol, respectively.Others:
- Nicholas P. Wynn, Yu Huang, Masakatsu Urairi, Richard W Baker, United States Patent US 8,128,787 B2, MEMBRANE-AUGMENTED DISTILLATION WITH PRESSURE CHANGE TO SEPARATE SOLVENTS FROM WATER (filed August 27, 2008) Describes processes for removing water from organic solvents, such as ethanol. The processes include distillation in two columns operated at sequentially higher pressure, followed by treatment of the overhead vapor by one or two membrane separation steps, indicated in process scheme below. Although the second distillation column is operated under pressure, mentioned up to 10 bar, and this is coupled to membrane vapor permeation – this unit is reported to typically operate at 3 bar with a permeate pressure of 0.5 or 0.25 bar.- David Lee, Adam Orosco, Lincoln Sarager, Victor J. Johnston, United States Patent US2013/0035519 A1, REDUCING ACENTALS AND/OR ESTERS DURING ETHANOLSEPARATION PROCESS (filed August 3, 2011)The purpose of the patent is to reduce the need to separate acetal and/or esters from the crude product. Some of the apparatus is similar, but again the same pressurized column and membrane is lacking. The return flow to the column must be pressurized to equal pressure to the incoming main stream, see Figure 5.The present invention has been developed to be industrial robust and reduce overall costs. Hence, it deviates from most other methods, including those from David Lee et al., by being simplified with as few steps as possible and having a clear concept for the process.The main conclusion from this evaluation of the mentioned process with pressurized distillation followed by membrane vapour permeation is that it is patentable in our opinion.The combination of distillation and membrane vapour separation is already patented and there are also reports of pressurized membrane vapour separation as well as pressurized distillation. However, the combination of one pressurized distillation column followed by a membrane separation step operated under the same pressurized conditions, are not found in any of the screened patents.Summary of the inventionThe most important operation conditions of the process were that the feed stream with 10 bar(a) and 150 ºC was fed into the distillation column; ethanol rich vapor was passed to the membrane module which operated with 1 bar(a) at the permeate side.The complete set-up was assembled according to the sketch in Figure 1. A pump feeded the water/ethanol solution ("broth") into an evaporator prepared using an aluminum block integrating heating elements with feed pipeline coiled around covered with heating tape and insulation. The outlet of the evaporator was connected to the membrane module which was packed in heating tape/insulation material in a similar way. The membrane module included a tubular hybrid-silica membrane. The membrane module was connected to a pump creating a slight under pressure (0.8 bara) on the
permeate side and a pressure gauge was manually regulated to create the retentate pressure (2-11 bara).Water cooling was used for condensation of both permeate and retentate and flow meters were used to quantify the non- condensed retentate gas (and permeate, if any) flow, when CO2 gas flow was applied in the feed. After construction, possible risks were evaluated in order to prepare procedures for safe operation. Tuning of operation parameters (temperatures, pressures, flow rates) were then performed using low concentration water/ethanol mixture (5 wt% ethanol) before testing of higher concentrations was initiated. These conditions varied considerably from today's operation conditions of distillation-based processes.The benefits of the inventionMain advantages of the given process vs. the industry reference:One columnLower energy consumption than by distillation to 94 wt% ethanol while yielding giving 99.5 wt% ethanolLimited amount of capital and operational cost intensive machinery, e.g.compressors. Only liquid pumps providing differential pressure of 9 barBy using higher pressure on the retentate side a costly low temperature cooling system becomes obsolete, and replaced by a conventional system with e.g.20 ºC cooling waterNo need for complicated vacuum systems with increased risk of leaksThe energy consumption is primarily dictated by the distillation column reboiler duty with a value of 6 MJ per kilogram of product. The typical industry standard reference case (Humbird et al., NREL/TP-5100-47764, May 2011) energy consumption is 7 MJ per kilogram of product. Although the proposed process uses medium pressure (MP) steam (10 bar) at a 6% higher cost per energy content than that of low pressure (LP) steam (5 bar) in the reference case, the 17% reduction in energy demand more than outweighs the higher energy cost for MP steam. There is also a higher potential for heat integration compared to the reference case due to the higher temperature of the column top product.The invention was integrated into a process model to calculate the benefits which were favorable.Detailed description of the inventionThe detailed process modelA more detailed description of the process follows and is depicted in the flowsheet in Figure 2. An aqueous feed of 5 wt % ethanol enters an off-gassing vessel VA001 for removal of CO2. The gas is scrubbed by water in a contacting column VG001 and ethanol is returned
to VA001. From VA001, the liquid phase is pressurized by means of a pump PA001 to 10 bar(a) and heated by the economizing heat exchanger HA001 before entering the distillation tower VE001. The distillation tower contains a partial condenser, the condenser liquid is returned to the column as reflux, the vapor stream from the condenser enters a membrane unit VB001 with water selective hybrid-silica membranes. The membrane modules operate with a differential pressure of 9 bar across the membrane. The water rich permeate is partially condensed with liquid returning to the distillation column after recompression in PA002, and permeate vapor is returned to VA001 for recirculation. The ethanol rich membrane retentate is depressurized and partially condensed (HA003), sending the CO2 rich vapor with ethanol in VA003 to VA001 for recirculation. The remaining liquid from VA003 is now high purity ethanol.For a production rate of 30000 kg/h of 99.5% pure ethanol product a membrane area of2300 m2 would be required, based on the experimental results obtained for the hybrid-silica membranes under the given operating conditions. This is significant less than expected.Further work on process modelling using the experimental results available from the study indicated that the ethanol concentration out of the distillation column and into the membrane module was ∼90 wt%, significantly higher than the feed concentrations used for most of the performed vapor permeation experiments. More accurate measures of high ethanol concentrations were also required in order to obtain reliable figures for permeance of each component as input to membrane module in the process modelling. Additionally, it was determined that a feed pressure of 5 bar(a) would not be sufficient to obtain the required purity of the final product (min 99.5 wt% EtOH). Furthermore, while a feed pressure of 6 bar(a) would be sufficient it was encouraged to investigate further the stability of the membranes under 10 bar(a) feed pressure as this would give a significant reduction in required membrane surface area and hence, also the respective capital costs.Experimental set-upThe set-up for vapor permeation measurements is described above in “Summary of the invention” and illustrated in Figure 1. A new tubular hybrid-silica membrane (active surface area ∼50 cm2) was used for this study. The membrane properties were investigated through batch experiments using ethanol feed concentrations of 90-98 wt% in applied feed pressures of 6.0±0.1 bar(a) and 10.0±0.2 bar(a). For each feed pressure a series of measurements were performed using the retentate from a batch of 90wt% EtOH in the feed, as a new feed in a subsequent experiment. This was repeated until a final concentration of minimum 99wt% EtOH was obtained in the final retentate. The feed flow rate for the liquid into the evaporator was kept constant at of 5 mL/min for all experiments (typically increased at start-up for quicker pressure buildup). The temperature of the feed (ethanol/water vapor) was also kept constant at 130-150°C (depending on feed pressure) and the permeate pressure was kept at 0.8±0.1 bar(a) ensuring good flow of condensed permeates out of the membrane module. Condensed permeate and retentate solutions are collected over a period of at least 1 hour for each experiment.
For investigation of the possible degradation effects long term experiments were performed using water as the feed into the evaporator. This provided increased safety and allowed for unsupervised overnight operation of the set-up. Additionally, a feed of water vapor is more corrosive than that of a high ethanol concentration mixture and hence, may speed up the degradation of the hydrophilic membrane. In total the membrane investigated was operated around 550 hours, of which 300 hours were performed with an applied feed pressure of 10±2 bar(a). Batch tests, using 90 wt% EtOH in the feed, were performed during and after the long term test.Ethanol concentration of all feed, retentate and permeate samples were measured by a hydrometer (relative density). Additionally, high performance liquid chromatography (HPLC), combined gas chromatography and mass spectrometry (GC-MS) as well as by densitometry were performed on selected samples.The alcohol recovery is expressed in form of the performance of the membrane in terms of flux, N, reported as permeated mass per time and membrane area [kg/(hr*m2)] and separation factor, α, defined as:where y and x are the molar fractions in the permeate and retentate, respectively.Permeance of component i, is defined as the flux (Ni) divided by the partial pressure difference (ΔPi) over the membrane:In the present work the vapor mixtures on both permeate and retentate side of the membrane is for simplicity assumed to follow the ideal gas law, and hence, the partial pressure of each component (Pi) is expressed by:where xi is the molar fraction of component i and Ptot is the total pressure.Results and discussionFlux, selectivity and permeance as a function of feed pressureThe main results for measurements with a feed pressure of 6 bar(a) is presented in
Figure 3. Four subsequent measurements, using the retentate as new feed in the next experiment, was required to obtain a product of ∼99 wt% EtOH.Total flux, selectivity (separation factor, α) and permeance through hybrid-silica membrane from vapor permeation using 90 wt% EtOH concentration in the feed and the resulting retentate as feed solution in subsequent measurements. Feed pressure: 6±0.1 bar(a), feed temperature: ∼130°C, permeate pressure: 0.8±0.1 bar(a). All concentrations and corresponding selectivity, permeability are derived from densitometer measurements with numbers in brackets derived from hydrometer measurements.With increased feed pressure to 10 bar(a), only three subsequent measurements, using the retentate as new feed in the next experiment, was required in order to obtain a product of ∼99 wt% EtOH. Figure 4 shows the total flux, selectivity (separation factor, α) and permeance through hybrid-silica membrane from vapor permeation using 90 wt% EtOH concentration in the feed and the resulting retentate as feed solution in subsequent measurements. Feed pressure was: 10±0.2 bar(a), feed temperature: ∼150°C, permeate pressure: 0.8±0.1 bar(a). All concentrations and corresponding selectivity, permeability are derived from densitometer measurements with numbers in brackets derived from hydrometer measurements.There is a clear increase in the separation factor when the feed pressure is increased from 6 to 10 bar(a) which is the opposite of what was previously observed for feed concentrations of 54 wt% EtOH. Densification of the hybrid silica membrane layer as a possible result of the increased feed pressure can however, explain the increase in selectivity as observed in the present study. As a comparison to the obtained separation factord (12-20 for 6 bar(a) and 21-32 for 10 bar(a) feed pressure), Vane reports an initial separation factor of 19 for vapor permeation with a hybrid-silica membrane at 115°C using an ethanol feed concentration of 31-39 wt%, ∼1.0 bar feed pressure and ∼0.2 bar permeate pressure. The corresponding water permeance reported by Vane is 1.1⋅10−6 kmol/(m 2 ⋅s⋅kPa) as compared to 0.69 – 1.36⋅10−6 kmol/(m 2 ⋅s⋅kPa) from the present study. This indicates that the membrane performance remains good under pressurized conditions. The permeance is reduced when the feed pressure is increased from 6 to 10 bar(a) due to the fact that the total flux does not increase, and the increase in separation factor only partly compensates for this.Long term degradation testsThe flux and selectivity as a function of time from long term degradation test at 10 bar(a) feed pressure is presented in the table shown in Figure 6 along with the corresponding permeance for water and ethanol. The flux is reduced by ∼20% after ∼125 hours, however, the separation factor is increased and the water permeance remains the same or is increased while the ethanol permeation is reduced. After ∼290 hours a significant change was observed as the selectivity of the membrane apparently was lost and the high flux detected after ∼295 hours indicated that the membrane had failed. However, when the membrane module was opened and the membrane examined, it became clear that the
membrane layer coated inside the porous tubular support was intact while the glass seal/glazing at the end of the tube was shown to be cracking.Likely the combination of high temperature, high pressure and many start-ups/shut downs caused the end seals/glazing to fail. Attempts of re-sealing the ends using available high temperature epoxy were successful but this material also failed when applying ethanol/water feed stream at 150°C and 10 bar(a). The issue needs to be addressed and resolved before upscaling is considered, however, this is not a critical point from a membrane point of view, as more stable seal material is available, e.g. glass or glass ceramic seals. Investigation of new seal materials for the commercial hybrid-silica membranes is out of the scope of this project.Figure 6 shows the total flux and separation factor for long term vapor permeation through hybrid-silica membrane using a feed with 90 wt% EtOH at 10±0.2 bar(a) and ∼150°C with a permeate pressure of 0.8±0.1 bar(a). Total operation (with average feed pressure <10 bar(a)) is ∼500 hours. Selectivity and permeance is derived from ethanol concentration measured by hydrometer.Conclusion of the detailed experimentsThe performance of hybrid-silica membranes under an applied feed pressure of 10 bar(a) is found to be comparable to or slightly reduced to that observed for 6 bar(a) feed pressure. The flux is slightly lower but the separation factor is higher for 10 bar(a) feed pressure, making the corresponding reduction in permeance smaller for water than for ethanol. The experimental data have been used as input for a membrane module in process modelling which indicate a possibility of significant energy savings by use of a hybrid process with a pressurized distillation/stripper column followed by vapor permeation module under the same pressurized conditions.While the active membrane layer appears to be rather stable over a time frame of 300 hours under 10 bar(a) feed pressure (500 hours total) the end seals break up during long term experiments. This is an important point for improvement prior to possible up-scaling. There is however, some degradation in flux observed with a corresponding increase in separation factor leaving the water permeance more or less unchanged.Densitometer measurements were found to give the highest accuracy and precision for measuring the high ethanol concentrations, while a simple hydrometer was found to be satisfactory for most measurements when calibrated towards densitometer measurements. It is recommended to also test out alternative membranes, i.e. zeolite membranes, for operation under the proposed conditions to evaluate the most suited membranes for the application.Possible challenges could be degradation of membrane module over time when operating under 9 bar pressure difference and 150 ºC. For the hybrid-silica membranes reduced total flux is reported over time, however, the corresponding increase in selectivity may as well result in a better separation process. There are also alternative membranes, e.g. zeolite type membranes and polymeric membrane, available that may be suitable for this type of separation and operational conditions.
Claims
ClaimsWe claim:1: An apparatus for separating a liquid mixture of two or more solvents using pressurized distillation followed by membrane vapor permeation under same pressure characterized by a compressor PA001 for pressurizing the liquid mixture to at least 2 bar(a), and are followed by an economizing heat exchanger HA001, a stripping column VE001 having at the bottom of said column an inlet for a stream containing a liquid mixture and an outlet for a stream of liquid that has passed through said stripping column and an outlet at the top of said column for a stream containing at least one vapor where components arranged in sequence with at least one being a water selective membrane VB001 and a condenser HA003 and these components are arranged so that said stream passes through said membrane and condenser before exiting the apparatus.2: A method for high purity ethanol production of anhydrous or hydrous alcohol through combined pressurized distillation and membrane permeation under the same pressurized condition and where the liquid returned from the membrane to the column will be recompressed to the same pressure which preferably should be in the range of 2 – 20 bar(a)a and the temperature preferably in the range of 120 – 160 ° C, characterized by the liquid solution (26) enters an off-gassing vessel VA 001 for removal of CO2 and where the gas (22) is scrubbed by water VG001 in a contacting column, and ethanol (25) is returned to the off-gassing vessel, and thereafter the liquid phase (1) is pressurized by means of a pump PA001 to 2 - 10 bar(a) and heated by the economizing heat exchanger HA001 and thereafter the heated stream (3) enters a distillation tower VE001+HE001 which tower contains a partial condenser HE001, from where the condenser liquid (8) is returned to the column as reflux and thereafter the vapor stream (9) from the condenser enters a membrane unit with water selective membranes VB001 where the membrane modules operate with a differential pressure of 2 - 19 bar(a) across the membrane, and thereafter the water rich permeate (14) is partially condensed with liquid (16) returning to the distillation column after recompression (17), and permeate vapor (20) is returned to the off-gassing vessel for recirculation and thereafter the ethanol rich membrane retentate (11) is depressurized and partially condensed, sending the CO2 rich vapor with ethanol (13) back to the off-gassing vessel for recirculation and the remaining liquid (21) from the condenser will be anhydrous ethanol.3: A method according to claim 2 where some or all of the membranes are hybridsilica type membranes.4: A method according to claim 2 where some or all of the membranes are polymeric type membranes.5: A method according to claim 2 where some or all of the membranes are zeolite type membranes.6: A method according to claim 2 where the origin is aqueous feed of 3 – 7 wt % alcohol from a water/ethanol solution from fermented organic particles after hydrolysis and after the solution has been filtered.Etanol – kombinert trykksatt destillasjon og membranfiltreringPatentkrav:1: Et apparat for å separere en flytende blanding av to eller flere løsninger med trykksatt destillering etterfulgt av membran damppermeasjon under det samme trykket karakterisert ved en kompressor PA001 for å trykksette den flytende miksturen til minst 2 bar(a) og er etterfulgt av en økonomiserende varmeveksler HA001, en strippingkolonne VE001 som ved bunnen av kolonnen har en inngang for en strøm bestående av en flytende blanding og en utgang for væskestrøm som har gått gjennom strippingkolonnen og en utgang på toppen av denne kolonnen for en strøm som inneholder minst en dampkomponent og denne strømmen blir behandlet videre av komponenter i sekvens med minst en vann selektiv membran VB001 og en kondensator HA003 og disse komponentene er satt opp slik at denne strømmen passere gjennom nevnte membran og kondensator før den forlater apparatet.2: En metode for produksjon av høy renhets etanol av anhydrert og hydrert alkohol gjennom kombinert trykkdestillasjon og membran permeat under de samme trykksatte forholdene og hvor væsken returnert fra membranen til kolonnen vil bli trykksatt til det samme trykket, som helst vil være i område 2-20 bara og ved temperaturer helst i området 120-160 grader celsius, karakterisert ved at den flytende løsningen (26) går inn i en avgasser for fjerning av CO2 og hvor VA001 og gassen (22) er renset med vann VG001 i en kontaktkolonne og etanol (25) er returnert til avgasser, og deretter en flytende fase (1) som er trykksatt av en pumpe PA001 til 2- 10 bar(a) og varmet med en økonomiserende varmeveksler HA001 og deretter sendes den varmet strøm (3) inn i et destillasjonstårn VE001+HE001 hvor tårnet inneholder en del-kondensator HE001 hvor den kondenserte væsken (8) blir returnert til kolonnen som refluks og hvor etter damp strømmen fra kondensatoren går inn i en membran enhet med vann selektive membraner VB001 hvor membranmodulene opererer med et differensial trykk på 2-19 Bara over membranen, og deretter den vannrike permatet (14) er delvis kondensert med væske (16) returnerer til destillasjons kolonnen etter rekompresjon (17), og permeat-dampen (20) er returnert til avgassings for resirkulering og hvoretter den etanolrike membran-retentatet (11) er trykkutjevnet og delvis kondensert, sender den CO2 rike dampen med etanol (13) tilbake til avgassing for resirkulering mens den gjenværende væsken (21) fra kondensatoren vil være anhydrert etanol.3: En metode ifølge krav 2 hvor noen eller alle membranene er av membraner av typen hybrid silika membraner.4: En metode ifølge krav 2 hvor noen eller alle membranene er av membraner av typen polymere membraner.5: En metode ifølge krav 2 hvor noen eller alle membranene er av membraner av typen zeolitt membraner.6: En metode ifølge 2 hvor opprinnelsen av den vandige innmatingen er 3-5 vektprosent alkohol fra en vann/ etanol løsning fra fermenterte organiske partikler etter hydrolyse og etter at løsningen har blitt filtrert.