FAQ’s – Membrane Degassing

Please provide water temperature, which gas needs to be removed (CO2 or O2), what is the CO2 (or O2) level in the feed and the level you need, and (if CO2) what is the pH of the feedwater to the Liqui-Cel.

SnowPure will need the “G-number” which is a number beginning with G and followed by 3 digits and sometimes a final letter. This will help us to ensure a good replacement match.

Please refer to this page of the SnowPure website and this 3M page.

“X40” membrane was developed first, prior to X50.  The “X50” membrane was then developed to allow a much higher amount of gas to pass through; this is why it is best used for CO2 reduction.  X40 can reduce CO2; it is just not as efficient (but may be evaluated on a case-by-case basis). X50 is optimized to remove a larger amount of CO2;  this membrane has a higher porosity and the wall is slightly thinner. This allows higher transport of soluble gas (in X50) through and out of the hollow fiber. In X50 the Henry’s Law ‘gradient’ is reestablished incrementally faster than in X40. The X50 was developed to deal with high levels of dissolved gases such as CO2, which is usually much higher than typical dissolved oxygen levels. SP (superphobic) membranes are non-porous but gas-permeable, and are used for degassing and debubbling low surface tension fluids, like inks.

Both X50 and X40 will remove O2 (DO) and CO2.  Removing dissolved O2 takes a stronger vacuum. If you are using a strong vacuum with X50 you will pull a greater amount of water vapor though the membrane. The thicker X40 membrane helps minimize the amount of vapor from coming through the membrane while still removing DO. The X50 membrane fiber has a thinner more porous wall for faster gas transfer than X40.  X50 is normally used for CO2 removal since the concentrations are often many times higher than DO.

Please see our downloads section for Liqui-Cel Inlet Recommendations and for “3M Liqui-Cel Extra-Flow Membrane Contactors Design and Operating Guidelines” or contact us.

SnowPure can provide a general GTM P&ID, and is working on detailed “Best Practices Liqui-Cel” P&ID package for OEMs. SnowPure can review your proposed P&ID and make guidance on improvements.

SnowPure can run a GASCAD projection for you if we have information on fluid inlet conditions, dissolved gas levels, and required dissolved gas targets. We can recommend best operating conditions, and vacuum and N2 sweep gas purity for you.

Typical water pressures are in the 30-60 psig range – allowing for a 5 psi pressure drop is normally a good number to use. Refer to individual product datasheets for recommended upper limits for water pressure.  Please contact SnowPure to discuss your specific design requirements and for assistance in selecting the best design for your conditions.

The system should be monitored for: online dissolved gas outlet, pressure drop across the system, water flow rates, vacuum level, and if used, N2 pressure and flow rate.  The water trap must be checked periodically and, if required, drained.

Water vapor (a gas) goes through the membrane pores and may condense. If the vacuum pump is a “liquid-ring” type this is no problem. However, diaphragm pumps and regen blowers can be damaged by condensed water. In the latter cases, we recommend using a water trap.

As a general rule each contactor in series can remove up to about 90% of the gas. The liquid flow will be through the two contactors in series and the air or N2 flow will be drawn through the contactors in parallel.

If using only Nitrogen Sweep mode, then the nitrogen will saturate the water according to Henry’s Law.  However, if using a combo mode (N2 + vacuum) the N2 will be removed and there will be no addition of dissolved N2 to water at contactor effluent. In fact, under vacuum with the addition of N2 sweep gas (aka combo mode) both N2 and O2 are removed from the process water. The technology is based on the partial pressure of the gases in the water phase, Henry’s Law, and the partial pressure of the N2 in the vacuum side. Since the N2 partial pressure on the vacuum side is small, and the vacuum deep, both N2 and O2 transfer from the water phase into the vacuum. The system performance (O2 removal) is enhanced significantly by the presence of the sweep gas (high purity N2 in this case). Additionally, the N2 sweep gas helps to keeps the water vapor that transfers into the gas phase from condensing and causing pump problems. 

No. For O2 removal (X40 type) the most efficient method to reach low dissolved oxygen levels is “combo mode”: high purity N2 gas plus vacuum. With Liqui-Cel GTM for CO2 removal (X50 type) you do not need to use nitrogen; use a filtered air gas sweep either in “sweep only” or “combo” mode. Depending upon what level of reduction you need to achieve, and the operating mode selected, you may be able to use filtered air, nitrogen, a combination of vacuum with a sweep gas, or vacuum only (with no sweep gas).

For O2 removal, N2 is strongly recommended.  Vacuum alone will not achieve the best O2 removal results. Further, using only vacuum will result in increased condensation of water vapor in the pump/water trap (the sweep gas helps to dilute the water vapor and reduce condensation).

Stopping and starting isn’t a problem. But, as you would expect, you need to avoid sudden valve closures (slamming shut or throwing open) that result in ‘water hammer’ (pressure surges).  If the water is to be static for a significant period it is recommended that the contactor is drained (and dried) to minimize the potential for biological growth.

As far as sending the degassed water to a pressurized tank, there shouldn’t be any issue as long as it is isolated from reintroduction of the gases you just removed.  However, if there is to be a headspace maintained in the ‘pressurized’ tank, that headspace should be filled with a vacuum or with an inert gas (blanketed) that is of no consequence to the process. Note:  the tank should be properly designed/rated for the pressure/vacuum applied.

In the case of 4×28 vs. 4×13 there is a longer contact time due to the length of the module and thus a higher efficiency. However putting two 4×13 in series can provide a better removal of than one single 4×28 module.

Yes, this may be an option depending upon the model you have. Please note that a field changeout would also require replacement of the o-rings and gaskets.  The reassembly is greatly enhanced with the specially designed Liqui-Cel changeout tools; especially the insertion tools. Please contact SnowPure for a quotation and for the Liqui-Cel “Assembly & Disassembly” procedure for that specific contactor.It should be noted that 3M does not offer a certificate of conformance for field-installed cartridges.

Sanitary flanges are available on 4×13 (PP & SS), 4×28 (PP & SS), 8×20 (SS), and 10×28 (SS).  The 4×13 size, with the minimum feed flow at 2 gpm, is the smallest unit with sanitary flanges.

3M does NOT recommend that users operate at flows lower than the range specified in the corresponding datasheet.  The reason is at lower flow rates there is a potential for channeling to occur – where the fluid can find a ‘path of least resistance’ or channel that does not flood the entire shell volume. This in turn will not provide the expected/optimized gas-liquid interface/surface area… resulting in lower than expected gas transfer performance. This effect can be somewhat mitigated by orienting the contactor in a vertical position and flowing the fluid from bottom to top or vertical up direction, but we would still recommend staying within the flow range guidelines.

Operating continuously at a high temperature is possible if you use a stainless steel housing, however it is not recommended. The membrane may oxidize and the contactor service life is reduced. There is a larger quantity of water vapor formed which crosses the membrane to the gas side. The vapor will condense as liquid water on the gas side or in the vacuum line and potentially block the flow path to vacuum pump.  You may also need to consider a higher N2 rates and a larger vacuum pump for DO applications at higher temperatures.

Plant air in sweep mode is not recommended for these reasons: (a) often plant air can heat-up and if it goes over 25 deg. C, it begins to oxidize the Liqui-Cel membrane (b) a slight vacuum from the blower gives a performance boost over plant air, and (c) If there is a restriction in air flow through the contactor (for example someone shuts-off the outlet valve), the Liqui-Cel will no longer work and will actually turn into a bubbler, and d) compressed air has entrained water and oil.  SnowPure recommends using the suction side of a regen blower instead of compressed plant air. We can help size and source the blower.

SnowPure offers Liqui-Cel mounting components and options.  Please contact us for more details.

No. Allow the module to thaw at room temperature and there should not be any issue with quality according to performance tests by 3M.  3M dries Liqui-Cel contactors before packaging and shipping.

Yes, SnowPure offers small pilot systems (for flows up to 10 GPM) for both CO2 and DO reduction and these are normally in stock. Find our “DPC-series” system datasheets on our website and contact us for pricing and lead time. 

SnowPure works with pre-qualified OEMs throughout the world.  Please send us the details of your application and we will be happy to match you with the right trained OEM for your application in your area.

Dry storage for contactors and GTM systems is the best option, but please refer to the guidelines for other options. This information is taken from the most updated 3M Cleaning Guide, which may be found in our downloads section of our website.

SnowPure and 3M recommend that all oxidizing agents be removed prior to feeding your Liqui-Cel. Ozone, chloramines, and chlorine are oxidizing disinfecting agents.  While chloramine is not as aggressive as free chlorine, oxidation is still likely.  Concentration of the oxidizer is also a factor in how quickly the Liqui-Cel degrades. 

Yes. There are some OEM pre-qualification requirements for specialized applications such as this. The OEM should have experience/expertise with filtration and acid/caustic handling. 3M has written a Technical Brief regarding the ammonia recovery process; it is found in our download section.  Please note that: (a) Liqui-Cel end caps need to be Noryl material. (b) Liqui-Cel is best used in a rough reduction of NH3 and will not provide a high reduction. (c) The residual NH3 would need to be removed by ion exchange resin. (d) Sulfuric acid is typically used in the Liqui-Cel module for this application which requires special design/materials of construction. If the end use is to precipitate out the ammonium sulfate for use in sales as fertilizer, a final precipitation step is needed.

Yes. H2S removal is analogous to ammonia NH3 removal, but with a different solution on lumen side of contactor.  Necessary pretreatment to reduce potential fouling of/by process water contaminants needs to be considered.  Please provide SnowPure with the H2S levels (incoming and target).

There will be some water vapor produced by the process but if the amount is larger than expected this can be caused by “fully wet-out” membranes. Normally the PP hollow fibers are wet on the shell side and dry on the lumen side, and the pores are dry. The pores can wet-out if cleaned with a surfactant (not recommended).  They can also wet-out if shell and lumen are both full of water.  We recommend that the Liqui-Cel be dried out over a 24-hour period.  Please find the procedure “3M-Liqui-Cel Cleaning Guidelines” in the downloads section of our website.  This procedure mostly is draining the module, running the sweep gas (or filtered air) at the gas inlet, and pulling a vacuum to pull all moisture out of both sides and out of all the pores.  We also recommend that a small N2 sweep gas is used.

The best way to clean 3M Liqui-Cel membranes is to use 4 to 5% of NaOH solution in the water at 30 to 40ºC, circulate the cleaning solution counter current of the original process for 2 hours. Water flush the module, and use 2% HCl acid in the water at room temperature for one hour. Water flush, then run N2 or CO2 in the gas side to flush any condensation. This dries the gas side of the membrane. We have a simplified cleaning procedure for the beverage industry. Please don’t use H2O2 for a long time, in order to avoid any oxidation of the membrane.  3M provides Cleaning Guidelines for typical maintenance and more challenging cleaning procedures – however, if fouling is substantial, it may be necessary to replace the contactor(s).   Consider the source of fouling and install prefiltration/treatment to prevent future fouling.

Depending on the service and the water quality feeding the contactor, the estimated life for Liqui-Cel is approximately 1 to 8 years. Please note that Liqui-Cel is very sensitive to chlorine and other oxidizers, and can be severely damaged as well as lifetime is significantly reduced if oxidizers are present.  Please refer to the section on expected service life in the Liqui-Cel Design & Operating Guidelines (applicable to EXF-series and assuming RO permeate quality water).