Search for: All records

An air gap membrane distillation (AGMD) module was developed by incorporating a poly(etheretherketone) (PEEK) hollow fiber membrane (HFM) having a nonporous wall. This PEEK HFM was placed inside a polyvinylidene fluoride (PVDF) hydrophobic porous wall HFM with a larger bore diameter. The outside diameter (OD) of PVDF HFM is 925 μm, small enough to be capable of achieving a high surface area packing density of 1297 m2/m3. The air gap thickness was very small, 121 μm. Hot brine flowed on the outside of the PVDF HFM; the colder liquid was passed through the lumen of the PEEK-based condenser hollow fibers. Water vapor condensed in the air gap formed between the inner surface of the porous PVDF HFM and the outer surface of the nonporous condenser PEEK fiber. With 85o C hot brine flowing at 40 mL•min􀀀1 and 5o C coolant flowing at 8 mL•min􀀀1, the water vapor flux was 9.05 kg/m2•h with a salt rejection of 98.7 %. Simulation by COMSOL Multiphysics predicted water flux and interfacial temperature of HFM, which supported the experimental observations. Moreover, the influence of module geometry, membrane characteristics and internal flow configuration on permeate flux, thermal efficiency, gained output ratio (GOR), and temperature and concentration polarization were evaluated. Principal component analysis (PCA) was used to illustrate the interconnections among various parameters and their respective contributions to water flux and other performance indicators. Air gap thickness had the strongest influence on temperature polarization. 

Purification of IgG from residual host cell proteins (HCPs) in post-Protein A chromatography is important since some HCPs bind with Protein A and elute with the monoclonal antibody (mAb); removal of HCPs from CHO cell lines is essential. To that end, an advanced separation and purification technique in biopharmaceutical manufacturing, namely, internally staged ultrafiltration (ISUF), is investigated here. Choosing BSA as a model for HCPs in post-protein A eluate, separation of a binary mixture of IgG and BSA containing 1.0 mg/ml IgG and 0.1 mg/ml BSA is successfully demonstrated here using a modified ISUF technique: two Omega 100 kDa membranes on top followed by one Omega 70 kDa membrane at the bottom. This modified configuration demonstrated exceptional performance with almost complete rejection, 99 % purity, and 99.5 % retention of IgG, along with 96.5 % recovery of BSA over 10 diavolumes. This modified membrane stacking resulted from strategic considerations of membrane stacking and careful selection of molecular weight cutoffs and materials, and performance analysis of different membranes and stacking configurations using rejection behaviors, purity levels, and recovery rates under varying diavolume and pressure differential. The approach adopted here enhances flexibility in membrane choices in ISUF and provides valuable insights for optimizing membrane-based biopharmaceutical separation techniques. 

Sometimes NH₃ is stripped from process/effluent streams through hydrophobic porous hollow-fiber-membranes (HFMs) via a supported-gas-membrane (SGM) process and recovered in concentrated H₂SO₄ solution as (NH₄)₂SO₄. To recover relatively purified (NH₄)₂SO₄, one can avoid excess H₂SO₄ with a more dilute H₂SO₄ strip solution. Neglect of strip-side mass-transfer resistance for low-pH strip H₂SO₄ solutions is not desirable with higher-pH H₂SO₄ strip solutions. Small hollow-fiber membrane modules (HFMMs) were used with a higher-pH H₂SO₄ strip solution. Mass transfer was successfully modeled using reaction-enhanced mass transport in higher-pH H₂SO₄ solution. Employing larger-scale crossflow HFMMs, time-dependent ammonia removal from a large tank having ammonia-containing process effluent was modeled for batch recirculation operation. The larger-scale modules employ shell-side feed liquid in crossflow with an overall countercurrent flow pattern and acid flow in the tube side. Modeling ammonia transport without water vapor transfer can cause substantial errors in batch recirculation method. Water vapor transport was considered here for low-pH and high-pH H₂SO₄ strip solutions for ammonia-containing feed in a large tank. Model results describe literature-based experimentally observed mass transfer behavior in industrial-treatment systems well. Model calculations were also made for continuous ammonia recovery from industrial effluents by a number of series-connected HFMMs without any batch recirculation. 

Polymeric membranes for separation of pharmaceutical intermediates/products by organic solvent nanofiltration (OSN) have to be highly resistant to many organic solvents including high-boiling polar aprotic ones, e.g., N- methyl-2-pyrollidone (NMP), dimethylsulfoxide (DMSO), dimethylformamide (DMF). Unless cross-linked, few polymers resist swelling or dissolution in such solvents; however particular perfluoropolymers are resistant to almost all solvents except perfluorosolvents. One such polymer, designated AHP1, a glassy amorphous hydrophobic perfluorinated polymer, has been studied here. Additional perfluoropolymers studied here are hydrophilically modified (HMP2 and HMP3) versions to enhance the flux of polar aprotic solvents. OSN performances of three types of membranes including the hydrophilically modified ones were studied via solvent flux and solute rejection at pressures up to 5000 kPa. The solutes were four active pharmaceutical ingredients (APIs) or pharmaceutical intermediates having molecular weights (MWs) between 432 and 809 Da and three dyes, Oil Blue N (378 Da), Sudan Black B (456 Da), Brilliant Blue R (826 Da). Solvents used were: ethyl acetate, toluene, n- heptane, iso-octane, DMSO, tetrahydrofuran (THF), DMF, acetone, NMP, methanol. Test cells included stirred cells and tangential flow cells. Pure solvent fluxes through three membrane types were characterized using a particular parameter employing various solvent properties. All three membranes achieved high solute rejections around 91–98% at ambient temperatures. HMP2 membrane achieved 95% solute rejection for an API (809 Da) in DMSO at a high temperature, 75 ◦C. A two-stage simulated nanofiltration process achieved 99%+ rejection of a pharmaceutical intermediate (MW, 432 Da) in 75v% NMP-25v% ethyl acetate solution. 

Defense against small molecule toxic gases is an important aspect of protection against chemical and biological threat as well as chemical releases from industrial accidents. Current protective respirators/garments cannot effectively block small molecule toxic gases and vapors and retain moisture transmission capability without a heavy burden. Here, we developed a nanopacked bed of nanoparticles of UiO-66-NH₂ metal organic framework (MOF) by synthesizing them in the pores of microporous expanded polytetrafluoroethylene (ePTFE) membranes. The submicron scale size of membrane pores ensures a large surface area of MOF nanoparticles which can capture/adsorb and react with toxic gas molecules efficiently. It was demonstrated that the microporous ePTFE membrane with UiO-66-NH₂ MOF grown inside and around the membrane can defend against ammonia for a significant length of time while allowing passage of moisture and nitrogen. It was also demonstrated that the MOF-loaded ePTFE membrane could provide significant protection from Cl₂ intrusion as well as intrusion from 2-chloroethyl ethyl sulfide (CEES) (a simulant for sulfur mustard). Such MOF-filled membranes exhausted by NH₃ breakthrough experiments were regenerated conveniently by heating at 60 °C for one week under vacuum for further/repeated use; a single regenerated membrane could block NH₃ for 200–300 min. The moisture permeability of such a membrane/nanopacked bed was considerably above the breathability threshold value of 2000 g/m² -day. The results suggest that microporous membranes filled with reactive MOF nanoparticles could be designed as protective barriers against toxic gases/vapors, e.g., NH₃ and Cl₂ and yet be substantially permeable to H₂O and air.