Ultrafiltration Membrane Reactor Research Articles

Emerging investigator series: photocatalytic treatment of PFAS in a single-step ultrafiltration membrane reactor

Combining ultrafiltration and advanced oxidation processes, a single-step reactor removes aqueous PFAS and destroys them in situ, allowing long-term reuse of the effective photocatalyst.

Electrocatalytic ultrafiltration membrane reactors designed from dry-spun self-standing carbon nanotube sheets

The development of electrochemically active ultrafiltration membrane reactors offers promising perspectives to achieve simultaneous separation and degradation of persistent organic pollutants and support triggered self-cleaning of membrane materials upon surface fouling. Here, electro-responsive ultrafiltration membranes were synthesised from drawable carbon nanotubes (CNT) dry-spun as ultra-thin sheets onto preformed carbon nanofibre (CNF) supports to generate a unique class of electrically conductive and flexible ultrafiltration membranes. The pore size of the CNT-based membranes, on the order of ∼ 28 nm, was fine-tuned by controlling the dry layering and orientation of the CNT sheets to manage the membrane selectivity. The CNT-based membranes were used as effective conductive platforms to promote charge transfer during electrocatalytic degradation of acetaminophen, as a model contaminant. The CNT-based membranes, besides offering water permeance up to 2.77 × 103 L.m−2.h−1.bar−1, yielded electrocatalytic kinetic constant up to 46.5 × 10−3 min−1 during combined electrochemical reaction and ultrafiltration process, which is 1.4 to 39 times larger than previously reported values. Such high performance was maintained quite stable even after 8 reuse cycles. These results demonstrate the potential of CNT dry spinning technology for the scalable fabrication of highly permeable, but selective CNT-based membranes with remarkable electrochemical properties towards cost-effective water treatment at an exceptional reaction rate.

Comparing the effects of pre-deposited and pre-mixed powdered activated carbons on algal fouling during ultrafiltration

This investigation compared the effects of pre-deposited and pre-mixed powdered activated carbons (PACs) in submerged ultrafiltration membrane reactors on the distribution of extracellular organic matter (EOM) and its contribution to algal fouling during the filtration of algae-rich water. Experimental results indicate that pre-mixed PAC mitigated algal fouling, whereas pre-deposited PAC exhibited the opposite effect. The porous fouling layer in the pre-mixed case favored the retention of rejected EOM in the concentrated solution owing to a high backward migration rate. Conversely, the dense fouling layer in the pre-deposited case impeded the backward diffusion of EOM, which caused the majority of the rejected EOM to accumulate on the surface and at the bottom of the fouling layer. This difference in EOM distribution led to different cleaning efficiencies. Physical cleaning easily removed both algal cells and pre-mixed PAC from the membrane surfaces, whereas physical and chemical cleanings were both ineffective for pre-deposited PAC, owing to the stronger PAC-EOM attractive force and the NaClO-induced change in the chemical bonds of the PAC and even the membranes. Moreover, decreased algae growth rates were observed in the reactors with PAC as a result of changes in the illumination and the cake layer structure, which could further intensify the contrast between the effects of pre-deposited and pre-mixed PACs on fouling.

A hydraulically optimized fluidized bed UF membrane reactor (FB-UF-MR) for direct treatment of raw municipal wastewater to enable water reclamation with integrated energy recovery

A hydraulically optimized pilot-scale fluidized bed ultrafiltration membrane reactor (FB-UF-MR) for water reclamation with integrated energy recovery was designed and operated for 576 days with raw municipal wastewater from a local wastewater treatment plant (WWTP). The design of the reactor and the fluidized bed flow regime were optimized by a computational fluid dynamic (CFD) model. By adopting a customized impact plate in the design, the recirculation flow rate could be reduced by approximately 38% lowering the energy demand significantly. Granular activated carbon (GAC) was used as fluidized medium and proved as successful fouling mitigation strategy, even at low concentrations of 1.7–3.3 g/L (based on total reactor volume) or 21–41 g/L (based on fluidized bed volume) as compared to previous studies. The optimal operational conditions of the FB-UF-MR were identified at (i) a sustainable flux of 8–10 LMH, (ii) an optimal backwash flux of 1.2 times of the permeate flux, (iii) an operational cycle of 10 min consisting of 9 min filtration and 30–60 sec backwashing and optional 30 sec relaxation. Chemical cleaning was applied every 4–6 weeks during long-term operation. Main foulants of the wastewater matrix were identified by 3D-fluorescence excitation-emission matrix (3D-EEM) measurements coupled with parallel factor (PARAFAC) analysis suggesting that protein-like organic matter were the main constituent of the cake layer. The results of this study indicate that direct treatment of raw municipal wastewater with UF is a promising option. Thus, it is possible to concentrate C and N (in a subsequent treatment step) for improved energy recovery while facilitating water reclamation.

Impact of the fouling mechanism on enzymatic depolymerization of xylan in different configurations of membrane reactors

In order to maximize enzymatic xylan depolymerization while simultaneously purifying the resulting monosaccharide (xylose), different ultrafiltration (UF) membrane reactor configurations were evaluated. Initial results showed that the two hydrolytic enzymes required for complete depolymerization of xylan, endo-1,4-β-xylanase and β-xylosidase, promoted different types of fouling, which had a direct impact on the extent of xylan hydrolysis achieved during reaction. Endo-1,4-β-xylanase generated DP 1–6 xylo-oligomers. These products contributed to partial pore blocking of the 1 kDa polysulfone membrane and caused irreversible flux loss (∼20%). The presence of β-xylosidase could not prevent deposition of xylan and xylooligomers on the UF membrane surface. Mulder’s modelling of the filtration parameters affirmed that this xylan and xylooligosaccharide deposition formed a cake layer on the membrane which hindered enzymatic attack in addition to fouling. Reaction with both enzymes followed by UF was found to be the optimal configuration, providing at least 40% higher xylan hydrolysis than the cascade configuration (involving sequential reaction with each of the enzymes separately) and the simultaneous reaction-filtration with both enzymes, respectively. This study thus confirmed that the reactor configuration has a crucial impact on the performance of both the reaction and the separation process of xylose during enzymatic xylan degradation, and that the type of fouling mechanism varies in response to the type of enzyme treatment.

Recycling Cellulase from Enzymatic Hydrolyzate of Laser-Pretreated Corn Stover by UF Membrane

The ultrafiltration membrane reactor, utilizing a membrane module with a suitable molecular weight alleyway, retains the larger cellulase components. Smaller molecules, such as the fermentable reducing sugars and water, pass through the membrane. The purpose of this work was to investigate the capability of recycling cellulase in the UF membrane. PS30 hollow fiber membrane, an ultrafiltration method using internal pressure, was found to be an ideal membrane separation device, allowing re-use of the enzyme. A Box-Behnken experimental design (BBD) established the following optimum pretreatment parameters: operation pressure at 1.73 bar, temperature at 36.38 °C, and a pH of 5.92. Under these conditions, the model predicted a membrane flux yield of 2.3174 L/(m2•h). The rejection rate of the UF membrane was over 95%.

Enzymatic hydrolysis of polyethylene terephthalate films in an ultrafiltration membrane reactor

Synthetic aromatic polyesters such as polyethylene terephthalate (PET) are degraded by microbial polyester hydrolases at mild reaction conditions compared to chemical processes. PET films were hydrolyzed by the polyester hydrolase TfCut2 from Thermobifida fusca KW3 in an aqueous reaction system contained in an ultrafiltration membrane reactor. The continuous removal of inhibitory hydrolysis products by the ultrafiltration membrane increased the efficiency of the biocatalytic hydrolysis of PET by 70% over a reaction time of 24h compared to a hydrolysis performed in batch mode. A model describing the biocatalytic hydrolysis of PET films with simultaneous ultrafiltration to remove the reaction products was in good agreement with the experimental data obtained. The ionic strength of the buffer as well as the mechanical agitation of the reaction mixture markedly influenced the enzymatic PET hydrolysis rates.

Continuous ultrafiltration membrane reactor coupled with nanofiltration for the enzymatic synthesis and purification of galactosyl-oligosaccharides

Continuous enzymatic production of galactosyl-oligosaccharides (GOS) from a lactose substrate in an ultrafiltration membrane reactor (UMR) coupled with a nanofiltration separation system (CPNSS) was investigated. The overall rate of production over 4h of continuous production was 80–104mg GOS formed/U with an average residence time of 66min, an initial lactose concentration of 300g/L, an inlet pressure of 2.0bar inlet pressure, and an outlet pressure of 1.5bar. In the continuous diafiltration (CD) process, the concentration of various sugars and the relationship between the yield and purity of oligosaccharides was well predicted by mathematical models, the increased rate of sugar rejections was less than 10%, and the decreased rate of concentrations in the tank was less than 15%. In the CPNSS, 33.4wt.% GOS was obtained in the UMR, and 1.24kg of high-purity GOS (specifically, GOS purity of 57.2% and lactose content less than 20%) was achieved. This final yield was 80.1% GOS, which meets industry standards.

Continuous production of cyclodextrins in an ultrafiltration membrane reactor, catalyzed by cyclodextrin glycosyltransferase from Bacillus circulans DF 9R.

Cyclodextrins (CDs) are cyclic oligosaccharides of wide industrial application, whose synthesis is catalyzed by Cyclodextrin glycosyltransferase (CGTase) from starch. Here, CDs were produced using CGTase from Bacillus circulans DF 9R in continuous process and an ultrafiltration membrane reactor. The batch process was conducted as a control. This method allowed increasing the yield from 40 to 55.6% and the productivity from 26.1 to 99.5 mg of CD per unit of enzyme. The method also allowed obtaining a high-purity product. The flow rate remained at 50% of its initial value after 24 h of process, improving the results described in the literature for starch hydrolysis processes. CGTase remained active throughout the process, which could be explained by the protective effect of the substrate and reaction products on CGTase stability. In addition, batch processes were developed using starches from different sources. We concluded that any of the starches studied could be used as substrate for CD production with similar yields and product specificity.

English

Operational stability of glucoamylase was studied at 45, 60 and 70 °C in the reaction of maltose hydrolysis. Experiments were carried out in the continuously operated ultrafiltration membrane reactor (UFMR) at constant residence time of 176.5 minutes. The rate of enzyme operational stability decay increased with temperature. This could be quantitatively observed from the measurements of volume activity during the experiments which were used to estimate enzyme operational stability decay rate constants. Results have shown that stationary conditions in UFMR can be maintained if sufficiently high enzyme concentration is used in the reactor, regardless of the enzyme operational stability decay that occurs. This was shown by the experiment carried out at 45 °C where it was proved that the enzyme operational decay occurs even though maltose conversion was at maximum during the entire experiment. Thus, the operational stability decay can be masked.

Vanadyl acetylacetonate filled PVDF membranes as the core of a liquid phase continuous process for pure phenol production from benzene

A study of the liquid phase hydroxylation of benzene to phenol under mild conditions using vanadyl acetylacetonate as vanadium(IV) source filled in PVDF membranes in an ultrafiltration membrane reactor is reported. Catalytic membranes, with different catalyst contents, were prepared and characterized. All of them showed an asymmetric structure, with a uniform catalyst distribution, and an average pore diameter in the range 26–32nm.Permeability tests at different transmembrane (TMP) pressures were carried out on the catalytic membranes to determine the contact time between the catalyst and the reacting mixture, allowing to obtain different contact times (range from 5.3 to 0.4s by changing the TMP from 0.01 to 0.03 MPa and the amount of entrapped catalyst from 0.1 to 0.6g).The catalytic tests evidenced that the system performance was highly influenced from both the amount of entrapped catalyst into the polymer matrix and the contact time. The best results were obtained by using the catalytic membrane made by PVDF containing 0.3g of vanadyl acetylacetonate (VAAC) entrapped inside (PVDF/VAAC0.3), which corresponds to 3% w/w of the membrane. By using this membrane a 100% phenol selectivity and 3.6% phenol yield were obtained. The overall results evidenced that such approach is suitable for operating on a continuous system to obtain pure phenol from benzene.

Continuous enzymatic carboligation of benzaldehyde and acetaldehyde in an enzyme ultrafiltration membrane reactor and laminar flow microreactors

Abstract The synthesis of (S)-2-hydroxypropiophenone ((S)-2-HPP) from benzaldehyde and acetaldehyde catalyzed by benzoylformate decarboxylase (BFD) from Pseudomonas putida was studied in an enzyme ultrafiltration membrane reactor (UFMR) and in three different microreactors (MRs). The aim was to compare the volume productivity (QP) as well as biocatalyst productivity number (BPN) of these two continuous reactor systems. Influence of geometry, surface roughness of the microchannel walls, and inner mixing in the microreactors on conversion were also tested. In all used microreactors full conversion of benzaldehyde was achieved at residence times longer than 4 min. This was not the case in the UFMR where BFD was “soluble” immobilized because of the enzyme instability under the working conditions. The microreactor with inner mixing (micromixers) showed higher conversion at residence times shorter than 2 min (higher flow rates); while at longer residence times there were no significant differences between the different microreactor types. The biocatalyst productivity numbers of the microreactors were higher in comparison to the UFMR. The highest BPN was achieved in a microreactor with smooth surface microchannel walls (1.55 g g−1) at residence time of 5 min while the highest BPN in the UFMR (0.21 g g−1) was achieved at residence time of 30 min and 48 min of processing time.

Mathematical model for aldol addition catalyzed by two d-fructose-6-phosphate aldolases variants overexpressed in E. coli

Two d-fructose-6-phosphate aldolase variants namely, single variant FSA A129S and double variant FSA A129S/A165G, were used as catalysts in the aldol addition of dihydroxyacetone (DHA) to N-Cbz-3-aminopropanal. Mathematical model for reaction catalyzed by both enzymes, consisting of kinetic and mass balance equations, was developed. Kinetic parameters were estimated from the experimental data gathered by using the initial reaction rate method. The model was validated in the batch and continuously operated ultrafiltration membrane reactor (UFMR). The same type of kinetic model could be applied for both enzymes. The operational stability of the aldolases was assessed by measuring enzyme activity during the experiments. FSA A129S/A165G had better operational stability in the batch reactor (half-life time 26.7h) in comparison to FSA A129S (half-life time 5.78h). Both variants were unstable in the continuously operated UFMR in which half-life times were 1.99 and 3.64h for FSA A129S and FSA A129S/A165G, respectively.

Screening of three commercial plant peroxidases for the removal of phenolic compounds in membrane bioreactors

A comparative study of three plant peroxidases, horseradish (HRP), soybean (SBP) and artichoke (AKPC), was carried out to select the most appropriate one for 4-chlorophenol treatment in an ultrafiltration membrane reactor. Soybean peroxidase showed the highest enzymatic activity, followed by HRP and AKPC. The same tendency was observed in a discontinuous tank reactor, where SBP attained more than 90% of 4-chlorophenol removal within the pH range tested. The optimum temperature was 30 °C, with SBP showing highest thermostability. With the ultrafiltration membrane reactor, SBP attained the highest operational stability, with 4-chlorophenol conversions of around 90% in the permeate stream for up to 200 minutes. Finally, permeate samples were analysed and no significant amount of enzyme was detected, so the observed loss of activity, less pronounced with SBP, was attributed to enzyme adsorption on the polymeric products deposited on the membrane surface. Soybean peroxidase was selected as the most appropriate peroxidase for future research.

Experimental Study of Lactose Hydrolysis and Separation in Continuous Stirred Tank -Ultrafiltration Membrane Reactor

In this study, lactose was hydrolyzed using β-galactosidase enzyme in a continuous stirred tank -Ultrafiltration (CSTR-UF) to produce galactose and glucose. The UF membranes of Molecular Weight Cut of 3 kDa of regenerated cellulose material were used to separate enzyme from products. Experiments were performed with 0.139 molar aqueous solution of lactose as feed. The effect of operating pressure ranging between 2 and 5 bar and time on rejection and permeate flux were studied. Results showed that the UF membrane rejects the enzyme completely. Also according to experimental data lactose concentration in permeate decrease with time due to concentration polarization and hydrolysis. It was found that rejection factor of lactose increases from 33 to77%, with time from 5 to 85 min.

Application of continuous stirred membrane reactor to 3-cyanopyridine bioconversion using the nitrile hydratase–amidase cascade system of Microbacterium imperiale CBS 498-74

The bioconversion of 3-cyanopyridine using the in situ nitrile hydratase–amidase cascade system of resting Microbacterium imperiale CBS 498-74 cells was investigated in an ultrafiltration-membrane reactor, operated in either batch or continuous mode. The effects of operating conditions such as the amount of biocatalyst, substrate concentration, substrate feeding rate, mean residence time, and enzyme-to-substrate ratio, were investigated with the aim of achieving almost 100% substrate conversion and high reactor productivity. As a result, it was found that the NHase–AMase cascade system could be adequately exploited in a continuous reactor configuration. The differing temperature dependence of nitrile hydratase and amidase kinetics enabled the operational parameters to be module d to ensure (i) nitrile hydratase operational stability (at 5 °C), and (ii) 100% conversion of 3-cyanopyridine into nicotinic acid, or, alternatively, (iii) enrichment of the effluent stream with the intermediate nicotinamide (up to 80% conversion). It was possible to select operating conditions that allowed long periods of operation (at least 100 h) at a constant flow-rate without enzyme activity loss.

Fe-zeolites filled in PVDF membranes in the selective oxidation of benzene to phenol

Benzene conversion to phenol has been studied in liquid-phase in a flow-trough ultrafiltration membrane reactor in mild conditions using Fe-zeolite catalysts filled in a PVDF (polyvinylidenefluoride) polymeric matrix. The catalysts and the catalytic membranes were prepared, morphologically characterized and catalytically tested. The affinity between the membrane system and reagents and/or products was evidenced by contact angles, sorption and permeation flux measurements. From sorption measurements it was found a greater affinity of benzene reagent for the membrane material (sorption = 0.37g benzenel/g membrane) with respect to a negligible sorption of phenol. Permeations flux measurements evidenced an increased permeate flux for the membrane filled with the zeolite catalysts: the enhanced flux allowed to get contact times between products and/or reagents with the catalyst in the range 0.4–12 s. Reactivity tests, performed at various contact time, showed the highest phenol yield (5.5%) in a single pass using the PVDF membrane filled with Na–Fe-Silicalite-1 at 35 °C. The use of different catalyst forms (free or entrapped in the polymeric matrix) in an open batch reactor evidenced the best performance in terms of selectivity for the entrapped catalyst (97% vs. 69%).

Continuous enzymatic production of 5′-nucleotides using free nuclease P1 in ultrafiltration membrane reactor

In this paper, production of 5′-nucleotides using a free enzyme membrane reactor was presented. The enzyme membrane reactor consists in a coupling of a membrane separation process with an enzymatic reaction. Retention of substrate (RNA) and catalyst (nuclease P1) by the ultrafiltration membranes was observed. The best enzyme membrane reactor configuration was achieved with a polyethersulfone membrane of 30 kDa molecular weight cut-off, since the substrate and biocatalyst were retained with no loss of enzyme activity. Enzyme stability and optimum temperature were investigated. Other optimized operating conditions were also obtained: working temperature 65 °C, flow rate 2.0 mL/min (transmembrane pressure 0.050 MPa), and reactor volume 100 mL. The results demonstrated that enzymatic production of 5′-nucleotides from RNA was efficiently conducted by employing an ultrafiltration membrane reactor.

Production of high degree polymerized chitooligosaccharides in a membrane reactor using purified chitosanase from Bacillus cereus

Crude chitosanase from Bacillus cereus NTU-FC-4 was separated by a cation exchanger to three fractions named CBCI, CBCII, and CBCIII. The CBCI hydrolyzed chitosan to yield dimers. The primary hydrolytic products of CBCII were low degree polymerized (DP) chitooligosaccharides. The CBCIII had the fastest reaction rate and yielded high DP chitooligosaccharides (heptamer and higher DP oligomers). When CBCIII was used in the ultrafiltration membrane reactor with enzyme/substrate ratio 0.06 unit/mg and 100 min of residence time (RT), the concentration of high DP oligomers was 9.78 mg/mL which occupied ca. 48% of total oligomers in the final product as compared to ca. 29% resulted from the crude enzyme. Decrease of RT to 50 min and 33 min, the high DP oligomers in the products were ca. 61% and 69%, respectively. This system could be operated for at least 24 h and kept a constant permeate flux and product output rate.

Cyclodextrin glycosyltransferase from Bacillus circulans DF 9R: Activity and kinetic studies

Activity characteristics and kinetic aspects of a cyclodextrin glycosyltransferase (CGTase) from Bacillus circulans DF 9R were studied. A mixture of α-, β- and γ-cyclodextrins (CDs), glucose, maltose and negligible amounts of longer linear dextrins were produced from gelatinized amylose, amylopectin and starch from different sources. In the coupling reaction, CDs were the substrates in the presence of acceptors such as maltose and/or longer oligosaccharides. From oligosaccharides formed by three or more glucose units, this enzyme produced linear chains of several lengths which were then cyclized. CGTase catalytic efficiency was compared employing an analytical grade starch and cassava starch for food use. Since the results obtained were similar for both starches, the use of an economic starch is an advantage. CGTase was inhibited by the substrate and its own products. Starch concentrations over 20 mg/mL inhibited the cyclizing activity. CDs behaved as competitive inhibitors and maltose as an uncompetitive inhibitor while maltotriose showed a mixed inhibition pattern. Limit dextrins showed a scarce inhibitory effect on enzyme activity. CD production could be improved with an ultrafiltration membrane reactor for continuous removal of the products; the starch concentration should be maintained below an inhibitory concentration and limit dextrins would remain in the reactor without affecting enzyme activity.