Membrane applications for biogas production and purification processes: an overview on a smart alternative for process intensification

Abstract

Biogas is the result of a complex conversion process that takes place because of the metabolic activity of various types of bacteria. The anaerobic digestion (AD) plants are characterized by many different criticisms, which risk their failure. One of these is the washout phenomena that imply a premature removal of the active biomass, owing to a vigorous addition of organic matter. There is also the possibility of generating an excess of digestate with high nitrogen and phosphorous content that can induce water eutrophication if left freely in nature. In this sense, membranes can be useful; with their high separation power, they can be employed for both the stabilization of the exhausted digestate and the enhancement of the solid retention time (SRT). Membranes are promising, even in the field of final biogas separation for bio-methane production. To date, various types of setups have been tested for capturing CO₂, and the results indicate a possible stable application for anaerobic digestion plants. Therefore, membranes are a good choice for the development of advanced processes optimized for both gas and semi-liquid phase handling.

---

Eros Rosalbino Minardi

Ing. Eros R Minardi studied at the University of Calabria, where he graduated in Chemical Engineering in 2013 and specialized in modeling and simulation of anaerobic digestion processes using waste materials for biogas production. He is currently working as a PhD student at the Department of Informatics, Modeling, Electronics and Systems Engineering (D.I.M.E.S.), University of Calabria, and is performing his research on biogas production from waste biomass.

Sudip Chakraborty

Dr. Sudip Chakraborty received a PhD in Chemical Engineering in 2010 from the University of Calabria. He is working as a post-doctoral researcher at the Department of Informatics, Modeling, Electronics and Systems Engineering (D.I.M.E.S.). His PhD thesis was about the development of submerged biocatalytic membrane reactors for innovative production systems. His main research activities relate to membrane bioreactor technology, membranes for water treatment and energy production. He is co-author of more than 25 scientific papers published in scientific journals, several book chapters and more than 20 contributions published in conference proceedings, as well as a member of the editorial board of several journals.

Vincenza Calabrò

Prof. Vincenza Calabrò is a full professor at the Department of Informatics, Modeling, Electronics and Systems Engineering (D.I.M.E.S.), University of Calabria. She graduated in Industrial Technologies Engineering in 1984 and received a PhD in Chemical Science and Technologies in 1989. She was the winner of the prestigious fellowship of ENI – P.F.E. Energyin in 1985. She also worked as a Researcher at ENIRICERCHE, as well as at ITM-CNR. She became a permanent researcher in Chemical Engineering at the University of Calabria, Faculty of Engineering, before being promoted to full professor at the same university. She was a visiting professor at Slovak Technical University & Nicolaus Copernicus University and co-authored more than 80 publications on scientific journals, proceedings and books, 1 Italian patent and 100 conference or congress proceedings.

Stefano Curcio

Dr. Stefano Curcio is an Assistant Professor of Transport Phenomena at the Department of Informatics, Modeling, Electronics and Systems Engineering (D.I.M.E.S.), University of Calabria. He graduated in Chemical Engineering in 1995 and received a PhD in Chemical Technologies and Novel Materials, University of Calabria, in 1998. His primary research interests are on transport phenomena, fluid mechanics, heat and mass transfer, biochemical reactors and dynamics and control of chemical processes. He authored 46 publications in international journals, seven book chapters, including one Italian patent, and more than 100 presentations held in international conferences and published in conference proceedings. He is also serving as an editor for several journals.

Enrico Drioli

Prof. Enrico Drioli, Emiritus Professor at the School of Engineering of the University of Calabria. Founding director of the Institute on Membrane Technology. Since 2010, WCU Distinguish Visiting Professor, Department of Energy Engineering, Hanyang University, Seoul Korea. His research activities focus on membrane science and engineering. He is involved in many international societies, scientific committees, editorial boards, and international advisory boards. Currently Chairman of the Section on “Membrane Engineering” of the European Federation of Chemical Engineering, he has been coordinator of several projects author of more than 600 scientific papers and 18 patents,and editor of 19 books on Membrane Science and Technology.

---

1. Introduction

Biochemical conversion processes for the development of gaseous and liquid fuels are an attractive solution to waste accumulation as a use of waste material and the production of useful resources.

In particular, AD is based on the activity of a group of numerous bacteria capable of exerting their metabolic action on various types of substrates. This biochemical process is comprised of various steps, which should not be seen as consecutive phases, but as different actions of the same metabolic chain.

Like any other natural spontaneous process, AD presents many difficulties which can greatly affect the results of the operations. First of all, there is the problem of temperature, therefore, producing an environment compatible with the bacteria is fundamental.

Low temperature determines low rates for the metabolic reactions, which become more difficult; this is why psychrophilic AD is not often pursued, except in cold regions.^1,2 Instead, with increased temperature there is the possibility of obtaining higher chemical oxygen demand (COD) removal and biogas production, while methane concentration remains the same.³ Favorable pH and volatile fatty acids (VFA) concentration are also important for good results. The optimal pH interval is 7.0–8.0. Below 6.5, the acidity of the medium is too high for methanogenic activity. Above 8.5, there is clear accumulation of ammonia with inhibition phenomena.

Another problematic parameter is the SRT, which can be defined as the time spent by solid particles inside the digester. When material is subtracted from the fermenting device, a part of the active biomass is removed and this implies the necessity of a recovery process. If the new biomass formation is not enough to compensate the previous removal, the entire procedure fails. Beyond this extreme condition, the interval of possible choices for the value of SRT is wide, as visible in Fig. 1.⁴

Fig. 1 Specific biogas production as a function of solid retention time.⁴

A longer retention time induces stronger alteration of the organic matter, and thus a smaller final volume of residuals and bigger gas production. Furthermore, with a long SRT it is possible to guarantee a type of acclimation to toxic compounds. However, it is fundamental to react for above 10 days;⁴ this value can be deduced considering the characteristic time of methanogenic bacteria activity. In fact, it is known that they present higher generation times than acidogenic bacteria, therefore, a short SRT can only induce a methanogen washout and a bigger VFA production. It is evident that the SRT enables shifting of the composition of the microbial population, when acting on the efficiency of volatile solid destruction.⁵ The typical design values of the SRT as a function of temperature are reported in Table 1.⁶

Table 1 SRT values for the design of a mixed digester⁶

Operating temperature [°C]	Minimum SRT [days]	Maximum SRT [days]
18	11	28
24	8	20
30	6	14
35	4	10
40	4	10

---

It is immediately clear that in the case of the absence of recycling or liquid withdrawal, the hydraulic retention time (HRT) is same as SRT. This is not the case with membrane coupled anaerobic bioreactors (AnMBR), which exploit membrane device action to selectively remove just the liquid phase.

In this manner, it is possible to operate with different values of HRT and SRT at the same time, providing the possibility of reduction in volume and an organic loading rate (OLR) increase.

This type of configuration is common for both aerobic and anaerobic treatment of wastewater,^7,8 but in the second case there is a real possibility of developing a gaseous fuel capable of fulfilling the thermal requirement of the whole process.

The application of membranes for biogas plants does not stop here. A persistent problem is finding a possible method to discharge and treat the digestate. In fact, even if it is true that it has reviving properties for the soil,⁹ there is also the problem of high total nitrogen content that goes beyond agricultural purposes. This is why many attempts have been made up to now to reduce this concentration.¹⁰ Membrane filtration, in particular, is good for the recovery of macronutrients from various types of wastes,¹¹ thus it can also be employed for the post-treatment of digestate, in order to obtain an easily discharged residual.

The necessity of pursuing an enhancement in biogas production must be coupled with the possibility of its easy application. Moreover, as stated for the digestate treatment, it is fundamental to find sustainable solutions even for product refining. Biogas is known to be made of various compounds: CH₄ (40–70 vol%), CO₂ (30–60 vol%) and trace amount of NH₃, H₂O and H₂S. Lower concentrations of diluting components also imply higher heats of combustion. By adding close to 803 × 10⁶ J kmol⁻¹, that is, the specific heat capacity of pure methane, it may be possible to produce bio-methane, useful for the natural gas grid or vehicle application.

Even in this field, membranes are promising and have stimulated a large amount of curiosity.^12,13

Therefore, the possibility of developing advanced membrane biogas plants, capable of good performances and easy final separations is real and will be critically investigated in this paper.

In particular, the first part will be devoted to the use of membrane filtration for conditioning the hydrodynamic behavior of a typical digester. Then, the nutrient membrane recovery techniques for liquid management will be investigated, and finally, attention will be focused towards gaseous product purification.

2. Biogas from wastewater: an alternative to solid biomass

Despite the big nomenclature variability present in the literature, the different types of anaerobic digesters can be simply classified in terms of solid content, as in Table 2.^14,15

Table 2 A possible classification of anaerobic digestion processes according to solid content (the reported ranges were deducted taking into account the variability presents in literature)

Solid content [%]	Wet anaerobic digestion	Semi-dry anaerobic digestion	Dry anaerobic digestion
	7 : 15	15 : 20	25 : 40

---

Even in the case of low solid concentration (e.g. in various types of wastewater), there is the possibility of having a highly energetic medium.

This explains why anaerobic digestion of wastewater has gained such success.¹⁶ Moreover, many different industrial sectors are responsible for large productions of liquid waste with high organic content. This problem of stabilization and reuse becomes nearly vital in the Mediterranean area, which combines deep industrial development with a traditional heritage. Slaughterhouses, cheese production centers, distilleries and olive oil mills are just a few of all the possible providers of nutrient wastewaters.

Therefore, in every case, in which there is production activity with contingent formation of wastewater (compatible with biological activity), there is also the chance of using it for biogas production.

The different technological solutions that have been employed in the last years for wastewater treatment are summarized in Fig. 2 and 3; some of them will be further explored in the next sections.¹⁷

Fig. 2 Anaerobic reactor technologies as monitored during the period 1981–2007 (2266 total plants analyzed).¹⁵

Fig. 3 Anaerobic reactor technologies as monitored during the period 2002–2007 (610 total plants analyzed).¹⁵

3. Anaerobic high rate reactors: main criticisms

Generally in the field of biogas production, the technological possibilities are of three types: standard rate digesters, high rate digesters and two-stage digesters.¹⁸ In the first case, the fermenting medium is not mixed or heated; there is a natural stratification that is far from allowing homogeneity. In the high rate digesters, instead, an external action of mixing and heating gives a great increase in the process rate and the possibility of a volume decrease. Sometimes a second vessel follows the high rate digester and gives the chance of promoting a decanting action of the outlet stream and recycling of the active sludge.¹⁸

The undifferentiated removal of material from the digester cannot prevent active biomass withdrawal. This is why a possible coupling with a membrane has been studied.^19,20,21 The possibility is very interesting, especially in the cases in which the classical technique of natural biofilm formation and granulation do not provide acceptable results.

In fact, one of the most common choices for municipal wastewater treatment is the Upflow Anaerobic Sludge Blanket (UASB).²² With this type of configuration, the possibility of separating the HRT and SRT is partially explored, because biomass agglomerates in the form of bigger particles or flocs go to the bottom of the vessel. This produces a clarified zone from which it is possible to withdraw a high value of SRT and a low value of HRT with certainty.

Actually, the process of granulation is not so easy to describe,^23,24,25 and it can be greatly influenced by the operative conditions. For example, it was found that natural agglomeration is favored by high liquid upflow velocity and short HRT.^26,27,28 With different conditions, there is the risk of producing a low, dense and porous biofilm. In contrast, a high shear strength is capable of inducing a thinner and more compact layer.²⁸ Even the value of the OLR must be chosen properly, because a high increase in the start-up phase can induce a massive formation of biogas that can alter the granulation process, owing to the rising bubbles. It is also easy to understand that the substrate composition can influence the results, because with its variation there is alteration of the cellular activities and the nature of the extracellular polymers (ECP) produced.^29,30

All these different aspects, combined with the necessity of avoiding extreme temperature conditions,³¹ make UASB a better alternative.

The biggest problem is that of the start-up, because it requires strict control and longer time. In the first phase of utilization, a badly developed active biomass does not allow efficient COD removal and there is the necessity of increasing the HRT in order to avoid a washout of the microbial seeds. For example, for the treatment of municipal wastewater at 20 °C, a start-up period of 60 days was found to be enough for a COD removal of 80–85%.³² A longer period, of exactly 120 days, was employed, instead, for the treatment of Lurgi coal gasification wastewater.³³ Even though the periods were always long, there is a certain difference between the various time choices reported in the literature. This discrepancy can be justified by considering the different operative conditions and nutritional capacities of the various substrates. For example, in the case of the Lurgi coal gasification wastewater, the need for a long start-up time is due to the low biodegradability and high inherent toxicity to make acclimation harder.³³

This difficulty arises with every type of wastewater characterized by high solid concentration. In fact, there could be an accumulation of inert particles and slow biodegradable solids, which lead the bacteria concentration to be at a low level, causing their inability to achieve good COD removal. For example, in the case of potato-maize wastewater treatment,³⁴ this accumulation was observed with a high OLR (>10 g_COD per L per days) and low HRT (>1 day).

As already underlined, temperature is one of the most problematic parameters, not only for its effect on the metabolic rate, but also for the influence on the macroscopic properties of the fermenting mixture, which can determine the bad output of a UASB. In particular, temperature is able to affect the rheology of the ECP. In the case of extracellular polysaccharides produced by a Pseudomonas oleovorans NRLL B-14682, a decrease in the steady shear and dynamic rheological parameters was detected with the increase in temperature.³⁵ This is the result of a greater mobility of the polymeric chains blocking bio-flocculation and settling ability. Therefore, enhancement of biological reaction rate is accompanied by an increase in the turbidity of the outlet stream and excessive biomass removal.^36,37

Settling can also be worrisome in the case of high lipid content, because they can produce a hydrophobic layer around the biomass particles, inducing their floatation, and hence the biomass discharge from the UASB. This is not the only problem that arises with this type of material. In fact, excessive scum formation with the problem of a plug at the gas outlet was observed.³⁸ Actually, the difficulties with high grease and fat content do not stop here because of the inhibitory effect they can induce. It is known that lipids can be easily hydrolyzed in long chain fatty acids with high toxicity for Gram-positive microorganisms,³⁹ like the methanogens.

Another problem related to the use of UASB, which can be partially solved with the use of a membrane, is that of the salinity.^40,41 A high concentration of salts induces a strong osmotic pressure that can force water out of the cells, producing deep distress for the microorganisms.^42,43 This is why the residues that comes from seafood processing are very difficult to handle^44,45,46 without considering the possibility of high calcium concentration. In fact, a low concentration of calcium can be useful for the sludge settling, but beyond a certain limit (about 200 mg L⁻¹), there arises a series of complications, such as precipitation of CaCO₃, cementing of the blanket and reduction in methanogenic activity.^47,48

Considering all the information reported to date, it is possible to understand the main criticisms of classical anaerobic high rate processes.

4. Anaerobic membrane reactors: state of the art

AnMBR appeared for the first time in 1978 (ref. 49) as a combination of anaerobic digestion and a cross-flow membrane. The idea was developed in order to obtain a process capable of major stabilization of wastewater. In fact, in its origins, the anaerobic treatment of organic waste employed just a vessel with short residence time. Under this condition, performances were limited and did not allow a drastic reduction of the polluting power from that of the starting wastewater.

Since this first attempt, it was clear that with a membrane module the global rate of the process could be enhanced by 3–4 times, with a biochemical oxygen demand (BOD) removal of 85–95% and an effluent with low nitrate concentration.⁴⁹ With this work, huge speculation about the theme of anaerobic membrane reactors started. Actually, in the first step of the research the main problem was ascribed to the cost of the membrane unit, which was high for large-scale application. This is why the membrane anaerobic reactor system (MARS) developed by Dorr-Oliver never overcame the pilot scale.⁵⁰

A good enhancement in knowledge was obtained with the help of several different research projects. Aqua Renaissance, in particular, was an intense attempt in Japan for promoting an anaerobic membrane reactor as a useful tool for wastewater treatment.⁵¹ All the works included in this project were barely contemporary to the development of the ADUF (anaerobic digestion ultrafiltration) process in South Africa.⁵² This was the first time in which self-made membranes were employed for the development of the ultrafiltration unit connected to the anaerobic digester. This interest in membranes is not surprising, because by acting on a feature of the membrane (pore size, thickness, smoothness, etc.), it is possible to exert a certain effect (flux, extent of fouling, durability of the system, etc.).

All the first studies were based just on the search for a good setting and an evaluation of the enhancement of the anaerobic treatment produced by ultrafiltration or microfiltration units.

This led to a configuration similar to the one in Fig. 4, which was found in various examples.^53,54,55

Fig. 4 A typical arrangement for anaerobic membrane reactors.

Contrary to popular belief, it is not obligatory to operate under pressure; in fact, the permeation can be induced with suction through a submerged membrane unit (SAnMBR). A typical scheme is that in Fig. 5.

Fig. 5 A typical arrangement for a submerged anaerobic membrane reactor.

These types of systems were systematically studied starting from the end of the 90's,⁵⁶ and they continue to attract a great deal of attention.⁵⁷ Innovation is also possible in recirculating the gaseous stream for generating cleaning action and fouling reduction in the membrane module. The concept directly comes from the processes of aerobic wastewater treatment in which airflow is employed to partially remove caking on the membrane surface. Even if the biogas recirculating stream could not be regular, as an air stream, it gave promising results.⁵⁸

In fact, the problem of membrane fouling has gained importance in the last decade and it has been treated by different points of view.^59,60 The rising transport resistance is one of the major difficulties that limits the use of this type of technology. Moreover, there is also the necessity of guaranteeing proper conditions for the anaerobic digestion process that can be easily altered. This is why AnMBR are not so commonly employed as a full-scale wastewater treatment. One exception is the Kubota's submerged anaerobic membrane reactor process that was used for the stabilization of all the wastes that come from the agro-industry.⁶¹ This patented process is based on the contemporary use of a solubilization tank, followed by a digester with submerged membranes, and it was used for the construction of less than 20 full-scale plants based in Japan and America.⁶¹ In doing so, it is possible to retain methanogenic bacteria, removing all the inhibitors of the anaerobic fermentation process. The results obtained are positive, with a total volume reduction of three to five times.

Considering all the materials that can be found in the literature, it is possible to summarize them as in Table 3.

Table 3 A brief review of the scientific production about AnMBR

Type of wastewater	Type of membrane	Volume [L]	Efficiency	Reference
Raw wastewater from alcohol fermentation	Tubular ceramic membrane (0.45 μm)	5	COD removal = 90–95% (OLR = 2–7 kgCOD per m^3 per days)	60
Synthetic wastewater (glucose = 10 000 mg L^−1)	Polypropylene membranes modified by ozone treatment (0.2 μm)	4.5	—	61
Raw wastewater from alcohol fermentation	Hydrophobic polypropylene membrane (0.2 μm), zirconia skinned inorganic membrane (0.14 μm)	5	COD removal > 90% (OLR = 3–3.5 kgCOD per m^3 per days)	62
Artificial wastewater; slaughterhouse effluent; sauerkraut brine	Ceramic cross-flow membrane (0.2 μm)	7	COD removal > 97 (OLR = 6–8 gCOD per L per days)	63
Wastewater from food processing	Polyethersulfone ultrafiltration membranes (20 000 to 70 000 Da)	0.5	COD removal > 90% (OLR < 2 kgCOD per m^3 per days), COD removal > 80% (OLR = 2–4.5 kgCOD per m^3 per days)	64
Sewage sludge	Polysulphone membrane (0.1 μm)	—	COD removal = 98.8% (OLR = 0.1 kgCOD per m^3 per days)	65
Synthetic substrate (460 ± 20 mgCOD L^−1)	Polyethylene membranes (0.4 μm)	3	COD removal = 90% (HRT = 3 h)	56, 66 and 67
Synthetic wastewater	Polysulphone microfiltration membranes (0.2 μm)	3.7	—	68–72
Raw domestic wastewater	Stork WFFX 0281 (100 kDa)	50	COD removal > 76% (OLR = 0.8–1.2 gCOD per L per days)	73
Cheese whey	Ceramic membrane (0.2 μm)	27	COD removal = 98.5% (OLR = 19.78 gCOD per L per days)	74
Swine manure	Polyethersulfone membrane (20 000 Da)	6	COD removal > 86% (1–2 gVS per L per days)	75
Municipal wastewater	Stork WFFX 0281 membrane (100 000 Da)	50	COD removal > 88% (OLR = 0.23–2 gCOD per L per days)	76
Landfill leachate	Capillary ultrafiltration module (polymeric membrane 0.1 μm)	29	COD removal = 90% (OLR of 2.5 kgCOD per m^3 per days)	77
Sucrose-based synthetic feed (4 gCOD L^−1)	Polyethylene membrane (0.4 μm)	3	COD removal = 98% (OLR = 16 gCOD per L per days)	78
Synthetic saline sewage (465 ± 20 mgCOD L^−1)	Polyethylene membrane (0.4 μm)	3	DOC removal = 99% (with 35 g NaCl per L)	79
Kraft evaporator condensate (COD : N : P = 100 : 2.6 : 1)	Polyvinylidene fluoride membrane (70 000 Da)	10	COD removal = 97–99% (OLR = 3.1 kgCOD per m^3 per days in thermophilic condition, OLR = 12.2 kgCOD per m^3 per days in mesophilic condition)	80 and 81
Synthetic substrate (10–17 gCOD L^−1)	Ceramic Al2O3 membrane (0.2 μm)	2	—	82
Synthetic sewage	Polyvinylidene fluoride ultrafiltration membrane coated with 1% PEBAX 1657 (100 000 Da)	10	COD removal > 96% (OLR = 5 kgCOD per m^3 per days)	83
Synthetic municipal wastewater	Poly-tetrafluoroethylene membrane (1 μm)	4	COD removal > 95% at 25 °C, COD removal > 85% at 15 °C (OLR = 1 kgCOD per m^3 per days)	84
Dilute municipal wastewater	Membrane filter by PCI membrane systems, Inc., Milford, OH (0.1 μm)	10	COD removal = 64% (OLR = 0.03–0.11 kgCOD per m^3 per days)	85
Sucrose	Flat sheet Kubota membrane (0.4 μm)	11	—	86
Municipal wastewater	Polyvinylidene fluoride membrane (140 000 Da)	80	COD removal = 90% (0.26 LCH4 per gCOD-removal)	87
Municipal wastewater (COD = 630 ± 82 mg L^−1)	Polyether sulfone ultra-filtration membrane (38 nm)	350	COD removal = 90% (OLR = 0.6–1.1 gCOD per L per days)	88
Suspended marine microalga P. tricornutum	Kubota microfiltration membrane module (0.4 μm)	8	COD removal = 52.2% (OLR = 1–6 gCOD per L per days)	89
Ethanol thin stillage	Polyvinilydene fluoride membrane (0.08 μm)	12 000	COD removal = 98% (OLR = 4.5–7 kgCOD per m^3 per days)	90
Synthetic wastewater (cheese whey : sucrose = 1 : 1)	Kubota microfiltration membrane module (0.4 μm)	11	COD removal = 94% (OLR = 1.5–10 gCOD per L per days), COD removal = 33% (OLR = 1.5–10 gCOD per L per days)	91
Domestic wastewater	Polyethersulfone membranes (0.2 μm)	7	COD removal = 92 ± 5% (OLR = 170–660 mgCOD per L per days)	92
Sludge and coffee grounds (85 : 15)	Chlorinated polyethylene (0.2 μm)	15	COD removal = 44.5–66.8% (OLR = 2.2–23.6 kgCOD per m^3 per days)	93
Synthetic wastewater	Cellulose triacetate membrane	3.6	TOC removal > 96%	55

---

5. Type of possible wastewater

As apparent from Table 3, AnMBRs were tested for the treatment of various types of wastewaters. In fact, in order to assess the feasibility of any type of technology, it is fundamental to show its capacity of producing a beneficial effect with moderate cost. In this case, the problem is even more complex, because there is the double necessity of guaranteeing a high rate process and a good energetic integration with biogas production.

An MBRs showed great performances with various types of substrates; the only doubts are connected to the possible treatment of wastewaters with high organic strength and low particulate content.^94–96 In this case, membrane application is not so advantageous, because the bigger capital cost and technical difficulties make the traditional high rate processes (UASB, EGSB) the best solution. In fact, it is inexpensive to promote a process based on a spontaneous flocculation even if the start-up time is longer. Thus, it is possible to have at the end of the initial phase a performance that is comparable with an AnMBR.

Most of the lab-scale work was performed using synthetic wastewaters in order to reveal possible anomalies before any further investigation.

For example, the effluents released by tapioca starch factories were easily simulated using an aqueous solution made of starch and additional salts in order to reproduce the salinity.⁹⁷ In that case, it was fundamental to try to understand the process for the possibility of incorporating the anaerobic digestion process into the economies of tropical countries, which highly exploit this crop.

Obviously, a synthetic mixture is only useful for a preliminary examination of the potentiality of a certain waste, because it cannot completely describe the complexity of the original scenario, which is perturbed by many variables. This is why in most of the cases, the attention was focused on waste materials directly produced by the process industry.

There are many possibilities which come from food and industrial processing, in general. These materials are often pollutants and can obtain a second life with a natural degradation process. For example, food wastes usually have high organic content (1000–85 000 mg_COD L⁻¹) and high solid concentration (50–17 000 mg L⁻¹).⁷ They are massively produced in the global economy, and they can be transformed into useful resources more easily than raw materials.⁹⁸ The huge nutritional power can sustain anaerobic digestion, and it should allow high OLR. Considering some of the works found in the literature^76,99,100 dealing with AnMBRs for food waste, it is possible to find out a suggested OLR that is bigger than 10 kg_COD per m³ per days with an HRT of few hours. These results must be compared with the ones obtained using traditional processes. In particular, the OLR that can be tolerated with high rate processes is similar to the one used for AnMBR.^101,102 The big difference is in the HRT, which is of many days. The possibility of treating food-processing residuals with low HRT indicates a faster process that is therefore more captivating.

Many other production activities generate highly useful wastewater. It is not easy to summarize because the typical features depend on the type of activity considered.

For example, there is the case of textile wastewaters, which come from fiber treatments. Considering the large amount of water used by this sector, it is not surprising that the problem of waste valorization has become important. Moreover, there is a big variability of the main characteristics associated with the type of transformation, as apparent from Table 4.¹⁰³

Table 4 Main characteristics of the typical wastewaters of textile industry¹⁰¹

Process	COD [g L^−1]	BOD [g L^−1]	TS [g L^−1]	TDS [g L^−1]	pH [-]	Colour (ADMI)	Water usage [L kg^−1]
Desizing	4.6–5.9	1.7–5.2	16.0–32.0	—	—	—	3–9
Scouring	8.0	0.1–2.9	7.6–17.4	—	10–13	694	26–43
Bleaching	6.7–13.5	0.1–1.7	2.3–14.4	4.8–19.5	8.5–9.6	153	3–124
Mercerizing	1.6	0.05–0.1	0.6–1.9	4.3–4.6	5.5–9.5	—	232–308
Dyeing	1.1–4.6	0.01–1.8	0.5–14.1	0.05	5–10	1450–4750	8–300

---

One of the most attractive possibilities is the reduction of various types of dyes. In particular, azo dyes are the most worrying because of their aromatic nature.

It is not possible to find a single type of microorganism or enzyme that is capable of reducing every azo dye. This is why attention is directly directed to the natural processes based on a metabolic chain of various microorganisms, like anaerobic digestion.

In particular, using a SAnMBR it was possible to promote immediate azo dye removal (more than 90%), even with high influent concentration.¹⁰⁴ The only problem observed was connected with the considerable amount of VFA in the effluent stream, together with the aromatic amines derived by azo dye cleavage. This result was justified by assuming the inability of methanogens to metabolize the aromatic amines and by assuming moderate inhibition at high azo dye concentration of the influent.¹⁰⁴

Another case that has been tested with anaerobic membrane reactors is that of wastewater generated by paper treatment. Even in this case there is a huge variability regarding the type of process and wood employed.¹⁰⁵ An interesting resource is the evaporator condensate of a kraft pulp mill, which has a high methanol concentration.¹⁰⁶ Good overall performances were obtained employing a SAnMBR equipped with a flat sheet microfiltration membrane;¹⁰⁷ COD removal reached an average level of 93–99%, without particular toxic reactions. Moreover, there is a possibility of using the residuals from pulping and paper mill operations, which have captured certain attention for their high nutritional capacities.^108,109

Actually, in the field of AnMBRs, the most exploited substrates are municipal wastewaters, which are streams with low organic content and high particulate concentration. Anaerobic treatments have obtained great attention in this field because of the possibility of removing any air addition and promoting the formation of a gaseous product capable of sustaining part of the costs. In particular, municipal wastewaters have been tested with CSTR, UASB and EGSB membrane coupled reactors. The results obtained with short HRT (0.8–120 h) and good OLR (0.23–12.5 kg_COD per m³ per days) are promising in terms of COD removal.¹¹⁰ This implies the possibility of generating a permeate stream of non-potable water that can be used for other purposes.

Despite the beneficial effect of the membranes, there are still some problems related to the use of municipal wastewaters. In particular, there is the possibility of a high pharmaceutical content that can inhibit microbial activity.¹¹¹ Antibiotics, in particular, are responsible for a bactericidal effect. However, this possibility is negligible in a large-scale process. In fact, comparing the experimental data obtained with and without pharmaceutical traces, just a slight drop in methane biogas concentration was observed.¹¹²

Another problem that is typical of municipal and industrial wastewater is that of the heavy metal content. Two-year monitoring of a biological wastewater treatment system in Gdansk allowed the finding of a fluctuating concentration of cadmium, copper, lead and zinc (0.01–0.8 mg L⁻¹) in the influent wastewater stream.¹¹³ Moreover, this concentration tends to increase during anaerobic digestion owing to the degradation of the organic compounds and the removal of material in the form of methane and carbon dioxide.¹¹³ The major problem associated with heavy metals is that of irreversible binding with enzymes produced by bacteria, which become incapable of acting their usual role.

Considering all the topics analyzed to date, it is clear that membrane coupled anaerobic digesters combine all the capabilities and difficulties of a traditional fermenter with the power and criticisms of membranes. Their optimization is even more complex because of all the possible substrate mixtures and technical arrangements.

6. Types of filtration and fouling problems

In the field of AnMBRs, the most common solutions are essentially two: microfiltration membranes or ultrafiltration membranes.

Small pores with a mean diameter in the range of 10–1000 Å characterize ultrafiltration membranes,¹¹⁴ which are capable of retaining even colloids and macromolecules according to their cut-off diameters. Generally, to qualify the performance of a membrane, it is necessary to estimate the flux that undergoes a rapid decline when used for wastewater. There is an initial drop that is caused by the formation of a cake layer and pore blocking action. Then, with the passing of days, there is a slow further drop caused by a densification effect.

The resulting gel layer is made of solid particles, macromolecules, microorganisms and their ECPs. In general, once the gel layer has appeared, it is impossible to further increase the flux through the ultrafiltration membrane with any operative conditions. Therefore, in order to avoid any useless energy consumption, it is better to work with a pressure drop capable of guaranteeing a flux near the critical value.

Actually, the situation is even more complex because of the possibility of gel layer removal due to eddies near the membrane surface, producing flux recovery and a result that is better than the predicted one.¹¹⁵

Microfiltration, instead, is based on the use of membranes with bigger pores (0.1–10 μm),¹¹⁴ and it is traditionally employed for the removal of bacteria from water. Even in this case there is a progressive decline in flux determined by membrane fouling that has a similar mechanism as the one previously described.¹¹⁶ The problem of water flux is of crucial importance because it determines the practical feasibility of an AnMBR. For example, using a commercial ultrafiltration membrane in a lab-scale AnMBR, a flux decline of 90% was observed after 20 days of usage.⁵³ This troubling fouling action was due to both inorganic and organic materials. In particular, struvite (MgNH₄PO₄·6H₂0) was identified as the major inorganic foulant.⁵³ Its formation is generally induced by the combined presence of magnesium, ammonium and phosphate ions in the broth, and it is responsible for a strong fouling action.

This is why some attempts have been tried in order to reduce this action. In particular, there is the possibility of employing a type of backfeeding.⁶² This technique was considered because in long time operations backflushing with permeate was not capable of recovering water flux to the original value.⁶² In particular, the operation of backfeeding consists in the sporadic addition of an acidic feed solution to the membrane module, which induces a pH drop and a deeper cleaning action. A lower pH inhibits struvite precipitation and helps in removing this solid phase that has grown inside the pores. The big difference from backflushing is in the possibility of feeding the bioreactor using the membrane module, which undergoes a recovery effect owing to the weak acidity of the solution.

Actually, the scenario is highly influenced by the constructive materials of the membrane. This is why, under the same conditions, inorganic and organic membranes gave different results.⁶⁴ For example, comparing the performance of a hydrophobic polypropylene membrane with a zirconia skinned inorganic membrane, a different behavior of the flux was observed. An organic membrane gives a stronger flux reduction with faster reaching of a steady state condition. An inorganic membrane, instead, after a consistent initial drop, does not reach a steady state flux in a period of 150 hours and guarantees a better overall performance.

The differing behavior was explained in term of surface smoothness. The hydrophobic polypropylene membrane had a rough surface that ensured strong adhesion of solid particles. This effect was absent in the case of the inorganic membrane, which did not give rise to a cake layer (Fig. 6).

Fig. 6 SEM analysis of the surface of the organic (a) and inorganic (b) membrane before use; cross sectional view of the organic (a′) and inorganic (b′) membrane after 70 days of use.⁶²

It was estimated that the biggest foulant of the inorganic membrane was again struvite, which was not so abundant in the organic membrane.⁶⁴ Therefore, precipitation is easier on surfaces which have a type of chemical similarity with the precipitate. In exactly the same manner, the hydrophobic flocs tend to accumulate more easily on a similar surface.

The problematic interaction between particles and surfaces was studied even in the treatment of pulping residues.¹¹⁷ In particular, it was found, with a polyvinylidene fluoride membrane of 70 000 Da, that the cake layer was primarily made of the smallest flocs present in the fermenting medium with higher bound ECP and even the microorganisms were different from the ones in the bulk. This is related to the concept that a potential gradient between the bulk and the surface membrane is the basis of the separation process and determines this diversification inside the system.

A composite structure¹¹⁷ was also identified in the growing cake, which is generally made by a dense layer strictly attached to the membrane, a porous layer and a loose upper part made of gel-like substances. The elemental analysis of the cake layer demonstrated, even in the case of pulping wastewater, the possible presence of struvite,¹¹⁷ which has already been identified as an aggressive foulant.

As it was briefly said, ECPs are another important agent that can promote the formation of an adhesive layer. They are essentially made by proteins and polysaccharides, and the ratio between these compounds can influence the adhesive properties even more than their actual quantities.¹¹⁸ A bigger protein concentration is responsible for a stronger hydrophobicity of microbial flocs. Owing to this condition, the flocs tend to more easily organize themselves in a sticky layer against the membrane.

It was also found that the microbial communities in the bulk phase and in the cake layer are not completely the same.¹¹⁷ Considering that different microorganisms can produce different types and quantities of ECPs, it is easy to realize that there must be some specific bacterial population involved in the starting of the biofouling action.¹¹⁹

Even temperature can affect the magnitude of biofouling phenomena. In particular, thermophilic digestion generates larger amounts of fine flocs (<15 μm), biopolymers and soluble products.⁸² This situation causes bigger difficulties in process management, with the impossibility of operating at the same water flux as in the mesophilic case, and with the necessity of frequent readjustment. First of all, with a higher temperature there is stronger production of soluble microbial products, which can alter membrane performance.^120,121 This effect is also accompanied by stronger production of biopolymer clusters, that is to say, substances with bigger dimensions than those of the soluble microbial products and independent from bacterial activity.^82,122 An even higher ratio between proteins and polysaccharides was detected in the thermophilic sludge⁸² along with all the problematic effects already described. However, the most significant and alarming effect related to thermophilic AnMBRs is the stronger production of small flocs. A bimodal particle size distribution was detected in several different works.^{19,82,123,124} Therefore, in the case of thermophilic AnMBRs, the bulk sludge is made of dispersed aggregates and big flocs. The smallest particles (1–10 μm) are considered responsible for the formation of a rather dense layer against the membrane surface, acting a key role in the flux decline mechanism. The formation of these fine particles is induced and favored by two coexisting effects that determine the disruption of the microbial aggregates: the proceeding of the demolition process and the action of the shear stress in the membrane module.¹²⁵ All this generates a progressive reduction of the mean diameter during the entire operation.¹⁹

The themes discussed so far are sufficient for understanding the complexity of the hydrodynamic behaviour of an AnMBR. In this case, the typical problems of membrane management are deeply connected with the biological nature of the operation, which evolves in respect of its microbial vocation.

7. Control of membrane fouling

The possible implementation of a membrane on a large-scale anaerobic digestion process is dependent on the possibility of controlling the flux. In fact, when appreciable fluxes require excessive transmembrane pressure, there is the necessity of recovering membrane performance with proper cleaning action.

To overcome the fouling problems, there are different routes which can be followed:

• change in the operation strategy;

• inhibition of foulant production;

• alteration of membrane surface properties;

• modification of broth characteristics.

Acting on the hydrodynamic behavior of the filtration module, it is possible to induce a reduction of transport resistance, and thus recover the flux. This is the meaning of a change of operative strategy, and it can be pursued in various manners. First, there is the already cited possibility of backflushing, which consists in a temporary inversion of the permeation through the membrane owing to a high pressure on the permeate side.

This cleaning action was highly studied for both ultrafiltration and microfiltration in anaerobic and aerobic reactors. Generally, the system is backflushed for a few seconds with a certain frequency (for 5–10 s every 3–15 min).¹²⁶ The main effect is a drastic reduction of the concentration polarization resistance, while just a small decrease of the internal fouling can be produced with this type of action.¹²⁶

Moreover, there is a certain dependency of the result from the nature of the foulant agents. This is why backflushing showed better performances with yeasts than with bovine serum albumins.¹²⁷ In fact, the effect of the recovering procedure is deeply linked with the nature of the interactions between the membrane and foulants. Some compounds like albumins are capable of widespread adhesion, and they do not allow proper and complete restoration. Considering that wastewater for biogas production is made of many constituents, it is easy to understand that there could be some limitations on the beneficial effects of backflushing.

Another possibility that is very similar to the one previously described is backpulsing. In this case, the inversion of the filtration through the membrane is considerably more frequent and faster, like a pulse. The results are absolutely positive, with good flux enhancement.¹²⁸ Even in this case, the results indicate the impossibility of complete internal fouling removal.¹²⁸ As it was for backflushing, there is the necessity of optimizing the operation, remaining inside the bonds imposed by the material resistance. In fact, it is easy to understand that the long exposure to the intermittent stresses can seriously damage the membrane used.

Another possibility is based on the use of gaseous streams other than air or oxygen to alter the anaerobic environment required for the biochemical conversion. For example, in the case of ceramic anaerobic membranes, which are able to tolerate high pressure, the possibility of employing N₂ backflushing was explored.¹²⁹ More specifically, with a pore membrane size of 1 μm, a filtration time of 16 min with backflushing intervals of 40 s and a transmembrane pressure of 1.0 kg_f cm⁻² demonstrated that it was the best solution.¹²⁹ The results of the optimization are highly dependent on the type of the membrane employed, and it is not surprising that smaller pores lead to the necessity of more frequent recovering activity.¹²⁹

The possibility of using inert gases for membrane care is also at the base of all reactors, which employ biogas sparging systems.^82,83,117

The rising bubbles are capable of promoting a certain removal of the foulant layer, and, above all, with their mixing action, they can cause a drop of the concentration polarization resistance.¹³⁰ This type of arrangement is not defect-free. In fact, some problems can arise.

For example, recycling the biogas inside the fermenter means increasing the CO₂ concentration in the liquid, and the resulting pH drop can induce some inhibition effects.¹³¹ Methane, instead, does not produce negative effects because of the law of solubility in the liquid phase. Therefore, it is right to assume that there is no accumulation in the broth.

Acting on the recycled gas flow means acting on the intensity of the shear against the membrane, and therefore on the ECP production. Long exposure to high shear stress induces a smaller production of floc-associated ECPs. This can be justified not only in terms of a worse generation of total ECPs, but also in terms of a smaller production of floc-ECPs resistant to the erosion effect.¹³² Obviously, there is a type of adaptive mechanism of the biomass that is able to change its activity according to the shear condition of the surrounding media.

A scouring action of the membrane surface can be produced with the addition of powdered activated carbon (PAC) to the mixture to be filtered.^80,133,134 Not only are these particles able to produce a mechanical scouring action, but they are also responsible for a buffer effect against VFA and they can adsorb fine colloids onto their surfaces, avoiding their adhesion to the membrane.⁸⁰ Even an increase in the resistance to compression has been shown.¹³⁵ This implies the possibility of having a porous cake with a weaker resistance. Considering all the aspects observed so far, it is clear that this technique (based on broth alteration) improves the performance of a membrane reactor owing to a double action that is both mechanical (erosion of the cake layer) and physical (adsorption of possible adhesive compounds). Obviously, the optimal concentration of particles in the liquid to be treated is also affected by economical estimation; one solution has to be considered costly until it can produce the same beneficial effect with a smaller cost.

There is also the possibility of employing zeolites for conditioning the behavior of the fermentation broth.¹³⁶ In fact, they are capable of reducing the ammonium concentration, preventing struvite precipitation. In particular, the experiments conducted with clinoptilolite (a natural zeolite) showed a flux enhancement of 15–20% in the case of a ceramic membrane.¹³⁶ This fact is not surprising, because these natural materials are well known for their ability of ion exchanging. Obviously, the overall performances are greatly affected by the physical and chemical properties of the system. The overall performance is a combination of two effects: the cation exchange and the adsorption onto pore walls. In particular, the ion exchange mechanism prevails at lower ammonium concentration, that is to say, when this amount is comparable to that of the cations entrapped inside the zeolite. Adsorption instead begins to prevail at higher concentrations.

As proved by many studies,^137,138,139 the development of a cake layer is connected to the morphology of the membrane surface. This explains why many attempts have been made in order to develop a surface conditioning technique capable of promoting slower flux decrease. In this sense, the case of graft polymerization inducing the attachment of functional groups to the membrane substrate is remarkable. The performance of the functionalized membrane could be promoted by the use of UV rays,^140,141,142 plasmas^143,144 or ozone,¹⁴⁵ and the result was the induction of hydrophilic behavior that could limit the hydrophobic interaction between the proteins and membrane.

Obviously, different techniques can be combined together to promote better membrane recovery. For example, graft polymerization and backpulsing were tested together with a polypropylene membrane.¹⁴⁶ In this particular case, the drop in interactions between the foulants and membrane produced a higher efficiency of the backpulsing action. As already underlined, a big contribution to cake layer formation comes from microbial activity. This is why efforts to limit this activity have been made.^147,148 The main objective is to try to disrupt the quorum sensing mechanism, which also contributes to biofilm formation. This method can give good results, even with only the addition of free¹⁴⁹ or immobilized¹⁵⁰ Acylase I to the fermentation broth in the membrane bioreactors (MBR). This enzyme can destroy N-acetyl homoserine lactone, an autoinducer of Gram-negative bacteria, providing a beneficial effect in terms of biofouling reduction. Considering all the problems of cost and stability connected to the use of free enzymes, the strategy of bacterial quorum quenching has also been tried.¹⁵¹ In this second case, recombinant microorganisms (capable of high enzyme production) are entrapped in a precise region of the MBR in order to prevent their dispersion. Therefore, using a membrane and developing a “microbial-vessel”, it is possible to put the broth in contact with microorganisms responsible for anti-fouling enzyme production.

To our knowledge, application of this technique to AnMBRs has not yet been tested. Even in anaerobic conditions there is the development of a quorum sensing mechanism, therefore, even in this case an enzymatic or bacterial quenching action could be helpful.

Finally, it is important to underline that a big effort has been made in the last few years to develop a procedure for the production of membranes capable of good performances without complex treatments.

Patterned membranes are the main results of this approach. They are prepared by inducing the formation of precise patterns on the flat surface of a membrane with soft lithographic techniques. Pyramid type and prism type membranes (Fig. 7) have been recently tested and show an enhanced flux in comparison with traditional flat-sheet membranes.¹⁵²

Fig. 7 SEM images of two different patterned membranes.¹⁵⁰

It is also possible to produce patterned hollow fiber membranes, which adopt a patterned nozzle during fabrication.¹⁵³

One fundamental aspect is the “fidelity of the pattern”, that is to say, the similarity of the master and replica mold. Obviously, it should be the highest possibility because only in this manner is there a chance of predicting the performance of the separation process. There are many variables and operative aspects that can greatly affect the fidelity of the pattern; the molecular weight and the concentration of the polymer in the casting dope are probably the most important, and they can be easily controlled during the preparation phase.¹⁵⁴ Instead, by acting on the height and the shape of the pattern, it is possible to produce a certain effect on the water flux and biofouling. Experimental evidence shows that, by increasing the height of the pattern (in the case of a prism-patterned membrane) it is possible to produce a higher flux and a smaller extent of biofouling.¹⁵⁴ While the first result can be easily justified, considering that a greater height means a larger area, the evidence for the second is connected to the intensity of the shear stress produced against the patterned wall.

Another option that is also based on micro-fabrication techniques is a micro-sieve membrane. The most captivating feature is in the possibility of obtaining huge fluxes (two or three orders of magnitude larger than those of traditional membranes) with moderate operative conditions. The non-tortuous pores and their uniform and geometrically organized distribution on the layer (Fig. 8) is responsible for such a good performance.¹⁵⁵

Fig. 8 SEM image of a microsieve membrane.¹⁵³

Moreover, the possibility of having a strict narrow pore size distribution implies the chance of real resolution of the complex mixtures fed to the membrane device.

The application of this alternative to wastewater treatment showed great results in terms of biofouling management. More specifically, by adopting UV-curable polyurethane acrylate as the starting material for the preparation of a micro-sieve membrane, it is possible to develop a layer with a low surface energy, which implies a lower biofouling tendency.¹⁵⁶ Obviously, with the passing of time, there is a certain pore blocking effect, but despite this, the huge increase of the flux is enough to justify the interest in this alternative in the field of membrane bioreactor management.

8. Membrane process for recovery of useful resources from digestate

Once the substrate is completely exhausted, it should be removed from the anaerobic digester and handled in proper manner. Considering that active plants are quite numerous nowadays, it is easy to understand that there is a real need of developing simple processes for the reduction of the residual and a possible recovery of useful resources from them. In this sense, membranes represent a valid alternative.

Conventionally, the digestate is just left in storage tanks and periodically used for fertilizing. There are also other options briefly explained in Fig. 9.¹⁵⁷

Fig. 9 Conventional treatments of anaerobic digestate.¹⁵⁵

Essentially, the two things that could be important to separate from a digestate are phosphorous and nitrogen. In particular, the latter is more problematic because of the gaseous loss of ammonia and the possibility of eutrophication of the aquifers. This risk is less in the case of phosphorous because of its ability to adhere to the soil particles.¹⁵⁸

A membrane process should be able to create a permeate, made of water, useful for other applications, and a concentrated retentate rich with the nutrients cited above, suitable as fertilizer and easy to storage. In this sense, ultrafiltration and microfiltration can be used for the removal of solid particles and macromolecules, while nanofiltration and reverse osmosis allow further purification of the water.

The first critical issue is related to the quality of the permeate. It is necessary to obtain water for agricultural or potable use. This last option requires extreme purification in accordance with the international and local standards. In the European Community, these limits are clearly explained in the text of the directive 98/83/EC, whose statements are briefly summarized in Table 5.

Table 5 European limitation for potable water (directive 98/83/EC)

Biological parameters
Parameter	Value
Escherichia coli (E. coli)	0 in 250 mL
Enterococci	0 in 250 mL
Pseudomonas aeruginosa	0 in 250 mL
Colony count 22 °C	100 in 1 mL
Colony count 37 °C	20 in 1 mL

---

Chemical parameters
Parameter	Limit concentration
Nitrate	50 mg L^−1
Nitrite	0.50 mg L^−1
Arsenic	10 μg L^−1
Benzene	1.0 μg L^−1
Nickel	20 μg L^−1
Chromium	50 μg L^−1
Mercury	1.0 μg L^−1
Cadmium	5.0 μg L^−1
Hydrogen ions	6.5–9.5 pH unit

---

Experimental evidence shows that E. Coli can be almost completely removed both with ultrafiltration and microfiltration membranes.¹⁵⁹ This capacity is mainly due to the dimension of the bacteria being bigger than the average pore size of both types of membranes. Even various types of parasites, like Nematode, can be completely excluded from the permeate.¹⁵⁹

Not only should membranes be capable of good retention, but also great attention should be paid to the contamination of the permeate side, which could be reached by undesirable microorganisms.¹⁶⁰ This, combined to the possible presence of nutrient traces, can induce the formation of a thick biofilm, even on the permeate side of the membrane, with the generation of unacceptable water.

The problem looks even more complex if it is analyzed in terms of finding the optimal solution for the simultaneous removal of not only the microorganisms and inert solid particles, but also of the heavy metals. Actually, their concentration in a well-functioning anaerobic digester cannot be too high, in order to prevent inhibition phenomena.¹⁶¹ More specifically, the results obtained for the co-digestion of sunflower hulls and poultry manure show a limited concentration of heavy metals (Ni, Zn, Cu, Pb, Cr, Cd, and Hg),¹⁶² which is, however, high if compared to the bounds imposed for purified water. It is possible to state that this content is mainly due to the manure and the seed sludge.¹⁶² In fact, different metals are often used to enrich the feedstuff of the animals.¹⁶³

Generally, the most exploited solution for the removal of metals from wastewater and liquid digestates is represented by nanofiltration and reverse osmosis.^164,165,166 The separation power of this type of membrane can be enhanced with the use of surfactants, additives or water-soluble polymers capable of complexing the metal ions. More specifically, beyond a critical concentration, the surfactants aggregate in micelles, which capture the metal ions, making their transport through the membrane more difficult.^167,168 Similarly, polymer-enhanced filtrations use the capacity of polymer ligands to create complexes with a size bigger than the cut-off of the chosen membrane. A number of different polymers have been tried in the last few years. Chitosan,^169,170 polyethylenimine¹⁷¹ and polyacrylamide¹⁷² are just some of the possible choices.

It is not infrequent to also use electrodialysis, a membrane process based on the use of an electric field and an ion-exchange membrane. Originally employed for the treatment of seawater and the production of salt, electrodialysis gave promising results even for the recovery of chromium,^173,174 copper^175,176 and lead¹⁷⁷ ions.

As previously stated, water is not the only good resource that can be obtained from the treatment of digestate. The interest in the production of fertilizers, based on phosphorous and nitrogen, easy to store and capable of avoiding the unconditioned discharge of anaerobic digestate, is demonstrated by the intensive research in the field of struvite crystallization.^178,179,180 This salt is made by the combination of magnesium, nitrogen and phosphorous based ions, and it can be used in slow-release fertilizers.

The possible application of membranes for this scope is not free of difficulties. For example, using a reverse osmosis membrane, helped by a vibratory shear action, for a clarified digestate, a removal of 93% and 59% for nitrogen and phosphorous, respectively, was achieved.¹⁸¹ This not so high performance was improved with a second filtration stage that allowed reaching a final removal of 95% for total nitrogen and 69% for phosphorous. Furthermore, this type of arrangement gave proof of inconsistent behavior, with the risk of not respecting continuously the limitation imposed for water reuse.¹⁸¹

The results cited above are less encouraging than the ones found by another research group.¹⁸² More specifically, with a single stage vibratory enhanced filtration, it was possible to provide the 96.4% removal of total nitrogen and 98.1% removal of phosphorous from the permeate.¹⁸² This discrepancy is not surprising because of the big amount of variables that can greatly perturb the results of such an operation. One of these is temperature. In fact, it is capable of modifying the viscosity of the feed affecting the rate of filtration, and it also alters the dissociation equilibria. For example, after increasing the temperature, there is a bigger concentration of free ammonia than ammonium, and this produces a bigger loss of ammonia as gas, owing to the reduced solubility. The equilibria are also highly affected by pH and dilution. This is enough for supposing that different conditions can produce apparently unmatchable experimental results.

Actually, very high performances of nitrogen recovery were also obtained using submerged hollow fiber ultrafiltration membranes.¹⁸³ Lab-scale experiments with high ammonia starting concentrations gave great removal efficiencies, up to 98.5%.¹⁸³

The physical conditions of the digestate are also capable of affecting phosphorous and metal recovery. For example, prior acidification of the sludge can provide a great increase in phosphorous extraction, with the possibility of obtaining a higher concentration on the permeate side of a microfiltration system.¹⁸⁴ With this type of system, the nutrient recovery and concentration do not concern the retentate side, but only the permeate. This is why it is necessary to provide an extraction agent. Deionized water is enough for guaranteeing stronger nitrogen recovery, while it is not good for enhancing phosphorous recovery, which prefers lower pH conditions.¹⁸⁴ A similar trend can be observed even for metals, which are bound to solid particles and can be released more easily with the help of an acidic solution. In fact, hydrogen ions can replace metal ions on the solid surface, amplifying the extraction effect.¹⁸⁴

A smarter technique, originally used with MBRs, is based on forward osmosis.^185,186

As shown in Fig. 10, the action of such a device is based on the use of a drawn solution with a lower chemical potential and the wastewater to be treated with a higher chemical potential. This gradient induces a water flux from the wastewater side to the drawn solution side, therefore, there is concentration of the sludge and dilution of the drawn solution occurring at the same time. The outputs of the whole process are a concentrated sludge stream rich in nutrients and a purified water stream useful for agricultural purposes.

Fig. 10 A schematic representation of an OsMBR.¹⁸³

More specifically, the concentration of the drawn solution is based on reverse osmosis or a membrane distillation unit.^187,188

The results of this type of operation are promising, even in the field of digestate treatment.¹⁸⁹ The major advantage is the possibility of managing the fouling phenomena in a better way. In fact, the forward osmosis stage preserves the reverse osmosis membrane of the second stage from a rapid decline in flux. This possibility is even more appreciable considering that the enhancement of specific energy consumption, induced by the forward osmosis pretreatment, is modest.¹⁸⁹

Even the recovery of volatile fatty acids from a discharged digestate can be attempted with membrane technology.^190,191 Considering the low molecular weight of these compounds (butyric acid, propionic acid, acetic acid, valeric acid, etc.), it is normal to think of nanofiltration as a suitable route for separation. These compounds are highly attractive because they can be obtained from a residual of another process and they can find application in many fields of the petrochemical and fuel industries.¹⁹²

The experimental results obtained from starting with an anaerobic digestate (pretreated with a microfiltration membrane for solid removal) pushed thinking about a possible application of the concept to real anaerobic digesters. In fact, a retention up to 72.23% was achieved with the production of a retentate highly rich in volatile fatty acids. This promising result, obtained with a polyvinyl alcohol-aromatic cross-linked polyamide composite membrane, was gained at a pH value of 9. At this condition, the acids are almost completely present in dissociated form, thus it is plausible to think about a charge effect blocking permeation of acids.¹⁹³

In the specific case of a membrane coupled anaerobic digester, which is already equipped with a microfiltration or ultrafiltration module for the retention of the active biomass, it is reasonable to think of a successive nanofiltration or reverse osmosis module capable of giving almost clear water and a retentate rich in nutrients. In this specific case, is even easier because there is not the real necessity of thinking about pretreatment of the exhausted digestate, which already comes from a separation device.

9. Methods for biogas separation

The only component that makes biogas an interesting resource is methane. This is why a huge attempt has been made over the years to find a cheap purification system.

The removal of CO₂ is generally conducted using one of these technologies: water absorption, amine absorption, pressure swing adsorption, cryogenic technologies or membrane technologies.

Absorption is probably the most common solution. It requires a solvent capable of capturing the components of interest (CO₂ and H₂S in the case of biogas separation) and two columns. One is for the absorption action and one for the recovery of the solvent. A possible choice is presented by water, which has a certain capacity of capturing carbon dioxide, as demonstrated by the solubility data reported in Table 6.¹⁹⁴

Table 6 Carbon dioxide solubility in water expressed as weight of CO₂ in 100 weights of H₂O¹⁹²

Pressure [atm]	Temperature [°C]
	18	25	35	50	75	100
25	3.86	—	2.56	1.92	1.35	1.06
50	6.33	5.38	4.39	3.41	2.49	2.01
75	6.69	6.17	5.51	4.45	3.37	2.82
100	6.72	6.28	5.76	5.07	4.07	3.49

---

Generally, the water scrubbers are operated at room temperature and at a pressure higher than atmospheric (less than 15 bar^12,195,196). This choice is determined by the thermodynamic constraints reported above. In the first column, water falls from above and cleans biogas from the acid compounds. The second column instead promotes CO₂ desorption with just a pressure drop. Considering that a complete removal of H₂S from water cannot be provided in the regeneration column, it is impossible to recycle all of the water stream while avoiding accumulation in the recycling loop.¹²

This option can be enhanced using an amine solution. In particular, monoethanolamine and diethanolamine have been widely employed for this purpose. Triethanolamine did not gained the same success because of the high steric hindrance that determines a less effective action.

Despite the greater purity of the resulting methane, there are some problems related to the use of this type of process. First, there is the risk of contamination of the environment because of the possible loss of these chemical agents. Then, there is the problem of pressure. Generally, these scrubbers work at atmospheric pressure, and this produces the necessity of an intense compression of the final product to satisfy the gas grid standard.

Pressure swing adsorption has become an even more common research theme in the last few decades. It is based on the use of non-toxic microporous or mesoporous solids, and it can induce diminished capital costs.¹⁹⁷ Moreover, by choosing a proper solid adsorbent, it is possible to obtain the simultaneous removal of water, hydrogen sulfide, carbon dioxide and other impurities, creating an excellent purification of methane. Natural and synthetic zeolite are the most typical choices owing to their huge selectivity. In some cases separation is promoted by stronger interaction of carbon dioxide with the solid substrate. In other circumstances, it is not an interaction that promotes separation, but by a difference in the diffusion rate inside the microporous or mesoporous structure.¹⁹⁸ Most of the works found in the literature propose a moderate overpressure for the adsorption phase and atmospheric pressure for the desorption stage.¹⁹⁹ This indicates that the resulting methane stream needs a compression even in this case.

There is also the possibility of applying cryogenic techniques, which exploit a temperature drop at high pressure for promoting the condensation of carbon dioxide.¹⁹⁴

The boiling point temperatures reported in Table 7 are enough for understanding that a good separation can be obtained by reaching the condensation temperature of carbon dioxide.

Table 7 Boiling temperature of the compounds that can be present in biogas¹⁹²

Compounds	Boiling temperature [°C]
	p = 1 atm	p = 10 atm	p = 20 atm
Hydrogen	−252.5	−241.8	—
Nitrogen	−195.8	−169.8	−157.6
Carbon monoxide	−191.3	−161.0	−149.7
Oxygen	−183.1	−153.2	−140.0
Methane	−161.5	−124.8	−108.5
Carbon dioxide	−78.2	−39.5	−18.9
Hydrogen sulfide	−60.4	−0.4	25.5
Ammonia	−33.6	25.7	50.1
Water	100	180.5	213.1

---

In this way, there is a production of a liquid phase made of carbon dioxide, water, hydrogen sulfide and ammonia. Moreover, the gas phase contains methane and small traces of hydrogen and other compounds (nitrogen, oxygen, and carbon monoxide), which can be accidentally present in the biogas.

All the technologies analyzed so far have been developed in the past few years for gas upgrading in various fields of the petrochemical and chemical industries.

In the case of a biogas plant, which is finalized except for the feed of a power station, it is not important to reach such high methane purity, and there is a problem of fluctuating behavior with sometimes-modest production. In such a situation, membranes may be the most appropriate solution. First, there is the possibility of developing modular separation systems capable of satisfying different requests. They are also perfect for a case in which there is not so high gas production and big carbon dioxide formation.²⁰⁰ In fact, the size and the cost of a traditional scrubber are more influenced by the carbon dioxide percentage in the feeding than the gas flow rate.²⁰⁰

Actually, this is not a defect-free alternative and many doubts continue to push stronger research. For example, there is the problem of product purity that is sometimes less than the one obtained with the absorption system. Another topic that causes many debates is that of plasticization,²⁰¹ a swelling effect induced by CO₂ sorption and accompanied by a loss in selectivity. Summarizing, it is possible to state that a good membrane for CO₂ separation and biogas upgrading must have the following features:²⁰²

• high carbon dioxide permeability and selectivity;

• moderate cost;

• certain resistance to plasticization and aging effects;

• good resistance to any type of alteration produced by chemical agents or high temperature conditions.

10. Polymeric membranes: an overview of the commonly used materials and main features

Membrane purification of biogas is generally conducted using all the options that have been tested for the removal of carbon dioxide from various types of flue gases and natural gas.

The first attempts for gas deacidification were tried with cellulose membranes.²⁰³ Cellulose is a polymer that can be found even in living systems. It is made of a linear sequence of glucosyl residues connected with β1-4 linkages. Beyond these first approaches, the real beginning was represented by cellulose acetate membranes.^204,205,206 In a dense cellulose acetate layer, the transport can be described with a solution diffusion mechanism, typical of all dense polymeric membranes. According to this explanation, the gas molecules dissolve in the polymer, and then diffuse owing to a concentration gradient, and this is why permeability depends from solubility and diffusivity at the same time (eqn (1)).

P_i = D_iS_i (1)

Good membrane separation requires that the permeability of the target component is higher than the permeability of all the others. This indicates having a pronounced selectivity, defined as in eqn (2).

Generally, the diffusivity of the component i is described with an Arrhenius dependency on temperature (eqn (3)).

In eqn (3), the two parameters (the activation energy and the front factor) are related in some way with the characteristics of the penetrant and the polymer, and they can be connected to each other with the relationship expressed by eqn (4).

In eqn (4), a always has a value of 0.64 and b depends just on the state of the polymer (9.2 for polymers in rubbery state, 11.5 for glassy polymers).

Moreover, the activation energy depends from the size of the penetrants and the nature of the polymer; a valid description of this dependence is provided by eqn (5).

E_Ai = cd_i² − f (5)

d_i is a diameter representative of the penetrant (e.g. the kinetic diameter), and c and f change their value according to the nature of the base polymer.

Combining all the equations seen so far, it is possible to develop the following expression for selectivity.

Eqn (6), deeply analyzed elsewhere,²⁰⁷ is a clear demonstration of the strong connection between selectivity and permeability, which are involved in a negative correlation.

The dependency of the solubility on the operative conditions can be expressed using the so-called dual-mode sorption model (reported in eqn (7)).

It is like assuming that the sorption of the gas in the cellulose acetate membrane (and also in a generic polymeric layer) is made of two parts: the simple dissolution (S_D) and the Langmuir adsorption term (S_H) typical of a porous solid.

The weight of the two terms changes according to the distance from the glass transition temperature of the polymer. If the working conditions are below the glass transition temperature, the polymeric chains are rigid and essentially incapable of constant movements. In this case, the polymer is in the glassy state. Beyond the glass transition temperature, the thermal energy is enough for allowing movement of the chains, producing a rubbery state.

Generally, this rubbery state is assumed to be an equilibrium condition, while a glassy polymer can be considered a hypothetical liquid matrix with a dispersed solid inside.

The experimental results for cellulose acetate membranes are perfectly compatible with this type of description,²⁰⁶ and they also underline a certain sensibility to the problem of plasticization. In fact, the micro-voids, which are present in the glassy polymer structure, trap some CO₂ molecules, producing a swelling action of the membrane matrix that is damaged. This negative action is empirically demonstrated by the selectivity loss that appears with considerably increasing the CO₂ concentration in the feeding side.²⁰⁶

Despite these problems, which are common to other materials, cellulose acetate membranes gave good performances with carbon dioxide and hydrogen sulfide removal higher than 90% (ref. 208) and a CO₂/CH₄ selectivity higher than 20.²⁰²

Silicone rubber membranes represent another commercially available solution. Nowadays they are employed for the treatment of the gas that comes from oil production.²⁰⁹ More particularly, these membranes allow the movements of bigger hydrocarbons, blocking the passage of methane and ethane. The high permeability that they show is connected to the large free volume among the chains.^203,210 A very low glass transition temperature and a good flexibility also characterize all of this type of polymer.

In order to get a bigger mechanical resistance, it is useful to adopt a copolymerization process. In this way, the other monomers can be chosen to enhance the stability of the base polymer and provide a material capable of higher performance of industrial processes.

However, it is clear that by pushing rubber behavior, it is possible to promote flux enhancement, while a glassy behavior is responsible for major selectivity.

Another option is represented by polysulphone membranes, which are probably one of the most exploited solutions.^211,212,213 The interest for this type of material was stimulated by its mechanical and thermal stability, which also determines the possibility of an easy spinning action for hollow fiber production.

Even for polysulphone membranes, a dual-mode sorption model can be adopted.²¹⁴ Empirical analysis also shows that the selectivity is strongly linked with the thickness of the layer. A higher thickness can induce a higher selectivity, with an upper limit of 22.4 for CO₂/N₂.²¹⁴ Actually, it is not so easy to make any kind of prediction because of the big variability in structure that can be produced by changing even the additives for the preparation. More particularly, non-solvent additives increase the non-miscibility of the spinning impurities and accelerate their precipitation, producing a structure with finger-like cavities. By reducing these agents, instead, the final structure will be more similar to a sponge.²¹² This mechanism is not just typical of polysulphone membranes, but of all polymeric membranes in general.

Another interesting type of material is represented by polycarbonates, which are synthetized by the reaction of carbonic acid and bisphenol-A. Considering the difficulty of the polymeric chains to organize into a compact structure, it is not surprising that polycarbonate membranes are characterized by considerable CO₂ permeability (around 40 barrers), combined with high selectivity of between 15 and 25.²¹⁵

Polyimides, instead, are obtained from a condensation reaction of a polyamic acid. By acting on the nature of the precursor, it is always possible to produce particular feature in the resulting polymer, and thus in the final membrane. In this sense, 6-FDA is probably one of the most studied precursors because of its ability of inducing high permeability and good separation.²¹⁵ Polyimides are generally characterized by high glass transition temperatures and certain hydrophobicity. This implies a perfect match with an operation like biogas treatment. In fact, the presence of moisture cannot affect considerably the hydrophobic polyimide membrane.²¹⁶

In addition, the interest incited by polyarylates is remarkable. They are produced by a reaction of a diol with a chloride. The performances obtained are not among the best possible, even if a permeability of 85 barrers is sometimes reported in the literature.²¹⁵

Beyond any empirical and phenomenological approach to the problem of polymer design for membrane production, some principles are clear by now. One of these is the importance of having polar groups in the polymer structure. First of all, there is the possibility of CO₂ solubility enhancement, in fact, considering the series polybutadiene, polytetramethylene oxide, polyethylene oxide, the CO₂ solubility goes from 0.89 to 1.4 cm³ (STP) per cm³ atm.²¹⁷ The stronger interactions of polar groups with carbon dioxide are also responsible for delayed CO₂ diffusion in the polymeric bulk. For the same reason, there is also a reduction of CO₂ permeability.²¹⁷ The total effect on the selectivity, instead, is quite difficult to explain. The effect of enhanced solubility, accompanied by a not so high delay effect on CO₂ diffusivity, should guaranty good selectivity. The polar groups are also capable of inducing a packed glassy structure with inherent sieving power. Despite this, a depressing action on selectivity was discovered in the case of benzylic amine substituents on a polysulfone structure. More specifically, the CO₂/CH₄ permselectivity of a polysulfone–CH₂–NH₂ (51%) membrane, was found to be noticeably lower than that of a non-substituted polysulfone membrane.²¹⁸

One exception is represented by an ether oxygen, which combines the positive actions already mentioned of polar groups with a less depressing action on CO₂ permeability.²¹⁷ This explains why polyethylene oxide membranes have attracted such considerable attention in the field of gas separation.^219,220,221 Moreover, other trends of research are in the direction of the use of low molecular weight polyethylene oxide or polyethylene glycol. In fact, by choosing a small chain polymer it is possible to create thin supported films that are liquid at moderate temperatures and capable of high performances.^222,223

Another chance for overcoming the high crystallinity problem is using copolymers in which polyethylene oxide segments are combined with hard fragments that inhibit crystallization, avoiding a compact arrangement of the chains.^224,225

It is fundamental to remember the chance of crosslinking that is used not just for polyethylene oxide membranes, but in all cases in which there is conditioning of the original properties of the material for inducing anti-plasticization behavior or better durability. Crosslinking implies a connection of different polymeric chains with the use of chemical agents, UV light or physical methods.²⁰³

Actually, it is not so easy to control such an operation. It is fundamental to guaranty a certain extent of control but, at the same time, this should be not too great, in order to prevent excessive brittleness of the material. Even the result in terms of permselectivity can be unexpected. For example, with a modified polyimide membrane, by increasing the crosslinking extent, it is possible to have a big increase of He/N₂ selectivity and a considerable decrease of CO₂/N₂ selectivity.²²⁶

A special role in the conditioning techniques for membrane capacity improvement is represented by thermal rearrangement reactions. This type of treatment is a post fabrication alternative induced by high temperatures (350–450 °C).^227,228

Starting from a hydroxyl-containing polyimide membrane, it is possible to induce a rearrangement of the chains to a fully aromatic benzoxazole product, according to the mechanism reported in Fig. 11.²²⁷

Fig. 11 A proposed mechanism for thermal rearrangement of hydroxyl-containing polyimides.

This transformation is responsible for not only macroscopic effects, such as reduction of the density, but also of microscopic changes, mainly due to the formation of a product with a stiff structure. This stiffness is mainly induced by the high energy barriers to rotation of the two aromatic fragments.

A possible application for biogas upgrading is plausible because of the great increase of CO₂ permeability obtained. It is like developing a structure full of micro-voids appropriate for molecular separation, which drastically reduces the transport resistance. The effect on the permeability is obviously accompanied by a simultaneous effect on selectivity, which is strongly connected with the extent of the thermal treatment.²²⁷ In general, the performances are surprising, with CO₂/CH₄ selectivity and CO₂ permeability that go beyond the upper limit of a typical polymeric membrane (Fig. 13).²²⁹ It is appropriate to assume that the big permeability is the effect of huge porosity, while the selectivity is the result of the shape of the cavities, which show a narrow neck region.²²⁹ Moreover, no effects of plasticization are monitored, even at high CO₂ partial pressure.

Fig. 12 The transport mechanism through a facilitated transport membrane.²³³

Fig. 13 Correlation of selectivity and permeability for polymeric membranes.²³⁸

All these particular features make thermally rearranged membranes a promising alternative to traditional glassy polymeric membranes.

11. Carbon molecular sieve membranes

A carbon molecular sieve membrane is produced by the pyrolysis of a polymeric precursor already available in membrane form. Despite the big separation potential, these materials are not so frequent in industrial use.^202,230 The major drawbacks are excessive brittleness, which implies the inability to use high pressure; an excessive sensibility to water and oxidizing agents; and the possibility of aggressive pore blocking action.²³⁰

Obviously, a long series of positive aspects makes these membranes interesting for researchers. First, they are characterized by strong thermal stability. This is not a secondary aspect, especially for a possible application in the petrochemical industry.²³¹ Then, they are almost inert to many of the chemical agents typical of industrial processes, and they can be easily produced in the form of flat sheets or fibers (depending on the shape of the precursor).

Probably the most interesting aspect is the possibility of acting on the permeability and selectivity of the layer, changing the pyrolysis conditions.^232,233 In particular, starting with a hollow fiber polyimide membrane, it was possible to see that an inert gas is capable of accelerating the carbonization of the substrate, also inducing a higher permeability than the one obtained with vacuum pyrolysis.²³³ More recent results, show the small effect of inert gas flow rate, indicating that oxygen plays the most important role.²³² Therefore, by changing the oxygen exposure conditions, it is possible to manipulate the CO₂ separation power.

This is just a brief overview on the big potential effects of carbon molecular sieve membranes, which could be applied even for biogas purification. Actually, to our knowledge, there is not a systematic study about the potential of this type of membrane for biogas upgrading. Probably this gap stems from the fact that it is still a new concept that has been implemented only in lab-scale applications for general estimations.

12. Facilitated transport membranes

The research for new possibilities for overcoming the tradeoff of selectivity and permeability led to increasing attention towards the so-called facilitated transport membranes. They are made of a polymeric structure with an insert, which acts as a carrier. More specifically, this carrier is capable of reacting, in a reversible way, with carbon dioxide. Therefore, the regular transport is accompanied by the facilitated transport mechanism, which is allows for only the target compound.

It is clear that the reversible interaction with the carrier is just a tool for increasing the selectivity of the membrane without compromising the permeability.

In Fig. 12, a possible transport mechanism for a facilitated transport membrane is proposed.²³⁴

It is clear from the figure that there could also be a certain amount of movement of methane through the polymeric base of the membrane, but it essentially remains in the retentate.

Considering that for biogas application the main objective is removing CO₂ and H₂S, the carrier should be able to interact with an acidic compound. In this sense, it is interesting to consider all that works about facilitated transport membranes which incorporate basic carriers, such as amines and imines.

In particular, in the case of an amine carrier, the interaction with CO₂ can be described with a zwitterion mechanism.²³⁵ First, CO₂ reacts with the amine, giving the zwitterions. Then, there is a deprotonation that leads to the formation of a carbamate ion, which in the presence of water can form a bicarbonate ion. Carbonate and carbamate ions are the forms in which CO₂ can move through the layer, reaching the other side.

If the membrane is just made of a solution in a microporous structure, then there is risk of a washout of the carrier or its evaporation. This is why immobilization by electrostatic forces is more common than liquid membranes. In this way, the carrier cannot move freely and CO₂ passes from one functional group of the polymeric chain to another, as in Fig. 7.

The results obtained with these membranes are promising for biogas upgrading. For example, with a cross-linked polyvinyl alcohol membrane with both mobile and fixed amines, it is possible to reduce the CO₂ concentration from 17% to 100 ppm (with a flow rate of 60 mL min⁻¹).²³⁵ It is a drastic reduction, and above all, the high initial concentration is perfectly compatible with a scenario in which the raw mixture is biogas.

Another solution that was tested directly with biogas is represented by a polyvinylamine-polyvinyl alcohol blend membrane.²³⁶ The carrier is fixed in the polymeric structure, and CO₂ reversibly reacts with the amino groups. The results obtained using low pressure (2 atm) and room temperature are good, with a CO₂/CH₄ selectivity of 40 and a permeance of 0.55 m³ per m² per h per bar.

The possibility of operating at nearly ambient conditions is of primary importance for biogas treatment, because most of the anaerobic digesters operate in conditions of small overpressure, so developing a system with a compression in the middle is of primary importance. Actually, despite the good selectivity at low pressure, there is a problem with the surface area, which tends to increase rapidly, for a certain amount of biogas to treat, with the decreasing of pressure. This is why the optimization process indicates that the best scenario is the alternative based on two modules in series, equipped with polyvinylamine–polyvinylalcohol blend membranes and operating at a pressure of 20 and 10 bar, respectively.

Another competitive option that has been recently tested in conditions of high pressure is based on the use of a water insoluble amino-starch derivative for the preparation of a polyether sulfone membrane.²³⁷ The amino-starch content is of primary importance for the regulation of the overall performance. In fact, its increase can help CO₂ sorption, reducing the CH₄ sorption at the same time. The presence of the amine group suggests a facilitated transport, exactly as in the cases already mentioned. The membranes were synthetized in the form of dense or anisotropic layers, but in both the cases, they gave remarkable performances, with a selectivity of 58.23 and 61.2 for asymmetric and dense membranes, respectively, at a pressure of 27 bar and with a synthetic mixture similar to biogas (30% CO₂, 70% CH₄).²³⁷

While most of these options are not commercially available, another alternative is produced on a large scale. In particular, nanofiltration membranes for water treatment showed good performance in CO₂ separation from humidified gaseous streams.²³⁸ This good behavior is supposed to be the result of the presence of piperazine groups and free amines within the membrane structure.

All the discussion above is a clear demonstration of the big potential of facilitated transport membranes for a problem like biogas upgrading, which remains a complex issue.

13. Mixed matrix membranes

Mixed matrix membranes try to combine the positive features of polymers with the high selectivity of other materials. This is why attention has been focused mainly on materials that have highly inherent separation power, such as zeolites, carbon nanotubes, metal–organic frameworks or covalent organic frameworks.

In particular, it is well-known that polymeric membranes presents a negative connection between selectivity and permeability.²³⁹ Choosing a material with high permeability generally means obtaining not as high a selectivity.²³⁹

By coupling polymers with inorganic fillers, it is possible to surpass that limit, which is clearly visible in Fig. 8, and obtain better performances.

The improved behavior of a membrane after the addition of inorganic fillers can be estimated using the simplified model of Maxwell.²⁴⁰

In the previous equation P_c is the permeability of the composite membrane, P_m is the permeability of the polymeric matrix, P_f is the permeability of the filler material, φ_f is the volume fraction of the filler and n is the shape factor of the particles.

This equation comes from the analogous problem of current circulation, generated by a field, in a dielectric material made of different phases. It shows good agreement with experimental data in the case of low filler concentration. In the case of high concentration, instead, it is possible to make an estimation using the Bruggeman equation.²⁴⁰

Both equations predict an increase in the maximum selectivity with filler concentration. This explains why their presence is so useful.

Handling a high solid concentration can be quite difficult, especially in the preparation phase. In fact, there is always the risk of obtaining a non-homogenous layer with particle aggregates in the inner part of the layer. The solutions to the problems are essentially three:²⁴¹

• increasing the base polymer viscosity to reduce the sedimentation rate of solid particles;

• reducing drastically the preparation time of the composite layer;

• adopting nanoparticles that are not able to precipitate in a rapid way.

There is also the risk of accumulation of the particles at the surface. This happens because of the convective motion, which appear to operate at high temperature.²⁴²

Another aspect that makes preparation of mixed matrix membranes not so easy is related to the interactions between the fillers and the polymer matrix. If there is no perfect contact and some gaps remain there is risk of producing a drop in selectivity because obviously the gases will choose the easiest way to reach the other side of the membrane.²⁴³ There is also the possible risk of matrix rigidification around the filler particles, with consequent permeability decrease. Obviously, the non-homogeneity can be a characteristic of not just the matrix, but also of the filler, therefore it is possible to have a reduced permeability region within the surface of particles or, in the opposite way, there could a portion highly permeable owing to bigger porosity.²⁴³

In all these cases, the prediction power of the Maxwell model appears moderate and various corrections can be tried in order to improve its predictions.²⁴¹

In the field of CO₂ removal, zeolites are one of the first materials to be tested for the creation of mixed matrix membranes. In particular, by adopting hydrophilic zeolites with high Al concentration, it is possible to adsorb polar compounds such as H₂S, while hydrophobic zeolites are more appropriate for polar compounds like CO₂.²⁴⁴

The possible choices that have been tested for the creation of mixed matrix membrane are numerous, and they can be summarized as in Table 8.^13,245

Table 8 Main features of the commonly used zeolites for mixed matrix membranes^13,245

Type of zeolite	Pore size [nm]	Si/Al	Pore volume [cm^3 g^−1]
3A	0.29	1	0.197
4A	0.4	1	0.2
5A	0.4–0.5	1	0.28
13X	0.74	1.2	0.36
KY	0.74	2.6	0.47
Silicalite 1	0.53–0.56	>500	0.18
SSZ-13	0.38	11.8	—

---

Despite the great flexibility of zeolites, certain attention has been gained in the last few years from MOFs (metal–organic frameworks). These materials can have an incredibly high surface area (>6000 m² g⁻¹),²⁴⁶ and their features are also tunable with respect to the nature of the organic ligands employed.²⁴⁶ Essentially, they can be described as coordination polymers comprising metal ions with organic linkers. These MOFs can also provide a greater stability of the composite layer because of the major affinity with the polymeric bulk.

The first big result connected to the application of MOFs, that is interesting even for biogas application, is from the use of copper(II) biphenyl dicarboxylate-triethylenediamine.²⁴⁷ The increased mobility of CO₂ across the membrane was inspiring for research in this field, which now also comprises newly established choices, such as MOF-5 (ref. 248) and zeolitic imidazolate framework (ZIF).^249,250

This research trend parallels the use of carbon nanotubes, that is to say, graphitic sheets organized into tubes with diameters of a few nanometers. Obviously, the presence of channels inside the polymer matrix presents an improvement of the permeability, while the difference in adsorption energy between different compounds is responsible for the great selectivity.

Some of the experimental results collected in the last few years are summarized elsewhere²⁴⁷ and greatly show the big potential of these composite materials, which have all the elements for application to mixtures like biogas.

14. Membranes for biogas refining: still an open research field

Despite the big amount of theoretical and experimental study regarding the feasibility of membranes for carbon dioxide removal from various types of gases, there is still a certain gap in the field of biogas application.

In the literature, there are some works which attest the promising potential of membranes for this type of application.^251,252,253 Most of the studies are based on lab-scale devices, which were tested with synthetic mixtures.

Actually, there is a kind of mismatch between the evolution of the study about the materials to be used and the actual technological development of a fully organized process. What appears immediately clear is the lack of a specific study about the monitoring, control and maintenance of membrane modules for biogas upgrading. Probably this gap is connected to the relatively young origin of this type of separation in such an area of study. It is not a negligible aspect, because absorption technologies are based on well-established knowledge that allows a perfect understanding of any of their aspects. Therefore, absorption will continue to be the most exploited solution for large scale processes until a user-friendly configuration for membrane separation is reached.

The most desired advances in the field of biogas purification can be summarized as follows:

• major research of cheaper materials;

• the development of simple modules, which can be easily guided by new users;

• greater limitation of negative phenomena, such as plasticization;

• greater attention to the problems of monitoring and controlling large-scale purification plants;

• appreciable improvement of the trade-off between permeability and selectivity;

• development of consolidated protocols for the preparation of well performing polymeric blends and composite membranes.

15. Cost estimation and economical analysis

The importance of biogas in the energy scene of highly industrialized and still developing countries is well established nowadays. It has been estimated that in the EU-15, the use of renewable energies has grown from 300 TW h in 1995 to 400 TW h in 2005, with biogas as the most popular alternative derived from biomass use.²⁵⁴ Moreover, the provisions for the future are quite generous, with an estimated biogas production for the European Member States of approximately 756 000 × 10⁹ kJ in 2020.²⁵⁴

The big success of such an alternative is mainly due to the possible easy integration with an already existing production framework. The starting raw materials are essentially wastes coming from other activities. Therefore, by only employing them in the immediate neighborhood, it is possible to obtain an appreciable gain. For example, the biogas potential of the Long Island (New York) area has been estimated as 12% of the total energy produced, starting from traditional sources in the same area, and this total amount can be directly obtained from local wastewater treatment plants, solid waste management centers and agricultural activities.²⁵⁵

Despite these general considerations, there is still some difficulties in the estimation of the real economic potential of an anaerobic digestion plant. First of all, the operative and capital costs strongly depend on the final use of the biogas, which can be directly burned in a stationary plant for electricity generation or refined for gas grid injection. In the second case, separation devices induce an increase in the capital and operative cost, but they simultaneously generate a final product with greater value.

The choice of upgrading technique is connected to different aspects, as shown in Fig. 14.

Fig. 14 A brief and schematic resume of some possible aspects to be considered in the choice and development of a biogas separation system.

The energy requirements are essentially related to all the operations comprised in the refining process, which are fundamental for the transformation of the raw stream in a biomethane stream. A good review of energy requirements and total costs for different alternatives is already present in the literature,²⁵⁶ and it can be summarized as in Fig. 15.

Fig. 15 A graphical review of the energy requirements and costs for different alternatives of biogas upgrading.

Despite of the lower average value that can be found for amine scrubbing, there is a wide discrepancy in the results of different authors. This is quite common, especially in the case of membrane processes, because different modules, equipped with different membranes, can produce completely different results. Changing the type of membrane implies acting on the permeability, and therefore on the pressure necessary for the separation. Considering that the compression costs are one of the most significant elements in the development of a chemical process, it is not surprising to have such a wide range of values.

The specific membrane features are also relevant for the quantification of methane loss, which represents a waste of a useful product, and thus, an indirect cost. Acting on the type of membrane means changing the selectivity and the overall performance of the device. Therefore, in the case of a simple polymeric membrane, it is not surprising to find that methane concentration in the permeate is not very low. Adopting extremely selective materials implies a reduction of this undesirable effect but also an exponential increase of the capital costs, without considering that most of the newest highly selective membranes have only been tested in lab scale applications.

Moreover, a membrane separation for biogas upgrading can be organized according to different process schemes (single-stage separation, single-stage with recycling, multistage processes, etc.).¹² By changing the main structure, it is possible to act on the compression cost, the total area required and the CH₄ recovery for a certain desired value of CH₄ purity.

Even for the digestate treatment, the cost problem is not easy to handle, with membranes being the effective solution in some situations and not in others. A widely used correlation for the estimation of the capital costs of a membrane for wastewater treatment is given by eqn (10).²⁵⁷

The previous amount is always accompanied by other costs related to all the other components generally present in a full-scale process.

In the case of biogas refining, a big contribution was the compression costs. In the case of the digestate treatment, a certain amount is related to the use of pumps. Actually, a pump is cheaper if compared with a compressor. Therefore, its effect on the capital costs is generally negligible; the operative costs, are strictly dependent on the flow rate to be used.

Finally, there is the problem of maintenance and membrane recovery, the second of which is quite troublesome. Actually, it is not so easy to make some general considerations, because, as already shown, different methods can be implemented for the fouling reduction. Even the frequency of the recovery action is variable with the nature of the wastewater to be treated. All this variability is probably the reason why a systematic economical assessment for AnMBRs is still missing in our knowledge.

16. Conclusion and final remarks

This review paper provides deep insights into the field of membrane separation for biogas production processes.

All the considerations start from the material that has been produced in the last few years on MBRs, AnMBRs and membranes for CO₂ separation. Even if these themes are quite common in the scientific literature, a systematic study of their combination for the improvement of anaerobic digestion processes is still missing.

Developing a digester equipped with membranes for both liquid and gas treatment can solve the problems typical of the traditional processes.

First, there is the possibility of conditioning the hydrodynamic behavior of a high rate digester, acting on the critical aspects that limit a UASB, especially in the start-up phase. The improvement of the solid retention time also induces a further increase of the overall rate, allowing a higher OLR.

Membranes also provide an easy way to manage the digestate. It is possible to recover water for potable or agricultural purposes and produce a concentrated sludge rich in nutrients. The main objective is to induce a reduction of the volume required for the storage of the digestate, while producing a final product with a high concentration of phosphorous and nitrogen that is useful as a fertilizer.

The semiliquid digestate produced by anaerobic digesters can be treated with ultrafiltration or microfiltration membranes. A more intense purification can be realized with a reverse osmosis membrane combined with a forward osmosis membrane.

The main limitation is due to fouling action. Many alternatives can be found for the reduction of this natural phenomenon, which is due not only to the deposition of inert solids, but also to microbial activity.

The progressive reduction of the flux is the most worrying aspect, because in a full-scale process the rate should always be at an acceptable level. Therefore, a kind of compromise between separation power and rate should be reached.

The anaerobic digester can be equipped with a successive membrane module for gas separation. The possible choices are numerous: polymeric membranes, mixed matrix membranes, carbon molecular sieves, facilitated transport membranes, etc.

Even in this case there is a certain lack of knowledge, because most of the works found in the literature deal with lab-scale applications. Despite this, it is clear that membranes can represent a green alternative for biogas purification because of the possibility of avoiding all the chemical agents often employed for absorption.

This paper is just an offer of a complete view of all the possible applications of membranes in an application in which their use is not very well established. Despite this, a further and more extensive analysis should be realized in order to verify the economical and technical profitability of this membrane-based transformation for a full-scale process.

References

1. R. L. Droste and D. I. Masse, Comprehensive model of anaerobic digestion of swine manure slurry in a sequencing batch reactor, Water Res., 2000, 34, 3087–3106.
2. C. Côté, D. I. Massé and S. Quessy, Reduction of indicator and pathogenic microorganisms by psychrophilic anaerobic digestion in swine slurries, Bioresour. Technol., 2006, 97, 686–691,  DOI:10.1016/j.biortech.2005.03.024.
3. D. Bolzonella, C. Cavinato, F. Fatone, P. Pavan and F. Cecchi, High rate mesophilic, thermophilic, and temperature phased anaerobic digestion of waste activated sludge: a pilot scale study, Waste Manage., 2012, 32, 1196–1201,  DOI:10.1016/j.wasman.2012.01.006.
4. L. Appels, J. Baeyens, J. Degrève and R. Dewil, Principles and potential of the anaerobic digestion of waste-activated sludge, Prog. Energy Combust. Sci., 2008, 34, 755–781,  DOI:10.1016/j.pecs.2008.06.002.
5. M. A. de la Rubia, M. Perez, L. I. Romero and D. Sales, Effect of solids retention time (SRT) on pilot scale anaerobic thermophilic sludge digestion, Process Biochem., 2006, 41, 79–86,  DOI:10.1016/j.procbio.2005.03.073.
6. L. Metcalf, H. P. Eddy and G. Tchobanoglous, Wastewater engineering : treatment, disposal, and reuse, McGraw-Hill, New York, 2004.
7. B.-Q. Liao, J. T. Kraemer and D. M. Bagley, Anaerobic Membrane Bioreactors: Applications and Research Directions, Crit. Rev. Environ. Sci. Technol., 2006, 36, 489–530,  DOI:10.1080/10643380600678146.
8. M. Gander, B. Jefferson and S. Judd, Aerobic MBRs for domestic wastewater treatment: a review with cost considerations, Sep. Purif. Technol., 2000, 18, 119–130,  DOI:10.1016/s1383-5866(99)00056-8.
9. F. Tambone, P. Genevini, G. D'Imporzano and F. Adani, Assessing amendment properties of digestate by studying the organic matter composition and the degree of biological stability during the anaerobic digestion of the organic fraction of MSW, Bioresour. Technol., 2009, 100, 3140–3142,  DOI:10.1016/j.biortech.2009.02.012.
10. X. Flotats and A. Magrì, Manure treatment strategies: an overview, II Int. Symp. Agric. Agroindustrial Waste Manag. SIGERA, Foz Do Iguacu, PR, 2011, vol. 1, pp. 21–28.
11. C. M. Galanakis, Recovery of high added-value components from food wastes: Conventional, emerging technologies and commercialized applications, Trends Food Sci. Technol., 2012, 26, 68–87,  DOI:10.1016/j.tifs.2012.03.003.
12. M. Scholz, T. Melin and M. Wessling, Transforming biogas into biomethane using membrane technology, Renewable Sustainable Energy Rev., 2013, 17, 199–212,  DOI:10.1016/j.rser.2012.08.009.
13. S. Basu, A. L. Khan, A. Cano-Odena, C. Liu and I. F. J. Vankelecom, Membrane-based technologies for biogas separations, Chem. Soc. Rev., 2010, 39, 750–768.
14. D. Bolzonella, P. Pavan, S. Mace and F. Cecchi, Dry anaerobic digestion of differently sorted organic municipal solid waste: a full-scale experience, Water Sci. Technol., 2006, 53, 23–32.
15. D. Bolzonella, L. Innocenti, P. Pavan, P. Traverso and F. Cecchi, Semi-dry thermophilic anaerobic digestion of the organic fraction of municipal solid waste: focusing on the start-up phase, Bioresour. Technol., 2003, 86, 123–129.
16. K. Rajeshwari, M. Balakrishnan, A. Kansal, K. Lata and V. V. Kishore, State-of-the-art of anaerobic digestion technology for industrial wastewater treatment, Renewable Sustainable Energy Rev., 2000, 4, 135–156,  DOI:10.1016/s1364-0321(99)00014-3.
17. J. B. Van Lier, N. Mahmoud and G. Zeeman, Anaerobic wastewater treatment, Biol. Wastewater Treat, 2008.
18. L. Metcalf, H. P. Eddy and G. Tchobanoglous, Wastewater engineering: treatment, disposal, and reuse, McGraw-Hill, New York, 2004.
19. K.-H. Choo and C.-H. Lee, Hydrodynamic behavior of anaerobic biosolids during crossflow filtration in the membrane anaerobic bioreactor, Water Res., 1998, 32, 3387–3397,  DOI:10.1016/s0043-1354(98)00103-1.
20. C. Wen, X. Huang and Y. Qian, Domestic wastewater treatment using an anaerobic bioreactor coupled with membrane filtration, Process Biochem., 1999, 35, 335–340,  DOI:10.1016/s0032-9592(99)00076-x.
21. A. Pierkiel and J. Lanting, Membrane-coupled anaerobic digestion of municipal sewage sludge, Water Sci. Technol., 2005, 52, 253–258.
22. J. E. Schmidt and B. K. Ahring, Granular sludge formation in upflow anaerobic sludge blanket (UASB) reactors, Biotechnol. Bioeng., 1996, 49, 229–246.
23. G. González, F. Ramírez and O. Monroy, Development of biofilms in anaerobic reactors, Biotechnol. Lett., 1992, 14, 149–154,  DOI:10.1007/bf01026243.
24. N. Kosaric and R. Blaszczyk, Microbial aggregates in anaerobic wastewater treatment, in Bioprocesses Appl. Enzymol. SE – 4, Springer, Berlin, Heidelberg, 1990, pp. 27–62.  DOI:10.1007/bfb0000730.
25. Y. Liu, H.-L. Xu, S.-F. Yang and J.-H. Tay, Mechanisms and models for anaerobic granulation in upflow anaerobic sludge blanket reactor, Water Res., 2003, 37, 661–673,  DOI:10.1016/s0043-1354(02)00351-2.
26. M. M. Alves, A. J. Cavaleiro, E. C. Ferreira, A. L. Amaral, M. Mota, M. da Motta, et al., Characterisation by image analysis of anaerobic sludge under shock conditions, 2000.
27. P. Arne Alphenaar, A. Visser and G. Lettinga, The effect of liquid upward velocity and hydraulic retention time on granulation in UASB reactors treating wastewater with a high sulphate content, Bioresour. Technol, 1993, 43, 249–258,  DOI:10.1016/0960-8524(93)90038-d.
28. Y. Liu and J.-H. Tay, The essential role of hydrodynamic shear force in the formation of biofilm and granular sludge, Water Res., 2002, 36, 1653–1665,  DOI:10.1016/s0043-1354(01)00379-7.
29. Y.-Q. Liu, Y. Liu and J.-H. Tay, The effects of extracellular polymeric substances on the formation and stability of biogranules, Appl. Microbiol. Biotechnol., 2004, 65, 143–148,  DOI:10.1007/s00253-004-1657-8.
30. M. Veiga, M. Jain, W. Wu, R. Hollingsworth and J. Zeikus, Composition and Role of Extracellular Polymers in Methanogenic Granules, Appl. Envir. Microbiol., 1997, 63, 403–407.
31. M.-K. H. Winkler, J. P. Bassin, R. Kleerebezem, R. G. J. M. van der Lans and M. C. M. van Loosdrecht, Temperature and salt effects on settling velocity in granular sludge technology, Water Res., 2012, 46, 5445–5451,  DOI:10.1016/j.watres.2012.07.022.
32. K. S. Singh and T. Viraraghavan, Start-up and operation of UASB reactors at 20 °C for municipal wastewater treatment, J. Ferment. Bioeng., 1998, 85, 609–614,  DOI:10.1016/s0922-338x(98)80014-7.
33. W. Wang, W. Ma, H. Han, H. Li and M. Yuan, Thermophilic anaerobic digestion of Lurgi coal gasification wastewater in a UASB reactor, Bioresour. Technol., 2011, 102, 2441–2447,  DOI:10.1016/j.biortech.2010.10.140.
34. S. Kalyuzhnyi, L. E. de los Santos and J. R. Martinez, Anaerobic treatment of raw and preclarified potato-maize wastewaters in a USAB reactor, Bioresour. Technol., 1998, 66, 195–199,  DOI:10.1016/s0960-8524(98)00061-3.
35. V. D. Alves, F. Freitas, N. Costa, M. Carvalheira, R. Oliveira and M. P. Gonçalves, et al., Effect of temperature on the dynamic and steady-shear rheology of a new microbial extracellular polysaccharide produced from glycerol byproduct, Carbohydr. Polym., 2010, 79, 981–988,  DOI:10.1016/j.carbpol.2009.10.026.
36. J. Vogelaar, J. van Lier, B. Klapwijk, M. de Vries and G. Lettinga, Assessment of effluent turbidity in mesophilic and thermophilic activated sludge reactors – origin of effluent colloidal material, Appl. Microbiol. Biotechnol., 2002, 59, 105–111,  DOI:10.1007/s00253-002-0977-9.
37. B. Q. Liao, H. J. Lin, S. P. Langevin, W. J. Gao and G. G. Leppard, Effects of temperature and dissolved oxygen on sludge properties and their role in bioflocculation and settling, Water Res., 2011, 45, 509–520,  DOI:10.1016/j.watres.2010.09.010.
38. Y. Zhang, L. Yan, L. Chi, X. Long, Z. Mei and Z. Zhang, Startup and operation of anaerobic EGSB reactor treating palm oil mill effluent, J. Environ. Sci., 2008, 20, 658–663,  DOI:10.1016/s1001-0742(08)62109-9.
39. J. J. Kabara, R. Vrable and M. S. F. L. K. Jie, Antimicrobial lipids: natural and synthetic fatty acids and monoglycerides, Lipids, 1977, 12, 753–759.
40. P. L. McCarty and R. E. McKinney, Salt toxicity in anaerobic digestion, J.-Water Pollut. Control Fed., 1961, 399–415.
41. Y. Chen, J. J. Cheng and K. S. Creamer, Inhibition of anaerobic digestion process: a review, Bioresour. Technol., 2008, 99, 4044–4064.
42. L. N. Csonka, Physiological and genetic responses of bacteria to osmotic stress, Microbiol. Mol. Biol. Rev., 1989, 53, 121–147.
43. M. Roeßler and V. Müller, Osmoadaptation in bacteria and archaea: common principles and differences, Environ. Microbiol., 2001, 3, 743–754,  DOI:10.1046/j.1462-2920.2001.00252.x.
44. F. Omil, R. Méndez and J. M. Lema, Anaerobic treatment of saline wastewaters under high sulphide and ammonia content, Bioresour. Technol., 1995, 54, 269–278.
45. R. Mendez, J. M. Lema and M. Soto, Treatment of seafood-processing wastewaters in mesophilic and thermophilic anaerobic filters, Water Environ. Res., 1995, 67, 33–45.
46. P. Sridang, A. Pottier, C. Wisniewski and A. Grasmick, Performance and microbial surveying in submerged membrane bioreactor for seafood processing wastewater treatment, J. Membr. Sci., 2008, 317, 43–49,  DOI:10.1016/j.memsci.2007.11.011.
47. E. P. A. van Langerak, G. Gonzalez-Gil, A. van Aelst, J. B. van Lier, H. V. M. Hamelers and G. Lettinga, Effects of high calcium concentrations on the development of methanogenic sludge in upflow anaerobic sludge bed (UASB) reactors, Water Res., 1998, 32, 1255–1263,  DOI:10.1016/s0043-1354(97)00335-7.
48. D. Batstone, J. Landelli, A. Saunders, R. Webb, L. Blackall and J. Keller, The influence of calcium on granular sludge in a full-scale UASB treating paper mill wastewater, Water Sci. Technol., 2002, 45, 187–193.
49. H. E. Grethlein, Anaerobic digestion and membrane separation of domestic wastewater, J.-Water Pollut. Control Fed., 1978, 754–763.
50. P. M. Sutton, A. Li, R. R. Evans, S. R. Korchin, Dorr-Oliver's fixed-film, suspended-growth anaerobic systems for industrial wastewater treatment and energy recovery [Grain processing and dairy industries], in Proc. Ind. Wastes Conf, Purdue Univ., 1983.
51. A. Scheidegger, Biotechnology in Japan – towards the year 2000 … and beyond, Trends Biotechnol., 1991, 9, 183–190,  DOI:10.1016/0167-7799(91)90061-L.
52. W. R. Ross, J. P. Barnard, J. Le Roux and H. A. De Villiers, Application of ultrafiltration membranes for solids-liquids separation in anaerobic digestion systems: The ADUF process, Water SA, 1990, 16, 85–91.
53. K.-H. Choo and C.-H. Lee, Membrane fouling mechanisms in the membrane-coupled anaerobic bioreactor, Water Res., 1996, 30, 1771–1780,  DOI:10.1016/0043-1354(96)00053-x.
54. A. Beaubien, M. Bâty, F. Jeannot, E. Francoeur and J. Manem, Design and operation of anaerobic membrane bioreactors: development of a filtration testing strategy, J. Membr. Sci., 1996, 109, 173–184,  DOI:10.1016/0376-7388(95)00199-9.
55. O. Ince, G. K. Anderson and B. Kasapgil, Composition of the microbial population in a membrane anaerobic reactor system during start-up, Water Res., 1997, 31, 1–10,  DOI:10.1016/s0043-1354(96)00206-0.
56. T. Ueda and K. Hata, Domestic wastewater treatment by a submerged membrane bioreactor with gravitational filtration, Water Res., 1999, 33, 2888–2892,  DOI:10.1016/s0043-1354(98)00518-1.
57. L. Chen, Y. Gu, C. Cao, J. Zhang, J.-W. Ng and C. Tang, Performance of a submerged anaerobic membrane bioreactor with forward osmosis membrane for low-strength wastewater treatment, Water Res., 2014, 50, 114–123,  DOI:10.1016/j.watres.2013.12.009.
58. A. Y. Hu and D. C. Stuckey, Treatment of dilute wastewaters using a novel submerged anaerobic membrane bioreactor, J. Environ. Eng., 2006, 132, 190–198.
59. W. Guo, H.-H. Ngo and J. Li, A mini-review on membrane fouling, Bioresour. Technol., 2012, 122, 27–34,  DOI:10.1016/j.biortech.2012.04.089.
60. A. Drews, Membrane fouling in membrane bioreactors—Characterisation, contradictions, cause and cures, J. Membr. Sci., 2010, 363, 1–28,  DOI:10.1016/j.memsci.2010.06.046.
61. M. Kanai, V. Ferre, S. Wakahara, T. Yamamoto and M. Moro, A novel combination of methane fermentation and MBR — Kubota Submerged Anaerobic Membrane Bioreactor process, Desalination, 2010, 250, 964–967,  DOI:10.1016/j.desal.2009.09.082.
62. S. H. Yoon, I. J. Kang and C. H. Lee, Fouling of inorganic membrane and flux enhancement in membrane-coupled anaerobic bioreactor, Sep. Sci. Technol., 1999, 34, 709–724.
63. A. Sainbayar, J. S. Kim, W. J. Jung, Y. S. Lee and C. H. Lee, Application of Surface Modified Polypropylene Membranes to an Anaerobic Membrane Bioreactor, Environ. Technol., 2001, 22, 1035–1042,  DOI:10.1080/09593332208618212.
64. I.-J. Kang, S.-H. Yoon and C.-H. Lee, Comparison of the filtration characteristics of organic and inorganic membranes in a membrane-coupled anaerobic bioreactor, Water Res., 2002, 36, 1803–1813,  DOI:10.1016/s0043-1354(01)00388-8.
65. W. Fuchs, H. Binder, G. Mavrias and R. Braun, Anaerobic treatment of wastewater with high organic content using a stirred tank reactor coupled with a membrane filtration unit, Water Res., 2003, 37, 902–908,  DOI:10.1016/s0043-1354(02)00246-4.
66. Y. He, P. Xu, C. Li and B. Zhang, High-concentration food wastewater treatment by an anaerobic membrane bioreactor, Water Res., 2005, 39, 4110–4118.
67. A. G. Liew Abdullah, A. Idris, F. R. Ahmadun, B. S. Baharin, F. Emby and M. J. Megat Mohd Noor, et al., A kinetic study of a membrane anaerobic reactor (MAR) for treatment of sewage sludge, Desalination, 2005, 183, 439–445,  DOI:10.1016/j.desal.2005.03.044.
68. S. F. Aquino, A. Y. Hu, A. Akram and D. C. Stuckey, Characterization of dissolved compounds in submerged anaerobic membrane bioreactors (SAMBRs), J. Chem. Technol. Biotechnol., 2006, 81, 1894–1904,  DOI:10.1002/jctb.1622.
69. A. Y. Hu and D. C. Stuckey, Activated carbon addition to a submerged anaerobic membrane bioreactor: effect on performance, transmembrane pressure, and flux, J. Environ. Eng., 2007, 133, 73–80.
70. D. Jeison and J. B. van Lier, Cake layer formation in anaerobic submerged membrane bioreactors (AnSMBR) for wastewater treatment, J. Membr. Sci., 2006, 284, 227–236,  DOI:10.1016/j.memsci.2006.07.035.
71. D. Jeison and J. B. van Lier, On-line cake-layer management by trans-membrane pressure steady state assessment in Anaerobic Membrane Bioreactors for wastewater treatment, Biochem. Eng. J., 2006, 29, 204–209,  DOI:10.1016/j.bej.2005.11.017.
72. D. Jeison and J. B. Van Lier, Bio-layer management in anaerobic membrane bioreactors for wastewater treatment, Water Sci. Technol., 2006, 54, 81–86.
73. D. Jeison and J. Vanlier, Cake formation and consolidation: Main factors governing the applicable flux in anaerobic submerged membrane bioreactors (AnSMBR) treating acidified wastewaters, Sep. Purif. Technol., 2007, 56, 71–78,  DOI:10.1016/j.seppur.2007.01.022.
74. D. Jeison and J. B. van Lier, Feasibility of thermophilic anaerobic submerged membrane bioreactors (AnSMBR) for wastewater treatment, Desalination, 2008, 231, 227–235,  DOI:10.1016/j.desal.2007.11.048.
75. A. Saddoud, M. Ellouze, A. Dhouib and S. Sayadi, A Comparative Study on the Anaerobic Membrane Bioreactor Performance During the Treatment of Domestic Wastewaters of Various Origins, Environ. Technol., 2006, 27, 991–999,  DOI:10.1080/09593332708618712.
76. A. Saddoud, I. Hassaïri and S. Sayadi, Anaerobic membrane reactor with phase separation for the treatment of cheese whey, Bioresour. Technol., 2007, 98, 2102–2108,  DOI:10.1016/j.biortech.2006.08.013.
77. S. I. Padmasiri, J. Zhang, M. Fitch, B. Norddahl, E. Morgenroth and L. Raskin, Methanogenic population dynamics and performance of an anaerobic membrane bioreactor (AnMBR) treating swine manure under high shear conditions, Water Res., 2007, 41, 134–144,  DOI:10.1016/j.watres.2006.09.021.
78. A. Saddoud, M. Ellouze, A. Dhouib and S. Sayadi, Anaerobic membrane bioreactor treatment of domestic wastewater in Tunisia, Desalination, 2007, 207, 205–215,  DOI:10.1016/j.desal.2006.08.005.
79. J. Bohdziewicz, E. Neczaj and A. Kwarciak, Landfill leachate treatment by means of anaerobic membrane bioreactor, Desalination, 2008, 221, 559–565,  DOI:10.1016/j.desal.2007.01.117.
80. A. Akram and D. C. Stuckey, Flux and performance improvement in a submerged anaerobic membrane bioreactor (SAMBR) using powdered activated carbon (PAC), Process Biochem., 2008, 43, 93–102,  DOI:10.1016/j.procbio.2007.10.020.
81. I. Vyrides and D. C. Stuckey, Saline sewage treatment using a submerged anaerobic membrane reactor (SAMBR): effects of activated carbon addition and biogas-sparging time, Water Res., 2009, 43, 933–942,  DOI:10.1016/j.watres.2008.11.054.
82. H. J. Lin, K. Xie, B. Mahendran, D. M. Bagley, K. T. Leung and S. N. Liss, et al., Sludge properties and their effects on membrane fouling in submerged anaerobic membrane bioreactors (SAnMBRs), Water Res., 2009, 43, 3827–3837,  DOI:10.1016/j.watres.2009.05.025.
83. H. J. Lin, K. Xie, B. Mahendran, D. M. Bagley, K. T. Leung and S. N. Liss, et al., Factors affecting sludge cake formation in a submerged anaerobic membrane bioreactor, J. Membr. Sci., 2010, 361, 126–134,  DOI:10.1016/j.memsci.2010.05.062.
84. D. Jeison, C. M. Plugge, A. Pereira and J. B. van Lier, Effects of the acidogenic biomass on the performance of an anaerobic membrane bioreactor for wastewater treatment, Bioresour. Technol., 2009, 100, 1951–1956,  DOI:10.1016/j.biortech.2008.10.028.
85. D.-W. Gao, T. Zhang, C.-Y. Y. Tang, W.-M. Wu, C.-Y. Wong and Y. H. Lee, et al., Membrane fouling in an anaerobic membrane bioreactor: differences in relative abundance of bacterial species in the membrane foulant layer and in suspension, J. Membr. Sci., 2010, 364, 331–338,  DOI:10.1016/j.memsci.2010.08.031.
86. J. Ho and S. Sung, Methanogenic activities in anaerobic membrane bioreactors (AnMBR) treating synthetic municipal wastewater, Bioresour. Technol., 2010, 101, 2191–2196,  DOI:10.1016/j.biortech.2009.11.042.
87. S. Baek, K. Pagilla and H.-J. Kim, Lab-scale study of an anaerobic membrane bioreactor (AnMBR) for dilute municipal wastewater treatment, Biotechnol. Bioprocess Eng., 2010, 15, 704–708,  DOI:10.1007/s12257-009-0194-9.
88. A. Spagni, S. Casu, N. A. Crispino, R. Farina and D. Mattioli, Filterability in a submerged anaerobic membrane bioreactor, Desalination, 2010, 250, 787–792,  DOI:10.1016/j.desal.2008.11.042.
89. H. Lin, J. Chen, F. Wang, L. Ding and H. Hong, Feasibility evaluation of submerged anaerobic membrane bioreactor for municipal secondary wastewater treatment, Desalination, 2011, 280, 120–126.
90. D. Martinez-Sosa, B. Helmreich, T. Netter, S. Paris, F. Bischof and H. Horn, Anaerobic submerged membrane bioreactor (AnSMBR) for municipal wastewater treatment under mesophilic and psychrophilic temperature conditions, Bioresour. Technol., 2011, 102, 10377–10385,  DOI:10.1016/j.biortech.2011.09.012.
91. C. Zamalloa, J. Vrieze, N. Boon and W. Verstraete, Anaerobic digestibility of marine microalgae Phaeodactylum tricornutum in a lab-scale anaerobic membrane bioreactor, Appl. Microbiol. Biotechnol., 2012, 93, 859–869,  DOI:10.1007/s00253-011-3624-5.
92. R. K. Dereli, D. R. Urban, B. Heffernan, J. A. Jordan, J. Ewing and G. T. Rosenberger, et al., Performance evaluation of a pilot-scale anaerobic membrane bioreactor (AnMBR) treating ethanol thin stillage, Environ. Technol., 2012, 33, 1511–1516,  DOI:10.1080/09593330.2012.665491.
93. S. Casu, N. A. Crispino, R. Farina, D. Mattioli, M. Ferraris and A. Spagni, Wastewater treatment in a submerged anaerobic membrane bioreactor, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng., 2012, 47, 204–209,  DOI:10.1080/10934529.2012.640562.
94. A. L. Smith, S. J. Skerlos and L. Raskin, Psychrophilic anaerobic membrane bioreactor treatment of domestic wastewater, Water Res., 2013, 47, 1655–1665,  DOI:10.1016/j.watres.2012.12.028.
95. W. Qiao, K. Takayanagi, Q. Niu, M. Shofie and Y. Y. Li, Long-term stability of thermophilic co-digestion submerged anaerobic membrane reactor encountering high organic loading rate, persistent propionate and detectable hydrogen in biogas, Bioresour. Technol., 2013, 149, 92–102,  DOI:10.1016/j.biortech.2013.09.023.
96. H. Lin, W. Peng, M. Zhang, J. Chen, H. Hong and Y. Zhang, A review on anaerobic membrane bioreactors: applications, membrane fouling and future perspectives, Desalination, 2013, 314, 169–188.
97. S.-H. Roh, Y. N. Chun, J.-W. Nah, H.-J. Shin and S.-I. Kim, WastewaterTreatment by Anaerobic Digestion Coupled with Membrane Processing, J. Ind. Eng. Chem., 2006, 12, 489–493.
98. B. Digman and D.-S. Kim, Review: Alternative energy from food processing wastes, Environ. Prog., 2008, 27, 524–537,  DOI:10.1002/ep.10312.
99. A. Torres, A. Hemmelmann, C. Vergara and D. Jeison, Application of two-phase slug-flow regime to control flux reduction on anaerobic membrane bioreactors treating wastewaters with high suspended solids concentration, Sep. Purif. Technol., 2011, 79, 20–25.
100. G. K. Anderson, B. Kasapgil and O. Ince, Microbial Kinetics of a Membrane Anaerobic Reactor System, Environ. Technol., 1996, 17, 449–464,  DOI:10.1080/09593331708616407.
101. J.-R. S. Ventura, J. Lee and D. Jahng, A comparative study on the alternating mesophilic and thermophilic two-stage anaerobic digestion of food waste, J. Environ. Sci., 2014, 26, 1274–1283,  DOI:10.1016/s1001-0742(13)60599-9.
102. M. A. Latif, A. Ahmad, R. Ghufran and Z. A. Wahid, Effect of temperature and organic loading rate on upflow anaerobic sludge blanket reactor and CH₄ production by treating liquidized food waste, Environ. Prog. Sustainable Energy, 2012, 31, 114–121.
103. A. B. dos Santos, F. J. Cervantes and J. B. van Lier, Review paper on current technologies for decolourisation of textile wastewaters: Perspectives for anaerobic biotechnology, Bioresour. Technol., 2007, 98, 2369–2385,  DOI:10.1016/j.biortech.2006.11.013.
104. A. Spagni, S. Casu and S. Grilli, Decolourisation of textile wastewater in a submerged anaerobic membrane bioreactor, Bioresour. Technol., 2012, 117, 180–185,  DOI:10.1016/j.biortech.2012.04.074.
105. D. Pokhrel and T. Viraraghavan, Treatment of pulp and paper mill wastewater—a review, Sci. Total Environ., 2004, 333, 37–58,  DOI:10.1016/j.scitotenv.2004.05.017.
106. M. Badshah, W. Parawira and B. Mattiasson, Anaerobic treatment of methanol condensate from pulp mill compared with anaerobic treatment of methanol using mesophilic UASB reactors, Bioresour. Technol., 2012, 125, 318–327,  DOI:10.1016/j.biortech.2012.08.109.
107. K. Xie, H. J. Lin, B. Mahendran, D. M. Bagley, K. T. Leung and S. N. Liss, et al., Performance and fouling characteristics of a submerged anaerobic membrane bioreactor for kraft evaporator condensate treatment, Environ. Technol., 2010, 31, 511–521,  DOI:10.1080/09593330903527898.
108. R. Grover, S. S. Marwaha and J. F. Kennedy, Studies on the use of an anaerobic baffled reactor for the continuous anaerobic digestion of pulp and paper mill black liquors, Process Biochem., 1999, 34, 653–657,  DOI:10.1016/s0032-9592(98)00138-1.
109. G. Thompson, J. Swain, M. Kay and C. F. Forster, The treatment of pulp and paper mill effluent: a review, Bioresour. Technol., 2001, 77, 275–286,  DOI:10.1016/s0960-8524(00)00060-2.
110. H. Ozgun, R. K. Dereli, M. E. Ersahin, C. Kinaci, H. Spanjers and J. B. van Lier, A review of anaerobic membrane bioreactors for municipal wastewater treatment: Integration options, limitations and expectations, Sep. Purif. Technol., 2013, 118, 89–104,  DOI:10.1016/j.seppur.2013.06.036.
111. T. Heberer, Occurrence, fate, and removal of pharmaceutical residues in the aquatic environment: a review of recent research data, Toxicol. Lett., 2002, 131, 5–17,  DOI:10.1016/s0378-4274(02)00041-3.
112. D. I. Massé, D. Lu, L. Masse and R. L. Droste, Effect of antibiotics on psychrophilic anaerobic digestion of swine manure slurry in sequencing batch reactors, Bioresour. Technol., 2000, 75, 205–211,  DOI:10.1016/s0960-8524(00)00046-8.
113. K. B. Chipasa, Accumulation and fate of selected heavy metals in a biological wastewater treatment system, Waste Manage., 2003, 23, 135–143,  DOI:10.1016/s0956-053x(02)00065-x.
114. R. W. Baker, Membrane technology, Wiley Online Library, 2000.
115. G. Belfort, R. H. Davis and A. L. Zydney, The behavior of suspensions and macromolecular solutions in crossflow microfiltration, J. Membr. Sci., 1994, 96, 1–58.
116. L. Song, Flux decline in crossflow microfiltration and ultrafiltration: mechanisms and modeling of membrane fouling, J. Membr. Sci., 1998, 139, 183–200,  DOI:10.1016/s0376-7388(97)00263-9.
117. H. Lin, B.-Q. Liao, J. Chen, W. Gao, L. Wang and F. Wang, et al., New insights into membrane fouling in a submerged anaerobic membrane bioreactor based on characterization of cake sludge and bulk sludge, Bioresour. Technol., 2011, 102, 2373–2379,  DOI:10.1016/j.biortech.2010.10.103.
118. W. Lee, Sludge characteristics and their contribution to microfiltration in submerged membrane bioreactors, J. Membr. Sci., 2003, 216, 217–227,  DOI:10.1016/s0376-7388(03)00073-5.
119. K. Zhang, H. Choi, D. D. Dionysiou, G. A. Sorial and D. B. Oerther, Identifying pioneer bacterial species responsible for biofouling membrane bioreactors, Environ. Microbiol., 2006, 8, 433–440,  DOI:10.1111/j.1462-2920.2005.00909.x.
120. X. Huang, R. Liu and Y. Qian, Behaviour of soluble microbial products in a membrane bioreactor, Process Biochem., 2000, 36, 401–406,  DOI:10.1016/s0032-9592(00)00206-5.
121. S. Liang, C. Liu and L. Song, Soluble microbial products in membrane bioreactor operation: behaviors, characteristics, and fouling potential, Water Res., 2007, 41, 95–101,  DOI:10.1016/j.watres.2006.10.008.
122. X.-M. Wang and X.-Y. Li, Accumulation of biopolymer clusters in a submerged membrane bioreactor and its effect on membrane fouling, Water Res., 2008, 42, 855–862,  DOI:10.1016/j.watres.2007.08.031.
123. I. Martin-Garcia, V. Monsalvo, M. Pidou, P. Le-Clech, S. J. Judd and E. J. McAdam, et al., Impact of membrane configuration on fouling in anaerobic membrane bioreactors, J. Membr. Sci., 2011, 382, 41–49,  DOI:10.1016/j.memsci.2011.07.042.
124. Y. Liu, H. Liu, L. Cui and K. Zhang, The ratio of food-to-microorganism (F/M) on membrane fouling of anaerobic membrane bioreactors treating low-strength wastewater, Desalination, 2012, 297, 97–103,  DOI:10.1016/j.desal.2012.04.026.
125. C. Wisniewski and A. Grasmick, Floc size distribution in a membrane bioreactor and consequences for membrane fouling, Colloids Surf., A, 1998, 138, 403–411,  DOI:10.1016/s0927-7757(96)03898-8.
126. E. Gabrus and D. Szaniawska, Application of backflushing for fouling reduction during microfiltration of yeast suspensions, Desalination, 2009, 240, 46–53,  DOI:10.1016/j.desal.2008.02.025.
127. V. T. Kuberkar and R. H. Davis, Microfiltration of protein–cell mixtures with crossflushing or backflushing, J. Membr. Sci., 2001, 183, 1–14,  DOI:10.1016/s0376-7388(00)00577-9.
128. H. Ma, Factors affecting membrane fouling reduction by surface modification and backpulsing, J. Membr. Sci., 2001, 189, 255–270,  DOI:10.1016/s0376-7388(01)00422-7.
129. J. Yong Park, S. Jin Choi and B. Reum Park, Effect of N₂-back-flushing in multichannels ceramic microfiltration system for paper wastewater treatment, Desalination, 2007, 202, 207–214,  DOI:10.1016/j.desal.2005.12.056.
130. Z. Cui, S. Chang and A. Fane, The use of gas bubbling to enhance membrane processes, J. Membr. Sci., 2003, 221, 1–35,  DOI:10.1016/s0376-7388(03)00246-1.
131. G. Hansson, Effects of carbon dioxide and methane on methanogenesis, Eur. J. Appl. Microbiol. Biotechnol., 1979, 6, 351–359,  DOI:10.1007/bf00499165.
132. A. Menniti, S. Kang, M. Elimelech and E. Morgenroth, Influence of shear on the production of extracellular polymeric substances in membrane bioreactors, Water Res., 2009, 43, 4305–4315,  DOI:10.1016/j.watres.2009.06.052.
133. H. Park, K.-H. Choo and C.-H. Lee, Flux Enhancement with Powdered Activated Carbon Addition in the Membrane Anaerobic Bioreactor, Sep. Sci. Technol., 1999, 34, 2781–2792,  DOI:10.1081/ss-100100804.
134. C. Aun Ng, D. Sun and A. G. Fane, Operation of Membrane Bioreactor with Powdered Activated Carbon Addition, Sep. Sci. Technol., 2006, 41, 1447–1466,  DOI:10.1080/01496390600634632.
135. Y.-Z. Li, Y.-L. He, Y.-H. Liu, S.-C. Yang and G.-J. Zhang, Comparison of the filtration characteristics between biological powdered activated carbon sludge and activated sludge in submerged membrane bioreactors, Desalination, 2005, 174, 305–314,  DOI:10.1016/j.desal.2004.10.005.
136. K. H. Choo, I. J. Kang, S. H. Yoon, H. Park, J. H. Kim and S. Adiya, et al., Approaches to membrane fouling control in anaerobic membrane bioreactors, Water Sci. Technol., 2000, 41, 363–371.
137. J. Y. Kim, H. K. Lee and S. C. Kim, Surface structure and phase separation mechanism of polysulfone membranes by atomic force microscopy, J. Membr. Sci., 1999, 163, 159–166,  DOI:10.1016/s0376-7388(99)00164-7.
138. T. Knoell, Biofouling potentials of microporous polysulfone membranes containing a sulfonated polyether-ethersulfone/polyether sulfone block copolymer: correlation of membrane surface properties with bacterial attachment, J. Membr. Sci., 1999, 157, 117–138,  DOI:10.1016/s0376-7388(98)00365-2.
139. M. Elimelech and A. E. Childress, Role of membrane surface morphology in colloidal fouling of cellulose acetate and composite aromatic polyamide reverse osmosis membranes, J. Membr. Sci., 1997, 127, 101–109,  DOI:10.1016/s0376-7388(96)00351-1.
140. S. Hu, X. Ren, M. Bachman, C. E. Sims, G. P. Li and N. Allbritton, Surface Modification of Poly(dimethylsiloxane) Microfluidic Devices by Ultraviolet Polymer Grafting, Anal. Chem., 2002, 74, 4117–4123,  DOI:10.1021/ac025700w.
141. J. Pieracci, D. W. Wood, J. V Crivello and G. Belfort, UV-Assisted Graft Polymerization of N-vinyl-2-pyrrolidinone onto Poly(ether sulfone) Ultrafiltration Membranes:  Comparison of Dip versus Immersion Modification Techniques, Chem. Mater., 2000, 12, 2123–2133,  DOI:10.1021/cm9907864.
142. M. Taniguchi, J. E. Kilduff and G. Belfort, Low fouling synthetic membranes by UV-assisted graft polymerization: monomer selection to mitigate fouling by natural organic matter, J. Membr. Sci., 2003, 222, 59–70,  DOI:10.1016/s0376-7388(03)00192-3.
143. M. Suzuki, A. Kishida, H. Iwata and Y. Ikada, Graft copolymerization of acrylamide onto a polyethylene surface pretreated with glow discharge, Macromolecules, 1986, 19, 1804–1808,  DOI:10.1021/ma00161a005.
144. D. S. Wavhal and E. R. Fisher, Hydrophilic modification of polyether sulfone membranes by low temperature plasma-induced graft polymerization, J. Membr. Sci., 2002, 209, 255–269,  DOI:10.1016/s0376-7388(02)00352-6.
145. Y. Wang, J.-H. Kim, K.-H. Choo, Y.-S. Lee and C.-H. Lee, Hydrophilic modification of polypropylene microfiltration membranes by ozone-induced graft polymerization, J. Membr. Sci., 2000, 169, 269–276,  DOI:10.1016/s0376-7388(9910.1016/s0376-7388(99)00345-2.
146. H. Ma, C. N. Bowman and R. H. Davis, Membrane fouling reduction by backpulsing and surface modification, J. Membr. Sci., 2000, 173, 191–200,  DOI:10.1016/s0376-7388(00)00360-4.
147. S.-R. Kim, H.-S. Oh, S.-J. Jo, K.-M. Yeon, C.-H. Lee and D.-J. Lim, et al., Biofouling control with bead-entrapped quorum quenching bacteria in membrane bioreactors: Physical and biological effects, Environ. Sci. Technol., 2013, 47, 836–842.
148. B. Lee, K.-M. Yeon, J. Shim, S.-R. Kim, C.-H. Lee and J. Lee, et al., Effective antifouling using quorum-quenching acylase stabilized in magnetically-separable mesoporous silica, Biomacromolecules, 2014, 15, 1153–1159.
149. K.-M. Yeon, W.-S. Cheong, H.-S. Oh, W.-N. Lee, B.-K. Hwang and C.-H. Lee, et al., Quorum Sensing: A New Biofouling Control Paradigm in a Membrane Bioreactor for Advanced Wastewater Treatment, Environ. Sci. Technol., 2008, 43, 380–385,  DOI:10.1021/es8019275.
150. J.-H. Kim, D.-C. Choi, K.-M. Yeon, S.-R. Kim and C.-H. Lee, Enzyme-immobilized nanofiltration membrane to mitigate biofouling based on quorum quenching, Environ. Sci. Technol., 2011, 45, 1601–1607.
151. H.-S. Oh, K.-M. Yeon, C.-S. Yang, S.-R. Kim, C.-H. Lee and S. Y. Park, et al., Control of Membrane Biofouling in MBR for Wastewater Treatment by Quorum Quenching Bacteria Encapsulated in Microporous Membrane, Environ. Sci. Technol., 2012, 46, 4877–4884,  DOI:10.1021/es204312u.
152. Y.-J. Won, J. Lee, D.-C. Choi, H. R. Chae, I. Kim and C.-H. Lee, et al., Preparation and application of patterned membranes for wastewater treatment, Environ. Sci. Technol., 2012, 46, 11021–11027,  DOI:10.1021/es3020309.
153. P. Z. Çulfaz, S. Buetehorn, L. Utiu, M. Kueppers, B. Bluemich and T. Melin, et al., Fouling behavior of microstructured hollow fiber membranes in dead-end filtrations: critical flux determination and NMR imaging of particle deposition, Langmuir, 2011, 27, 1643–1652,  DOI:10.1021/la1037734.
154. Y.-J. Won, D.-C. Choi, J. H. Jang, J.-W. Lee, H. R. Chae and I. Kim, et al., Factors affecting pattern fidelity and performance of a patterned membrane, J. Membr. Sci., 2014, 462, 1–8,  DOI:10.1016/j.memsci.2014.03.012.
155. G. Brans, J. Kromkamp, N. Pek, J. Gielen, J. Heck and C. J. M. Van Rijn, et al., Evaluation of microsieve membrane design, J. Membr. Sci., 2006, 278, 344–348.
156. D.-C. Choi, Y.-J. Won, C.-H. Lee, S. Lee, M.-H. Lee and D.-Y. Khang, Tunable pore size micro/submicron-sieve membranes by soft lithography, J. Mater. Chem. A, 2013, 1, 12448,  10.1039/c3ta12490h.
157. T. Rehl and J. Müller, Life cycle assessment of biogas digestate processing technologies, Resour., Conserv. Recycl., 2011, 56, 92–104,  DOI:10.1016/j.resconrec.2011.08.007.
158. L. Thörneby, K. Persson and G. Trägårdh, Treatment of Liquid Effluents from Dairy Cattle and Pigs using Reverse Osmosis, J. Agric. Eng. Res., 1999, 73, 159–170,  DOI:10.1006/jaer.1998.0405.
159. J. Arévalo, L. M. Ruiz, J. A. Parada-Albarracín, D. M. González-Pérez, J. Pérez and B. Moreno, et al., Wastewater reuse after treatment by MBR. Microfiltration or ultrafiltration?, Desalination, 2012, 299, 22–27,  DOI:10.1016/j.desal.2012.05.008.
160. K. Zhang and K. Farahbakhsh, Removal of native coliphages and coliform bacteria from municipal wastewater by various wastewater treatment processes: implications to water reuse, Water Res., 2007, 41, 2816–2824,  DOI:10.1016/j.watres.2007.03.010.
161. Y. Chen, J. J. Cheng and K. S. Creamer, Inhibition of anaerobic digestion process: a review, Bioresour. Technol., 2008, 99, 4044–4064,  DOI:10.1016/j.biortech.2007.01.057.
162. B. Demirel, N. Göl and T. Onay, Evaluation of heavy metal content in digestate from batch anaerobic co-digestion of sunflower hulls and poultry manure, J. Mater. Cycles Waste Manage., 2013, 15, 242–246,  DOI:10.1007/s10163-012-0107-4.
163. P. Mantovi, G. Bonazzi, E. Maestri and N. Marmiroli, Accumulation of copper and zinc from liquid manure in agricultural soils and crop plants, Plant Soil, 2003, 250, 249–257,  DOI:10.1023/a:1022848131043.
164. R. Molinari, T. Poerio and P. Argurio, Selective separation of copper(II) and nickel(II) from aqueous media using the complexation–ultrafiltration process, Chemosphere, 2008, 70, 341–348.
165. R. Molinari, T. Poerio and P. Argurio, Chemical and operational aspects in running the polymer assisted ultrafiltration for separation of copper(II)–citrate complexes from aqueous media, J. Membr. Sci., 2007, 295, 139–147.
166. H. Ozaki, K. Sharma and W. Saktaywin, Performance of an ultra-low-pressure reverse osmosis membrane (ULPROM) for separating heavy metal: effects of interference parameters, Desalination, 2002, 144, 287–294.
167. P. Yenphan, A. Chanachai and R. Jiraratananon, Experimental study on micellar-enhanced ultrafiltration (MEUF) of aqueous solution and wastewater containing lead ion with mixed surfactants, Desalination, 2010, 253, 30–37.
168. J. Landaburu-Aguirre, E. Pongrácz, A. Sarpola and R. L. Keiski, Simultaneous removal of heavy metals from phosphorous rich real wastewaters by micellar-enhanced ultrafiltration, Sep. Purif. Technol., 2012, 88, 130–137.
169. R.-S. Juang and R.-C. Shiau, Metal removal from aqueous solutions using chitosan-enhanced membrane filtration, J. Membr. Sci., 2000, 165, 159–167.
170. J. Llorens, M. Pujolà and J. Sabaté, Separation of cadmium from aqueous streams by polymer enhanced ultrafiltration: a two-phase model for complexation binding, J. Membr. Sci., 2004, 239, 173–181,  DOI:10.1016/j.memsci.2004.02.034.
171. R. Camarillo, Á. Pérez, P. Cañizares and A. de Lucas, Removal of heavy metal ions by polymer enhanced ultrafiltration: batch process modeling and thermodynamics of complexation reactions, Desalination, 2012, 286, 193–199,  DOI:10.1016/j.desal.2011.11.021.
172. K. R. Desai and Z. V. P. Murthy, Removal of Ag(I) and Cr(VI) by Complexation–Ultrafiltration and Characterization of the Membrane by CFSK Model, Sep. Sci. Technol., 2014, 49, 2620–2629.
173. A. Tor, T. Büyükerkek, Y. Çengelolu and M. Ersöz, Simultaneous recovery of Cr(III) and Cr(VI) from the aqueous phase with ion-exchange membranes, Desalination, 2005, 171, 233–241.
174. S. K. Nataraj, K. M. Hosamani and T. M. Aminabhavi, Potential application of an electrodialysis pilot plant containing ion-exchange membranes in chromium removal, Desalination, 2007, 217, 181–190.
175. T. Mohammadi, A. Moheb, M. Sadrzadeh and A. Razmi, Separation of copper ions by electrodialysis using Taguchi experimental design, Desalination, 2004, 169, 21–31,  DOI:10.1016/j.desal.2004.08.004.
176. S. L. Vasilyuk, T. V. Maltseva and V. N. Belyakov, Influence of water hardness on removal of copper ions by ion-exchange-assisted electrodialysis, Desalination, 2004, 162, 249–254,  DOI:10.1016/s0011-9164(04)00048-7.
177. T. Mohammadi, A. Razmi and M. Sadrzadeh, Effect of operating parameters on Pb₂₊ separation from wastewater using electrodialysis, Desalination, 2004, 167, 379–385.
178. S. I. Lee, S. Y. Weon, C. W. Lee and B. Koopman, Removal of nitrogen and phosphate from wastewater by addition of bittern, Chemosphere, 2003, 51, 265–271.
179. N. O. Nelson, R. L. Mikkelsen and D. L. Hesterberg, Struvite precipitation in anaerobic swine lagoon liquid: effect of pH and Mg : P ratio and determination of rate constant, Bioresour. Technol., 2003, 89, 229–236.
180. M. Latifian, J. Liu and B. Mattiasson, Struvite-based fertilizer and its physical and chemical properties, Environ. Technol., 2012, 33, 2691–2697.
181. C. Vaneeckhaute, E. Meers, E. Michels, P. Christiaens and F. M. G. Tack, Fate of Macronutrients in Water Treatment of Digestate Using Vibrating Reversed Osmosis, Water, Air, Soil Pollut., 2011, 223, 1593–1603,  DOI:10.1007/s11270-011-0967-6.
182. G. Johnson, B. Culkin and L. Stowell, Membrane filtration of manure wastewater, A Comp. Conv. Treat. Methods VSEP, a Vibratory RO Membr. Syst. Emeryv. New Log. Res, 2004.
183. F. Wäeger-Baumann and W. Fuchs, The application of membrane contactors for the removal of ammonium from anaerobic digester effluent, Sep. Sci. Technol., 2012, 47, 1436–1442.
184. M. L. Gerardo, M. P. Zacharof and R. W. Lovitt, Strategies for the recovery of nutrients and metals from anaerobically digested dairy farm sludge using cross-flow microfiltration, Water Res., 2013, 47, 4833–4842.
185. A. Achilli, T. Y. Cath, E. A. Marchand and A. E. Childress, The forward osmosis membrane bioreactor: A low fouling alternative to MBR processes, Desalination, 2009, 239, 10–21,  DOI:10.1016/j.desal.2008.02.022.
186. E. R. Cornelissen, D. Harmsen, K. F. De Korte, C. J. Ruiken, J.-J. Qin and H. Oo, et al., Membrane fouling and process performance of forward osmosis membranes on activated sludge, J. Membr. Sci., 2008, 319, 158–168.
187. C. R. Martinetti, A. E. Childress and T. Y. Cath, High recovery of concentrated RO brines using forward osmosis and membrane distillation, J. Membr. Sci., 2009, 331, 31–39,  DOI:10.1016/j.memsci.2009.01.003.
188. Q. Ge, P. Wang, C. Wan and T.-S. Chung, Polyelectrolyte-Promoted Forward Osmosis–Membrane Distillation (FO–MD) Hybrid Process for Dye Wastewater Treatment, Environ. Sci. Technol., 2012, 46, 6236–6243,  DOI:10.1021/es300784h.
189. R. W. Holloway, A. E. Childress, K. E. Dennett and T. Y. Cath, Forward osmosis for concentration of anaerobic digester centrate, Water Res., 2007, 41, 4005–4014.
190. M. P. Zacharof and R. W. Lovitt, Recovery of volatile fatty acids (VFA) from complex waste effluents using membranes, Water Sci. Technol., 2014, 69, 495–503.
191. J.-H. Choi, K. Fukushi and K. Yamamoto, A study on the removal of organic acids from wastewaters using nanofiltration membranes, Sep. Purif. Technol., 2008, 59, 17–25,  DOI:10.1016/j.seppur.2007.05.021.
192. H. Chang, N.-J. Kim, J. Kang and C. Jeong, Biomass-derived volatile fatty acid platform for fuels and chemicals, Biotechnol. Bioprocess Eng., 2010, 15, 1–10,  DOI:10.1007/s12257-009-3070-8.
193. B. Van der Bruggen, J. Schaep, D. Wilms and C. Vandecasteele, Influence of molecular size, polarity and charge on the retention of organic molecules by nanofiltration, J. Membr. Sci., 1999, 156, 29–41,  DOI:10.1016/s0376-7388(9810.1016/s0376-7388(98)00326-3.
194. D. W. Green, Perry's chemical engineers' handbook, McGraw-hill New York, 2008.
195. S. S. Kapdi, V. K. Vijay, S. K. Rajesh and R. Prasad, Biogas scrubbing, compression and storage: perspective and prospectus in Indian context, Renew. Energy, 2005, 30, 1195–1202.
196. C. Ofori-Boateng and E. M. Kwofie, Water scrubbing: a better option for biogas purification for effective storage, World Appl. Sci., 2009, 5, 122–125.
197. M. T. Ho, G. W. Allinson and D. E. Wiley, Reducing the Cost of CO₂ Capture from Flue Gases Using Pressure Swing Adsorption, Ind. Eng. Chem. Res., 2008, 47, 4883–4890,  DOI:10.1021/ie070831e.
198. M. P. S. Santos, C. A. Grande and A. E. Rodrigues, Pressure Swing Adsorption for Biogas Upgrading. Effect of Recycling Streams in Pressure Swing Adsorption Design, Ind. Eng. Chem. Res., 2010, 50, 974–985,  DOI:10.1021/ie100757u.
199. E. Ryckebosch, M. Drouillon and H. Vervaeren, Techniques for transformation of biogas to biomethane, Biomass Bioenergy, 2011, 35, 1633–1645,  DOI:10.1016/j.biombioe.2011.02.033.
200. R. W. Baker and K. Lokhandwala, Natural Gas Processing with Membranes:  An Overview, Ind. Eng. Chem. Res., 2008, 47, 2109–2121,  DOI:10.1021/ie071083w.
201. A. Bos, I. G. M. Pünt, M. Wessling and H. Strathmann, Suppression of CO2-plasticization by semiinterpenetrating polymer network formation, J. Polym. Sci., Part B: Polym. Phys., 1998, 36, 1547–1556,  DOI:10.1002/(sici)1099-0488(19980715)36:9<1547::aid-polb12>3.0.co;2–5.
202. A. Brunetti, F. Scura, G. Barbieri and E. Drioli, Membrane technologies for CO₂ separation, J. Membr. Sci., 2010, 359, 115–125,  DOI:10.1016/j.memsci.2009.11.040.
203. S. Sridhar, B. Smitha and T. M. Aminabhavi, Separation of Carbon Dioxide from Natural Gas Mixtures through Polymeric Membranes—A Review, Sep. Purif. Rev., 2007, 36, 113–174,  DOI:10.1080/15422110601165967.
204. M. D. Donohue, B. S. Minhas and S. Y. Lee, Permeation behavior of carbon dioxide–methane mixtures in cellulose acetate membranes, J. Membr. Sci., 1989, 42, 197–214,  DOI:10.1016/s0376-7388(00)82376-5.
205. S. A. Stern and A. H. De Meringo, Solubility of carbon dioxide in cellulose acetate at elevated pressures, J. Polym. Sci., Polym. Phys. Ed., 1978, 16, 735–751,  DOI:10.1002/pol.1978.180160415.
206. A. Y. Houde, B. Krishnakumar, S. G. Charati and S. A. Stern, Permeability of dense (homogeneous) cellulose acetate membranes to methane, carbon dioxide, and their mixtures at elevated pressures, J. Appl. Polym. Sci., 1996, 62, 2181–2192,  DOI:10.1002/(sici)1097-4628(19961226)62:13<2181::aid-app1>3.0.co;2-f.
207. B. D. Freeman, Basis of Permeability/Selectivity Tradeoff Relations in Polymeric Gas Separation Membranes, Macromolecules, 1999, 32, 375–380,  DOI:10.1021/ma9814548.
208. W. H. Mazur and M. C. Chan, Membranes for natural gas sweetening and CO₂ enrichment, Chem. Eng. Prog., 1982, 78, 38–43.
209. R. W. Baker, Future Directions of Membrane Gas Separation Technology, Ind. Eng. Chem. Res., 2002, 41, 1393–1411,  DOI:10.1021/ie0108088.
210. J. C. Lötters, W. Olthuis, P. H. Veltink and P. Bergveld, The mechanical properties of the rubber elastic polymer polydimethylsiloxane for sensor applications, J. Micromech. Microeng., 1997, 7, 145.
211. A. F. Ismail and W. Lorna, Suppression of plasticization in polysulfone membranes for gas separations by heat-treatment technique, Sep. Purif. Technol., 2003, 30, 37–46,  DOI:10.1016/s1383-5866(02)00097-7.
212. A. Mansourizadeh and A. F. Ismail, Effect of additives on the structure and performance of polysulfone hollow fiber membranes for CO₂ absorption, J. Membr. Sci., 2010, 348, 260–267.
213. A. F. Ismail and N. Yaacob, Performance of treated and untreated asymmetric polysulfone hollow fiber membrane in series and cascade module configurations for CO₂/CH₄ gas separation system, J. Membr. Sci., 2006, 275, 151–165.
214. C. A. Scholes, G. Q. Chen, G. W. Stevens and S. E. Kentish, Plasticization of ultra-thin polysulfone membranes by carbon dioxide, J. Membr. Sci., 2010, 346, 208–214.
215. C. E. Powell and G. G. Qiao, Polymeric CO₂/N₂ gas separation membranes for the capture of carbon dioxide from power plant flue gases, J. Membr. Sci., 2006, 279, 1–49,  DOI:10.1016/j.memsci.2005.12.062.
216. L. S. White, T. A. Blinka, H. A. Kloczewski and I. Wang, Properties of a polyimide gas separation membrane in natural gas streams, J. Membr. Sci., 1995, 103, 73–82,  DOI:10.1016/0376-7388(94)00313-n.
217. H. Lin and B. D. Freeman, Materials selection guidelines for membranes that remove CO₂ from gas mixtures, J. Mol. Struct., 2005, 739, 57–74.
218. K. Ghosal, R. T. Chern, B. D. Freeman, W. H. Daly and I. I. Negulescu, Effect of Basic Substituents on Gas Sorption and Permeation in Polysulfone, Macromolecules, 1996, 29, 4360–4369,  DOI:10.1021/ma951310i.
219. H. Lin and B. D. Freeman, Gas solubility, diffusivity and permeability in poly(ethylene oxide), J. Membr. Sci., 2004, 239, 105–117,  DOI:10.1016/j.memsci.2003.08.031.
220. M. Yoshino, K. Ito, H. Kita and K.-I. Okamoto, Effects of hard-segment polymers on CO₂/N₂ gas-separation properties of poly(ethylene oxide)-segmented copolymers, J. Polym. Sci., Part B: Polym. Phys., 2000, 38, 1707–1715,  DOI:10.1002/1099-0488(20000701)38:13<1707::aid-polb40>3.0.co;2-w.
221. A. Tena, A. E. Lozano, L. Palacio, A. Marcos-Fernández, P. Prádanos and J. de Abajo, et al., Gas separation properties of systems with different amounts of long poly(ethylene oxide) segments for mixtures including carbon dioxide, Int. J. Greenhouse Gas Control, 2013, 12, 146–154,  DOI:10.1016/j.ijggc.2012.10.014.
222. P. Cserjési, N. Nemestóthy and K. Bélafi-Bakó, Gas separation properties of supported liquid membranes prepared with unconventional ionic liquids, J. Membr. Sci., 2010, 349, 6–11.
223. P. Li, D. R. Paul and T.-S. Chung, High performance membranes based on ionic liquid polymers for CO₂ separation from the flue gas, Green Chem., 2012, 14, 1052–1063,  10.1039/c2gc16354c.
224. V. I. Bondar, B. D. Freeman and I. Pinnau, Gas sorption and characterization of poly(ether-b-amide) segmented block copolymers, J. Polym. Sci., Part B: Polym. Phys., 1999, 37, 2463–2475.
225. M. G. Buonomenna, W. Yave and G. Golemme, Some approaches for high performance polymer based membranes for gas separation: block copolymers, carbon molecular sieves and mixed matrix membranes, RSC Adv., 2012, 2, 10745–10773,  10.1039/c2ra20748f.
226. Y. Liu, R. Wang and T.-S. Chung, Chemical cross-linking modification of polyimide membranes for gas separation, J. Membr. Sci., 2001, 189, 231–239,  DOI:10.1016/s0376-7388(01)00415-x.
227. H. B. Park, S. H. Han, C. H. Jung, Y. M. Lee and A. J. Hill, Thermally rearranged (TR) polymer membranes for CO₂ separation, J. Membr. Sci., 2010, 359, 11–24.
228. J. N. Barsema, S. D. Klijnstra, J. H. Balster, N. F. A. Van der Vegt, G. H. Koops and M. Wessling, Intermediate polymer to carbon gas separation membranes based on Matrimid PI, J. Membr. Sci., 2004, 238, 93–102.
229. H. B. Park, C. H. Jung, Y. M. Lee, A. J. Hill, S. J. Pas and S. T. Mudie, et al., Polymers with cavities tuned for fast selective transport of small molecules and ions, Science, 2007, 318, 254–258.
230. M.-B. Hägg, J. O. N. A. Lie and A. Lindbråthen, Carbon Molecular Sieve Membranes, Ann. N. Y. Acad. Sci., 2003, 984, 329–345,  DOI:10.1111/j.1749-6632.2003.tb06010.x.
231. H. Suda and K. Haraya, Alkene/alkane permselectivities of a carbon molecular sievemembrane, Chem. Commun., 1997, 93–94.
232. M. Kiyono, P. J. Williams and W. J. Koros, Effect of pyrolysis atmosphere on separation performance of carbon molecular sieve membranes, J. Membr. Sci., 2010, 359, 2–10,  DOI:10.1016/j.memsci.2009.10.019.
233. V. C. Geiszler and W. J. Koros, Effects of Polyimide Pyrolysis Conditions on Carbon Molecular Sieve Membrane Properties, Ind. Eng. Chem. Res., 1996, 35, 2999–3003,  DOI:10.1021/ie950746j.
234. T.-J. Kim, B. Li and M.-B. Hägg, Novel fixed-site–carrier polyvinylamine membrane for carbon dioxide capture, J. Polym. Sci., Part B: Polym. Phys., 2004, 42, 4326–4336,  DOI:10.1002/polb.20282.
235. J. Huang, J. Zou and W. S. W. Ho, Carbon Dioxide Capture Using a CO₂-Selective Facilitated Transport Membrane, Ind. Eng. Chem. Res., 2008, 47, 1261–1267,  DOI:10.1021/ie070794r.
236. L. Deng and M.-B. Hägg, Techno-economic evaluation of biogas upgrading process using CO₂ facilitated transport membrane, Int. J. Greenhouse Gas Control, 2010, 4, 638–646.
237. S. Saedi, S. S. Madaeni, F. Seidi, A. A. Shamsabadi and S. Laki, Synthesis and application of a novel Amino-Starch derivative as a new polymeric additive for fixed facilitated transport of carbon dioxide through an asymmetric polyether sulfone (PES) membrane, Int. J. Greenhouse Gas Control, 2013, 19, 126–137,  DOI:10.1016/j.ijggc.2013.06.019.
238. S. Andrew Lee, G. W. Stevens and S. E. Kentish, Facilitated transport behavior of humidified gases through thin-film composite polyamide membranes for carbon dioxide capture, J. Membr. Sci., 2013, 429, 349–354,  DOI:10.1016/j.memsci.2012.11.047.
239. L. M. Robeson, The upper bound revisited, J. Membr. Sci., 2008, 320, 390–400,  DOI:10.1016/j.memsci.2008.04.030.
240. R. H. B. Bouma, A. Checchetti, G. Chidichimo and E. Drioli, Permeation through a heterogeneous membrane: the effect of the dispersed phase, J. Membr. Sci., 1997, 128, 141–149,  DOI:10.1016/s0376-7388(96)00303-1.
241. T.-S. Chung, L. Y. Jiang, Y. Li and S. Kulprathipanja, Mixed matrix membranes (MMMs) comprising organic polymers with dispersed inorganic fillers for gas separation, Prog. Polym. Sci., 2007, 32, 483–507,  DOI:10.1016/j.progpolymsci.2007.01.008.
242. R. Mahajan, R. Burns, M. Schaeffer and W. J. Koros, Challenges in forming successful mixed matrix membranes with rigid polymeric materials, J. Appl. Polym. Sci., 2002, 86, 881–890,  DOI:10.1002/app.10998.
243. T. T. Moore and W. J. Koros, Non-ideal effects in organic–inorganic materials for gas separation membranes, J. Mol. Struct., 2005, 739, 87–98,  DOI:10.1016/j.molstruc.2004.05.043.
244. P. Cosoli, M. Ferrone, S. Pricl and M. Fermeglia, Hydrogen sulphide removal from biogas by zeolite adsorption: Part I. GCMC molecular simulations, Chem. Eng. J., 2008, 145, 86–92,  DOI:10.1016/j.cej.2008.07.034.
245. C. Baerlocher, L. B. McCusker and D. H. Olson, Atlas of zeolite framework types, Elsevier, 2007.
246. H. B. Tanh Jeazet, C. Staudt and C. Janiak, Metal–organic frameworks in mixed-matrix membranes for gas separation, Dalton Trans., 2012, 14003–14027,  10.1039/c2dt31550e.
247. P. S. Goh, A. F. Ismail, S. M. Sanip, B. C. Ng and M. Aziz, Recent advances of inorganic fillers in mixed matrix membrane for gas separation, Sep. Purif. Technol., 2011, 81, 243–264,  DOI:10.1016/j.seppur.2011.07.042.
248. Y. Yoo, Z. Lai and H.-K. Jeong, Fabrication of MOF-5 membranes using microwave-induced rapid seeding and solvothermal secondary growth, Microporous Mesoporous Mater., 2009, 123, 100–106,  DOI:10.1016/j.micromeso.2009.03.036.
249. Y. Dai, J. R. Johnson, O. Karvan, D. S. Sholl and W. J. Koros, Ultem®/ZIF-8 mixed matrix hollow fiber membranes for CO₂/N₂ separations, J. Membr. Sci., 2012, 401, 76–82.
250. M. J. C. Ordonez, K. J. Balkus Jr, J. P. Ferraris and I. H. Musselman, Molecular sieving realized with ZIF-8/Matrimid® mixed-matrix membranes, J. Membr. Sci., 2010, 361, 28–37.
251. A. Molino, M. Migliori, Y. Ding, B. Bikson, G. Giordano and G. Braccio, Biogas upgrading via membrane process: modelling of pilot plant scale and the end uses for the grid injection, Fuel, 2013, 107, 585–592,  DOI:10.1016/j.fuel.2012.10.058.
252. A. G. Chmielewski, A. Urbaniak and K. Wawryniuk, Membrane enrichment of biogas from two-stage pilot plant using agricultural waste as a substrate, Biomass Bioenergy, 2013, 58, 219–228,  DOI:10.1016/j.biombioe.2013.08.010.
253. P. Izak, M. Kárászová, J. Vejrazka, K. Friess, A. Randová and J. C. Jansen, The Effective Upgrading of Raw Biogas to Methane by Selective Membranes, Procedia Eng., 2012, 44, 429–431,  DOI:10.1016/j.proeng.2012.08.438.
254. C. Tricase and M. Lombardi, State of the art and prospects of Italian biogas production from animal sewage: technical-economic considerations, Renewable Energy, 2009, 34, 477–485,  DOI:10.1016/j.renene.2008.06.013.
255. S. Patel, D. Tonjes and D. Mahajan, Biogas potential on Long Island, A quantification study, J. Renew. Sustain. Energy, New York, 2011, vol. 3, p. 043118,  DOI:10.1063/1.3614443.
256. T. Patterson, S. Esteves, R. Dinsdale and A. Guwy, An evaluation of the policy and techno-economic factors affecting the potential for biogas upgrading for transport fuel use in the UK, Energy Policy, 2011, 39, 1806–1816,  DOI:10.1016/j.enpol.2011.01.017.
257. G. Owen, M. Bandi, J. A. Howell and S. J. Churchouse, Economic assessment of membrane processes for water and waste water treatment, J. Membr. Sci., 1995, 102, 77–91,  DOI:10.1016/0376-7388(94)00261-v.

---

Footnote

† These authors contributed equally.

---