Next Article in Journal
Optical Waveguide Refractive Index Sensor for Biochemical Sensing
Next Article in Special Issue
Copolymerization of Ethylene with Alpha-Olefins over Supported Titanium–Magnesium Catalysts Containing Titanium Compounds in Different Oxidation and Coordination States
Previous Article in Journal
Characterization of Pterygomaxillary Suture Morphology: A CBCT Study
Previous Article in Special Issue
The Effect of Local Doping of the Polymer–Polymer Interface Using Cu2O Particles
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Fouling of Polyalkylmethylsiloxane Composite Membranes during Pervaporation Separation of ABE-Fermentation Mixtures

1
A.V. Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, Leninsky Prospect 29, 119991 Moscow, Russia
2
Biological and Environmental Science, and Engineering Division (BESE), Advanced Membranes and Porous Materials Center (AMPM), King Abdullah University of Science and Technology, Thuwal 23955, Saudi Arabia
*
Authors to whom correspondence should be addressed.
Appl. Sci. 2023, 13(6), 3827; https://doi.org/10.3390/app13063827
Submission received: 20 February 2023 / Revised: 13 March 2023 / Accepted: 15 March 2023 / Published: 16 March 2023
(This article belongs to the Special Issue Functional Polymers: Synthesis, Properties and Applications)

Abstract

:
Production of bio-alcohols is one of the approaches used in the development of alternative energy. Pervaporation is a promising option for the separation of bio-alcohols from the fermentation mixture. A serious problem in the process of continuous extraction of biobutanol from the fermentation broth is the contamination of the membrane, which leads to a decrease in its permeability over time. In this work, the transport properties of composite membranes based on polyheptylmethylsiloxane (PHeptMS), polydecylmethylsiloxane (PDecMS), and a commercial membrane MDK-3 were studied during separation of a real ABE-fermentation broth in vacuum pervaporation mode. The study was performed before and after continuous contact of the membranes with the fermentation broth for one month. Visually and by scanning electron spectroscopy, the presence of membrane surface residue and its effect on the wettability of the membrane selective layer by the components of the ABE broth were determined. The sediment composition was evaluated by energy dispersive analysis and infrared spectroscopy. According to the pervaporation separation of the ABE-broth using PHeptMS, PDecMS, and MDK-3 membranes, the butanol flux was 0.029, 0.012, and 0.054 kg/(m2·h), respectively. The butanol-water partition factor was 41, 22, and 13 for PHeptMS, PDecMS, and MDK-3, respectively. After one month of incubation of the membranes in ABE-fermentation broth during the separation of the model mixture, a decrease of 10 and 5% in permeate flux and separation factor, respectively, was observed for all membranes. Temperature dependences (30–60 °C) of permeate flux, permeability, and selectivity were obtained for the membranes after clogging. The most promising in terms of minimal negative changes as a result of fouling was demonstrated by the PHeptMS membrane. For it, the clogging dynamics during separation of the real fermentation broth for 216 h were investigated. Two characteristic steps of decrease in transport and separation properties were observed, after 28 and 150 h of the experiment. After 216 h of experiment, a 1.28-fold decrease in total flux through the membrane, a 9% decrease in butanol permeability, and a 10% decrease in n-butanol selectivity were found for PHeptMS.

1. Introduction

Biotechnology is one of the most popular and widely developing technologies. Due to the increasing needs of the growing population, significant attention is being paid to the development of new and alternative routes for the production of chemicals from renewable feedstock [1,2,3,4,5]. One of the areas of development of green energy is the production of biofuels, such as bio-alcohols (biomethanol, bioethanol, and biobutanol), based on renewable raw materials. Biomass is one of the most promising sources for the production of liquid biofuels, and can provide up to 14% of global energy consumption [6,7]. The most promising bio-alcohol in terms of compatibility with internal combustion engines is 1-butanol. Moreover, biobutanol can be mixed with gasoline in any ratio and has a high octane number. Compared to methanol and ethanol, butanol also has a higher combustion heat and lower volatility [8,9,10,11].
Biobutanol is produced by acetone-butanol-ethanol (ABE) fermentation of a plant substrate (for example, lignin cellulosic biomass (LCB)) under the action of microorganisms [12]: Clostridium acetobutylicum, Clostridium beijerinckii, and Clostridium saccharobutylicum and saccharoperbutylacetonicum [13,14,15,16]. Using non-LCB as feedstock requires pretreatment and hydrolysis to release fermentable sugars from the biomass [17]. One of the main problems of fermentation is the low yield of butanol (up to 2 wt. %) caused by the inhibitory effect of alcohol on producing microorganisms [18]. This significantly reduces the efficiency of raw material use and increases the cost of its separation from the fermentation mixture [19]. The solution of this problem is to remove the target products from the fermentation mixture and return it to the bioreactor for further fermentation. Such an approach will make it possible to obtain a concentrated product as the output as well as to reduce the inhibitory effect of alcohols for deeper processing of raw materials. The main requirements in the process of butanol extraction from fermentation mixture are low energy consumption and the absence of negative influences of the conditions and materials used on microorganisms. To date, the following methods of biobutanol extraction, which can be integrated with ABE fermentation, have been proposed: distillation [20], adsorption [21], liquid extraction [22], reverse osmosis [23], and pervaporation [24,25,26,27,28,29,30,31]. Among the presented methods, pervaporation separation is of particular interest: separation at fermentation temperatures (35–40 °C), isolation of microorganisms from a solution with a higher concentration of n-butanol by means of a membrane [32,33,34], and preservation of the productivity of microorganisms by withdrawing alcohols from the broth [35]. The combined scheme of fermentation and hydrophobic pervaporation processes makes it possible to concentrate butanol from the fermentation mixture by ~10 times [31]. Low separation temperatures in the process of pervaporation result in low energy consumption for concentrating butanol. Thus, in previous work [36] the comparison of energy consumption for the separation of a butanol-water mixture was carried out, and it was shown that less energy is required for the pervaporation separation of the mixture (3.7 kJ/g) in comparison with adsorption (5.4 kJ/g). Extraction of bio-alcohols from fermentation mixtures using pervaporation through hydrophobic membranes is economically competitive with distillation for small to medium volumes of extraction (~2–20 million liters per year) [24].
The traditional membrane material for hydrophobic pervaporation is polydimethylsiloxane (PDMS) [24]. Based on PDMS, most commercial pervaporation membranes are made, such as Pervap (Sulzer Chemtech, Winterthur, Switzerland), Pervatech PDMS (Pervatech, Rijssen, The Netherlands), and PolyAn (PolyAn GmbH, Berlin, Germany) [31]. This material allows selective extraction of n-butanol from a mixture with water due to increased sorption of n-butanol. Lee et al. [37] measured the sorption of water and butanol by PDMS membranes at 37 °C. The sorption of water was 0.6 g/g, whereas the sorption of butanol was 16.3 g/g, indicating that the more hydrophobic substance penetrates the PDMS membrane more easily and causes its swelling. Due to this effect, the polymer chains become more flexible and the transport of substances through the siloxane membrane increases. In [38] it was shown that when separating the ABE-fermentation broth through a PDMS membrane at 37 °C, the butanol-water separation factor reaches a value of 16.6. Zhu H. et al. [39] investigated the separation of the fermentation broth through a ceramic tubular membrane with an inner selective layer of PDMS and it was shown that at 37 °C the butanol/water separation factor is 24.7.
The selectivity of membranes for biobutanol is a key factor influencing the economics of its extraction from fermentation mixtures. Accordingly, the task of developing pervaporation membranes based on polymers with increased selectivity for n-butanol is relevant. Chemical modification of the polysiloxane chain allows improving the separation properties of pervaporation membranes [40]. The main approaches to such modifications are the introduction of functional groups into the main or side chain, as well as chemical cross-linking. For example, the use of trifunctional cross-linking agents in the synthesis of PDMS to create composite PDMS/PVDF membranes made it possible to increase the butanol/water separation factor to a value of 51 [41]. A wide range of functional polysiloxanes has been synthesized and studied in the process of the pervaporation separation of organic substances from water [42]. In particular, it was demonstrated that the introduction of 10% tridecyl substituents into the side chain leads to an almost twofold increase in the methyl isobutyl ketone/water separation factor. Functionalized membranes in comparison with PDMS show an increase in sorption of organic substances and a decrease in sorption of water [42]. The introduction of alkyl side substituents having a length of 7–10 carbon atoms into the siloxane chain makes it possible to increase the selectivity of the membrane for organic components. The polyheptylmethylsiloxane-based membrane (PHeptMS) exhibits a record separation factor of n-butanol-water binary mixture of 97, which is almost three times higher than the separation factor of PDMS [43]. Increasing the hydrophobicity of the membrane selective layer makes it possible to reduce water transfer and thus increase the concentration of n-butanol in the permeate. For example, it was shown in [44] that butanol/water selectivity for the PDecMS/MFFK-1 membrane was 3–7 times higher than that for commercial membranes. At a selective layer thickness of the PDecMS/MFFK-1 composite membrane of about 4.5 μm, the 1-butanol permeability comparable to that of commercial composite membranes (5.2 mol·m−2·h−1·kaPa−1) can be achieved.
Nevertheless, in addition to a high separation capacity, the membrane must demonstrate stable performance during the pervaporation process. During the separation of butanol from the ABE-fermentation mixture, membrane clogging by components of the fermentation mixture is observed, resulting in reduced transport and separation capacity of the membrane. Qureshi N. et al. [45] reported a threefold decrease in the butanol selectivity of the PDMS membrane during pervaporation separation of the ABE-fermentation broth compared with the model mixture. The researchers attributed the decrease in the transport characteristics of the PDMS membrane mainly to the activity of active microbial cells from the fermentation medium, which are adsorbed on the membrane surface [46]. Reduction in membrane transport properties in the process of continuous extraction of biobutanol from the fermentation broth due to membrane fouling by microorganisms is a factor limiting the industrial application of pervaporation [47]. Previous researchers [48] studied the effect of fouling of a PDMS-based membrane during ABE fermentation integrated with pervaporation separation. The method of scanning electron microscopy revealed that the membrane surface after contact with the fermentation broth, in addition to microbial cells, contains adsorbed extracellular polymeric substances (EPSs), which is a sign of surface fouling. Researchers suggest membrane washing [49], as well as preliminary removal of active cells from fermentation mixtures by various filtration methods, as ways to solve this problem [50].
Another way to avoid membrane fouling is to preliminarily remove cells from fermentation mixtures using different filtration methods [51]. However, this approach complicates the separation process and increases capital costs. Introduction of fluorine-containing fragments into the material structure of polysiloxane membranes significantly reduces their fouling by components of the fermentation mixture and increases the stability of the membrane transport properties during butanol production using a pervaporation membrane bioreactor [52]. Moreover, at the introduction of fluorine derivatives, hydrophobicity and lipophobicity of a surface of selective layer of a membrane increases [53]. In [54], it was shown that when fluorine derivatives were introduced into PDMS during the pervaporation separation of butanol-water mixture, there was a decrease in total flux compared to a pure PDMS membrane. However, the partial flux of butanol remained unchanged. Thus, such a modification of the membrane material can prevent the growth without compromising the transport performance.
Thus, the problem of membrane fouling during the pervaporation separation of fermentation mixtures has a significant impact on the efficiency of the separation process. The effect of fouling and biofouling of membranes based on highly selective polyalkylmethylsiloxanes has not been previously studied. Therefore, in the present work the effect of long-term contact of polyalkylmethylsiloxane-based membranes with real fermentation mixtures was studied for the first time.

2. Materials and Methods

2.1. Materials

For the synthesis of polydecylmethylsiloxane (PDecMS) and polyheptylmethylsiloxane (PHeptMS), we used: polymethylhydrosiloxane (PMHS) (Mn = 1900 g/mol, ABCR, Karlsruhe, Germany); 1-decene (95% wt., Sigma-Aldrich, St. Louis, MO, USA); isooctane (Chemical grade, Component Reagent, Moscow, Russia); 1.7-octadiene (95%, Sigma-Aldrich, St. Louis, MO, USA); 1.3-divinyl-1.1.3. 3-tetramethyldisiloxane platinum complex (0), solution in xylene (Sigma-Aldrich, St. Louis, MO, USA); polydimethylsiloxane vinyl terminated (PDMS) (Mn = 25,000 g/mol, Sigma-Aldrich, St. Louis, MO, USA); 1-heptene (95% wt., Sigma-Aldrich, St. Louis, MO, USA). A porous microfiltration support MFFK-1 based on fluoroplastic F-42L (Vladipor, Vladimir, Russia) was chosen as a support.
A commercial gas separation composite membrane MDK-3 (Vladipor, Vladimir, Russia) was used as a comparison sample. The membrane is a porous polymeric film material based on fluoroplastic F42L on a support of nonwoven materials (polypropylene, lavsan) with a thin separating layer based on Carbosil© (Polymersintez, Vladimir, Russia).
The real ABE-fermentation broth was produced at the Kurchatov Institute-GosNIIgenetika according to the method described in [55]. The bacterial strain Clostridium B-10939 was used as a biobutanol producer in flour medium, containing 100 g/L rye flour and 2 g/L CaCO3. The concentrations of acetone, butanol, and ethanol in the fermentation mixture were 0.56, 1.26, and 0.26% wt., respectively.
The model separable mixture was prepared by dissolving acetone (A grade, 99.75% wt., Vekton, St. Petersburg, Russia), ethanol (A grade, Component Reactiv, Moscow, Russia), and butanol (A grade, ECOS-1, Moscow, Russia) in distilled water. The model mixture contained 0.76% wt. acetone, 1.6% wt. butanol, 0.3% wt. ethanol, and 97.34% wt. water.

2.2. Membrane Materials Synthesis

Synthesis of polyalkylmethylmethylsiloxanes was carried out by hydrosilylation reaction, the scheme of which is shown in Figure 1. Polymethylhydrosiloxane was mixed with 15 wt. % solution of 1-alkene (1-heptene, 1-decene) in isooctane and with 10 wt. % solution of 1.7-octadiene in isooctane stirred for 2 h at 60 °C in the presence of Carsted catalyst (1,3-divinyl-1,1,3,3-tetramethyldisiloxane platinum (0) complex, a solution in xylene).
Then, an 11 wt. % solution of polydimethylsiloxane vinyl terminated with Mn = 25,000 g/mol (PDMS) was added to the solution in isooctane in the molar ratio of PDMS:PMHS = 0.1628.
Stirring of the resulting reaction mixture was continued for 1 h, after which a 3 wt. % solution of PMHS in isooctane was added to the reaction mixture to a stoichiometric ratio of 1-alkene:PMHS = 0.8317 and 1.7-octadiene:PMHS = 0.0519. The reaction mixture was stirred until a viscosity of 15–20 mPa·s was reached.

2.3. Development Flat Sheet Composite Membranes

The selective layer of polyalkylmethylsiloxanes was deposited from a solution of PDecMS and PHeptMS polymers in isooctane on the MFFK-1 microfiltration support. The composite membranes of PDecMS/MFFK and PHeptMS/MFFK were obtained by touching the porous microfiltration support MFFK-1 to the surface of the polymer solution, similar to previous work [56]. The touch method consists in pulling the substrate tape over the surface of the polymer solution with the formation of a meniscus at tapping. This method of application minimizes the polymer flow into the pores of the substrate. Hereinafter, composite membranes PDecMS/MFFK and PHeptMS/MFFK are designated as PDecMS and PHeptMS, respectively.

2.4. Prolonged Exposure of Composite Membranes in Fermentation Broth

To study the effect of long-term contact on the transport and separation properties of the composite membranes, they were placed in the fermentation broth for 1 month at room temperature. To exclude the penetration of contamination into the pores of the support, only the selective layer of membranes was in contact with the fermentation broth. For this purpose, the membrane was fixed in a filtration cell, over the membrane volume of which was filled with broth.

2.5. Scanning Electron Microscopy and Energy Dispersive X-ray Spectral Analysis (EDX)

Scanning electron microscopy (SEM) was used to characterize the structure and morphology of the membranes. SEM was carried out on a Thermo Fisher Phenom XL G2 Desktop SEM (Thermo Fisher Scientific, Waltham, MA, USA). Cross-sections of the membranes were obtained in liquid nitrogen after preliminary impregnation of the specimens in isopropanol. A thin (5–10 nm) gold layer was deposited on the prepared samples in a vacuum chamber (~0.01 mbar) using a “Cressington 108 auto Sputter Coater” desktop magnetron sputter (Cressington Scientific Instruments Ltd., Watford, UK). The accelerating voltage during image acquisition was 15 keV.

2.6. Contact Angle Measurement

The measurements of contact wetting angles were taken using the standard method of a lying drop using a LK-1 goniometer manufactured by RPC OpenScience Ltd. (Krasnogorsk, Russia). Data acquisition and subsequent digital processing of droplet images for the direct calculation of angles using the Young–Laplace equation were carried out using DropShape software v.1.0. The error of measurements was ±2°. The temperature at which the experiments were conducted was equal to the ambient temperature and was 21 ± 2 °C.

2.7. Infrared Spectroscopy

PDecMS, PHeptMS, and MDK-3 membranes were studied by ATR infrared spectroscopy before and after direct contact of the selective layer with the fermentation ABE medium for 1 month. The spectra of the membrane samples were recorded in ATR mode (ZnSe crystal, scan. −50, resolution 2 cm−1, range 600–4000 cm−1, spectrometer IFS-66v/s Bruker, Billerica, MA, USA).

2.8. Vacuum Pervaporation

The transport and separation properties of PHeptMS, PDecMS, and commercial MDK-3 membranes were studied in the mode of vacuum pervaporation when separating the ABE-fermentation broth at 30–60 °C.
Figure 2 shows the scheme of the vacuum pervaporation setup. From a thermostatically controlled container of 1 L with a stirrer (1), by means of an MV-Z gear pump (Ismatec, Zurich, Switzerland) (2), the initial separated mixture (flow I) is directed through the heat exchanger (3) to the membrane module (4) and then returns back to the container (1) by flow II. The effective membrane area in the membrane module is 13.5 cm2 and the volume flow rate of the separated mixture is 350 mL/min. The permeate passing through the membrane (flow III) in the vapor phase is condensed in glass traps (5) that are placed in Dewar flasks with liquid nitrogen (−196 °C). To achieve continuity of the pervaporation separation, the traps are arranged in parallel and operated alternately throughout the experiment. The LOIP LT-100 liquid thermostat (St. Petersburg, Russia) (6) ensures a constant temperature of the separated mixture with an accuracy of ±0.1 °C. Creation and maintenance of the difference of partial pressures of the mixture vapor was provided by vacuumization of the sub-membrane space using an Ebara PDV-250 vacuum pump (EBARA Corporation, Tokyo, Japan) (7). A safety trap (8) prevents permeate vapors from entering the vacuum pump.
Concentration of the initial solution, as well as retentate and permeate, was determined by gas chromatography on a Crystallux-4000M chromatograph (RPC “META-CHROM”, Co., Ltd., Yoshkar-Ola, Russia) equipped with a flame ionization detector. Analysis was performed using the following parameters: evaporation temperature—200 °C, column temperature—120 °C, and detector temperature—150 °C. A Phenomenex Zebron ZB-FFAP (Phenomenex, Torrance, CA, USA) capillary column (length 50 m, diameter 0.32 mm, phase thickness 0.50 µm) was used with the following phase composition: polymeric ester of 2-nitroterephthalic acid and polyethylene glycol. Water was added during the permeate analysis to homogenize the sample.
Total permeate flux J, kg/(m2·h), was determined using the weight method according to Formula (1):
J = m S · t ,
where m is the total mass of permeate (kg) permeated through the membrane of area S (m2) in time t, h.
The separation factor β was determined using Formula (2):
β = y o · x w y w · x o ,
where xo and xw are the mass fractions of the organic component and water in the separated mixture, and yo and yw are the mass fractions of the organic component and water in the permeate.
Mass fluxes of components in the permeate were determined as:
J i = J · y o ,
To describe the balance between permeability and selectivity of separation, a criterion of membrane quality was introduced—pervaporation separation index (PSI):
P S I = J ( β i j 1 ) ,
The permeability (P, mol/(m2·h·kPa)) for component i was calculated according to Equation (5):
P i = J i ( P i f P i p ) ,
where p i f and p i p are the vapor pressure of component i in the initial mixture and permeate (kPa), respectively.
The membrane selectivity ( α i j m ) was determined from the ratio:
α i j m = P i P j ,
To determine the vapor pressure of the permeate and the initial mixture, activity coefficients were calculated using the NRTL (Non-Random Two-Liquid) model using the Aspen Plus 10 software package.

3. Results

3.1. Composite Membrane Characterisation: Before and after Exposure to ABE-Fermentation Broth

The morphology of the flat composite membranes was studied using SEM microphotographs. As can be seen from Figure 3, the thickness of the selective layer for the PHeptMS membrane was 10 ± 3 μm, and that for PDecMS was 10 ± 3 μm. The thickness of the commercial MDK-3 membrane was 3.1 ± 2 μm.
SEM images of the membranes before and after contact with the fermentation mixture are shown in Figure 4. Before contact with the fermentation broth, the surfaces of the composite membranes were white. The selective polysiloxane layer was visible as a transparent glossy coating on the fluoroplastic surface (Figure 4a,c,e). After contact of the membranes with ABE-fermentation broth for one month, the formation of deposits was observed on the surface of the selective layer (Figure 4b,d,f). The PDecMS membrane (Figure 4b) was visually characterized by more severe contamination, expressed as a complete change in the color of the surface of the selective layer from white to pale brown with an “oil” cast in the light. PHeptMS (Figure 4d) and MDK-3 (Figure 4f) membranes were visually less contaminated after contact with the fermentation broth. Orange-brown stains and yellow patches in places were observed on the surface of the selective layer. SEM images of the deposits on the surface of the membranes are shown in Figure 5.
Before contact of the membranes with the ABE-fermentation mixture, the surfaces of the selective layer were smooth and homogeneous (Figure 5a,c,e). After one month of contact with the ABE-fermentation medium, deposits were observed on the surfaces of all the studied membranes (Figure 5b,d,f). To determine the elemental composition of the detected deposit, the surface of the membranes was examined using energy dispersive analysis (Table 1). The initial membranes contained silicon, oxygen, and carbon atoms on the surface, whereas on the surface of the membranes after contact with the broth, the presence of nitrogen, sulfur, and phosphorus atoms, and an increase in carbon and oxygen content compared with the initial membrane, was observed. Thus, for the PHeptMS membrane after contact with the fermentation broth, in addition to atoms of carbon, oxygen, and silicon, the presence of 30% wt. of nitrogen and 0.7% wt. of sulfur on the surface of the selective layer is characteristic. Moreover, for this membrane an increase in oxygen atoms content is observed—from 14% wt. for the initial membrane to 20% wt. for the membrane after fouling. For the PDecMS membrane after contact with the fermentation mixture, we observe an increase in the surface content of 0.7% wt. of phosphorus atoms, as well as an increase in oxygen content from 12.1 to 18.1% wt. and in carbon content from 54.2 to 56.3% wt. In the case of the MDK-3 membrane, the presence of fluorine atoms was observed on the surface of the initial membrane, which corresponds to the material of the MFFK-1 substrate (fluoroplastic). On the surface of MDK-3 membrane after fouling, the oxygen content increased by 2% wt. % and the number of fluorine atoms increased 2.25 times. Sulfur atoms also appeared with a concentration of 0.1% wt.
Thus, the change in the elemental composition of the selective layer of membranes after one month of contact with the fermentation broth indicates the deposition of various compounds that are products of the vital activity of microorganisms.
To determine the functional composition of the observed deposits, IR spectra of the membranes were obtained before and after their contact with the fermentation medium. Figure 6 and Figure 7 show a comparison of the IR spectra of the PDecMS, PHeptMS, and MDK-3 membranes before and after exposure to the fermentation mixture. The spectra of PDecMS and PHeptMS membranes fully correspond to the chemical structure of the polyalkylsiloxane polymer of the membrane selective layer [56].
In Figure 6 it is seen that the most intense split band at 1014 cm−1 refers to the valent vibrations of the -O-Si-O- bonds of the main chain of the polymer. The intense band at 795 cm−1 characterizes the valent vibrations of the Si-C bonds. The band at 1258 cm−1 is associated with deformation vibrations in the methyl group at the silicon atom, and bands at 2850–2960 cm−1 and 1460 cm−1 refer respectively to the valent and deformation vibrations in the decyl substituent. In the spectrum of MDK-3 membrane, which differs in chemical structure from PHeptMS and PDecMS, in addition to the siloxane bands described above, the bands of the aromatic ring (1602 and 1506 cm−1) and the C=O bond (1772 cm−1) are present. In addition, long alkyl chains at the MDK-3 silicon atom are absent; the relative intensity of the bands from the C-H alkyl bonds (2840–2960 cm−1) changes.
In the spectrum of PDecMS and PHeptMS, all bands characterizing the structure of polyalkylsiloxane are preserved after exposure in the fermentation broth, and only the intensities of almost all bands drop significantly (Figure 6). It can be assumed that the drop in the intensity of the main polysiloxane bands in the spectra of membranes after ABE is due to the appearance of a new thin layer of polypeptide from microorganisms on the membrane surface. The decrease in intensity of all siloxane bands after contact with the fermentation medium can be related to the deposition of a thin layer of dead microorganisms or their spores on the membrane surface.
In addition, new bands appear that are in good agreement with the spectrum of the polypeptide chain of protein molecules contained in the cell membranes of bacteria or their spores. The new bands in the spectra of the membranes after exposure include: (1) A broad mid-intensity band in the 3284 cm−1 region (Figure 7), which usually characterizes the hydrogen-bonded amino groups in the polypeptide chains of -C(O)-NH-protein molecules. In addition, this region can contain bands of valent vibrations of -NH-bonds in purine and pyrimidine nucleotide bases, as well as bands of valent vibrations of OH-water. (2) A broad split band of medium intensity in the range of 1650 cm−1 (Figure 7), which well reflects amide I (1653 and 1625 cm−1) and amide II (1553 cm−1) bands in the spectra of protein molecules. In the same region (1597, 1510 cm−1) lies the bands of skeletal vibrations of aromatic nitrogenous nucleotide bases. (3) A weak band in the region of 1408 cm−1 shifted relative to the 1410 cm−1 band from the spectrum of the original siloxane membrane also always appears in the IR spectra of polymeric amino acids. For the MDK-3 membrane, spectral signatures of a thin layer of protein and nucleic polymers on the surface are also observed.
Moreover, in the case of the MDK-3 membrane, there is a maximum loss of intensity of almost all bands in the initial membrane spectrum, and the band from the carbonyl group also loses more than half of its intensity, although the band from the polypeptide chain has the same intensity as in the spectra of PHeptMS and PDecMS membranes after fouling.
Thus, the spectra of all investigated membranes after exposure in the fermentation medium identify signs of polypeptide chain and nucleotides characterizing the structure of any microorganisms.
The affinity of a polymer to a liquid is determined by solubility parameters, which are a quantitative measure of intermolecular interactions and are calculated by the formula:
s p = ( δ d , s δ d , p ) 2 + ( δ p , s δ p , p ) 2 + ( δ h , s δ h , p ) 2 ,
where ∆sp is the Hansen solubility parameter for the organic solvents (s) and polymer (p) interactions, and δd, δp, and δh refer to the dispersion solubility parameter, polar solubility parameter, and hydrogen bonding solubility parameter, respectively.
Based on the data [57] and calculated group contributions for polyalkylsiloxanes [58], the parameters of the remote interaction were found (Table 2).
The distance parameters for PHeptMS and PDecMS polymers are quite close (Table 2). The least interaction with polyalkylsiloxanes in the studied series of solvents is observed for water. It is worth noting that the distance parameter of water-PHeptMS interaction is lower than that of water-PDecMS. The low solvent-polymer interaction distance parameter shows the high affinity of acetone to the membrane in the case of both PHeptMS and PDecMS. Because of the high affinity of acetone to the membrane, the membrane can swell. Thus, the polymer chains become more open to diffusion of molecules and an increase in the partial flux of acetone. Butanol is the next most compatible with PDecMS and PHeptMS. In addition, butanol is only partially miscible with water, unlike acetone and ethanol, which have unlimited solubility in water. This is caused by rather weak forces holding butanol molecules in water, which promote its penetration through a membrane [59]. This also results in a smaller contact wetting angle of the membrane by butanol in comparison with acetone. For ethanol, the distance interaction parameter is higher than that for butanol and acetone. Thus, the flow of ethanol across membranes will be less than that of acetone and n-butanol.
The obtained correlations of the solvent-polymer interaction parameters correlate well with the data of the contact boundary wetting angles of PHeptMS and PDecMS with respect to water, acetone, ethanol, and butanol (Table 3). Contact wetting angles of commercial MDK-3 membrane are also presented in the table.
As can be seen from the data presented in Table 3, the contact wetting angle by water for PHeptMS and PDecMS membranes was 103° and 106°, respectively. With the increase in the length of the side radical from C7 to C10, the value of the contact angle tends to increase. This indicates an increase in their hydrophobicity as well as the remote interaction parameter. The MDK-3 Carbosil© commercial membrane in the studied membrane series has the lowest hydrophobicity (88°); thus, one should expect the highest water flux through it and the lowest alcohol/water selectivity among the membranes under study. The membranes studied have a high degree of wetting of the target components (θ < 90°). It is worth noting that the lowest contact angles with acetone, ethanol, and butanol were shown by the PHeptMS membrane. The low contact angles of the PHeptMS membrane predict high separation factors in the ABE-fermentation broth pervaporation process. Moreover, the obtained data correlate well with the data of [43], which demonstrated high selectivity of separation of butanol from the mixture (94) with water. The PHeptMS membrane material in comparison with PDMS and polyoctylmethylsiloxane (POMS) demonstrated the highest selectivity of separation of C2–C4 alcohols from water.
After exposure of the selective layer of the membranes in the fermentation broth, a change in the contact wetting angle with respect to the studied substances (water, butanol, acetone, and ethanol) was observed. The general trend was a decrease in the contact wetting angle for all solvents. The largest relative drop in the angle was recorded for water: by 15, 12, and 17% for MDK-3, PHeptMS, and PDecMS, respectively. The observed phenomenon correlates with the presence of a peptide film on the surface of the membrane selective layer after contact with the broth, as shown by SEM, EDX, and IR data. Such a film appears to reduce the hydrophobicity of the surface of polysiloxane membranes and increase their affinity to the ABE mixture. Thus, the water contact angle for the MDK-3 membrane decreased by about 1.2 times compared to the value for the original membrane. For the PDecMS membrane, a stronger decrease in hydrophobicity was observed—the water contact angle decreased by 17% compared to the pure surface of the selective layer. More stable values of the contact angle of wetting after contact with the fermentation medium were characteristic of the PHeptMS membrane. This was characterized by a decrease in the value of the contact angle of wetting with water by about 12% compared to the original membrane.

3.2. Comparison of the Pervaporation Properties of Composite Membranes before and after Exposure to ABE-Fermentation Broth

To assess the effect of membrane clogging on its transport properties, the PHeptMS, PDecMS, and MDK-3 composite membranes were studied in the process of pervaporation separation of real fermentation broth before and after continuous contact with it for one month. The difference between the ABE-fermentation broth and the model solution lies not only in the different density, pH, and viscosity, but also in the presence or absence of inorganic salts, glucose, active and inactive microbial cells, and some other metabolic compounds. It was shown in [60,61] that the presence of such components in the fermentation broth contributes to an increase in the activity of the ABE-fermentation broth. A comparison of the partial fluxes of the components and separation factors at 30 °C for the initial and contaminated membranes in the separation of the real ABE-fermentation broth is shown in Figure 8.
The total permeate flux for the PHeptMS, PDecMS, and MDK-3 composite membranes at separation of the model mixture was 0.32, 0.44, and 0.77 kg/m2·h, respectively. The flux for MDK-3 was almost twice as high, which correlates well with the difference in the thicknesses of the selective layers of the membranes studied. It is worth noting that the greatest contribution to the total flux for all membranes was made by water. This fact is related to the low concentration of organic components in the model fermentation mixture. The partial flux of water for the MDK-3 membrane was 2–4 times higher than that for the PHeptMS and PDecMS membranes. In addition, the partial fluxes of acetone, ethanol, and n-butanol were comparable. The butanol flux showed the maximum values among the extracted organic components for all the membranes studied. This experimental picture is well reflected in the data of Figure 9b. Thus, acetone/water and n-butanol/water separation factors are 1.5–2 times higher for the initial PHeptMS and PDecMS membranes than for MDK-3. The ethanol/water partition factors are small and comparable for all membranes studied. The data obtained are in agreement with the values of contact wetting angles and distance interaction parameters.
When passing to contaminated membranes, the picture changes. Total (up to 0.12, 0.06, and 0.39 kg/m2·h for PHeptMS, PDecMS, and MDK-3, respectively) and partial fluxes for all membranes decrease (Figure 8a). This confirms the EDX and IR data indicating the formation of a film of proteins and peptides on the surface of the selective layer. Thus, microbial cells remaining in the fermentation broth were previously adsorbed on the surface of the selective membrane layer and prevented the transport of permeate through it. Moreover, some compounds from the yeast extract contained in the ABE-fermentation broth can be adsorbed on the surface of the composite membrane, which leads to a decrease in its hydrophobicity [62,63]. A similar assumption was made in the present article in the analysis of contact wetting angle data. Moreover, all the membranes were characterized by a decrease in water flow after their contamination (Figure 8a).
The negative effect of membrane contamination on its separation properties is mainly observed for the PDecMS membrane. It is characterized by the highest surface contamination (Figure 4b), a decrease in flux, and a decrease in the separation factor. Apparently, on this membrane due to its increased affinity for organic substances, the adsorbed protein layer was the largest, which significantly increased the hydrophilicity of the selective layer and hindered the diffusion transfer through the membrane as a whole. For the PHeptMS and MDK-3 membranes, the picture is significantly different.
For the PHeptMS membrane, a decrease in total and partial fluxes was observed. However, the ABE/water partition factors, on the contrary, increased. Thus, the n-butanol/water separation factor increased from 33 for the initial membrane to 41 for the membrane after clogging. By comparison, for PDecMS and MDK-3 membranes, a slight decrease in the butanol/water separation factor was observed: from 27 to 21 for PDecMS and from 14 to 13 for MDK-3. In the case of the MDK-3 membrane, there was an increase in the acetone/water separation factor for the membrane after exposure compared to the initial membrane. Apparently, differences in the chemical structure of the membranes studied strongly influence the nature of the substances adsorbed from the fermentation broth. This in turn changes the interaction between the contaminated surfaces to the components of the fermentation mixture.
The deterioration in the transport characteristics of the membranes after contact with the fermentation broth is clearly traced in the change in their PSI (Figure 9). PSI values decreased in the butanol ˃ acetone ˃ ethanol series. The PHeptMS membrane retained the maximum PSI value for n-butanol among the membranes studied, even after contact with the fermentation broth. The higher PSI value of acetone for the MDK-3 membrane after fouling, compared to the others, may be due to sorption of particles contributing to the membrane’s selectivity for ketone release.
The tendency of increasing PSI in the ethanol-acetone-butanol series was maintained with increasing separation temperature of the mixture. For example, the PSI values for butanol, acetone, and ethanol were 4.3, 1.5, and 0.03, respectively, when separating the ABE-fermentation broth at 60 °C using a PHeptMS membrane.

3.3. Study of the Effect of Temperature on the Primary Separation of ABE-Fermentation Broth

The temperature dependences of the permeate flux for PDecMS, PHeptMS, and MDK-3 membranes after contact with the ABE-fermentation broth during separation of the real ABE broth are shown in Figure 10. As the temperature of the separated mixture increased, the permeate flux tended to increase for all membranes, which was associated with an increase in the diffusion rate of penetrants through the selective layer. As a result of increasing temperature, the vapor pressure of organic substances and water increases, which leads to a higher driving force of permeate. In general, all membranes were characterized by a more than twofold increase in the flux of organic component when increasing the temperature from 30 to 60 °C. PDecMS was characterized by a slight increase in water flux from 0.04 to 0.15 kg/(m2·h) with increasing temperature, compared to the MDK-3 membrane, for which the water flux increased from 0.29 (30 °C) to 1.12 (60 °C). The MDK-3 membrane was characterized by a sharper (almost linear) increase in fluxes of both organic components and water.
The temperature dependences of the separation factors for the studied membranes after exposure to the fermentation broth do not look as unambiguous. The values of the separation factor at different temperatures are presented in Table 4.
In the case of PHeptMS and MDK-3 membranes, a gradual increase in the n-butanol/water and acetone/water separation factors with increasing temperature was observed, while the acetone/water separation factor passed through a minimum (40 °C). In the case of PDecMS membranes, a decrease in the separation factors was observed with increasing temperature, except for the ethanol/water separation factor, which passed through the minimum at 45 °C. Such differences may be related to the mutual influence of the peptide and selective layer of the membrane on the transport. It is known that diffusion increases with increasing temperature due to increased molecular mobility, but sorption decreases. Accordingly, the increase in the separation factors of PHeptMS and MDK-3 membranes may be related to the desorption of protein components from the surface of the selective layer. The maximum values of the organic component/water separation factor in the temperature range studied were observed for the PHeptMS membrane—74 for butanol, 61 for acetone, and 10 for ethanol at 60 °C. These values exceed the highest separation factors of the PDMS membrane obtained in [64], which were 33.36, 19.81, and 10.07 for acetone, butanol, and ethanol, respectively. When using a composite PDMS/ceramic membrane in the separation of the ABE-fermentation broth at 37 °C, the authors of [65] achieved a butanol/water separation factor of 27.3. The tendency of the permeate flow and the organic component/water separation factor to increase with increasing temperature was also observed in [66], where a POMS-based membrane was used in the separation of a model ABE solution.
In order to normalize the results obtained to the driving force of the process, the obtained results were recalculated using such quantities for permeability and selectivity. The permeability of n-butanol, acetone, and ethanol for PHeptMS and PDecMS membranes after contact with the ABE-fermentation broth tends to decrease with increasing temperature and then to increase (Figure 11). The increase in the permeability coefficient of the components at 60 °C correlates well with the increase in the partial fluxes at this temperature. A similar tendency can also be caused by entrainment of contaminants by raw material flow. The MDK-3 membrane was characterized by the opposite tendency: increase in permeability at 40 °C and then decrease at 60 °C.
Acetone was the most permeable of the separated components of the mixture. However, the permeability of water was higher for PDecMS and MDK-3 membranes with increasing temperature. This result indicates the predominant diffusive contribution to the total transport through the membrane at a given temperature. The permeabilities of acetone and butanol were fairly similar for the PHeptMS membrane, while the permeability of water, in turn, was lower and closer in value to the permeability of ethanol. Due to the observed changes in the permeability of the components, we can note the highest selectivity for acetone, and then for butanol and ethanol. Along with the increase in permeability at 60 °C, an increase in selectivity was also observed for the PHeptMS membrane, apparently caused by desorption of protein compounds and a decrease in membrane hydrophilicity. Such low permeability values in the series of all temperatures studied, 54 mol/m2·h·bar for butanol, 53 mol/m2·h·bar for acetone, and 24 mol/m2·h·bar for ethanol at 30 °C, and the decrease in selectivity with increasing temperature for the PDecMS membrane, may be due to the strongest fouling of this membrane (Figure 12).

3.4. Dynamics of Contamination of the PHeptMS Membrane during Separation of ABE-Fermentation Broth

Based on the results obtained above, the composite PHeptMS membrane was of the greatest interest, since it demonstrated not only a high separation ability of the ABE broth, but also was least susceptible to fouling. Nevertheless, it is interesting to trace the dynamics of changes in the transport and separation characteristics of this membrane in the process of pervaporation of the real ABE-fermentation broth under conditions of continuous contact of the PHeptMS membrane with the medium to be separated. Between experiments of the separation of the real ABE broth, the membrane was in the module under a layer of broth. Pervaporation separation of ABE-fermentation broth was carried out at 30 °C. The data obtained are shown in Figure 13.
When separating the real ABE-fermentation broth for 6 h, the total permeate flux decreased from 0.203 to 0.175 kg/(m2·h) and continued to decrease during the first 24 h of membrane operation. This behavior appears to be related to the beginning of the precipitation of protein on the membrane surface. This is also evidenced by the increased selectivity of the membrane for acetone. As shown earlier, clogging of the membrane surface with proteins leads to an increase in the contact wetting angle by acetone and leads to an increase in the selectivity of its transport across membranes. Figure 13a shows that the performance of the PHeptMS composite membrane in separating the ABE-fermentation mixture from 28 to 150 h was generally stable, and this membrane exhibited a high and constant total flux average of 0.162 kg/(m2·h) and average separation factors of 49.5, 44.8, and 6.3 for acetone, butanol, and ethanol, respectively. After 150 h of continuous contact of the membrane with the fermentation broth, a drop in the membrane selectivity for n-butanol and acetone was observed. This decrease is caused by a decrease in membrane permeability to n-butanol and acetone, as well as a slight increase in water permeability through the membrane. For ethanol, no noticeable changes in transport were observed. Apparently, after 150 h of contact, there is a significant increase in the thickness of the sludge layer, which directly affects the transport and separation characteristics of the membrane.
Nevertheless, the values obtained exceed the characteristics of the PDMS/PVDF membrane studied for 100 h of separation of the ABE-fermentation broth. Thus, it was shown in [67] that the PDMS-PVDF composite membrane exhibited an average total flux of 0.1105 kg/(m2·h) and average separation factors of 21.8, 19.1, and 5.2 for acetone, butanol, and ethanol, respectively. In [68], PDMS- and POMS-based membranes were used to separate the ABE-fermentation broth. For the PDMS membrane, the separation factors for acetone, butanol, and acetone at 37 °C were 21.6, 18.8, and 8.9, respectively. The POMS-based membrane at 32 °C showed separation factors for acetone, butanol, and ethanol of 27.2, 19.7, and 7.8, respectively. The dynamics of the effect of the contact time of the membrane with the fermentation medium on its transport properties were not studied. The PHeptMS membrane obtained in this work before contact with the broth in the separation of the ABE-fermentation broth showed better separation factor values for butanol and acetone, being 46.3 for butanol, 35.9 for acetone, and 6.0 for ethanol. The value of the total permeate flux for the PHeptMS membrane during the first 2 h of the experiment was 0.203 kg/(m2·h). After 216 h of separation of the ABE-fermentation mixture, the value of the total permeate flux decreased to 0.158 kg/(m2·h). Thus, the observed 1.28-fold decrease in the total flux during separation of the ABE-fermentation mixture suggests that its modification is needed to improve stability during long-term contact with the fermentation mixture. For example, fluorination of the surface of the selective layer leads to a decrease in the adsorption of the fermentation mixture components on the membrane surface. This approach can be considered further as potentially promising for stabilizing the pervaporation properties of polyalkylmethylsiloxanes over time.

4. Conclusions

In this work the effect of long-term contact with real ABE-fermentation broth of composite membranes made of PDecMS and PHeptMS, as well as the commercial membrane MDK-3, was studied for the first time. It was shown that for all investigated membranes after one month of contact with broth there was a clogging of the surface of the selective layer. Based on the data of elemental analysis and infrared spectroscopy, the protein nature of deposits (microorganisms, products of their life activity) was revealed. For all membranes, a decrease in contact angles for water, n-butanol, acetone, and ethanol was observed after fouling. The smallest change in wetting angles was observed for the PHeptMS membrane.
A comparison of the transport and separation properties of the membranes before and after a one-month exposure in the broth was studied in the pervaporation separation of the real ABE-fermentation mixture. It was noted that the flux of clogged membranes was lower than that of fresh membranes. The ABE/water separation factors of PDecMS and MDK-3 membranes decreased after clogging, which was due to an increase in the hydrophilicity of the surface of these membranes. The opposite trend was observed for PHeptMS. For all investigated membranes, a decrease in their PSI was observed after fouling as compared to the initial membranes. Nevertheless, the PHeptMS membrane before and after clogging retained the maximum PSI value for n-butanol among the studied membranes. As the temperature of the separated ABE mixture increased, an increase in permeate flux was observed for all membranes. The maximum values of the organic component/water separation factor were observed for the PHeptMS membrane—74 for butanol, 61 for acetone, and 10 for ethanol at 60 °C.
For the most selective PHeptMS membrane, the dynamics of its contamination during vacuum pervaporation of real ABE-fermentation broth for 216 h were investigated. During the first 28 h of the experiment, the partial fluxes of the mixture components decreased. Subsequent stabilization of water, n-butanol, and ethanol fluxes, as well as some increase in acetone flux, indicate contamination of the selective layer surface. For the next ~100 h, the performance of the composite membrane was stable and exhibited an average total flux of 0.16 kg/(m2·h) and average separation factors of 49.5, 44.8, and 6.3 for acetone, butanol, and ethanol, respectively. After 150 h of contact with the fermentation mixture, a 10% decrease in the selectivity of PHeptMS for all components was observed due to an increase in water permeability through the membrane. The overall decrease in the total flux during separation of the ABE-fermentation mixture by a factor of 1.28 indicates the necessity of its modification to increase its stability during long-term contact with the fermentation mixture.

Author Contributions

Conceptualization, E.A.G. and I.L.B.; Data curation, T.N.R., O.V.A. and G.N.B.; Formal analysis, T.N.R., E.A.G. and O.V.A.; Funding acquisition, G.S.G.; Investigation, T.N.R., E.A.G. and O.V.A.; Methodology, T.N.R., E.A.G., G.N.B., G.S.G. and I.L.B.; Project administration, G.S.G.; Visualization, T.N.R. and A.V.V.; Writing—original draft, T.N.R. and E.A.G.; Writing—review and editing, E.A.G., G.S.G. and A.V.V. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Russian Science Foundation, grant number 22-79-10332, https://rscf.ru/project/22-79-10332/ (accessed on 19 February 2023).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

This work was performed using the equipment of the Shared Research Center “Analytical center of deep oil processing and petrochemistry of TIPS RAS”. Authors thanks to Danila Bakhtin to provide SEM images. Alexey Volkov acknowledges KAUST for support.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Dürre, P. Biobutanol: An attractive biofuel. Biotechnol. J. Healthc. Nutr. Technol. 2007, 2, 1525–1534. [Google Scholar] [CrossRef]
  2. Maity, S.K. Opportunities, recent trends and challenges of integrated biorefinery: Part I. Renew. Sustain. Energy Rev. 2015, 43, 1427–1445. [Google Scholar] [CrossRef] [Green Version]
  3. Delidovich, I.; Hausoul, P.J.; Deng, L.; Pfützenreuter, R.; Rose, M.; Palkovits, R. Alternative monomers based on lignocellulose and their use for polymer production. Chem. Rev. 2016, 116, 1540–1599. [Google Scholar] [CrossRef]
  4. Akiyama, M.; Tsuge, T.; Doi, Y. Environmental life cycle comparison of polyhydroxyalkanoates produced from renewable carbon resources by bacterial fermentation. Polym. Degrad. Stab. 2003, 80, 183–194. [Google Scholar] [CrossRef]
  5. Wilke, D. Chemicals from biotechnology: Molecular plant genetics will challenge the chemical and the fermentation industry. Appl. Microbiol. Biotechnol. 1999, 52, 135–145. [Google Scholar] [CrossRef]
  6. Mahalingam, L.; Abdulla, R.; Sani, S.A.; Sabullah, M.K.; Faik, A.A.M.; Misson, M. Lignocellulosic Biomass–A Sustainable Feedstock for Acetone-Butanol-Ethanol Fermentation. Period. Polytech. Chem. Eng. 2022, 66, 279–296. [Google Scholar] [CrossRef]
  7. Sindhu, R.; Binod, P.; Pandey, A.; Ankaram, S.; Duan, Y.; Awasthi, M.K. Biofuel production from biomass: Toward sustainable development. In Current Developments in Biotechnology and Bioengineering; Elsevier: Amsterdam, The Netherlands, 2019; pp. 79–92. [Google Scholar]
  8. Lynd, L.R.; Cushman, J.H.; Nichols, R.J.; Wyman, C.E. Fuel ethanol from cellulosic biomass. Science 1991, 251, 1318–1323. [Google Scholar] [CrossRef]
  9. da Silva Trindade, W.R.; dos Santos, R.G. Review on the characteristics of butanol, its production and use as fuel in internal combustion engines. Renew. Sustain. Energy Rev. 2017, 69, 642–651. [Google Scholar] [CrossRef]
  10. Nanda, S.; Golemi-Kotra, D.; McDermott, J.C.; Dalai, A.K.; Gökalp, I.; Kozinski, J.A. Fermentative production of butanol: Perspectives on synthetic biology. N. Biotechnol. 2017, 37, 210–221. [Google Scholar] [CrossRef]
  11. Nakagawa, H.; Harada, T.; Ichinose, T.; Takeno, K.; Matsumoto, S.; Kobayashi, M.; Sakai, M. Biomethanol production and CO2 emission reduction from forage grasses, trees, and crop residues. JARQ 2007, 41, 173–180. [Google Scholar] [CrossRef] [Green Version]
  12. Kang, A.; Lee, T.S. Converting sugars to biofuels: Ethanol and beyond. Bioengineering 2015, 2, 184–203. [Google Scholar] [CrossRef] [Green Version]
  13. García, V.; Päkkilä, J.; Ojamo, H.; Muurinen, E.; Keiski, R.L. Challenges in biobutanol production: How to improve the efficiency? Renew. Sustain. Energy Rev. 2011, 15, 964–980. [Google Scholar] [CrossRef]
  14. Ezeji, T.C.; Qureshi, N.; Blaschek, H.P. Production of acetone, butanol and ethanol by Clostridium beijerinckii BA101 and in situ recovery by gas stripping. World J. Microbiol. Biotechnol. 2003, 19, 595–603. [Google Scholar] [CrossRef]
  15. Qureshi, N.; Blaschek, H.P. Recent advances in ABE fermentation: Hyper-butanol producing Clostridium beijerinckii BA101. J. Ind. Microbiol. Biotechnol. 2001, 27, 287–291. [Google Scholar] [CrossRef]
  16. Schiel-Bengelsdorf, B.; Montoya, J.; Linder, S.; Dürre, P. Butanol fermentation. Environ. Technol. 2013, 34, 1691–1710. [Google Scholar] [CrossRef]
  17. Geddes, C.C.; Nieves, I.U.; Ingram, L.O. Advances in ethanol production. Curr. Opin. Biotechnol. 2011, 22, 312–319. [Google Scholar] [CrossRef] [Green Version]
  18. Kujawska, A.; Kujawski, J.; Bryjak, M.; Kujawski, W. ABE fermentation products recovery methods—A review. Renew. Sustain. Energy Rev. 2015, 48, 648–661. [Google Scholar] [CrossRef]
  19. Liu, S.; Qureshi, N. How microbes tolerate ethanol and butanol. New Biotechnol. 2009, 26, 117–121. [Google Scholar] [CrossRef]
  20. Luyben, W.L. Pressure-swing distillation for minimum-and maximum-boiling homogeneous azeotropes. Ind. Eng. Chem. Res. 2012, 51, 10881–10886. [Google Scholar] [CrossRef]
  21. Levario, T.J.; Dai, M.; Yuan, W.; Vogt, B.D.; Nielsen, D.R. Rapid adsorption of alcohol biofuels by high surface area mesoporous carbons. Microporous Mesoporous Mater. 2012, 148, 107–114. [Google Scholar] [CrossRef]
  22. Evans, P.J.; Wang, H.Y. Enhancement of butanol formation by Clostridium acetobutylicum in the presence of decanol-oleyl alcohol mixed extractants. Appl. Environ. Microbiol. 1988, 54, 1662–1667. [Google Scholar] [CrossRef] [Green Version]
  23. Oudshoorn, A.; Van Der Wielen, L.A.; Straathof, A.J. Assessment of options for selective 1-butanol recovery from aqueous solution. Ind. Eng. Chem. Res. 2009, 48, 7325–7336. [Google Scholar] [CrossRef]
  24. Vane, L.M. A review of pervaporation for product recovery from biomass fermentation processes. J. Chem. Technol. Biotechnol. 2005, 80, 603–629. [Google Scholar] [CrossRef]
  25. Rozicka, A.; Niemistö, J.; Keiski, R.L.; Kujawski, W. Apparent and intrinsic properties of commercial PDMS based membranes in pervaporative removal of acetone, butanol and ethanol from binary aqueous mixtures. J. Membr. Sci. 2014, 453, 108–118. [Google Scholar] [CrossRef]
  26. Niemistö, J.; Kujawski, W.; Keiski, R.L. Pervaporation performance of composite poly (dimethyl siloxane) membrane for butanol recovery from model solutions. J. Membr. Sci. 2013, 434, 55–64. [Google Scholar] [CrossRef]
  27. Liu, Q.; Li, W.; Wang, H.; Liu, L. A facile method of using sulfobetaine-containing copolymers for biofouling resistance. J. Appl. Polym. Sci. 2014, 131, 40789-1–40789-8. [Google Scholar] [CrossRef]
  28. García, V.; Pongrácz, E.; Muurinen, E.; Keiski, R.L. Recovery of n-butanol from salt containing solutions by pervaporation. Desalination 2009, 241, 201–211. [Google Scholar] [CrossRef]
  29. Tan, H.; Wu, Y.; Li, T. Pervaporation of n-butanol aqueous solution through ZSM-5-PEBA composite membranes. J. Appl. Polym. Sci. 2013, 129, 105–112. [Google Scholar] [CrossRef]
  30. Tong, C.; Bai, Y.; Wu, J.; Zhang, L.; Yang, L.; Qian, J. Pervaporation recovery of acetone-butanol from aqueous solution and fermentation broth using HTPB-based polyurethaneurea membranes. Sep. Sci. Technol. 2010, 45, 751–761. [Google Scholar] [CrossRef]
  31. Borisov, I.L.; Golubev, G.S.; Vasilevsky, V.P.; Volkov, A.V.; Volkov, V.V. Novel hybrid process for bio-butanol recovery: Thermopervaporation with porous condenser assisted by phase separation. J. Membr. Sci. 2017, 523, 291–300. [Google Scholar] [CrossRef]
  32. Atlaskin, A.A.; Petukhov, A.N.; Yanbikov, N.R.; Salnikova, M.E.; Sergeeva, M.S.; Vorotyntsev, V.M.; Vorotyntsev, I.V. Evaluation of the absorbing pervaporation technique for ammonia recovery after the Haber process. Chem. Proc. Eng. 2018, 39, 323–333. [Google Scholar]
  33. Otvagina, K.V.; Penkova, A.V.; Dmitrenko, M.E.; Kuzminova, A.I.; Sazanova, T.S.; Vorotyntsev, A.V.; Vorotyntsev, I.V. Novel composite membranes based on chitosan copolymers with polyacrylonitrile and polystyrene: Physicochemical properties and application for pervaporation dehydration of tetrahydrofuran. Membranes 2019, 9, 38. [Google Scholar] [CrossRef] [Green Version]
  34. Liu, G.; Wei, W.; Jin, W. Pervaporation membranes for biobutanol production. ACS Sustain. Chem. Eng. 2014, 2, 546–560. [Google Scholar] [CrossRef]
  35. Nakayama, S.; Morita, T.; Negishi, H.; Ikegami, T.; Sakaki, K.; Kitamoto, D. Candida krusei produces ethanol without production of succinic acid; a potential advantage for ethanol recovery by pervaporation membrane separation. FEMS Yeast Res. 2008, 8, 706–714. [Google Scholar] [CrossRef] [Green Version]
  36. Guan, P.; Ren, C.; Shan, H.; Cai, D.; Zhao, P.; Ma, D.; Si, Z. Boosting the pervaporation performance of PDMS membrane for 1-butanol by MAF-6. Colloid Polym. Sci. 2021, 299, 1459–1468. [Google Scholar] [CrossRef]
  37. Li, S.Y.; Srivastava, R.; Parnas, R.S. Separation of 1-butanol by pervaporation using a novel tri-layer PDMS composite membrane. J. Membr. Sci. 2010, 363, 287–294. [Google Scholar] [CrossRef]
  38. Azimi, H.; Tezel, H.; Thibault, J. Optimization of the in situ recovery of butanol from ABE fermentation broth via membrane pervaporation. Chem. Eng. Res. Des. 2019, 150, 49–64. [Google Scholar] [CrossRef]
  39. Zhu, H.; Liu, G.; Yuan, J.; Chen, T.; Xin, F.; Jiang, M.; Jin, W. In-situ recovery of bio-butanol from glycerol fermentation using PDMS/ceramic composite membrane. Sep. Purif. Technol. 2019, 229, 115811. [Google Scholar] [CrossRef]
  40. Grushevenko, E.A.; Borisov, I.L.; Volkov, A.V. Highly selective polysiloxane membranes for the separation of gases and liquids (review). Pet. Chem. 2021, 61, 571–590. [Google Scholar] [CrossRef]
  41. Lee, J.Y.; Lee, J.S.; Lee, J.H. High performance and thermally stable PDMS pervaporation membranes prepared using a phenyl-containing tri-functional crosslinker for n-butanol recovery. Sep. Purif. Technol. 2020, 235, 116142. [Google Scholar] [CrossRef]
  42. Bennett, M.; Brisdon, B.J.; England, R.; Field, R.W. Performance of PDMS and organofunctionalised PDMS membranes for the pervaporative recovery of organics from aqueous streams. J. Membr. Sci. 1997, 137, 63–88. [Google Scholar] [CrossRef]
  43. Grushevenko, E.A.; Podtynnikov, I.A.; Golubev, G.S.; Volkov, V.V.; Borisov, I.L. Polyheptylmethylsiloxane-new material for pervaporation extraction of oxygenates from water. Membr. Membr. Technol. 2018, 8, 334–342. [Google Scholar]
  44. Grushevenko, E.A.; Podtynnikov, I.A.; Borisov, I.L. Highly selective pervaporation membrane for the isolation of 1-butanol from wastewater. J. Appl. Chem. 2019, 92, 1488–1496. [Google Scholar]
  45. Qureshi, N.; Blaschek, H.P. Production of acetone butanol ethanol (ABE) by a hyper-producing mutant strain of Clostridium beijerinckii BA101 and recovery by pervaporation. Biotechnol. Prog. 1999, 15, 594–602. [Google Scholar] [CrossRef]
  46. Qureshi, N.; Blaschek, H.P. Fouling studies of a pervaporation membrane with commercial fermentation media and fermentation broth of hyper-butanol-producing Clostridium beijerinckii BA101. Sep. Sci. Technol. 1999, 34, 2803–2815. [Google Scholar] [CrossRef]
  47. Dubreuil, M.F.; Vandezande, P.; Van Hecke, W.H.; Porto-Carrero, W.J.; Dotremont, C.T. Study on ageing/fouling phenomena and the effect of upstream nanofiltration on in-situ product recovery of n-butanol through poly [1-(trimethylsilyl)-1-propyne] pervaporation membranes. J. Membr. Sci. 2013, 447, 134–143. [Google Scholar] [CrossRef]
  48. Liu, G.; Wei, W.; Wu, H.; Dong, X.; Jiang, M.; Jin, W. Pervaporation performance of PDMS/ceramic composite membrane in acetone butanol ethanol (ABE) fermentation–PV coupled process. J. Membr. Sci. 2011, 373, 121–129. [Google Scholar] [CrossRef]
  49. Volkov, V.; Borisov, I.; Golubev, G.; Vasilevsky, V.; Matveev, D.; Bondarenko, G.; Volkov, A. Sorption-assisted thermopervaporation method for organics recovery from ABE fermentation broth. J. Chem. Technol. Biotechnol. 2020, 95, 40–51. [Google Scholar] [CrossRef]
  50. Qureshi, N.; Meagher, M.M.; Huang, J.; Hutkins, R.W. Acetone butanol ethanol (ABE) recovery by pervaporation using silicalite–silicone composite membrane from fed-batch reactor of Clostridium acetobutylicum. J. Membr. Sci. 2001, 187, 93–102. [Google Scholar] [CrossRef]
  51. Knozowska, K.; Kujawska, A.; Kujawa, J.; Kujawski, W.; Bryjak, M.; Chrzanowska, E.; Kujawski, J.K. Performance of commercial composite hydrophobic membranes applied for pervaporative reclamation of acetone, butanol, and ethanol from aqueous solutions: Binary mixtures. Sep. Purif. Technol. 2017, 188, 512–522. [Google Scholar] [CrossRef]
  52. Zhu, H.; Li, X.; Pan, Y.; Liu, G.; Wu, H.; Jiang, M.; Jin, W. Fluorinated PDMS membrane with anti-biofouling property for in-situ biobutanol recovery from fermentation-pervaporation coupled process. J. Membr. Sci. 2020, 609, 118225. [Google Scholar] [CrossRef]
  53. Lee, S.A.; Oh, S.H.; Lee, W. The effect of direct fluorination of polydimethylsiloxane films on their surface properties. J. Colloid Interface Sci. 2009, 332, 461–466. [Google Scholar] [CrossRef]
  54. Zheng, P.Y.; Li, X.Q.; Wu, J.K.; Wang, N.X.; Li, J.; An, Q.F. Enhanced butanol selectivity of pervaporation membrane with fluorinated monolayer on polydimethylsiloxane surface. J. Membr. Sci. 2018, 548, 215–222. [Google Scholar] [CrossRef]
  55. Borisov, I.L.; Golubev, G.S.; Patrusheva, E.V.; Sinekoy, S.P. Intensification of acetone-butanol-ethanol fermentation via products recovery: Thermopervaporation assisted by phase separation. Chem. Eng. Trans. 2018, 64, 43–48. [Google Scholar]
  56. Brunetti, A.; Zito, P.F.; Borisov, I.; Grushevenko, E.; Volkov, V.; Volkov, A.; Barbieri, G. CO2 separation from humidified ternary gas mixtures using a polydecylmethylsiloxane composite membrane. Fuel Process. Technol. 2020, 210, 106550. [Google Scholar] [CrossRef]
  57. Hansen, C.M. Hansen Solubility Parameters—A User’s Handbook; CRC Press: Boca Raton, FL, USA, 2007. [Google Scholar]
  58. Van Krevelen, D.W. Properties of Polymers: Their Correlation with Chemical Structure: Their Numerical Estimation and Prediction from Additive Group Contributions, 4th ed.; Elsevier: Oxford, UK, 2009; p. 1030. [Google Scholar]
  59. Liu, F.; Liu, L.; Feng, X. Separation of acetone–butanol–ethanol (ABE) from dilute aqueous solutions by pervaporation. Sep. Purif. Technol. 2005, 42, 273–282. [Google Scholar] [CrossRef]
  60. Li, J.; Polka, H.M.; Gmehling, J. A gE model for single and mixed solvent electrolyte systems: 1. Model and results for strong electrolytes. Fluid Phase Equilibria 1994, 94, 89–114. [Google Scholar] [CrossRef]
  61. Lipnizki, F.; Hausmanns, S.; Field, R.W. Influence of impermeable components on the permeation of aqueous 1-propanol mixtures in hydrophobic pervaporation. J. Membr. Sci. 2004, 228, 129–138. [Google Scholar] [CrossRef]
  62. Ikegami, T.; Negishi, H.; Nakayama, S.; Kobayashi, G.; Sakaki, K. Pervaporative concentration of biobutanol from ABE fermentation broths by Clostridium saccharoperbutylacetonicum using silicone rubber-coated silicalite-1 membranes. Sep. Purif. Technol. 2014, 132, 206–212. [Google Scholar] [CrossRef]
  63. Huang, J.; Meagher, M.M. Pervaporative recovery of n-butanol from aqueous solutions and ABE fermentation broth using thin-film silicalite-filled silicone composite membranes. J. Membr. Sci. 2001, 192, 231–242. [Google Scholar] [CrossRef]
  64. Van Hecke, W.; Hofmann, T.; De Wever, H. Pervaporative recovery of ABE during continuous cultivation: Enhancement of performance. Bioresour. Technol. 2013, 129, 421–429. [Google Scholar] [CrossRef]
  65. Wu, H.; Chen, X.P.; Liu, G.P.; Jiang, M.; Guo, T.; Jin, W.Q.; Zhu, D.W. Acetone–butanol–ethanol (ABE) fermentation using Clostridium acetobutylicum XY16 and in situ recovery by PDMS/ceramic composite membrane. Bioprocess Biosyst. Eng. 2012, 35, 1057–1065. [Google Scholar] [CrossRef]
  66. Rom, A.; Esteve, D.; Friedl, A. Organophilic pervaporation of butanol from an aqueous solution with POMS. Chem. Eng. Trans. 2013, 35, 1315–1320. [Google Scholar]
  67. Li, Y.; Shi, J.; Liu, X.; Yu, Q.; Qiu, W.; Liu, L. Pervaporation Performance of Polydimethylsiloxane-Polyvinylidene Fluoride Composite Membrane for Butanol Recovery from Model Solutions and Acetone-Butanol-Ethanol Fermentation Broth. J. Biobased Mater. Bioenergy 2020, 14, 294–302. [Google Scholar] [CrossRef]
  68. Van Hecke, W.; De Wever, H. High-flux POMS organophilic pervaporation for ABE recovery applied in fed-batch and continuous set-ups. J. Membr. Sci. 2017, 540, 321–332. [Google Scholar] [CrossRef]
Figure 1. Scheme of the hydrosilylation reaction to produce cross-linked polyalkylmethylsiloxanes.
Figure 1. Scheme of the hydrosilylation reaction to produce cross-linked polyalkylmethylsiloxanes.
Applsci 13 03827 g001
Figure 2. Scheme of the vacuum pervaporation setup: 1—a container with a mixing device; 2—a gear pump; 3—a heat exchanger; 4—a membrane module; 5—traps for collecting permeate, placed in Dewar vessels with liquid nitrogen; 6—thermostat; 7—vacuum pump; 8—safety trap; I—initial separable mixture; II—retentate; III—permeate; IV—coolant.
Figure 2. Scheme of the vacuum pervaporation setup: 1—a container with a mixing device; 2—a gear pump; 3—a heat exchanger; 4—a membrane module; 5—traps for collecting permeate, placed in Dewar vessels with liquid nitrogen; 6—thermostat; 7—vacuum pump; 8—safety trap; I—initial separable mixture; II—retentate; III—permeate; IV—coolant.
Applsci 13 03827 g002
Figure 3. SEM images of cross sections of initial membranes: (a) MDK-3; (b) PDecMS; (c) PHeptMS.
Figure 3. SEM images of cross sections of initial membranes: (a) MDK-3; (b) PDecMS; (c) PHeptMS.
Applsci 13 03827 g003
Figure 4. Photographs of membranes before: (a) PDecMS; (c) PHeptMS; (e) MDK-3 and after one month of contact with ABE-fermentation broth: (b) PDecMS; (d) PHeptMS; (f) MDK-3.
Figure 4. Photographs of membranes before: (a) PDecMS; (c) PHeptMS; (e) MDK-3 and after one month of contact with ABE-fermentation broth: (b) PDecMS; (d) PHeptMS; (f) MDK-3.
Applsci 13 03827 g004aApplsci 13 03827 g004b
Figure 5. SEM images of the surface of the selective membrane layer before: (a) PDecMS; (c) PHeptMS; (e) MDK-3 and after one month of contact with ABE-fermentation broth: (b) PDecMS; (d) PHeptMS; (f) MDK-3.
Figure 5. SEM images of the surface of the selective membrane layer before: (a) PDecMS; (c) PHeptMS; (e) MDK-3 and after one month of contact with ABE-fermentation broth: (b) PDecMS; (d) PHeptMS; (f) MDK-3.
Applsci 13 03827 g005
Figure 6. IR spectra of the membranes.
Figure 6. IR spectra of the membranes.
Applsci 13 03827 g006
Figure 7. IR spectra of the membranes in the region of 3700–1400 cm−1.
Figure 7. IR spectra of the membranes in the region of 3700–1400 cm−1.
Applsci 13 03827 g007
Figure 8. Comparison of partial fluxes (a) and separation factors (b) at 30 °C of PHeptMS, PDecMS, and MDK-3 membranes before and after membrane contact with ABE-fermentation broth.
Figure 8. Comparison of partial fluxes (a) and separation factors (b) at 30 °C of PHeptMS, PDecMS, and MDK-3 membranes before and after membrane contact with ABE-fermentation broth.
Applsci 13 03827 g008
Figure 9. Comparison of PSI at 30 °C of PHeptMS, PDecMS, and MDK-3 membranes before and after membrane contact with ABE-fermentation broth.
Figure 9. Comparison of PSI at 30 °C of PHeptMS, PDecMS, and MDK-3 membranes before and after membrane contact with ABE-fermentation broth.
Applsci 13 03827 g009
Figure 10. Temperature dependences of permeate flow for butanol (a), ethanol (b), acetone (c), and water (d) for PDecMS, PHeptMS, and MDK-3 membranes after contact with the fermentation broth.
Figure 10. Temperature dependences of permeate flow for butanol (a), ethanol (b), acetone (c), and water (d) for PDecMS, PHeptMS, and MDK-3 membranes after contact with the fermentation broth.
Applsci 13 03827 g010
Figure 11. Temperature dependences of permeability for PHeptMS (a), PDecMS (b), and MDK-3 (c).
Figure 11. Temperature dependences of permeability for PHeptMS (a), PDecMS (b), and MDK-3 (c).
Applsci 13 03827 g011
Figure 12. Temperature dependences of selectivity for PHeptMS (a), PDecMS (b), and MDK-3 (c).
Figure 12. Temperature dependences of selectivity for PHeptMS (a), PDecMS (b), and MDK-3 (c).
Applsci 13 03827 g012
Figure 13. Dependence of partial fluxes of components (a), permeability (b), and selectivity (c) on the membrane contact time with ABE-fermentation broth.
Figure 13. Dependence of partial fluxes of components (a), permeability (b), and selectivity (c) on the membrane contact time with ABE-fermentation broth.
Applsci 13 03827 g013
Table 1. EDX analysis data of PHeptMS, PDecMS, and MDK-3 membranes before and after contact with the fermentation broth.
Table 1. EDX analysis data of PHeptMS, PDecMS, and MDK-3 membranes before and after contact with the fermentation broth.
MembraneConditionElement Weight Concentration, %
COSiNSPF
PHeptMSinitial50.714.335.0----
after ABE24.620.024.730.00.7--
PDecMSinitial54.212.133.7----
after ABE56.318.124.9--0.7-
MDK-3initial52.418.627.9---0.5
after ABE52.320.625.5-0.1-0.9
Table 2. Hansen solubility parameters.
Table 2. Hansen solubility parameters.
Component/PolymerδdδpδhΔs-PHeptMSΔs-PDecMS
Acetone15.510.47.08.258.95
Butanol16.05.715.811.4412.15
Ethanol15.88.819.415.8816.64
Water15.516.042.339.8340.57
PHeptMS16.772.544.83--
PDecMS16.771.974.26--
Table 3. Contact wetting angles of the membranes before and after exposure to the ABE-fermentation broth (θ, °).
Table 3. Contact wetting angles of the membranes before and after exposure to the ABE-fermentation broth (θ, °).
MembraneConditionΘ by WaterΘ by AcetoneΘ by EthanolΘ by Butanol
MDK-3initial88 ± 241 ± 223 ± 226 ± 2
after ABE74 ± 1.832 ± 121 ± 220 ± 1.3
PHeptMSinitial103 ± 231 ± 237 ± 223 ± 2
after ABE91 ± 225 ± 1.634 ± 1.421 ± 1.1
PDecMSinitial106 ± 236 ± 241 ± 228 ± 2
after ABE88 ± 1.324 ± 2.135 ± 1.223 ± 1.7
Table 4. Separation factor (β) organic component/water for PHeptMS, PDecMS, and MDK-3 membranes at different temperatures.
Table 4. Separation factor (β) organic component/water for PHeptMS, PDecMS, and MDK-3 membranes at different temperatures.
Membraneβ Butanol/Waterβ Acetone/Waterβ Ethanol/WaterTemperature, °C
PHeptMS41.545.35.530
59.740.85.640
74.460.910.560
PDecMS22.220.93.730
19.317.72.645
16.914.64.260
MDK-312.826.62.730
14.720.63.540
18.421.45.160
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Rokhmanka, T.N.; Grushevenko, E.A.; Arapova, O.V.; Bondarenko, G.N.; Golubev, G.S.; Borisov, I.L.; Volkov, A.V. Fouling of Polyalkylmethylsiloxane Composite Membranes during Pervaporation Separation of ABE-Fermentation Mixtures. Appl. Sci. 2023, 13, 3827. https://doi.org/10.3390/app13063827

AMA Style

Rokhmanka TN, Grushevenko EA, Arapova OV, Bondarenko GN, Golubev GS, Borisov IL, Volkov AV. Fouling of Polyalkylmethylsiloxane Composite Membranes during Pervaporation Separation of ABE-Fermentation Mixtures. Applied Sciences. 2023; 13(6):3827. https://doi.org/10.3390/app13063827

Chicago/Turabian Style

Rokhmanka, Tatyana N., Evgenia A. Grushevenko, Olga V. Arapova, Galina N. Bondarenko, George S. Golubev, Ilya L. Borisov, and Alexey V. Volkov. 2023. "Fouling of Polyalkylmethylsiloxane Composite Membranes during Pervaporation Separation of ABE-Fermentation Mixtures" Applied Sciences 13, no. 6: 3827. https://doi.org/10.3390/app13063827

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop