Synthesis and characterization of brominated poly(2,6-diphenyl-p-phenylene oxide)/SiO₂ nanocomposite membrane for CO₂/N₂ separation

Abstract

Synthesized poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp) and brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) are identified as new membranes for CO₂/N₂ separation and characterized by SEM, FTIR, and ¹HNMR. BPPPO_dp is known as a flexible membrane with higher permselectivity ($\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$ = 30.53) than PPPO_dp membranes. BPPPO and SiO₂ nanocomposite membranes display improved CO₂ permeability and CO₂/N₂ selectivity compared to the pure PPPO_dp, and BPPPO_dp membranes. The CO₂/N₂ separation mechanism in the membranes is related to the amount of gas dissolution than the amount of gas diffusion.

Graphical abstract

Introduction

One of the straightest techniques to decrease the global warming effect due to atmospheric CO₂ increase is found in CO₂ removal from gaseous streams which is produced in the production of energy from industrial processes [1, 2]. Using membrane technology in gas separation has created great revolution compared to traditional separation processes [3, 4].That is owing to their low cost, excellent efficiency and being trustworthy.

Recently, three popular gas separation membranes including polymeric membrane, inorganic membrane and new block polymeric membrane are used extensively for CO₂ separation. However, all of them have some defects to industrialization [5]. Highly permeable inorganic membranes suffer from brittle structure, structure defects and also high costs. Therefore, preparation, characterization, and application of inorganic separation membranes were just limited to the laboratory. While, polymeric separation membranes attract more and more attention because of the no expensive and easily fabricated materials. The modified polymeric membranes display improved permeability and CO₂ selectivity. Polymeric membranes are commonly used for CO₂ separation because of their extensive applications: separation of CO₂/H₂ in gas refinement, separation of CO₂/N₂ in carbon capture, separation of CO₂/CH₄ in natural gas refinement, separation of CO₂/O₂ in food packaging [6].

The mechanism of gas separation is distinguished by size sieving because of the rigid chain structures of the polymer matrix [7]. Polymeric membranes with great permeability and selectivity are difficult to prepare.

Poly(2,6-dimethyl-p-phenyleneoxide) (PPPO_dm) as a glassy polymeric material is introduced suitably as a gas separation membrane, because of the great permeability and suitable selectivity [1]. In particular, PPPO_dm membrane with great CO₂ permeability (~ 40 bar) and great CO₂/N₂ selectivity (~ 15) is preferred to other glassy polymers [8]. Chemically modified membranes such as brominated poly(2,6-dimethyl-p-phenyleneoxide) (BPPPO_dm) could often improve the gas permeability and selectivity [9, 10].

Addition of nanosized fillers to the polymer matrix could improve the separation performance of polymeric materials through modifying the chain packing of the polymer [1, 11]. Nano-scaled fillers like to be adjusted to the chain packing of glassy polymers in a way that increases the permselectivity trough creating additional selective free holes.

The SiO₂, TiO₂ and MoS₂ nanoparticles through polymer chain interruption bring improvement in gas separation. Polymeric membrane with dispersing nanoparticles is usually challenging due to undesirable interactions between the polymeric matrix and inorganic nano fillers. The appropriate polymer selection and applying the useful inorganic nanoparticle are two essential issues to get useful polymeric membranes [12].

Among different inorganic nanoparticles, SiO₂ nanoparticles are identified as an appropriate modifier for gas separation membranes owing to their high specific surface area and the ability of chemical functionalization, high mechanical and thermal stabilities [12,13,14,15,16].

Merkel et al. introduced poly(4-methyl-2-pentyne)(PMP)/silica nanocomposite including silica NPs (30 wt%), as more effective membrane than pure PMP at gas separation [17]. Also poly(amide-6-b-ethylene oxide) (PEBAX) including 27 wt% silica NPs displayed 277 bar for CO₂ permeability of and 79 for $\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$ selectivity [18].

With regard to the previous literature, here, we synthesized poly(2,6-diphenyl-p-phenylene oxide) (PPPO), brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO), and BPPPO/SiO₂ nanocomposite as suitable porous membrane for CO₂ adsorption (Fig. 1).

Fig. 1

The monomer structure of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp) (a); and brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) (b); and three-dimensional model of poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) (c)

Experimental

Materials

PPO_dm (M_n 25,000, polydispersity 2.0), N,N,Nʹ,Nʹ tetramethylethylenediamine (TMEDA, 99%), 2,6-diphenylphenol (98%), bromine (Br₂, 99.5%), chloroform (CHCl₃, 99.8%), ethanol (99.8%), methanol (99.8%), anhydrous hydrazine (98%), 1,2-dichlorobenzene (98.5%), chlorosulfonic acid (99%), anhydrous magnesium sulfate (99.5%), glacial acetic acid (99.8%) and silica nanopowders (SiO₂, 99.5%, 10 nm) and copper (I) chloride (CuCl, 98.5%) from Aldrich.

Synthesis of PPPO_dp

The synthesis of PPPO_dp was performed according to reported procedure [19]. In a 100 cc flask, CuCl (0.041 g), TMEDA (0.031 g), and anhydrous magnesium sulfate (2 g) are added to 1,2-dichlorobenzene (35 cc). The mixture is heated in an oil bath at 65 °C under magnetic stirring. Oxygen is bubbled into the mixture for 15 min. After green color observation in solution, 40 cc of 2,6-diphenylphenol solution is added dropwise during 20 min and made a dark red color immediately. After 24 h, several drops of anhydrous hydrazine are added to the solution to remove byproducts. The inorganic solids are filtered, and the solution is added dropwise to methanol (400 cc) including hydrazine and the solution is shaken for several hours. Then produced polymer is filtered and resolved in chloroform (40 cc) and precipitated in methanol (400 cc). The 3 g of polymer with M_n ∼ 120,000 is collected by filtration and dried at 80 °C under vacuum for 24 h. The density of PPPO is ∼ 1.11 g cm⁻³ measured by an excluded volume method at room temperature.

Synthesis of BPPPO_dp

For synthesis of BPPPO, PPPO (5 g) and CHCl₃ (50 cc) are added to a 100 cc flask. The mixture is stirred with a magnetic stirrer. The 20 cc of bromine solution included Br₂ (10 cc) and CHCl₃ (10 cc), is added dropwise to the mixture during 30 min under Argon atmosphere. After 1 h stirring at room temperature, 800 cc ethanol is added under mechanical stirring and the BPPPO is precipitated out. The produced polymer is filtered and dried at room temperature under vacuum. The extent of bromination is 80–95% from NMR spectrum. The density of BPPPO is ∼ 1.43 g cm⁻³ measured by the excluded volume method at room temperature with M_n ∼ 180,000.

Membrane preparation

The PPPO_dp, BPPO_dp, and BPPPO_dp/SiO₂ nanocomposite membranes are fabricated. The solution of PPPO (3 wt%, in CHCl₃) is cast on glass plate at room temperature. Similarly, the BPPPO membrane is prepared.

The BPPPO/silica membrane is prepared as follows: 0.3 g BPPPO is dissolved in 5 cc CHCl₃ under stirring. Silica NPs (10 nm) is added slowly to the BPPPO solution. The mixture is vigorously stirred with 1150 rpm for 15 min to dispersion. Then mixture is cast onto a plate at room temperature to dry. After drying, the resulting membranes, with 55–85 µm thickness, are peeled off and stored in a desiccator for testing.

Adding nanoparticles to glassy polymer membranes increases the gas permeability through disturbed polymer chain packing and then increase free volume of membranes.

Gas separation experiments

The membranes gas permeability is examined using a constant-volume setup (Fig. 2). The test setup is made of two compartment stainless steel cells in a way that a flat circular membrane with an effective area of 10.45 cm², and a reticular steel support are placed between two flanges and held by O-rings to avoid gas leakage. The experiments are started by vacating the cell up to − 0.1 bar for sending out any remaining gases before using feeding gases, which is to avoid further measurement errors. The rate of permeation (Barrer) is calculated from the pressure changes on the permeate side versus time, according to the following equation (1 Barrer = 10⁻¹⁰ cm³ (STP)-cm/cm² s cmHg):

$$P = \frac{{273 \times 10^{10} }}{760}\frac{VL}{{AT\left( {\frac{{p_{2} \times 76}}{14.7}} \right)}}\left( {\frac{{{\text{d}}p}}{{{\text{d}}t}}} \right)$$

Fig. 2

Schematic illustration of the constant-volume variable-pressure permeation test setup

where V refers to the volume of the permeate side in cm, L and A represent membrane thickness (cm) and the effective area (cm²), and T is the absolute temperature. p₂ is the feed pressure of the upstream in bar and dp/dt is the slope of permeate pressure change versus time, in its steady state (linear) region.

Permeability data of CO₂ and N₂ is measured for three times, and the average values are reported. The membranes CO₂/N₂ ideal selectivity ($\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$) is determined as the ration of CO₂ permeation to N₂ permeation using the following equation: $\alpha = P_{{{\text{CO}}_{ 2} }} /P_{{{\text{N}}_{ 2} }}$; where ${\text{P}}_{{{\text{CO}}_{ 2} }}$ and P_B are related to the permeabilities of pure CO₂ and N₂, respectively [20,21,22].

Characterization

Scanning electron microscope (SEM) (Philips 505) is used to detect the morphology and particle dispersion in all three membranes. The SEM is used at 10 kV operation voltage. FTIR spectra of pure PPPO_dp, and BPPO_dp are obtained with a FTIR spectrometer (Thermoscientific Nicolet 6700) in the range of 4000–400 cm⁻¹. The solution of PPPO_dp, and BPPO_dp in CDCl₃ (~ 2% w) are used for ¹HNMR analyses on a Bruker Advance DRX-400 spectrometer.

Result and discussion

Characterization of PPPO, BPPPO, and BPPPO/SiO₂ nanocomposite and their membranes

Differential scanning calorimetry (DSC)

Thermal properties of the white crystalline PPPO_dp, and yellowish crystalline BPPPO_dp are investigated by DSC analysis (Table 1). To eliminate the influences of less favorable crystal phases are formed at 248 °C, 288 °C for PPPO, BPPPO, respectively. Higher Tg of BPPPO_dp is related to three bulky bromine groups in each monomer unit. The second heating run as their melting point are considered at 470 and > 500 °C for PPPO and BPPPO that confirm their crystalline structures.

Table 1 DSC testing results of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp); and brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp)

The scanning electron microscope image (SEM)

The gas transport properties strongly depend on the membranes morphology. The cross section morphology of the neat PPPO_dp, BPPPO_dp, and BPPPO_dp/silica nanocomposite membranes are investigated via SEM images (Fig. 3). Figure 3a show porous structure for PPPO_dp membrane while Fig. 3b displays optimized porous structure with fewer surface defects for its relatively brominated membrane (BPPPO_dp). Adding silica nanoparticles to the brominated membrane (BPPPO_dp) propel it to an agglomerated BPPPO_dp/silica nanocomposite membrane (Fig. 3c). The nanocomposite membranes with their hydrophilic characterization of silica nanoparticles prefer to agglomerate. Hence BPPPO_dp/silica nanocomposite membrane seems to appear more selective due to its high density.

Fig. 3

The SEM image of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp) (a), brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) (b), and BPPPOdp/SiO₂ nanocomposite membranes (c)

FT-IR

Functional groups on the surface of poly(2,6-diphenyl-p-phenylene oxide) PPPO_dp and brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) are investigated by means of FTIR spectra (Fig. 4). The observed peaks at 1400–1600 cm⁻¹ are assigned to the C–C stretching bands of the aromatic rings. PPPO_dp reveals that some of the CH stretching of aromatic rings at 3025 cm⁻¹ (Fig. 4a). In addition, the observed peaks at 695 cm⁻¹, and 1147 cm⁻¹ corresponds to out of plane bending aromatic CH and C–O stretching bond.

Fig. 4

The FT-IR of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp) (a), and brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) (b)

As mentioned before, the observed peaks at 1400, 1600 cm⁻¹ illustrate aromatic CC stretching, 3058 cm⁻¹ clarify aromatic CH stretching, and 1183 cm⁻¹ show out of plane bending aromatic CH bond for brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp). Also, monitored peaks around 1030–1050 and 1183 cm⁻¹ imply C–Br and C–O stretching bond, respectively.

Nuclear magnetic resonance (NMR)

In ¹HNMR spectrum of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp) seen some peaks at 6.28 (2H, C₆H₂O), 6.96, 7.1–7.3 ppm (10H, 2 C₆H₅) (Fig. 5a). While, brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) demonstrate some peaks at 6.32 (1H, C₆HOBr), 6.75 and 7.4–7.8 ppm (8H, 2 C₆H₄Br) (Fig. 5b). As seen, decrease of the number of hydrogen atoms in brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) could confirm the new structure from brominating (Fig. 5b).

Fig. 5

The NMR spectrum of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp) (a), and brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) (b)

Gas separation performance test

Three prepared membranes of PPPO_dp, BPPPO_dp, BPPPO_dp/SiO₂ nanocomposite are used in the gas separation test (Table 2). According to previous works, the permeability of pure gas is measured in a constant-volume variable-pressure unit and the permselectivity (α) is determined from the following formula: α = P_A/P_B; where P_A and P_B are related to the permeabilities of pure gases A and B, respectively (Fig. 2, Table 2) [20,21,22].

Table 2 Gas separation performance of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp), brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp), and BPPPOdp/SiO₂ nanocomposite

The brominated poly(2,6-dimethyl-1,4-phenylene oxide) (BPPPO_dp) flexible membrane is identified with lower permeability of CO₂ (${\text{P}}_{{{\text{CO}}_{ 2} }}$ = 58 Barrer) but higher selectivity ($\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$ = 30.53) than poly(2,6-dimethyl-1,4-phenylene oxide) (PPPO_dm) membranes with permeability of CO₂ (${\text{P}}_{{{\text{CO}}_{ 2} }}$ = 68 Barrer) and selectivity ($\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$ = 14.7) (Table 2, Fig. 6). Thus, BPPPO_dp is selected as a beginning material for manufacture process of polymer/SiO₂ nanocomposite membrane. The BPPPO_dp/SiO₂ nanocomposite membrane including 10 wt% silica NPs present a greatly enhanced CO₂ permeability in (${\text{P}}_{{{\text{CO}}_{ 2} }}$ = 85 Barrer) and ($\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$ = 40.48). Hence they are found more effective than poly(4-methyl-2-pentyne)(PMP)/silica nanocomposite with permeability and selectivity for large organic molecules [17]. Meanwhile poly(amide-6-b-ethylene oxide)/silica nanocomposite displayed 277 bar permeability of CO₂ and $\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$ = 79 [18].

Fig. 6

The permselectivity diagram ($\alpha_{{{\text{CO}}_{ 2} / {\text{N}}_{ 2} }}$) of poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp) (blue), brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp) (orange), and BPPPOdp/SiO₂ nanocomposite membranes (gray) (according to the calculation in Table 2)

The study of mechanism demonstrates the silica nanoparticles owing to their incompatibility with polymer chains produce some gaps in the structure of nanocomposite, which results in enhanced permeability [19]. The surface of silica NPs with polar silano groups displayed low compatibility with the BPPO_dp matrix that causes silica particles to aggregate in the membrane. A narrow gap around the silica NPs formed due to lack of tight interaction between polymer chains and silica NPs, which made high gas diffusivity and permeability (Fig. 7).

Fig. 7

Illustration of nanogap formation in the BPPOdp/silica nanocomposite membranes

Conclusion

Poly(2,6-diphenyl-p-phenylene oxide) (PPPO_dp), brominated poly(2,6-diphenyl-p-phenylene oxide) (BPPPO_dp), and BPPPOdp/SiO₂ nanocomposite are synthesized and compared together in CO₂ permeability and CO₂/N₂ permselectivity. BPPPO_dp membranes are known to be more efficient than PPPO_dp through higher CO₂/N₂ permselectivity. The BPPPO_dp/SiO₂ nanocomposite membrane with the highest CO₂ permeability and CO₂/N₂ permselectivity is introduced as the most effective membrane. The CO₂ permeability increase through silica addition to the membrane.