polybenzimidazole

0 downloads 0 Views 3MB Size Report
May 2, 2017 - rearranged (TR) polymers [3–5], polymers of intrinsic microporosity ... properties because it combines the advantages of each polymer to.
Journal of Membrane Science 537 (2017) 315–322

Contents lists available at ScienceDirect

Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci

Thermally treated polyaniline/polybenzimidazole blend membranes: Structural changes and gas transport properties

MARK



V. Giel , Z. Morávková, J. Peter, M. Trchová Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky Sq. 2, 16206 Prague 6, Czech Republic

A R T I C L E I N F O

A B S T R A C T

Keywords: Polybenzimidazole Polyaniline Thermal treatment Gas permeation Gas sorption

Polyaniline/polybenzimidazole (PANI/PBI) blend membranes with various ratios of PANI/PBI were prepared by the solution casting method. The resulting free-standing membranes were subsequently doped with hydrochloric acid. In the next step, the membranes were heated up to 600 °C in an inert nitrogen atmosphere. FTIR spectroscopy and Raman scattering show the conversion of the blend to a nitrogen-containing carbon-like material. The permeability and sorption properties with respect to H2, O2, N2, CH4, and CO2 were determined. The separation qualities of the thermally treated PANI/PBI blend membranes significantly exceed those of nonthermally treated membranes. The permeation properties show a strong dependence on the PANI concentration in the blend and on its doping. The best separation performance for H2/N2, CO2/N2, and CO2/CH4 is obtained with the thermally treated undoped PANI/PBI 20/80 blend membrane.

1. Introduction Separation of gases on the industrial scale has increasingly been accomplished in recent years by membrane technology, which offers size compactness, process simplicity, energy efficiency, and low environmental impacts compared to conventional separation techniques [1,2]. Different materials have been used for membranes manufactured for ammonia purge gas processing, natural gas dehydration, oxygen enrichment, syngas production, CO2 separation, and hydrogen purification [1,2], including zeolites [3], silica [3], metal organic frameworks (MOFs) [3], graphene-based materials [3], organic–inorganic hybrid materials [3–5], high-performance polyimides (PI) [3–5], thermally rearranged (TR) polymers [3–5], polymers of intrinsic microporosity (PIMs) [3,4], and ionic liquids (ILs) [3–5]. The choice of a membrane material depends on the specific requirements of the gas separation application. Up to now, particular attention has been paid to polymeric membranes because of their cost effectiveness, high chemical stability, and excellent processability [6–10]; however, their performance is circumscribed by a trade-off between permeability and selectivity [11–13]. Thermally treated membranes have attracted considerable attention because of their chemical stability in corrosive environments, applicability at high temperatures, and gas separation behavior [14–17]. The most notable advantages of these membranes have been recently reviewed [18] to demonstrate their attractive features in comparison to polymeric membranes and to depict their suitability for membrane



gas separation. Hitherto, numerous polymers have been investigated for the preparation of thermally treated membranes. However, only a few studies have applied the idea of blending for the development of thermally treated membranes [15,19–23]. Rather than synthesizing new polymers, polymer blending is a unique time- and cost-effective technique that can generate new polymeric materials with superior properties because it combines the advantages of each polymer to obtain a new material with synergistic properties [2,24]. Moreover, the new material can also offer other features that may not be found in either of the constituents. Therefore, blending suitably selected materials provides the opportunity to develop or improve membranes with different separation properties and physico-chemical characteristics to meet specific needs. To date, several polymeric materials have been tested for the preparation of thermally treated polymer blend membranes [15,18–23,25,26]. Recent developments have shown promising features of thermally treated blend membranes made from polybenzimidazole (PBI). PBI is known to be a suitable membrane material for harsh operating conditions due to its outstanding physical, thermal and chemical stability [27,28]. Given its very high Tg of approximately 420 °C [29], PBI is a promising polymer for the preparation of thermally treated membranes. With its excellent film-forming properties and compatibility with other polymers, e.g., poly(vinyl acetate-covinyl alcohol) [30], polyacrylate [31], polyvinylidene fluoride [32], and polyimide (PI) [29], the preparation of a stable polymer film is

Corresponding author. E-mail address: [email protected] (V. Giel).

http://dx.doi.org/10.1016/j.memsci.2017.04.062 Received 7 November 2016; Received in revised form 30 March 2017; Accepted 25 April 2017 Available online 02 May 2017 0376-7388/ © 2017 Elsevier B.V. All rights reserved.

Journal of Membrane Science 537 (2017) 315–322

V. Giel et al.

2.2. Membrane preparation

feasible. Hosseini and Chung [22] prepared thermally treated blend membranes from PBI and polyimides for N2/CH4 and CO2/CH4 separation and hydrogen purification. The thermal treatment increased the blend membranes’ permeability to gases. Later, the Hosseini group [23] reported remarkable selectivity values of various gas pairs for the thermally treated PBI/PI (Kapton) blend membranes. As the Hosseini group studies show, the composition plays a major role in governing the physical and chemical properties of blend systems. Moreover, their results demonstrate that PBI is an interesting candidate for the preparation of new materials with tailorable properties. Based on the observations of our previous work, the gas separation performance of PBI can be efficiently tailored by the incorporation of polyaniline (PANI) [33]. PANI has received great scientific attention because of the variety of morphologies it forms [34–36] and its high gas-pair selectivities of 3590 (H2/N2), 30 (O2/N2), and 336 (CO2/CH4) [37]. Of particular interest was the observation that the structure of PANI membranes can be controlled by acid doping [38]. Acid doping of PANI rearranges the molecular chains, which leads to an increase in the packing density of the chains and thus higher selectivities [38]. To date, most research about PANI as a membrane material has been directed towards the efficient preparation of PANI membranes and its doping/ dedoping process. However, another way to alter the structure of PANI and thus its gas transport properties is thermal treatment. Thermal treatment of PANI leads to the formation of nitrogen-containing carbonaceous material [39]. After the thermal treatment, PANI preserves its morphology [35,40–43], which is a key characteristic to obtain defect-free thermally treated membranes [44]. Although various studies have been carried out on PANI and PBI, there is to date no academic literature on thermally treated PANI/PBI blend membranes and their physico-chemical and gas transport properties, to the best of our knowledge. Therefore, we attempted to investigate in the present study the influence of thermal treatment, doping, and PANI/PBI blend composition on the resulting membrane microstructure, morphology and gas separation performance.

Dense PANI/PBI blend membranes were prepared by the solution casting method. PANI emeraldine base was first dissolved in 1methylimidazole (IO-LI-TEC) and stirred at room temperature for 72 h. Subsequently, the PBI solution in DMAc was added to the PANI solution and stirred overnight. Insoluble fractions of the blend solution were filtered off afterward. The resulting polymer blend solution was poured onto a clean glass surface and dispensed with a doctor blade (0.3 mm gap). The cast films were dried in an oven (24 h, 60 °C) and then stripped off from the glass in a tray of water. The resultant membranes were then dried under vacuum at 70 °C for 48 h. The blend ratios of PANI were 5, 10 and 20 wt% in PBI. The obtained dense films were named as ‘undoped PANI/PBI (blend ratio) blend membrane’, e.g., undoped PANI/PBI 5/95 blend membrane. The initial PANI/PBI blend membranes were the undoped version since those samples were not exposed to acid. Another set of blend membranes was prepared in this study, so-called ‘doped PANI/PBI blend membranes’. For these, the undoped PANI/PBI blend membranes were immersed into 0.1 M hydrochloric acid solution (Lach-Ner) for 24 h. The resultant doped PANI/PBI blend membranes were then washed with distilled water and dried in a vacuum oven at 70 °C for 24 h. Moreover, an additional membrane was prepared from neat PBI as a standard sample for comparison, which was not treated with hydrochloric acid. All polymer membranes were afterward thermally treated in a LAC LE05/11 furnace equipped with an Ht60B temperature controller. For this, the furnace was heated at a rate of 10 °C/min from room temperature to the predetermined temperatures of 100 °C, 200 °C, 300 °C, 400 °C, 500 °C, and 600 °C. Each heating step was executed for 3 h under N2 atmosphere with a flow rate of 100 ml/min. At the end of the process, the membranes were cooled steadily to room temperature under nitrogen flow. All thermally treated membranes are referred to in the text as ‘thermally treated membranes’, e.g., thermally treated undoped PANI/ PBI 5/95 blend membrane. 2.3. Membrane characterization

2. Experimental

Fourier-transform infrared (FTIR) measurement was performed in attenuated total reflection (ATR) mode using a Thermo Nicolet NEXUS 870 FTIR Spectrometer (Madison, WI, USA). The spectra of the membranes were measured with a Golden Gate™ Heated Diamond ATR Top-Plate (MKII single reflection ATR system; Specac; Orprington, UK). Raman spectra excited with a near-infrared diode 785 nm laser were collected on a Renishaw inVia Reflex Raman microspectrometer. A research-grade Leica DM LM microscope with a 50x objective magnification was used to focus the laser beam on the sample placed on an X–Y motorized sample stage. The scattered light was analyzed by the spectroscope with a holographic grating of 1200 lines mm–1. A Peltiercooled CCD detector (576×384 pixels) registered the dispersed light. The gas permeability through the membranes was determined using a laboratory time-lag method high vacuum apparatus at 30 °C with a static permeation cell, which possesses an effective membrane area of 1.24 cm2. The studied membrane was placed and sealed in a cell which was evacuated overnight at 30 °C in the ultra-high vacuum apparatus to degas the sample. The feed pressure pi was 1.5 bar. The permeability P was determined from the increase in pressure Δpp per time Δt in a calibrated volume Vp of the product part of the cell during steady-state permeation. For calculation of permeability, the following formula was used (Eq. (1)) [46]:

2.1. Materials Poly(5,5-bis-benzimidazole-2,2-diyl-1,3-phenylene) (PBI) was supplied by Hoechst Celanese and used as received as a 10% N,Ndimethylacetamide (DMAc) solution with a lithium chloride content of ≈2%. Polyaniline (PANI) was synthesized by oxidation of aniline hydrochloride (Sigma-Aldrich) with ammonium peroxidisulfate (Lach-Ner). The reaction scheme of the PANI synthesis is shown in Fig. 1. The details of the synthesis are described elsewhere [45]. The resulting PANI precipitate is in its protonated form and it is called PANI emeraldine salt. Subsequently, the PANI emeraldine salt was filtered off and deprotonated to PANI emeraldine base with a 0.1 M aqueous solution of ammonium hydroxide (Lach-Ner).

P=

Δpp Vp⋅l 1 . . Δt A⋅pi RT

(1)

where l is the membrane thickness, A is the area, T is the temperature and R is the gas constant. The permeabilities are reported in units of

Fig. 1. Synthesis of PANI.

316

Journal of Membrane Science 537 (2017) 315–322

V. Giel et al.

Barrer (1 Barrer=10−10 cm3 (STP) cm/(cm2 s cm Hg)). In our experiments, gases such as H2, O2, N2, CH4, and CO2 were studied. Each gas possessed a purity of 99.99% and was used as received from Messer Technogas s.r.o. (Czech Republic). The ideal selectivity αi/j of two gases i and j was determined by the following ratio (Eq. (2)) [46]:

αi / j = Pi / Pj

(2)

The accuracy of the measurement is given by the sum of the relative accuracies of each measured term of Eq. (1). The relative error of Δpp/Δt measured with MKS Baratron is smaller than 0.3% plus the inaccuracy attributed to the resolution of the pressure transducer which is 1/10 of mbar. The relative standard deviation of the calibrated volume is less than 0.1%, of the membrane area less than 0.5%, and of the feed pressure 0.2%. The thickness of the membrane can be measured as precise as 1 µm. Sorption studies were conducted on a gravimetric sorption balance, IGA-002, Hiden Isochema, UK, according to the procedure described in [33]. The sorption isotherms were measured by stepwise pressure changes (pressure increase rate 100 mbar/min) within the pressure range of 1–5 bar. The sorption balance consists of a large capacity microbalance (5 g) with a resolution of 0.1 μg and excellent long-term stability of ± 1 μg.

Fig. 3. Raman spectra of neat PBI membranes treated at different temperatures.

753 cm–1 (Fig. 3). After heating to 300, 400, 500, or 600 °C, the Raman spectra become featureless except for a very broad fluorescence band at 1400 cm–1 and an indistinct band at 300 cm–1. The Raman spectra indicate that neat PBI is stable up to approximately 200 °C (i.e., they indicate a slightly lower temperature than the infrared spectra). After blending neat PBI membranes with various amounts of PANI base and exposing them to hydrochloric acid, the infrared spectra of all blends were qualitatively identical. Therefore, only the spectra of a doped PANI/PBI 20/80 blend membrane is shown below. The infrared spectra of doped PANI/PBI blend membranes contain the peaks of pure PBI in all compositions (Fig. 4). Weak bands of PANI emeraldine salt at approximately 1558, 1498, 1301, 1232 and 1144 cm–1 can be detected in the spectra [47], especially for the highest amount of PANI emeraldine salt in PBI (doped PANI/PBI 20/80; Fig. 4). After heating to 400 °C the bands of PBI are preserved in the spectra, and at 500 °C the spectra are very close to the spectrum of the neat PBI membrane heated to 600 °C. After heating to 600 °C the spectra are transformed into a broad band with two maxima at approximately 1580 and 1238 cm–1 corresponding to the nitrogen-containing carbonized material. We suppose that this material is carbonized PANI emeraldine salt, which stabilizes the PBI membrane against heating. The peak at 2230 cm–1 that appeared in the spectrum corresponds to the presence of the nitrile N≡C bond in the structure of the PBI membrane. Raman spectra of the doped PANI/PBI blend membranes (Fig. 5) contain a fluorescence background of PBI with the bands of PANI emeraldine salt (1605 cm–1 with a shoulder at 1650 cm–1, 1513, 1455, 1393, 1337, 1229, 1175, 836, 815, 730, 572, 520, 440 and 415 cm–1)

3. Results and discussion 3.1. Infrared and Raman spectra Fig. 2 shows the infrared spectra of neat PBI, which was thermally treated at different temperatures. The peaks of PBI observed at 1601, 1524, a doublet at 1442/1400, 1280, 1102, and a sharp peak at 797 cm–1 remain well distinguished in the spectrum after heating to 300 °C (Fig. 2). After treating at 400 and 500 °C, the three main peaks are broadened into bands with maxima at 1580, 1410, and 1285 cm–1, and the sharp peak at 797 cm–1 decreased. After treatment to 600 °C, the three bands at 1580, 1410 and 1285 cm–1 dominate the spectrum, the peak at 787 cm–1 is still present, and a peak at 2226 cm–1 corresponding to the nitrile N≡C bonding stretching vibration appeared in the spectrum. As indicated by the infrared spectra, the neat PBI membrane preserves its molecular structure up to a temperature of 300 °C. Above 400 °C, a transformation into a cross-linked structure occurs, and at 600 °C a partly carbonized nitrogen-containing material including a nitrile bond is obtained. Fig. 3 shows the Raman spectra of the same material obtained with an excitation laser at 785 nm. Neat PBI treated up to 200 °C shows distinctive peaks at 1597 (with shoulders at 1618 and 1632 cm–1), 1543, 1507, 1453, 1425, 1383, 1298, 1275, 1233, 1183, and 1140 cm–1; two sharp peaks at 1000 and 962 cm–1; and a peak at

Fig. 4. FTIR spectra of doped PANI/PBI 20/80 blend membranes treated at different temperatures.

Fig. 2. FTIR spectra of neat PBI membranes treated at different temperatures.

317

Journal of Membrane Science 537 (2017) 315–322

V. Giel et al.

Fig. 5. Raman spectra of doped PANI/PBI 20/80 blend membranes treated at different temperatures.

Fig. 7. Raman spectra of undoped PANI/PBI 20/80 blend membranes treated at different temperatures.

[48]. After heating to 300 °C the bands of PANI emeraldine salt get weaker, and at higher temperatures, the shape of the fluorescence band changes slightly. At 600 °C, two bands at 1605 and 1337 cm–1 can be observed. These bands may be attributed to the disordered (D) and graphitic (G) bands of carbonaceous material [49]. By comparing the fluorescence, we can deduce that the PBI is preserved from thermal degradation by PANI emeraldine salt and that PANI emeraldine salt is transformed to a disordered nitrogen-containing carbon material. After blending the pure PBI membrane with various amounts of PANI emeraldine base, without exposing the blend membranes to hydrochloric acid, the infrared spectra of the undoped PANI/PBI blend membranes contain the peaks of neat PBI (1601, 1524, 1442/1400 (doublet), 1280, 1102, and a sharp peak at 797 cm–1) and of PANI emeraldine base (1588, 1497, and 1172 cm−1) (Fig. 6) [47]. During heating, the bands of PBI are preserved in the spectra up to 500 °C. At 600 °C, the spectrum is close to the spectrum of PBI heated to the same temperature. In contrast to the doped PANI/PBI blend membranes, the thermal treatment up to 600 °C does not result in the formation of carbonized PANI emeraldine base. In the Raman spectra of the undoped PANI/PBI blend membranes (Fig. 7), the bands of PANI emeraldine base (1605, 1462, 1385, 1225, 1175 cm–1, a group of bands at approximately 800 and 600 cm–1, 520, and 415 cm–1 with a shoulder at 440 cm–1) are observed on a fluorescence band of PBI. The band at 1337 cm–1 corresponds to protonated units and is strong in intensity, leading to the conclusion that the deprotonation to PANI emeraldine base was not completed.

After heating to temperatures above 300 °C, the shape of the fluorescence band changes in the same way as did that of neat PBI. Moreover, the formation of D and G bands is not observed. These results indicate that blending PBI with PANI emeraldine base is less effective in the preservation of PBI against thermal degradation than blending with PANI emeraldine salt, which is consistent with our previous observations [50]. 3.2. Gas sorption properties Fig. 8 shows pure gas sorption isotherms for O2 (Fig. 8a), N2 (Fig. 8b), CH4 (Fig. 8c), and CO2 (Fig. 8d) for thermally treated PBI and thermally treated undoped/doped PANI/PBI blend membranes. The sorption isotherms were in the same order as their condensability expressed by the critical temperature Tc of the gases (Tc(N2) =126 K < Tc(O2)=155 K < Tc(CH4)=191 K < Tc(CO2)=304 K). The sorption between the samples varies in the following order: doped PANI/PBI 10/90 (600 °C) > undoped PANI/PBI 10/90 (300 °C) > doped PANI/PBI 10/90 (300 °C) > PBI (300 °C). The sorption capacities of the blend membranes are much higher than that of neat PBI. This shows a positive impact of PANI blending on the porosity of the PBI matrix. At the same time, doping causes a decrease in the sorption capacities. It is concluded that the porous structure of the PBI matrix is reduced due to the rearrangement of the macromolecular chains upon doping. With the increase in temperature, the sorption capacities increase, which in turn indicate pore enlargement or an increase in the number of pores in the PBI matrix. In summary, the membrane structure of PBI can be effectively adjusted by incorporating PANI, pretreatment of PANI/PBI membranes, and increasing the temperature. 3.3. Gas transport properties To study the influence of thermal treatment on the gas separation behavior of PANI/PBI blend membranes, permeability measurements of H2, O2, N2, CH4 and CO2 were performed on a time-lag apparatus for both initial and thermally treated membranes. The permeabilities and ideal selectivities of the initial membranes are presented in Table 1 (membranes without thermal treatment), and those of the thermally treated membranes are presented in Table 2 (membranes treated at 300 °C). In general, all prepared membranes exhibited low permeabilities. In addition, their permeabilities depended on the kinetic diameter of the gas species: PH2 (2.89 Å) > PCO2 (3.30 Å) > PO2 (3.46 Å) > PN2 (3.64 Å) > PCH4 (3.80 Å), which is typical for many highly aromatic polymers and indicates a sieving mechanism [51].

Fig. 6. FTIR spectra of undoped PANI/PBI 20/80 blend membranes treated at different temperatures.

318

Journal of Membrane Science 537 (2017) 315–322

V. Giel et al.

Fig. 8. Comparison of gas sorption isotherms of prepared thermally treated PANI/PBI blends with thermally treated PBI for O2 (a), N2 (b), CH4 (c) and CO2 (d) measured at 30 °C.

smaller (N2 – 128%; CH4 – 75%). In the case of O2, a permeability increase of approximately 317% was observed. Moreover, the blend membranes treated at temperatures above 300 °C could not be tested for their gas transport properties as they were very brittle and could not be handled without being damaged. Compared to undoped PANI/PBI blend membranes and thermally treated undoped PANI/PBI blend membranes, doped membranes possess considerably lower permeability coefficients (Table 1 and Table 2). The highest permeability coefficients of thermally treated doped PANI/PBI blend membranes were observed for those with the highest content of PANI (Table 2), whereas for the non-thermally treated doped PANI/PBI blend membranes highest permeabilities were observed for those with the lowest PANI content (Table 1). Furthermore, Table 1 and Table 2 display the ideal selectivities of the membranes before (Table 1) and after thermal treatment (Table 2) for the gas pairs H2/N2, O2/N2, H2/CO2, CO2/N2, and CO2/CH4 in dependency of the membrane composition and doping of the membranes.

Furthermore, the addition of PANI to PBI caused a decrease in permeability coefficients in the case of the initial membranes (Table 1) since PANI is less permeable than PBI, thus decreasing the permeation properties of the membranes. However, a large difference in the permeability of the samples was observed after thermal treatment (Table 2). It is interesting to note that the gas permeability of the thermally treated membranes followed a trend opposite to that observed in the initial membranes. For instance, the PANI/PBI 20/80 blend membrane demonstrated the lowest gas permeability among the initial membranes (Table 1) but was found to exhibit the largest permeability upon thermal treatment (Table 2). It is assumed that the pore structure may have enlarged due to domains occupied by PANI or that PANI provides an enhanced diffusion pathway for gas species in PBI, enabling the gas molecules to pass more rapidly through the membrane. The largest permeability enhancement was obtained for CO2 (357%) with the addition of 20 wt% PANI. The permeability of a smaller gas species, H2, was enhanced by 327%, while for larger gas species such as N2 and CH4, the increments were much

Table 1 Gas permeability coefficients and ideal selectivities of the neat PBI, undoped and doped PANI/PBI blend membranes measured at 30 °C and 1.5 bar (abs). Sample (non-treated)

PANI content (wt%)

Permeability coefficients (Barrer)

Ideal selectivity

H2

CO2

O2

N2

CH4

H2/N2

O2/N2

H2/CO2

CO2/N2

CO2/CH4

neat PBI

0

2.17

0.53

0.11

0.062

0.031

35.0

1.7

4.1

8.5

17.1

Undoped PANI/PBI membrane

5 10 20

1.59 1.19 0.98

0.367 0.225 0.170

0.074 0.045 0.032

0.041 0.023 0.016

0.021 0.012 0.008

38.8 50.6 60.9

1.8 1.9 2.0

4.3 5.3 5.7

8.9 9.6 10.6

17.6 18.6 20.9

Doped PANI/PBI membrane

5 10 20

0.95 0.77 0.62

0.21 0.16 0.12

0.041 0.032 0.025

0.025 0.019 0.014

0.015 0.011 0.008

38.0 41.2 43.9

1.6 1.7 1.8

4.5 4.8 4.9

8.4 8.6 8.9

14.1 14.6 15.9

319

Journal of Membrane Science 537 (2017) 315–322

V. Giel et al.

Table 2 Gas permeability coefficients and ideal selectivities of neat PBI, undoped and doped PANI/PBI blend membranes thermally treated at 300 °C measured at 30 °C and 1.5 bar (abs). Sample (thermally treated)

PANI content (wt%)

Permeability coefficients (Barrer)

Ideal selectivity

H2

CO2

O2

N2

CH4

H2/N2

O2/N2

H2/CO2

CO2/N2

CO2/CH4

neat PBI

0

1.30

0.08

0.018

0.007

0.004

180.1

2.5

16.5

10.9

22.5

Undoped PANI/PBI membrane

5 10 20

2.18 3.55 5.56

0.14 0.23 0.36

0.030 0.048 0.075

0.009 0.012 0.016

0.004 0.005 0.007

242.2 295.5 340.3

3.3 4.0 4.6

15.9 15.8 15.4

15.2 18.7 22.1

33.9 44.1 51.5

Doped PANI/PBI membrane

5 10 20

1.89 2.61 3.69

0.121 0.170 0.242

0.026 0.036 0.050

0.009 0.011 0.014

0.005 0.007 0.008

210.0 231.5 261.3

2.9 3.2 3.5

15.6 15.3 15.2

13.4 15.1 17.1

24.2 26.1 28.6

decrease in selectivity (~7%). Even though more CO2 is able to absorb in the blend than in the neat thermally treated PBI, it seems, that the H2 transport is more promoted compared to CO2. For the non-thermally treated doped PANI/PBI blend membranes and thermally treated doped PANI/PBI blend membranes, there is a significant decline in gas permeability and ideal selectivity upon doping in comparison to their corresponding undoped counterparts. This trend can be ascribed to the difference in the microstructure properties due to the molecular rearrangement of the polymer chains through the acid treatment. The decrease in permeability and ideal selectivity of the thermally treated doped blend membranes corresponds with the sorption isotherms (Fig. 8). As the isotherms indicated, thermally treated doped blend membranes sorbed less gas than the thermally treated undoped ones. As mentioned previously, differences in the membrane structure cause differences in the sorption and gas transport properties and subsequently alter the gas selectivities. In summary, thermal treatment of PANI/PBI blend membranes yields superior materials with higher permeabilities and selectivities, even though the permeabilities and gas selectivities of the initial membranes may be low.

As Table 1 and Table 2 indicate the ideal selectivities are a strong function of the blend composition. The selectivities of all gas pairs increased with increasing amount of PANI. Usually, there is an increase in selectivity when there is a decrease in permeability and vice versa. Therefore, it was expected that the ideal selectivity of the nonthermally treated membranes would increase. In the case of the thermally treated membranes, the ideal separation factor increased even though the permeability of all gases increased. The increase in the ideal selectivity may be attributed to some created voids between the interface of PANI and PBI. Additionally, it appears that the addition of PANI tends to induce the formation of a porous structure in the PBI matrix. These voids and the additionally created pores in the PBI matrix influence the membrane structure and therefore cause differences in the gas transport properties. A similar observation about increased permeability and selectivity upon thermal treatment was made by Itta et al. [15], who were studying the blend material polyphenylene oxide/ polyvinylpyrrolidone (PPO/PVP). In comparison with the non-thermally treated blend membranes (Table 1), the thermally treated blend membranes (Table 2) showed the greatest selectivity for all gas pairs, especially for the undoped PANI/ PBI blend membranes. Before the thermal treatment, the selectivities of undoped PANI/PBI 20/80 blend membranes for H2/N2 and H2/CO2 were 60.95 and 5.75, respectively. After treatment, the selectivities of the thermally treated undoped PANI/PBI 20/80 blend membranes drastically increased to 340.35 for H2/N2 and to 15.41 for H2/CO2, which represent an increase in selectivity of 458% and 168%, respectively. In the case of O2/N2, CO2/N2, and H2/CH4, the selectivities were 2.01, 10.60, and 20.93, respectively. After the thermal treatment, the values increased to 4.59, 22.08, and 51.53, which represent an increase of 128%, 108%, and 146%, respectively, for the mentioned gas pairs. The high separation performance of the thermally treated undoped PANI/PBI blend membranes can be ascribed to the enhanced morphological properties caused by the partial contributions of rigid PBI chains with high packing density, which can have large effects on the chain configurations within the membrane structure, and by the chemically integrated PANI macromolecules that might have large effects on the pore formation process during the thermal treatment process. Moreover, as shown in the literature, the initial materials generally have a different pore structure than their thermally treated counterparts, which usually changes upon thermal treatment and decreases with the increase in temperature [15,23,52]. Comparing the permselectivity of the thermally treated neat PBI matrix with the thermally treated undoped PANI/PBI blend membranes, the addition of 20 wt% PANI to PBI resulted in increases of 89%, 102%, and 129% in the selectivities of H2/N2, CO2/N2, and CO2/ CH4, respectively. Similarly, the selectivity of O2/N2 increased by 84%, which is very interesting since both gases are non-polar, possess a minor difference in kinetic diameter, and sorb to almost the same extent in polymers. Such enhanced selectivity may imply the formation of finely tuned pores in the microstructure of the thermally treated PANI/PBI blend membranes that could either constrict or retard the transport of larger molecules such as N2. In the case of H2/CO2, there is a slight

3.4. Comparison of the separation performance of the current membranes with the Robeson upper bound To display the membrane performances of the present work for the gas pairs H2/N2, CO2/CH4, O2/N2, and CO2/N2, the separation data of selected membranes of this study are compared with respect to the Robeson upper bound (Fig. 9). As seen from Fig. 9, the membrane composition, the chemical pretreatment (doping), and the thermal treatment of the membranes play an important role. It is noteworthy that the prepared thermally treated membranes of the present work are closer to the upper bound than the initial membranes. As depicted in Fig. 9, with the addition of PANI, the transport properties of the respective thermally treated membranes shift towards the upper bound. In contrast, doping the membranes triggers a decline in the transport properties. Even so, the thermal treatment of PANI/PBI blend membranes induces a fair enhancement of the gas transport properties for all gases and gas pairs, additional studies on this material should be carried out to further improve the membrane performance, since none of the prepared materials are located above the Robeson upper bound. 4. Conclusions Thermally treated polymer blend membranes were developed from PANI and PBI. The effects of blend composition, doping, and thermal temperature on the structural change and gas separation performance of these membranes were investigated. FTIR and Raman spectroscopy show the gradual destruction of the PANI structure and the possible formation of intermediate oxime and nitrile groups, resulting in Ncontaining carbon materials. The single gas permeation tests for H2, O2, N2, CH4, and CO2 320

Journal of Membrane Science 537 (2017) 315–322

V. Giel et al.

Fig. 9. Robeson diagrams with the present upper bound for the following gas pairs: (a) H2/N2, (b) CO2/CH4, (c) O2/N2, and (d) CO2/N2.

[2] H.A. Mannan, H. Mukhtar, T. Murugesan, R. Nasir, D.F. Mohshim, A. Mushtaq, Recent applications of polymer blends in gas separation membranes, Chem. Eng. Technol. 36 (2013) 1838–1846, http://dx.doi.org/10.1002/ceat.201300342. [3] P. Li, Z. Wang, Z. Qiao, Y. Liu, X. Cao, W. Li, et al., Recent developments in membranes for efficient hydrogen purification, J. Membr. Sci. 495 (2015) 130–168, http://dx.doi.org/10.1016/j.memsci.2015.08.010. [4] P. Bakonyi, N. Nemestóthy, K. Bélafi-Bakó, Biohydrogen purification by membranes: an overview on the operational conditions affecting the performance of nonporous, polymeric and ionic liquid based gas separation membranes, Int. J. Hydrog. Energy 38 (2013) 9673–9687, http://dx.doi.org/10.1016/j.ijhydene.2013.05.158. [5] C.H. Lau, P. Li, F. Li, T.-S. Chung, D.R. Paul, Reverse-selective polymeric membranes for gas separations, Prog. Polym. Sci. 38 (2013) 740–766, http://dx. doi.org/10.1016/j.progpolymsci.2012.09.006. [6] D.F. Sanders, Z.P. Smith, R. Guo, L.M. Robeson, J.E. McGrath, D.R. Paul, et al., Energy-efficient polymeric gas separation membranes for a sustainable future: a review, Polymer 54 (2013) 4729–4761, http://dx.doi.org/10.1016/j.polymer. 2013.05.075. [7] W.J. Koros, D.R.B. Walker, Gas Separation Membrane Material Selection Criteria: weakly and Strong Interacting Feed Component Situation, Polym. J. 23 (1991) 481–490, http://dx.doi.org/10.1295/polymj.23.481. [8] P.M. Budd, N.B. McKeown, Highly permeable polymers for gas separation membranes, Polym. Chem. 9 (2010) 63–68, http://dx.doi.org/10.1039/ b9py00319c. [9] Y. Yampolskii, B.D. Freeman, Membrane Gas Separation, John Wiley & Sons Ltd, Chichester, United Kingdom, 2010. [10] Y. Yampolskii, Polymeric gas separation membranes, Macromolecules 45 (2012) 3298–3311, http://dx.doi.org/10.1021/ma300213b. [11] L.M. Robeson, Correlation of separation factor versus permeability for polymeric membranes, J. Membr. Sci. 62 (1991) 165–185, http://dx.doi.org/10.1016/03767388(91)80060-J. [12] L.M. Robeson, The upper bound revisited, J. Membr. Sci. 320 (2008) 390–400, http://dx.doi.org/10.1016/j.memsci.2008.04.030. [13] H. Lin, M. Yavari, Upper bound of polymeric membranes for mixed-gas CO2/CH4 separations, , J. Membr. Sci. 475 (2015) 101–109, http://dx.doi.org/10.1016/j. memsci.2014.10.007. [14] C.H. Jung, G.W. Kim, S.H. Han, Y.M. Lee, Gas separation of pyrolyzed polymeric membranes: effect of polymer precursor and pyrolysis conditions, Macromol. Res. 15 (2007) 565–574, http://dx.doi.org/10.1007/BF03218832. [15] A.K. Itta, H.-H. Tseng, M.-Y. Wey, Fabrication and characterization of PPO/PVP blend carbon molecular sieve membranes for H2/N2 and H2/CH4 separation, J.

revealed that the derived thermally treated blend membranes are more permeable than their respective non-thermally treated membranes. Furthermore, the increase in PANI content results in an increase in permeability and permselectivity. In addition, thermally treated undoped PANI/PBI blend membranes exhibit higher permeabilities than the doped ones. Sorption studies revealed that the sorption capacities increase with the addition of PANI. The thermally treated doped PANI/ PBI blend membranes exhibit lower sorption capacities than the undoped ones. The highest gas sorption (for all studied gases) is found for the doped PANI/PBI blend membrane thermally treated at 600 °C. In summary, this work provides a new strategy to design and fabricate new thermally treated membranes with improved gas separation performance which might be interesting for gas separation applications. Nevertheless, additional studies on mixed gases are required since this work focuses only on a detailed analysis of pure gases. Hence, the results obtained in the present work opens the possibility to study the impact of mixed gases on the gas separation performance. Acknowledgement We gratefully acknowledge the research funding provided by the Ministry of Education, Youth and Sports of Czech Republic within the National Sustainability Program I (NPU I), Project POLYMAT LO1507 and by the Czech Science Foundation (16-02787S). References [1] P. Bernardo, E. Drioli, G. Golemme, Membrane gas separation: a review/state of the art, Ind. Eng. Chem. Res. 48 (2009) 4638–4663, http://dx.doi.org/10.1021/ ie8019032.

321

Journal of Membrane Science 537 (2017) 315–322

V. Giel et al.

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

Membr. Sci. 372 (2011) 387–395, http://dx.doi.org/10.1016/j.memsci.2011.02. 027. V. Pirouzfar, A.Z. Moghaddam, M.R. Omidkhah, S.S. Hosseini, Investigating the effect of dianhydride type and pyrolysis condition on the gas separation performance of membranes derived from blended polyimides through statistical analysis, J. Ind. Eng. Chem. 20 (2014) 1061–1070, http://dx.doi.org/10.1016/j.jiec.2013. 06.043. B. Zhang, Y. Wu, Y. Lu, T. Wang, X. Jian, J. Qiu, Preparation and characterization of carbon and carbon/zeolite membranes from ODPA-ODA type polyetherimide, J. Membr. Sci. 474 (2015) 114–121, http://dx.doi.org/10.1016/j.memsci.2014.09. 054. W.N.W. Salleh, A.F. Ismail, Carbon membranes for gas separation processes: recent progress and future perspective, J. Membr. Sci. Res. 1 (2015) 2–15, http://dx.doi. org/10.22079/jmsr.2015.12301. W.N.W. Salleh, A.F. Ismail, Fabrication and characterization of PEI/PVP-based carbon hollow fiber membranes for CO2/CH4 and CO2/N2 separation, AIChE J. 58 (2012) 3167–3175, http://dx.doi.org/10.1002/aic.13711. V.M. Linkov, R.D. Sanderson, E.P. Jacobs, Carbon membranes from precursors containing low-carbon residual polymers, Polym. Int. 35 (1994) 239–242, http:// dx.doi.org/10.1002/pi.1994.210350304. H. Hatori, T. Kobayashi, Y. Hanzawa, Y. Yamada, Y. Iimura, T. Kimura, et al., Mesoporous carbon membranes from polyimide blended with poly(ethylene glycol), J. Appl. Polym. Sci. 79 (2001) 836–841, http://dx.doi.org/10.1002/10974628(20010131)79:53.0.CO;2-1. S.S. Hosseini, T.S. Chung, Carbon membranes from blends of PBI and polyimides for N2/CH4 and CO2/CH4 separation and hydrogen purification, J. Membr. Sci. 328 (2009) 174–185, http://dx.doi.org/10.1016/j.memsci.2008.12.005. S.S. Hosseini, M.R. Omidkhah, A. Zarringhalam Moghaddam, V. Pirouzfar, W.B. Krantz, N.R. Tan, Enhancing the properties and gas separation performance of PBI-polyimides blend carbon molecular sieve membranes via optimization of the pyrolysis process, Sep. Purif. Technol. 122 (2014) 278–289, http://dx.doi.org/10. 1016/j.seppur.2013.11.021. J. Parameswaranpillai, S. Thomas, Y. Grohens, Polymer bends: state of the art, new challenges, and opportunities, Characterization of Polymer Blends: Miscibility, Morphology and Interfaces, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2014, pp. 1–5, http://dx.doi.org/10.1002/9783527645602.ch01. W.N.W. Salleh, A.F. Ismail, T. Matsuura, M.S. Abdullah, Precursor selection and process conditions in the preparation of carbon membrane for gas separation: a review, Sep. Purif. Rev. 40 (2011) 261–311, http://dx.doi.org/10.1080/15422119. 2011.555648. S.M. Saufi, A.F. Ismail, Fabrication of carbon membranes for gas separation – a review, Carbon N. Y. 42 (2004) 241–259, http://dx.doi.org/10.1016/j.carbon. 2003.10.022. S.C. Kumbharkar, Y. Liu, K. Li, High performance polybenzimidazole based asymmetric hollow fibre membranes for H2/CO2 separation, J. Membr. Sci. 375 (2011) 231–240, http://dx.doi.org/10.1016/j.memsci.2011.03.049. S. Choi, J. Coronas, Z. Lai, D. Yust, F. Onorato, M. Tsapatsis, Fabrication and gas separation properties of polybenzimidazole (PBI)/nanoporous silicates hybrid membranes, J. Membr. Sci. 316 (2008) 145–152, http://dx.doi.org/10.1016/j. memsci.2007.09.026. E. Földes, E. Fekete, F.E. Karasz, B. Pukánszky, Interaction, miscibility and phase inversion in PBI/PI blends, Polymer 41 (2000) 975–983, http://dx.doi.org/10. 1016/S0032-3861(99)00236-0. G.W. Adams, J.M.G. Cowie, Blends of rigid and flexible macromolecules: poly (benzimidazole) mixed with poly(vinyl acetate-star-vinyl alcohol) copolymers, Polymer 40 (1999) 1993–2001, http://dx.doi.org/10.1016/S0032-3861(98) 00439-X. T.-S. Chung, P.N. Chen, Film and membrane properties of polybenzimidazole (PBI) and polyarylate alloys, Polym. Eng. Sci. 30 (1990) 1–6, http://dx.doi.org/10.1111/ j.0307-6962.2005.00454.x. N.A. Ahmad, C.P. Leo, A.L. Ahmad, A.W. Mohammad, Separation of CO2 from hydrogen using membrane gas absorption with PVDF/PBI membrane, Int. J. Hydrog. Energy 41 (2016) 4855–4861, http://dx.doi.org/10.1016/j.ijhydene.2015. 11.054. V. Giel, J. Kredatusová, M. Trchová, J. Brus, J. Žitka, J. Peter, Polyaniline/

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46] [47]

[48]

[49] [50]

[51]

[52]

322

polybenzimidazole blends: characterisation of its physico-chemical properties and gas separation behaviour, Eur. Polym. J. 77 (2016) 98–113, http://dx.doi.org/10. 1016/j.eurpolymj.2016.02.008. E.N. Konyushenko, M. Trchová, J. Stejskal, I. Sapurina, The role of acidity profile in the nanotubular growth of polyaniline, Chem. Pap. 64 (2010) 56–64, http://dx.doi. org/10.2478/s11696-009-0101-z. J. Stejskal, M. Trchová, J. Hromádková, J. Kovářová, A. Kalendová, The carbonization of colloidal polyaniline nanoparticles to nitrogen-containing carbon analogues, Polym. Int. 59 (2010) 875–878, http://dx.doi.org/10.1002/pi.2858. H.D. Tran, J.M. D’Arcy, Y. Wang, P.J. Beltramo, V.A. Strong, R.B. Kaner, The oxidation of aniline to produce “polyaniline”: a process yielding many different nanoscale structures, J. Mater. Chem. 21 (2011) 3534–3550, http://dx.doi.org/10. 1039/c0jm02699a. M.R. Anderson, B.R. Mattes, H. Reiss, R.B. Kaner, Conjugated polymer films for gas separations, Science 252 (1991) 1412–1415, http://dx.doi.org/10.1126/science. 252.5011.1412. Y.M. Lee, S.Y. Ha, Y.K. Lee, D.H. Suh, S.Y. Hong, Gas separation through conductive polymer membranes. 2. Polyaniline membranes with high oxygen selectivity, Ind. Eng. Chem. Res. 38 (1999) 1917–1924, http://dx.doi.org/10.1021/ie980259e. G. Ćirić-Marjanović, I. Pašti, N. Gavrilov, A. Janošević, S. Mentus, Carbonised polyaniline and polypyrrole: towards advanced nitrogen-containing carbon materials, Chem. Pap. 67 (2013) 781–813, http://dx.doi.org/10.2478/s11696-0130312-1. S. Mentus, G. Ćirić-Marjanović, M. Trchová, J. Stejskal, Conducting carbonized polyaniline nanotubes, Nanotechnology 20 (2009), http://dx.doi.org/10.1088/ 0957-4484/20/24/245601. M. Trchová, E.N. Konyushenko, J. Stejskal, J. Kovářová, G. Ćirić-Marjanović, The conversion of polyaniline nanotubes to nitrogen-containing carbon nanotubes and their comparison with multi-walled carbon nanotubes, Polym. Degrad. Stab. 94 (2009) 929–938, http://dx.doi.org/10.1016/j.polymdegradstab.2009.03.001. Z. Lei, M. Zhao, L. Dang, L. An, M. Lu, A.-Y. Lo, et al., Structural evolution and electrocatalytic application of nitrogen-doped carbon shells synthesized by pyrolysis of near-monodisperse polyaniline nanospheres, J. Mater. Chem. 19 (2009) 5985–5995, http://dx.doi.org/10.1039/b908223a. J. Stejskal, M. Trchová, J. Brodinová, I. Sapurina, Flame retardancy afforded by polyaniline deposited on wood, J. Appl. Polym. Sci. 103 (2007) 24–30, http://dx. doi.org/10.1002/app.23873. K. Briceño, D. Montané, R. Garcia-Valls, A. Iulianelli, A. Basile, Fabrication variables affecting the structure and properties of supported carbon molecular sieve membranes for hydrogen separation, J. Membr. Sci. 415–416 (2012) 288–297, http://dx.doi.org/10.1016/j.memsci.2012.05.015. J. Stejskal, R.G. Gilbert, Polyaniline preparation of a conducting ploymer (IUPAC Technical Report), Pure Appl. Chem. 74 (2006) 857–867, http://dx.doi.org/10. 1351/pac200274050857. J. Crank, The Mathematics of Diffusion, 2nd ed., Oxford University Press, Oxford, 1975. M. Trchová, Z. Morávková, I. Šeděnková, J. Stejskal, Spectroscopy of thin polyaniline films deposited during chemical oxidation of aniline, Chem. Pap. 66 (2012) 415–445, http://dx.doi.org/10.2478/s11696-012-0142-6. Z. Morávková, M. Trchová, E. Tomšík, J. Stejskal, Influence of ethanol on the chainordering of carbonised polyaniline, Chem. Pap. 67 (2013) 919–932, http://dx.doi. org/10.2478/s11696-013-0329-5. S. Van Dommele, Nitrogen Doped Carbon Nanotubes: Synthesis, Characterization and Catalysis, Utrecht University, 2008. Z. Morávková, M. Trchová, E. Tomšík, J. Čechvala, J. Stejskal, Enhanced thermal stability of multi-walled carbon nanotubes after coating with polyaniline salt, Polym. Degrad. Stab. 97 (2012) 1405–1414, http://dx.doi.org/10.1016/j. polymdegradstab.2012.05.019. B. Freeman, Y. Yampolskii, I. Pinnau, Materials Science of Membranes for Gas and Vapor Separation, John Wiley & Sons Ltd, Chichester, United Kingdom, 2006, http://dx.doi.org/10.1002/047002903X. P.S. Rao, M.-Y. Wey, H.-H. Tseng, I.A. Kumar, T.-H. Weng, A comparison of carbon/ nanotube molecular sieve membranes with polymer blend carbon molecular sieve membranes for the gas permeation application, Microporous Mesoporous Mater. 113 (2008) 499–510, http://dx.doi.org/10.1016/j.micromeso.2007.12.008.