Fabrication and characterization of a novel nanofiltration membrane by the interfacial polymerization of 1,4-diaminocyclohexane (DCH) and trimesoyl chloride (TMC)

Abstract

This study focuses on the preparation and nanofiltration properties of a novel thin-film composite polyamide membrane formed by the interfacial polymerization of 1,4-diaminocyclohexane (DCH) and trimesoyl chloride (TMC) on a porous polysulfone supporting membrane. At the same time, we found that the introduction of sodium N-cyclohexylsulfamate (SCHS) can improve to a degree the water flux and salt rejection. The active surface of the membrane was characterized by employing SEM and AFM. The performance of the nanofiltration membrane was optimized by considering the preparation conditions, including the monomer concentration, reaction time, curing conditions and SCHS concentration. The resulting NF membrane prepared under the optimum conditions exhibited a Na₂SO₄ rejection of 98.1% and a water flux of 44.6 L m⁻² h⁻¹ at 0.6 MPa. The pore size of the NF membrane was about 0.33–0.42 nm, which was calculated from the rejection of PEG and carbohydrates, respectively.

---

1. Introduction

With the current progress of urbanization and industrialization in the world, the scarcity of potable water has become an alarming question, especially in arid regions. To deal with water scarcity, many efforts have been made to remove heavy metals before its discharge and reuse wastewater. Nanofiltration (NF), as an important and environment-friendly separation technique between ultrafiltration and reverse osmosis, has attracted more and more research attention due to its low energy consumption,¹ higher rejection of multivalent salts and compounds with molecular weight >300, and broad applications in the desalination of brackish water and seawater,² wastewater treatment,³ and industrial substance separation,⁴ etc. Therefore, nanofiltration membranes with good separation performance and other excellent properties are required for more complicated applications.

At present, among the different successful NF membrane preparation techniques, interfacial polymerization is of particular interest because the selective layer and the porous support layer can be optimized separately.^5,6 Though the number of applications for NF by interfacial polymerization is increasing steadily, this technology still suffers from some drawbacks, such as membrane fouling and insufficient separation factor.⁷ Compared with membrane modification improvement, it is more accessible to ameliorate NF performance by finding a novel monomer and adding it to an outstanding additive. To address different separation requirements, a series of negatively charged nanofiltration membranes were prepared by exploiting new monomers or adding a new additive with special functional groups. These new monomers include polyhexamethylene guanidine hydrochloride,⁸ tannins,⁹ polyvinylamine,¹⁰ N-aminoethyl piperazine propane sulfonate and PIP mixtures,¹¹ 2,2′-bis(1-hydroxyl-1-trifluoromethyl-2,2,2-trifluoroethyl)-4,4′-methylene-dianiline,¹² disulfonated bis[4-(3-aminophenoxy)phenyl]sulfone,¹³ etc. Some of the additives are the following: silica,¹² poly(styrene sulfonic acid) sodium salt,¹⁴ TiO₂, Al₂O₃, ZrO₂, Al₂O₃ and TiO₂ mixtures,¹⁵ etc. Among all the preparation methods, there are few reports of aliphatic cyclic diamine as a monomer or amine salt as an additive.

In order to improve nanofiltration membrane performance, such as salt rejection, water flux and antifouling performance, information on the inherent material properties and surface structures of the active layer polymers is necessary. In this paper, we chose 1,4-diaminocyclohexane as a water-soluble monomer to prepare a novel NF membrane, considering that aliphatic amines have anti-fouling ability. Additionally, sodium N-cyclohexylsulfamate (SCHS) dissolved in the aqueous phase was introduced as the additive during the interfacial polymerization process. The resulting NF membranes were measured by a cross-flow nanofiltration system (Fig. 2), demonstrating their great potential. Remarkably, as an inorganic salt with a negative charge in solution, SCHS could not only improve the hydrophilicity and flux of the membrane, but also could enhance the anti-fouling properties of the membrane. In this work, the effects of different polymerization conditions on membrane performance were investigated, and the NF membrane performances were evaluated using different methods.

2. Experimental

2.1. Materials

1,4-Diaminocyclohexane (DCH, purity > 98%) was purchased from Tansoole; trimesoyl chloride (TMC, purity > 98%) was bought from Qingdao Ocean Chemical Company. Their chemical structures are shown in Fig. 1.

Fig. 1 Chemical structures of the monomers used: 1,4-diaminocyclohexane (DCH) and trimesoyl chloride (TMC).

Polysulfone ultrafiltration (PSF-UF) supporting membranes (MWCO = 70 000 Da) were fabricated in our lab. Sodium N-cyclohexylsulfamate (SCHS) was purchased from Tansoole; other reagents were purchased from the Sinopharm Chemical Reagent company, which were of analytical grade purity and used without further purification.

2.2. Preparation of composite nanofiltration membranes

The composite NF membranes were fabricated by an interfacial polymerization technique. PES ultrafiltration membranes were used as the porous support. The PES ultrafiltration membranes were washed thoroughly with water over 24 h before use as the porous support. Aqueous solutions of DCH and n-hexane solutions of TMC at different concentrations were prepared. The PSF membranes were first soaked in aqueous solutions for about 5 min to ensure that DCH monomers could diffuse into the porous support. The residual water on the surface was drained off by an air knife. Subsequently, the organic phase was poured on the top of the PSF-UF membrane for a predetermined time (10–60 s) for interfacial polymerization. The excess organic solution was removed from the surface, and the coated surfaces were air-dried in an oven at a certain temperature (post treatment temperature) and for a certain time (post treatment reaction time) for further polymerization reaction and hexane evaporation. Finally, the composite NF membranes were rinsed with deionized water and stored in 1% NaHSO₃ solution until they were tested.

2.3. Characterization of composite nanofiltration membranes

The membrane samples used for ATR-FTIR, SEM and AFM analysis were rinsed with de-ionized water several times and dried under a vacuum at 40 °C for 24 h.

The chemical structures of the TFC (thin-film composite) membranes were characterized using ATR-FTIR (Perkin-Elmer Spectrum 2000 FTIR spectrometer) to confirm the existence of SCHS in the membrane and the interfacial polymerization reactions.

Scanning electron microscopy (SEM, S-3400N, Hitachi) was used to analyze the surface and cross-sectional morphologies of the composite NF membranes. The membranes were fractured in liquid nitrogen to obtain clean cuts for cross sectional view. The samples were then gold sputtered for producing electrical conductivity.

Atomic force microscopy (AFM, Veeco, NanoScope III a Multimode AFM) was used to analyze the surface morphology and roughness of prepared membranes. The membrane surfaces were imaged with a scan size of 5 μm × 5 μm. The surface roughness was reported in terms of root mean square (RMS).

A contact angle measurement instrument (JC2000D, PowerEach, China) was utilized to determine the water contact angle of the NF membranes.

2.4. Permeability, rejection, pore size

The separation performance tests of the thin-film composite membranes were carried out at 0.6 MPa and 25.0 °C employing a cross-flow nanofiltration system. The system (Fig. 2) used for filtration contains a membrane cell, plunger pump, pressure gauge and solution vessel. Membranes with an effective area of 75 cm² were loaded into the cell for filtration. The feed for the permeation test was de-ionized water or de-ionized water with added solutes such as Na₂SO₄, MgSO₄, MgCl₂, NaCl, or polyethylene glycol (PEG) of different molecular weights. The permeability F was calculated using the equation:
where V (L) is the total volume of the permeate collected under a transmembrane pressure of 0.6 MPa at the time t (h) and A is the effective area of the membrane (m²). The solute rejection rate (R) was calculated using the equation:
where C_p and C_f are the solute concentrations in the permeate and feed solutions, respectively. The ion concentrations were measured using a DDS-11A conductance meter (Shanghai Neici Instrument Company). The salt rejection rate was calculated using the conductivity–concentration curves. The concentrations of organics were determined using a TOC analyzer (TOC-VCPH, SHIMADZU, JAPAN). The results presented are average data with standard deviation from at least three samples of each type of membrane.

Fig. 2 Nanofiltration assessment equipment. 1-Solution vessel, 2-heat exchanger, 3-conductance meter, 4-plunger pump, 5-crossflow cell containing the membrane, 6-permeate, 7-flowmeter, 8-bypass, 9-concentrate valve.

The pore size, pore size distribution and molecular weight cut-off (MWCO) of the SCHS/DCH–TMC NF membrane were calculated through two series permeation tests, of which one test used a group of five PEG with molecular weights of 200, 400, 600, 1000 and 2000 Da as model solutes¹⁶ and the other test used glucose, sucrose and raffinose as model solutes. The pore size distribution was calculated with the assumption of no gap and hydrodynamic interactions between the membrane material and the organic solutes. The MWCO value was taken as the molecular weight of the solute at which the membrane rejection rate was 90% . The mean effective pore radius of the membrane (r_p) was assumed to be the same as the geometric mean radius of the solute (r_s) when R equals 50%. The geometric standard deviation of the membrane (σ_p) was assumed to be the geometric standard deviation (σ_g), which is the ratio of the Stokes radius when R equals 84.13 to the Stokes radius when R equals 50%. Based on the hypothesis, the pore size distribution of membrane can be expressed as the probability density function^17–22

where r_p is the Stokes radii of the organic solutes.²³

The Stokes radii of the organic solutes used during the pore size distribution tests can be calculated from the following formulas:^18,24

For PEG, r_p = 16.73 × 10⁻¹² × M^0.557

For small molecules, ln r_p = −1.4962 + 0.4654 ln M

where M is the MW of the organic solutes.

2.5. Anti-biofouling performance assessment of the composite nanofiltration membranes

In the antifouling experiment, BSA was chosen as a representative protein in natural water sources to evaluate the antifouling properties of the NF membranes. 500 mg L⁻¹ BSA was forced to permeate through the membrane under an operation pressure of 0.6 MPa, and the flux was recorded as J_w. The antifouling experiment was carried out over a time period of 24 h. To analyze the antifouling properties in detail, several ratios were defined.^11,25–27 The flux decay ratio (DR) was calculated as follows:
where J_w0 and J_wt are the fluxes at the initial time and time t of the antifouling test, respectively. A lower DR value means better antifouling properties of the nanofiltration membrane, corresponding to only slight deposition or adsorption of fouling on the membrane surface. After the flux decay measurements, the solutions were then poured out and water was added to the filtration cell. The used nanofiltration membranes were cleaned directly in the cell for 30 min under magnetic stirring. Lastly, the cell was emptied and refilled with water again. The water flux (J_w2) of the cleaned membrane was measured. The flux recovery ratio (FRR) was calculated as follows:

The higher the FRR value, the better the antifouling properties of the nanofiltration membrane.

2.6. Long-term stability test of the composite nanofiltration membranes

A long-term test was conducted at a pressure of 0.6 MPa with a 2000 ppm Na₂SO₄ aqueous solution at pH 7.0 and 25.0 °C to investigate the durability and performance stability of the NF membrane. Periodical measurements were carried out to check the water flux and salt rejection of the membrane.

3. Results and discussion

3.1. Chemical structures of the membrane surface

The FTIR spectra of the DCH and SCHS samples and the ATR-FTIR spectra of the PSF support membrane, DCH–TMC composite membrane and SCHS/DCH–TMC composite membrane are presented in Fig. 3 to analyze the chemical structural changes of the composite membrane. In addition to the typical PSF bands of the substrate, the DCH–TMC composite membrane and SCHS/DCH–TMC composite membranes possessed additional peaks at 1651 cm⁻¹ (C=O stretch), 1619 cm⁻¹ (N–H stretch) and 1434 cm⁻¹ (C–N stretch) that corresponded to the amide group.^28–30 It can be seen from the figure that interfacial polymerization between DCH and TMC occurred and a polyamide active layer was formed. Moreover, compared with the DCH–TMC composite membrane, the SCHS/DCH–TMC composite membrane exhibited new peaks at 1111 and 951 cm⁻¹, which are the characteristic absorbances of the S=O and S–N bonds of the sulfonamide group.³¹ Based on the results of the ATR-FTIR analysis, the chemical structure of the active skin layer formed through the reaction of TMC with DCH and SCHS is shown schematically in Fig. 4, which also gives the formula for the polymerization step.

Fig. 3 FTIR spectra of (a) DCH and (b) SCHS, ATR-FTIR spectra of (c) the PSF support membrane; (d) the DCH–TMC composite membrane and (e) the SCHS/DCH–TMC composite membrane.

Fig. 4 Structure of the polyamide skin layer formed by the interfacial polymerization of CHD with TMC and SCHS.

3.2. Analysis of membrane morphology

The structure and surface roughness of the membranes were characterized by using SEM and AFM, respectively (Fig. 5 and Fig. 6). The surface and cross-section morphologies of the membranes were visualized by SEM. It can be seen clearly that the DCH–TMC composite membrane takes on a composite structure, namely a thin and dense active functional layer existing on the porous polysulfone supporting membrane, and that adding SCHS results in an increase in the skin layer density of the resulting membrane. After further measurement, the thickness of the dense layer was found to be about 200 nm. Quantitative analysis of the roughness of the surface area was made possible by AFM image statistics. The average roughness (RMS) is defined as the mean of the root of the deviation from the standard surface.²³ In Fig. 6, the RMS of various membranes are as follows: polysulfone support membrane is 5.57 nm, DCH–TMC composite membrane is 227.34 nm, and the SCHS/DCH–TMC composite membrane is 128.37 nm, which is in good agreement with the SEM results.

Fig. 5 SEM images of the membranes (left: surface; right: cross section): (a) polysulfone support membrane, (b) DCH–TMC composite membrane, (c) SCHS/DCH–TMC composite membrane.

Fig. 6 AFM images of the surface morphologies of the (a) polysulfone support membrane, (b) DCH–TMC composite membrane and (c) SCHS/DCH–TMC composite membrane.

3.3. Effects of preparation conditions on filtration performance of the nanofiltration membranes

It’s well-known that the performance of the composite nanofiltration membranes is determined by the chemistry and the preparation conditions of the thin selective layer.^32–38 In this section, the NF performance of DCH–TMC/PSF composite membranes prepared with variable conditions was investigated to determine the optimized fabrication parameters.

3.3.1. Monomer concentration. The influence of the DCH concentration on the filtration performance of the nanofiltration membranes was first investigated and is presented in Fig. 7. The DCH–TMC NF membranes were prepared using different concentrations of DCH under the conditions of TMC concentration = 0.15% (w/v), reaction time = 15 s, curing temperature = 80 °C and curing time = 5 min.

Fig. 7 Effect of DCH concentration on the salt rejection and water flux of the resulting membranes, tested with 2000 ppm Na₂SO₄ aqueous solution at 0.6 MPa, 25 °C and pH 7.0.

It can be seen clearly from the figure that when the DCH concentration of the aqueous phase was increased from 1.0 to 2.0% (w/v) at a fixed TMC concentration of 0.15% (w/v), the salt rejection of the membranes progressively increased, while the water flux decreased. When the DCH concentration exceeded 2.0% (w/v), the salt rejection changed only slightly.

Similarly, the effect of the concentration of TMC in the organic phase was also investigated to optimize the performance of the DCH–TMC NF membrane. Fig. 8 shows the performance of the composite membranes prepared under the conditions of DCH concentration = 2.0% (w/v), reaction time = 15 s, curing temperature = 80 °C and curing time = 5 min. The salt rejection of the membranes increased at first until the TMC concentration in the organic phase reached 0.25% (w/v), and then changed only slightly, while the water flux of the membranes decreased rapidly with increasing TMC concentration to 0.50% (w/v).

Fig. 8 Effect of TMC concentration on the salt rejection and water flux of the resulting membranes tested with 2000 ppm Na₂SO₄ aqueous solution at 0.6 MPa, 25 °C and pH 7.0.

This observation can be explained following the work of Freger on PA film formation kinetics and that of Nadler and Srebnik.^5,39–42 The concentrations of monomers in both the aqueous and organic phases have great effects on the rate of the interfacial polymerization. The interfacial polymerization occurring between a diamine and an acid chloride takes place on the organic side of the two phase interface. When DCH and/or TMC are at low concentration, the rate of the reactions is expected to be lower and the polyamide skin layer is formed by low molecular weight polymer, which causes lower salt rejection and higher water flux. With increasing DCH concentration and/or TMC concentration, the formation of the thin barrier layer tends to reach a maximum thickness and density since the film thickness remains almost unchanged. Thus, the water flux is decreased and the salt rejection is increased. Further increase in the concentration of DCH and/or TMC, however, tends to have little effect on the rate and extent of polymerization, so the water flux and salt rejection change only slightly.

3.3.2. Reaction time. The effect of the reaction time on the membrane performance is shown in Fig. 9. The membrane flux decreased quickly from 10 to 20 s, and then decreased slightly from 20 to 60 s. However, the membrane rejection increased quickly as the reaction time increased from 10 to 15 s, and increased slightly from 15 to 60 s.

Fig. 9 Effect of interfacial reaction time on the salt rejection and water flux of the resulting membranes tested with 2000 ppm Na₂SO₄ aqueous solution at 0.6 MPa, 25 °C and pH 7.0.

It’s well known that the interfacial polymerization between DCH and TMC occurs on the organic side of the aqueous–organic interface, and the reaction is diffusion-controlled and exists as a self-limiting phenomenon. The reaction time plays an important role in determining the extent of polymerization, and thereby the cross-linking degree and thickness of the top skin layer as well as the resulting membrane performance.^12,42–45 The thickness of the active layer of the NF membrane increases with increasing reaction time. When the thickness of the active layer is enough to prevent the DCH from diffusing from the aqueous phase into the organic phase, the top skin layer thickness will stop growing. Thus, the density of the active layer had no significant change as the reaction time was prolonged from 20 to 60 s. In this study, shorter reaction time lead to higher permeation of the water flux. As the reaction time increased, the water flux decreased and the salt rejection increased. Considering both the good salt rejection and high water flux, the reaction time of 15 s was selected as the optimum reaction time to prepare the membranes.

3.3.3. Curing temperature and time. The curing treatment of the nascent polyamide composite membrane was helpful to the diffusion of the monomers into the interface for polymerization, which increased the cross-linking degree of the polymer film. The denser cross-linking structures led to decreased mass transport across the membrane and better mechanical properties of the NF membrane.^5,46 Table 1 shows that with increasing curing temperature from 40 to 80 °C, two processes took place simultaneously: the pore size kept decreasing appreciably and the densification of the ultra-thin layer increased gradually, which resulted in higher rejection with a marginal decrease in water flux. However, further increases in curing temperature or time resulted in pore shrinkage of the support membranes and a much more compact structure of the skin layer with a consequent decrease in water flux,⁴⁶ which led to an overall decline in salt rejection. Considering the Na₂SO₄ rejection and water flux together, the optimal curing time is 5 min at 80 °C.

Table 1 Effect of the curing conditions on the performance of the DCH–TMC NF membranes

Curing temperature^a (°C)	Curing time^a (min)	Na2SO4 rejection^b (%)	Water flux^b (L m^−2 h^−1)
a Membrane preparation conditions: DCH concentration = 2.0% (w/v), TMC concentration = 0.25% (w/v), reaction time = 15 s.b Test conditions: feed = 2000 ppm Na2SO4 aqueous solution, pressure = 0.6 MPa, temperature = 25.0 °C and pH = 7.0.
40	5	75.62 ± 1.21	47.08 ± 1.23
60	5	80.78 ± 1.34	44.57 ± 2.78
80	3	90.73 ± 0.89	37.81 ± 1.97
80	5	96.83 ± 0.72	31.56 ± 2.73
80	10	97.31 ± 1.01	23.43 ± 1.99
100	5	85.39 ± 2.01	28.56 ± 2.33

---

3.3.4. Sodium N-cyclohexylsulfamate (SCHS) concentration. SCHS concentrations from 0 to 1.0% (w/v) of the composite NF membranes were investigated under the following membrane preparation conditions: 2% (w/v) DCH and a certain concentration of SCHS in the aqueous phase; 0.25 (w/v) TMC in the organic phase; reaction time of 15 s and curing temperature at 80 °C for 5 min. Fig. 10 shows the effect of SCHS concentration on the salt rejection and water flux. As the SCHS concentration increased from 0 to 0.07% (w/v), the salt rejection increased from 96.8 to 98.1% (w/v) and the water flux increased from 31.6 to 44.6 m⁻² h⁻¹. When the SCHS concentration was further increased from 0.07 to 1% (w/v), the salt rejection decreased whereas the water flux increased. It could be concluded that the proper SCHS concentration should be about 0.07%. This phenomenon can be explained as follows: the rejection rate of the charged nanofiltration membrane of salt is mainly determined by both the size and Donnan exclusion effects.⁴⁷ With increasing SCHS concentration, both the negative surface charge and pore size of the formed composite membrane increased; the increased pore size would weaken the size exclusion effect, while the increased surface charge would enhance the Donnan exclusion effect between the membrane surface and the anions. For the salt Na₂SO₄, the enhancement of the Donnan exclusion effect was dominant for the composite NF membranes prepared with SCHS concentrations lower than 0.07% (w/v). Such phenomena fit well with the rejection results of different inorganic salts for the NF composite membranes as shown in Fig. 11. As a result, the incorporation of SCHS led to the formation of composite NF membranes with improved Na₂SO₄ rejection. However, the reverse was true for the composite NF membranes prepared with SCHS concentrations higher than 0.07% (w/v), at which the weakening of the size exclusion effect would be dominant, resulting in a decline of decline of the rejection of Na₂SO₄.

Fig. 10 Effect of SCHS concentration on the salt rejection and water flux of the resulting membranes tested with 2000 ppm Na₂SO₄ aqueous solution at 0.6 MPa, 25 °C and pH 7.0.

Fig. 11 Rejection of different inorganic salts for the (A) DCH–TMC composite membrane and (B) SCHS/DCH–TMC NF membrane.

Furthermore, when comparing the results for the composite membranes with other nanofiltration membranes which were described by other references, like a PAMAM/PAN membrane⁴⁸ (15.3 m⁻² h⁻¹, Na₂SO₄ rejection 86.7%) and MPD membrane⁴⁹ (water flux: 22.8 m⁻² h⁻¹, Na₂SO₄ rejection 95.5%), the results for the composite membrane (water flux: 44.6 m⁻² h⁻¹, salt rejection 98.1%) have a higher water flux or Na₂SO₄ rejection.

3.4. Separation performance of the optimized composite nanofiltration membranes

In this section, DCH–TMC/PSF NF membranes prepared with the optimized conditions were investigated to evaluate their potential applications. The optimum conditions are as follows: 2.0% (w/v) DCH and 0.07% (w/v) SCHS in the aqueous phase; 0.25% (w/v) TMC in the organic phase; reaction time of 15 s and curing temperature at 80 °C for 5 min.

3.4.1. Inorganic salts and PEG rejection. The rejection rate of different inorganic salts was compared for membranes prepared from DCH–TMC and SCHS/DCH–TMC as shown in Fig. 11. Four kinds of salt solution were used in the experiment. It was seen that the salt rejection of the DCH–TMC or SCHS/DCH–TMC composite NF membrane decreased in the following order: Na₂SO₄ > MgSO₄ > MgCl₂ > NaCl, which demonstrated that the composite NF membranes were negatively charged membranes. In addition, the salt rejection of the SCHS/DCH–TMC composite NF membrane was higher than DCH–TMC composite NF membrane’s, which was more obvious for bivalent salts. The improvement in salt rejection arose from the introduction of strongly negatively charged functional groups on the active layer of the SCHS/DCH–TMC composite NF membrane.^50,51

3.4.2. Pore size, pore size distribution and molecular weight cut-off (MWCO). Table 2 shows the rejection of neutral solutes used during the pore size distribution evaluation. Based on this, the pore size, pore size distribution and molecular weight cut-off (MWCO) of the SCHS/DCH–TMC NF membrane were calculated and are shown in Fig. 12. The probability density function curves of the pore size distribution calculated from the rejections of PEG and carbohydrates are indicated in Fig. 12(a1) and (b1) respectively. It can be found that about 90% of the pores have a size of less than 1.5 nm. The pore size obtained from the rejections of PEG and carbohydrates are presented in Fig. 12(a2) and (b2) respectively. It was found that the pore size is about 0.33–0.42 nm. From the rejection behavior, it was found that the MWCO of the membrane is about 1000 Da. In addition to this, they also have small σ, indicating that they have narrow pore size distributions. Two tests using different solutes both proved it to be a good NF membrane in pore size, pore size distribution and MWCO.

Table 2 Rejection of neutral solutes used during the pore size distribution test

Solute	MW (g mol^−1)	Rejection (%)
PEG 200	200	64.52 ± 4.23
PEG 400	400	90.69 ± 2.37
PEG 600	600	94.72 ± 2.07
PEG 1000	1000	97.34 ± 1.56
PEG 2000	2000	97.69 ± 2.11
Glucose	180	63.56 ± 3.72
Sucrose	342	82.79 ± 2.83
Raffinose	504	92.50 ± 1.79

---

Fig. 12 (a1 and b1) Cumulative pore size distribution curves and (a2 and b2) probability density function curves of the DCH–TMC/PSF NF membranes.

3.4.3. Anti-biofouling performance of the composite nanofiltration membranes. Biofouling caused by bacterial film formation on the surface of a membrane is a severe problem during the nanofiltration process.^52,53 Foulants can absorb to the membrane surface due to hydrophobic interactions, hydrogen bonding, van der Waals attraction, and electrostatic interactions.⁵⁴ In this study, the effect of biofouling on the resulting NF membrane was evaluated by measuring the variation of the water flux. It is clearly seen from Fig. 13 that the flux declined notably within the first 1.5 h, which was caused by concentration polarization and membrane fouling. In the subsequent operation and re-suspension of BSA, equilibrium was reached due to the rigorous stirring near the membrane surface, so that a stable flux was obtained. At this point, the DR values of the SCHS/DCH–TMC composite membrane and the DCH–TMC composite membrane were 12.9% and 52.5% respectively. Finally, the flux recovered to a stable high level after simple water washing. The FRR values of the SCHS/DCH–TMC composite membrane and the DCH–TMC composite membrane reached 98.9% and 73.7%. In the above test, the resulting NF membranes exhibited excellent antifouling properties, which are attributed to the addition of SCHS.

Fig. 13 Time dependent fluxes of the NF membranes in the antifouling evaluation. The filtration operation was carried out at a temperature of 25 °C and operation at 0.6 MPa.

3.4.4. Dynamic water contact angle of the composite nanofiltration membranes. The dynamic water contact angles of the membranes (Fig. 14) demonstrate that the hydrophilicity of the PSF supporting membrane was improved. This is because the DCH–TMC composite membrane has many hydrophilic groups, namely carboxyl, amine and acylamino groups. Obviously, the hydrophilicity of the SCHS/DCH–TMC composite membrane is superior. This is attributed to the hydroxyl group of SCHS. This result also indicates that the active layer successfully formed on the surface of the NF membrane.

Fig. 14 Dynamic water contact angle of the membranes.

3.4.5. Stability of the composite nanofiltration membranes in long-time running. The long-time stability of the composite nanofiltration membranes was also very important for practical application. The water flux and salt rejection of the membranes during 10 days of filtration are presented in Fig. 15. The results showed only slight variation in the water flux and rejection of Na₂SO₄, which showed the excellent stability during 10 days of operation. The composite nanofiltration membrane with 0.07% (w/v) SCHS added maintained a high permeation flux (approximately 40 L m⁻² h⁻¹ pure water flux) and identical rejection (approximately 98% Na₂SO₄ rejection) during the whole testing period.

Fig. 15 Long-term testing of the resulting NF membranes during 10 days tested with 2000 ppm Na₂SO₄ at 0.6 MPa, 25 °C and pH 7.0.

4. Conclusions

In this work, a simple and effective approach for a novel NF membrane has been demonstrated. Thin-film composite polyamide nanofiltration membranes were successfully prepared by an interfacial polymerization technique from DCH and TMC, which was confirmed by ATR-FTIR and SEM images and measured by a NF system (Fig. 2).

The key finding is that the addition of SCHS has a significant influence on the NF performance of the resulting DCH–TMC composite membrane. The resulting NF membrane prepared under the optimum conditions exhibited a Na₂SO₄ rejection of 98.1% and a water flux of 44.6 L m⁻² h⁻¹ at 0.6 MPa. The rejections decrease in the order Na₂SO₄, MgSO₄, MgCl₂ and NaCl, with values of 98.1%, 92.0%, 80.6% and 27.0% respectively. The pore size of the NF membrane is about 0.33–0.42 nm, and the MWCO of the NF membrane is about 1000 Da, which was calculated with two different methods. In addition, the NF membrane also showed anti-fouling ability, hydrophilicity, and good stability.

Acknowledgements

The authors are thankful for the financial support received from the National Natural Science Foundation of China (51172144), Shanghai Union Program (LM201249), the Key Technology R&D Program of Shanghai Committee of Science and Technology in China (13521102000 and 14231201503), 2013 Year Special Project of the Development and Industrialization of New Materials of National Development and Reform Commission in China (20132548) and the Key Technology R&D Program of Jiangsu Committee of Science and Technology in China (BE2013031).

References

1. N. Hilal, H. Al-Zoubi, N. A. Darwish, A. W. Mohamma and M. Abu Arabi, Desalination, 2004, 170, 281–308.
2. B. V. Bruggen and C. Vandecasteele, Environ. Pollut., 2003, 122, 435–445.
3. L. Shu, T. D. Waite and P. J. Bliss, Desalination, 2005, 172, 235–243.
4. T. Gotoh, H. Iguchi and K.-I. Kikuchi, Biochem. Eng. J., 2004, 19, 165–170.
5. N. K. Saha and S. V. Joshi, J. Membr. Sci., 2009, 342, 60–69.
6. C. Wu, S. Zhang, D. Yang and X. Jian, J. Membr. Sci., 2009, 326, 429–434.
7. B. Van der Bruggen, M. Mänttäri and M. Nyström, Sep. Purif. Technol., 2008, 63, 251–263.
8. X. Li, Y. Cao, H. Yu, G. Kang, X. Jie, Z. Liu and Q. Yuan, J. Membr. Sci., 2014, 466, 82–91.
9. Y. Zhang, Y. Su, J. Peng, X. Zhao, J. Liu, J. Zhao and Z. Jiang, J. Membr. Sci., 2013, 429, 235–242.
10. M. Liu, Y. Zheng, S. Shuai, Q. Zhou, S. Yu and C. Gao, Desalination, 2012, 288, 98–107.
11. Q.-F. An, W.-D. Sun, Q. Zhao, Y.-L. Ji and C.-J. Gao, J. Membr. Sci., 2013, 431, 171–179.
12. D. Hu, Z.-L. Xu and Y.-M. Wei, Sep. Purif. Technol., 2013, 110, 31–38.
13. W. Xie, G. M. Geise, B. D. Freeman, H.-S. Lee, G. Byun and J. E. McGrath, J. Membr. Sci., 2012, 403–404, 152–161.
14. D. Hu, Z.-L. Xu and C. Chen, Desalination, 2012, 301, 75–81.
15. T. V. Gestel, C. Vandecasteele and A. Buekenhoudt, J. Membr. Sci., 2002, 207, 73–89.
16. N. Hilal, M. Al-Abri, H. Al-Hinai and M. Abu-Arabi, Desalination, 2008, 221, 284–293.
17. L.-B. Zhao, Z.-L. Xu, M. Liu and Y.-M. Wei, J. Membr. Sci., 2014, 454, 184–192.
18. S. Zhang, F. Fu and T.-S. Chung, Chem. Eng. Sci., 2013, 87, 40–50.
19. Y. Cui, H. Wang, H. Wang and T.-S. Chung, Chem. Eng. Sci., 2013, 101, 13–26.
20. S. P. Sun, T. A. Hatton, S. Y. Chan and T.-S. Chung, J. Membr. Sci., 2012, 401–402, 152–162.
21. P. H. H. Duong, T.-S. Chung, K. Jeyaseelan, A. Armugam, Z. Chen, J. Yang and M. Hong, J. Membr. Sci., 2012, 409–410, 34–43.
22. G. Han, S. Zhang, X. Li, N. Widjojo and T.-S. Chung, Chem. Eng. Sci., 2012, 80, 219–231.
23. J. Xiang, Z. Xie, M. Hoang, D. Ng and K. Zhang, J. Membr. Sci., 2014, 465, 34–40.
24. J. Xiang, Z. L. Xie and M. Hoang, J. Membr. Sci., 2014, 465, 34–40.
25. Y. Li, Y. Su, Y. Dong, X. Zhao, Z. Jiang, R. Zhang and J. Zhao, Desalination, 2014, 333, 59–65.
26. W. Chen, Y. Su, J. Peng, Y. Dong, X. Zhao and Z. Jiang, Adv. Funct. Mater., 2011, 21, 191–198.
27. P.-Y. Zhang, Z.-L. Xu, H. Yang, Y.-M. Wei, W.-Z. Wu and D.-G. Chen, Desalination, 2013, 319, 47–59.
28. B. A. Anderson, A. Literati, B. Ball and J. Kubelka, Spectrochim. Acta, Part A, 2015, 134, 473–483.
29. A. K. Ghosh, B.-H. Jeong, X. Huang and E. M. V. Hoek, J. Membr. Sci., 2008, 311, 34–45.
30. S. A. Sundet, W. A. Murphey and S. B. Speck, journal of polymer science, 1959, XL, 389–397.
31. S. Muthu, G. Ramachandran, E. Isac Paulraj and T. Swaminathan, Spectrochim. Acta, Part A, 2014, 128, 603–613.
32. Y. Ji, Q. An, Q. Zhao, H. Chen and C. Gao, J. Membr. Sci., 2011, 376, 254–265.
33. M. N. A. Seman, M. Khayet and N. Hilal, J. Membr. Sci., 2010, 348, 109–116.
34. Y. Liu, B. He, J. Li, R. D. Sanderson, L. Li and S. Zhang, J. Membr. Sci., 2011, 373, 98–106.
35. X.-Z. Wei, L.-P. Zhu, H.-Y. Deng, Y.-Y. Xu, B.-K. Zhu and Z.-M. Huang, J. Membr. Sci., 2008, 323, 278–287.
36. Y.-C. Chiang, Y.-Z. Hsub, R.-C. Ruaan, C.-J. Chuang and K.-L. Tung, J. Membr. Sci., 2009, 326, 19–26.
37. F. Yang, S. Zhang, D. Yang and X. Jian, J. Membr. Sci., 2007, 301, 85–92.
38. S. Yu, M. Ma, J. Liu, J. Tao, M. Liu and C. Gao, J. Membr. Sci., 2011, 379, 164–173.
39. R. Nadler and S. Srebnik, J. Membr. Sci., 2008, 315, 100–105.
40. S. S. Dhumal and A. K. Suresh, Polymer, 2009, 50, 5851–5864.
41. W. Xie, G. M. Geise and B. D. Freeman, J. Membr. Sci., 2012, 403, 152–161.
42. D. Hu, Z.-L. Xu and C. Chen, Desalination, 2012, 301, 75–81.
43. A. Ahmad and B. Ooi, J. Membr. Sci., 2005, 255, 67–77.
44. B. Su, T. Wang, Z. Wang, X. Gao and C. Gao, J. Membr. Sci., 2012, 423–424, 324–331.
45. A. Abidi, N. Gherraf, S. Ladjel, M. Rabiller-Baudry and T. Bouchami, Arabian J. Chem., 2011 DOI:10.1016/j.arabjc.2011.04.014.
46. P. Daraei, S. S. Madaeni, N. Ghaemi, H. Ahmadi Monfared and M. A. Khadivi, Sep. Purif. Technol., 2013, 104, 32–44.
47. J. M. M. Peeters, J. P. Boom and M. H. V. Mulder, J. Membr. Sci., 1998, 145, 199–209.
48. X.-X. Xu, C.-L. Zhou, B.-R. Zeng, H.-P. Xia, W.-G. Lan and X.-M. He, Sep. Purif. Technol., 2012, 96, 229–236.
49. Y. Li, Y. Su, J. Li, X. Zhao, X. Fan, J. Zhu, Y. Ma, Y. Liu and Z. Jiang, J. Membr. Sci., 2015, 476, 10–19.
50. B. B. Tang, Z. B. Huo and P. Y. Wu, J. Membr. Sci., 2008, 320, 198–205.
51. H. W. Sun, G. H. Chen, R. H. Huang and C. J. Gao, J. Membr. Sci., 2007, 297, 51–58.
52. Y.-C. Chiang, Y. Chang, C.-J. Chuang and R.-C. Ruaan, J. Membr. Sci., 2012, 389, 76–82.
53. Y.-L. Ji, Q.-F. An, Q. Zhao, W.-D. Sun, K.-R. Lee, H.-L. Chen and C.-J. Gao, J. Membr. Sci., 2012, 390–391, 243–253.
54. Y.-N. Wang and C. Y. Tang, J. Membr. Sci., 2011, 376, 275–282.

---