Novel mixed-matrix-membrane with activated carbon for steel-industry effluent treatment

Waste water effluents from various iron and steel industries contain a lot of harmful chemicals. Especially the by-products from coke-oven plant contain phenols, which is very toxic and carcinogenic [1, 2]. This group of compounds is in the 11th position out of 126 chemicals, which are termed as priority pollutants by United States Environmental Protection Agency. Exposure to phenol has many bad effects like high irritation to eyes, skin and mucous and also causes headache and dizziness. Long term exposure can even result in high blood pressure, liver and kidney damage. Present technologies applied for removal of
phenolic compounds are solvent extraction, adsorption, chemical oxidation, biological treatment and distillation [3-7]. However, the need for cost-effective, simple and high throughput technology still persists. It has been found that, activated carbon is very good adsorbent of phenolic compounds [8]. But, this process is slow and less economic. On the other hand, membrane separation technology is a physical separation process with high throughput and it is also external chemical free. Mixed matrix membrane combines the goodness of adsorptive property of inorganic additives and simplicity of membrane technology. Mixed matrix membrane of some base polymer and activated carbon can cater to the need of phenol removal technology. In this work, one flat sheet mixed membrane of cellulose acetate phthalate (CAP) and powdered activated carbon (PAC) has been fabricated. Various properties of this membrane are compared with the undoped CAP membrane. Effluent of a steel industy is filtered using the doped membrane. It has been found that the doped membrane has higher permeability, better hydrophilicity, lesser porosity and better phenol adsorption properties. This special MMM is found to be capable of reducing the phenol concentration to a great extent.
Flat sheet mixed matrix membrane of cellulose acetate phthalate and activated carbon was cast by phase inversion method. Casting solution was prepared by mixing PAC (25%) and CAP(15%) with solvent di-methyle formamide (DMF) (60%), and the solution was sonicated for one hour by probe-type sonicator. Then one nonwoven fabric was attached with a glass plate. Using a doctor’s knife, the membrane was cast by manual draw down at a speed of 20 mm/min. The membrane was immediately transferred to water bath. Then for complete phase inversion, the membrane was kept in water bath for 12 hours.
The morphology the membrane was analysed using scanning electron microscopy (SEM) and atomic force microscopy (AFM). SEM images of the top surface of the undoped and dopes membranes are given in Fig. 1(a) and 1(b), respectively. Presence of carbon particles on the top can be observed from SEM image of the top surface of the doped membrane, which is absent in the undoped membrane. Also, AFM image analysis confirmed that average surface roughness increases with addition of carbon. The average r.m.s. roughness of undoped membrane is 12 nm, where as that is 247 nm in dopes membrane. These findings corroborate that activated carbon particles are accumulated mainly on the surface of the membrane. PAC particles are hydrophilic in nature. During the phase inversion process, activated carbon particles come toward the surface. Cross-sectional SEM images of the undoped and the doped membranes are presented in Fig. 1(c) and 1(d), respectively. From the cross-sectional image, we can see that pores are more open for the doped membrane. As PAC is hydrophilic in nature, addition of PAC induces thermodynamic instability, delaying instantaneous demixing. Hence the pores are larger.




Fig 1: SEM image of (a) undoped membrane top surface, (b) doped membrane top surface; (c) undoped membrane cross-section, (d) doped membrane cross-section.
Porosity of the two membranes were measured by gravimetric analysis. The mixed- matrix membrane has lesser porosity (48%) than the undoped membrane (68%). Carbon particles sit inside the pores and block those. Hence, porosity is reduced. Hydrophilicity of the membrane increased significantly. Contact angle of distilled water with the membrane was 69o for the undoped membrane, which reduced to 56o for doped membrane. Activated carbons are accumulated on the surface of the doped membrane. As, PAC is hydrophilic, overall hydrophilicity of the membranes are increased. Distilled water permeability of the membrane increased because of higher hydrophilicty. The linear permeability increases fron
5.3 to 7.8 m3/m2.Pa.s for the composite membrane. Hydrophilic surface draws more water toward the membrane, as a result, permeability of the mbrane is increased. The mechanical
strength of the membrane was measured by Universal Testing Machine. The doped membrane has a tensile strength 19.5 MPa, indicating that it can with stand high pressure.
To study the effect of pH on phenol adsorption on activated carbon, phenol adsorption isotherms of different pH values, like 5.5,7,8 and 10 were prepared. It was observed that at lower pH, adsorption of phenol on membrane is higher. As the pH was increased, adsorption capacity goes down. Some basic sites on the activated carbons are developed because of the basic surface oxygen complexes and/or p-electron reach complexes at the basal plane. Among these basic sites and the aromatic rings of the phenols, electron donor-acceptor complexes are formed. At lower pH, this chemistry is favoured, hence higher adsorption is observed. However, at a pH, higher than pKa value of the phenol, phenol compounds dissociate in phenolate ion, having residual negative charge. Electrostatic repulsion between phenolate ions and negatively charged membrane results in lower adsorption capacity [9].
To test the applicability of the membrane in phenol removal process, real life effluent from TATA Steel India Limited was collected which contains phenol in significant amount (around 17-24 mg/l). The effluent stream was passed over the membrane in cross-flow filtration unit, where the fluid flow is in parallel with the membrane surface. The result of which is presented in Fig. 2.

4.5

4.0

3.5

3.0

2.5

2.0

1.5

0 90 180 270 360 450 540

Time (min)

200

180

160

140

120

100

80

Fig. 2: Permeate flux and concentration of effluent, after cross-flow filtration, by the doped membrane at 70 kPa and 20 lph flow-rate.
It can be observed that, in the initial time period of the filtration process, total phenol concentration in the permeate stream is rapidly reduced. After 240 minutes of filtration total phenol content in permeate reduces down to 1.8 mg/l, and remains almost steady there-after. Phenol compounds are adsorbed by the membrane, hence permeate contains less pheol. Initial
stage of rapid decline occurs because of unsteady state adsorption. After the steady state (dynamic equilibrium) is reached, steady concentration of phenol in permeate can be observed.
From Fig. 2, we can see that permeate flux declines rapidly from initial 195 l/m2h to
96 l/m2h, after 60 minutes. In the initial stage, solid particles rapidly form a cake layer on the membrane which offers additional resistance to permeate stream in its flow path. As more solids are filtered, the layer grows in depth. After sometime, (in this case, it is around 60 minutes) the cake layer does not grow further, as equilibrium is achieved. This cake layer, as well as the carbon in the membranes aids to the filtration of phenolic compounds. The phenolic compounds are adsorbed by the membrane and resisted by the cake layer. Once the cake layer stops growing, flux becomes almost steady at a certain value (in this case, around
96 l/m2 h).

Reference:

[1] Canadian Environmental Protection Act, Priority Substances List Assessment Report: Phenol, 1999, www.ec.gc.ca.
[2] R. J. Lifton, The Nazi Doctors, first ed., Basic Books, New York, 1986, pp. 254–268.
[3] Z. Li, M. Wu, Z. Jiao, B. Bao, S. Lu, Extraction of phenol from wastewater by N-octoyl- pyrolidone, J. Hazard. Mater. B114 (2004) 111–114.
[4] L. Damjanovic, V. Rakic, V. Rac, D. Stošic, A. Auroux, The investigation of phenol removal from aqueous solutions by zeolites as solid adsorbents, J. Hazard. Mater. 184 (2010)
477-484.
[5] Y. Yavuz, A. S. Koparal, Electrochemical oxidation of phenol in a parallel plate reactor using ruthenium mixed metal oxide electrode, J. Hazard. Mater., 136, (2006) 296-302.
[6] Y. Wang, J. Song, W. Zhao, X. He, J. Chen, M. Xiao , In situ degradation of phenol and promotion of plant growth in contaminated environments by a single Pseudomonas aeruginosa strain, J. Hazard. Mater. 192 (2011) 354-360.
[7] H. G. Franck, J. W. Stadelhofer, Industrial Aromatic Chemistry, Springer Verlag, Berlin, (1989) 148–157.
[8] A. Da’browski , P. Podkos´cielny, Z. Hubicki, M. Barczak, Adsorption of phenolic compounds by activated carbon—a critical review, Chemosphere 58 (2005) 1049–1070.
[9] C. Moreno-Castilla, J. Rivera-Utrilla, M. V. Lopez-Ramon, F. Carrasco-Marin, Adsorption of some substituted phenols on activated carbons from a bituminous coal, carbon
33 (1995) 841-851.