Irena Petrinic, Niels Peder Raj Andersen, Kristian Keiding, Mitja Kolar and Sonja Sostar-Turk

• Introduction

Around 25% of the dyes used in textile printing today are reactive dyes and they represent an increasing market share because they are used to dye cotton, which makes up about half of the world’s fibre consumption [1]. The fixation rate of reactive dyes on cotton is only 70% and reactive dye, therefore, contributes significantly to the overall colourization of the textile wastewater. As a result, dyeing/printing effluent are usually highly coloured, containing utilized hydrolyzed dye at concentrations ranging from 0.6 - 0.8g dye/L [2].

Wastewaters remaining from the washing and rinsing processes of printed textiles are very complex with high concentrations of not only dyes, but also of thickeners and auxiliary chemicals such as salts, urea and oxidizing agents (Rapidoprint), which help to increase the rate of fixation on the textiles. A thickener such as sodium alginate is widely used in reactive printing. Alginate serves as a protective agent for the various dyes and chemicals dissolved in the printing paste, and it ensures that the dye is transferred from the paste to the substrate. It contains hydroxyl groups, which at in the pH range 10.5 - 11.5 do not react with carboxyl or sulfonic groups at the dye due to mutual anion repulsion [3].

Thus, alginate does not stay in the fabric but is discharged into the wastewater. Salt is required to shift in the equilibrium position of the dye and its aggregates in solution in favour of the fibre surface, and to compensate for the unfavourable effects of the repulsions caused by the interaction of dye anions and the negatively charged cellulose [4].

Sodium bicarbonate/carbonate is added to give the printing paste sufficient stability with most reactive dyes [5]. Urea maintains the swelling of the fibre, keeps the dye in solution and increases the diffusion time during which the dye is delivered to the interior of the fibre [4]. The oxidizing agent, Sodium m-ni-trobenzene sulfonate, minimizes the reduction of the reactive dyes during the steaming process, and depends on the nature of the chromophoric group. Together with hydrolyzed dye all these compounds form an electrolyte rich, highly coloured wastewater.
A variety of physical-chemical methods have been studied for the removal of colour from textile wastewater effluent. These studies include the usage of oxidizing agents [6], membrane filtrations [7], electro-chemical [8] and adsorption [9]. The advantages and disadvantages of each technique have been reviewed recently [10]. The main issues for many of these treatments are the large cost and the large variation in concentration of waste streams [11]. The need for more efficient treatment processes has attracted attention towards pressure driven membrane techniques because the application of membrane filtration processes not only enables high removal efficiencies, but also allows the water and some of the valuable waste constituents to be reused [12]. Recently, nanofiltration and reverse osmosis systems were tested for reuse of rinsing water from reactive dyeing of cotton [13].  The experiments, carried out in laboratory and pilot scale, suggested that the water quality obtained from nanofiltration could be used for rinsing. Yet, the separation of salt/ dye with nanofiltration and reverse osmosis has only been carried out on rinsing water effluent from the reactive dyeing of cotton, and not printing effluent containing thickener. The purpose of this study is, therefore, to investigate the possible application of nanofiltration for treating textile wastewater from reactive dye printing and subsequent washing process. Synthetically-prepared wastewaters which were similar to the wastewater taken from a local textile factory in Slovenia were studied using two nanofiltration membranes, NFT-50 and DL.

The retention of compounds (reactive dye, sodium ion and COD) and a reduction in the conductivity were determined in order to monitor the membranes separation efficiency.

2- Materials and methods

2.1. Filtration unit

The experiments were carried out with a DSS Lab 20 unit in which four membranes were installed in series (Fig.1). The membrane area of each membrane sheet was 0.018 m2 and the technical characteristics for each membrane are given in Table-1.

All experiments were performed in batch circulation mode, in which both permeate and retentate were carried back to the feed vessel. The retentate outlet pressure was varied from 2-15 bar. The filtrations were carried at 0.4 m/s cross-flow velocity.

2.2. Composition of the synthetically prepared     textile wastewater

The studied wastewater was adapted from the washing process performed after reactive printing on textiles. The substances, which are removed whilst washing off the prints, are thickener, unfixed dye and auxiliaries applied to the material as part of the printing process. The concentration of these components in the wash-water depends on the component concentration in the printing paste formulation. Table-2 presents the composition of the synthetically prepared wastewater used

during the research, which was similar to the wastewater taken from a local textile factory in Slovenia

Two different synthetically prepared wastewaters were used according to the recipe in Table-2. The concentrations of the auxiliary chemical were kept constant, only the dye types (C.I. Reactive Red 24 and C.I. Reactive Black 5) in the concentration of o.47 g/L were varied. The solution contained one of the reactive dyes in each experiment.

C.I. Reactive Red 24 was supplied by CHT Bezema, Switzerland and C.I. Reactive Black 5 was supplied by DyStar, Germany (Fig.2).

CHT-Alginate MV and Rapidoprint XRN Pearls were supplied by CHT R. Beithlich GMBH, Germany.

The membranes separation efficiency was monitored by measuring the removal efficiency of conductivity, Na+, colour and COD. The removal efficiency equals the retention defined as:

Retention = 1-   Concentration in permeate
                       Concentration in retentate

2.3. Analytical methods

Feed and permeate colour measurements were performed using a Cary 50 Varian Spectrophotometer. Colour concentration was measured at a maximum absorbance wavelength of 535 nm for C.I. Reactive Red 24 and 599 nm for C.I. Reactive Black 5 at the original pH = 9.6.COD was measured according to the Standard methods [14]. The concentration of sodium ion was determined by atomic absorption spectrometry (Perkin Elmer 1100B).

3- Results and discussion

3.1. Water permeability

Fig.3 shows the permeate flux versus applied pressure for both membranes. The data obtained show a linear profile. The slope of the straight lines gives the water permeability value for each membrane and determines how fast the component is transported through the membrane. The slope is a measure of the resistance exerted by the membranes as a diffusion medium, when a given force (pressure) is acting on a component [15]. Table-3 lists the water permeability values in Lm-2h-1 or LMH. The obtained permeability values are in the range normally obtained with NF membranes available from the market, which range from 1 to 50 LMH/bar [16].

3.2. Membrane permeate flux

 Permeate flux is defined as the volume of permeate flowing through the membrane pr. Unit area and time, and is affected by several factors such as feed pressure, operating temperature, feed velocity and composition. Fig.4 shows the relationship between permeate flux and trans-membrane pressure for NTF- 50 and Fig.5 for DL with wastewater solutions containing C.I. Reactive Red 24 and C.I. Reactive Black 5, respectively. The figures also show the relationship for pure water, where permeate flux increases linearly with trans-membrane pressure. With wastewater the permeate flux do not increase linearly with pressure when filtrating wastewaters at high pressures. Cheryan (1998) [17] defined these two distinct as a pressure controlled region and mass-transfer controlled region. In the mass-transfer controlled region, increasing the pressure only results in a build up of a solute layer. During the filtration, solvent and solutes are transported to the membrane surface. This is the result of increasing concentration polarisation on the feed site of the membrane. Concentration polarisation is related to the transport of solvent and solutes to the membrane surface during the filtration. As the solvent and permeable solutes pass through the membrane, the concentration of the retained solutes increases until the convective transport rate of the solutes to the membrane surface equals the diffusion transport rate of the solutes out of the boundary layer (i.e. from the membrane surface to the feed solution). The thickness of the boundary layer decreases when the cross-flow velocity is increased whereby higher permeate flow is obtained in the mass-transfer controlled region.

The permeate flux values for NFT 50 for C.I. Reactive Red 24 and C.I. Reactive Black 5 wastewaters did not show significant difference. At 10 bar the permeate flux values were 56 and 47 LMH, respectively. However, while using the DL membrane a significant difference in the permeate flux values for C.I. Reactive Red 24 and C.I. Reactive Black 5 wastewaters was observed. Thus, at 10 bar the permeate flux values were 73 and 52 LMH, respectively. This coincides with the observation of the C.I. Reactive Black 5 dye’s greater tendency to be adsorbed on the membrane surface due to the greater molecular size. It was reported that flux decline occurs by a combination of two effects: molecular size and adsorption of the organic molecules [18].

However, after each experiment the membranes were rinsed with demineralised water and about 85% of the initial permeate flux was obtained for both membranes. This means that the observed flux decline due to adsorption is nearly reversible.

3.3. Permeate quality

The membranes separation efficiency was monitored by measuring the removal efficiency of conductivity, Na+, colour and COD. Table-4 and 5 show retention values of conductivity, Na+, colour and COD at a different trans-membrane pressure.

As shown in Table-4 the highest colour removal efficiency was obtained with the NFT-50 membrane. The retentions varied from 0.994 to 1 or practically colorless samples Fig.5. However, a decrease in colour retention could be seen in the DL permeate sample for all pressure range (Fig.6) where retentions varied from 0.90 to 0.996. The latter had a lower retention achieved with the C.I. Reactive Red 24 where all permeates were coloured and retention decreased from 0.98 (pressure 2 bar) to 0.90 (pressure 15 bar), but while using NFT-50 membrane no difference due to dye type used was experienced. The overall increase between NFT-50 and DL membrane was 10%.

For the NFT-50 membrane the retention of Na+ ions was higher than the DL membrane and the retention of conductivity followed a similar trend (Table-4 and 5). This was expected, since the pore structures of NFT-50 membrane are tighter than in the DL membrane, so together with dye the major part of the ions is removed. However, for both wastewaters (C.I. Reactive Red 24 and C.I. Reactive Black 5), and for membranes, Na+ and conductivity removal efficiency increased with the pressure up to 7/10 bar.

Above this pressure the removal efficiency decreased again.
The wastewater had a high COD ranging from 8700 mgO2/L (C.I. Reactive Red 24) to 7300 mgO2/L (C.I. Reactive Black 5). By using the NFT-50 membrane the CODwas reduced  to 3100  the NFT-50 membrane the COD was reduced to 3100  mgO2/L and 3800 mgO2/L for wastewater with C.I.Reactive Red 24 and C.I.Reactive Black 5, respectively. This corresponds to more than 64% removal. A much lower removal was achieved with DL membrane: 29% for ww-C.I. Reactive Red 24 and 44% for ww-C.I. Reactive Black 5, respectively. This is due to the fact that DL membrane has a more open structure which allows smaller molecules such as urea, rapidoprint (nitrobenzene sulphonate) to permeate through the membrane.

4. Conclusion

The separation performance of two nanofiltration membranes for treating coloured textile wastewater effluent was investigated. The filtration process was carried out on synthetically prepared textile wastewater simulating the wastewater obtained after reactive dye printing and subsequent washing. NFT-50 and DL nanofiltration membranes were used. Both membranes proved to be efficient for retaining conductivity. The overall removal efficiencies for conductivity, Na+ and colour was 86% 89%and 99.4% for NFT-50 membrane and 33.7%, 99.1% and 70.7% for the DL membrane, respectively. However, the COD removal efficiency was only 64% for the NFT-50 membrane and 44 for the DL-membrane.
It was observed that the DL membrane had better operational performance (higher permeate flux) than the NFT-50 membrane. In the case of multi-component mixtures, as this wastewater is, driving force and fluxes are coupled. Therefore, higher fluxes obtained with DL membrane resulted not only in a higher solvent flux but also in a higher mass flux of components through the membrane (lower retention values) and development of a concentration gradient. However, the better separation efficiency achieved with NFT-50, where the permeate samples were practically colorless, clearly compensate for the lower permeate flux. Thus, the work demonstrates that by using the proper membrane and proper membrane operation conditions it is possible to apply nanofiltration in the treatment of reactive dye printing textile wastewater.

References

1-   Phillips D.: Environmentally friendly, productive and reliable: priorities for cotton dyes and dyeing processes, J.Soc. Dyes and Color 112 (1996)183-186

2-      Vandivevere P.C. et al: Treatment and reuse of wastewater from the textile wet-processing industry: Review of emerging technologies, J.Chem.Technol.Biotechnol. 71 (1998) 4, 289-302

3-      Whistler RL and BeMiller JN.: Industrial Gums, Polysaccharides and their Derivatives, 3rd edition, Academic Press Limited, London, 1993

4-   Hamlin J.D. et al: UV/Visible spectroscopic studies of the effects of common salt and urea upon reactive dye solutions, Dyes and pigments 41 (1999)137-142

5-   Miles LWC.: Textile printing, Second edition, Society of Dyes and colourists, The Alden Press, Oxford, 1994
 
6-   Ince N.H., D.T. Gonenc: Treatability of a textile azo dye by UV H2O2,Environmental technology 18 (1997)2, 179-185

7-   Xu L., R.E. Lebrun: Treatment of textile dye plant effluent by nanofiltration membrane, Separation science and technology 34 (1999) 13,2501-2519

8-   Ding Z. et al: A study of the use of bipolar particles-electrode in the decolorization of dyeing effluent and its principle, Water Science and Technology 19 (1987) 39-44

9-   Walker G.M., L.R. Weatherley: Adsorption of acid dyes on to granular activated carbon in fixed beds. Water Research 31 (1997) 8, 2093- 2101

10-      Robinson T.et al: Remediation of dyes in textile effluent: a critical review on current treatment technologies with a proposed alternative, Bioresource Technology 77 (2001)   3, 247-255

11- Schafer A.I., A.G. Fane, T.D. Waite: Nanofiltration: Principles and Applications, Elsevier, Advanced Technology ISBN 1- 85617-405-0, Oxford, UK, 2005

12- Akbari A. et al: Treatment of textile dye effluent using a polyamide-based nanofiltration membrane, Chemical Engineering and Processing 41 (2002) 601-609

13- Sojka-Ledakowicz J. et al: Membrane filtration of textile dyehouse wastewater for technological water reuse. Desalination 119 (1998) 1-3, 1-9

14-      International Standard: Determination of the chemical oxygen demand, reference number ISO 6060, International Organization for Standardization, Geneve, Switzerland, 1989

15- Mulder M.: Basic Principles of Membrane Technology, Kluwer Academic, 1991

16- Bowen W.R., AW. Mohammad: A theoretical basis for specifying nanofiltration membranes - dye/ salt/water streams, Desalination 117 (1998) 257-264

17-      Cheryan M.: Ultrafiltration and Microfiltration Handbook, Technomic Publishing. Co. Inc., Lancaster. Pen., 1998

18-      Koyuncu I., D. Topacik: Effects of operating conditions on the salt rejections of nanofiltration membranes in reactive dye, salt mixtures. Separation and Purification Technology 33 (2003) 283-294.