The FF (flocculation–flotation) process

The FF (flocculation–flotation) process

Minerals Engineering 18 (2005) 701–707 This article is also available online at: The FF (flocculation–flotation) process...

351KB Sizes 9 Downloads 74 Views

Minerals Engineering 18 (2005) 701–707 This article is also available online at:

The FF (flocculation–flotation) process Jailton J. da Rosa, Jorge Rubio


Laborato´rio de Tecnologia Mineral e Ambiental, Departamento de Engenharia de Minas, PPGEM-Universidade Federal do Rio Grande do Sul., Av. Osvaldo Aranha 99/512, 90035-190 Porto Alegre, RS, Brazil Received 18 September 2004; accepted 19 October 2004

Abstract A new on-line flocculation system (FF) has been developed which is coupled with a rapid flotation to remove the aerated flocs (flocs with entrained and entrapped bubbles). These aerated flocs are formed only in the presence of high molecular weight polymers and bubbles and under high shearing (and head loss) in special ‘‘flocculators’’. The air excess air abandons the flotation tank (a centrifuge or a column) by the top and the flocs float after very short residence times (within seconds). The aerated flocs are large units (some millimetres in diameter) having an extremely low-density. Process efficiency was found, in all cases, to be a function of the trilogy, head loss, type (and concentration) of flocculants and air flow rate. Mechanisms involved appear to include small bubble formation and their rapid occlusion (entrapment) within flocs, nucleation of bubbles at floc/water interfaces, polymer coiling as a result of ‘‘salting out’’ effects at the aqueous/air interface and plug flow type of mixing (flocculation) instead of perfect. Successful examples of emulsified oil and solids removal from water are shown and because in all cases were obtained high efficiencies (>90% removal), at high hydraulic loadings (>130 m h 1) it is believed that this kind of flocculation–flotation appears to have a great potential in solid/liquid or liquid/liquid separation.  2004 Elsevier Ltd. All rights reserved. Keywords: Flocculation; Flotation; Wastewater treatment; Environmental

1. Introduction Flocculation and flotation processes in the mineral industry are primarily designed to separate one particle type from another. In contrast, for wastewater treatment, the flotation is designed to remove all particles—generally encountered as very fine emulsions, suspended solids, microorganisms and colloidal dispersions. Thus, processes are optimized by the maximum recovery of cleaned water with the lowest concentration of pollutants and sludge containing low percentage of solids (or oils) (Rubio, 2003).

* Corresponding author. Tel.: +55 51 3316 3540; fax: +55 51 3316 3530. E-mail address: [email protected] (J. Rubio). URL:

0892-6875/$ - see front matter  2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.mineng.2004.10.010

Industrial effluents and even flotation tailings commonly bear wastewater that contains a mixture of suspended particles and stable oil emulsions. It is well known that it is difficult to remove fine colloidal particles and highly emulsified oil from process wastewater. Oil can be present as a ‘‘free’’, non-dispersed surface layer, usually floating at the air/water interface. The oily layers can be readily separated off by gravity but the separation, as in the case of fine particle dispersions, is always very poor in the case of oil-in-water emulsions, especially if oil is present as a physically dispersed phase in the form of fine droplets, say <10 lm. The separation is even more difficult when emulsions are stabilized with surfactants or other emulsifying agents (Toyoda et al., 1999). Flotation of organic fluids such as oil spills, oily sewages or oil-in-water emulsions has been known for decades in various fields but is not commonly used in the


J.J. da Rosa, J. Rubio / Minerals Engineering 18 (2005) 701–707

mining and/or metallurgy industries. Oil in water may be dispersed, emulsified or soluble in concentrations usually up to 1000 mg L 1. Here, residual oily waste-waters are common in the form of flotation and solvent extraction reagents losses, free wasted oil and oil spills in process waters (Gu and Chiang, 1999; Capps et al., 1993; De Oliveira, 1995). The flotation separation of very fine oil droplets (<10 lm) or dispersed fine solids is even more complicated and usually requires fine bubbles, efficient flocculants, and quiescent hydrodynamic conditions in the cell separation zone (Gopalratnam et al., 1988). This is due to collection and adhesion factors, which makes the process very slow, especially when treating high flow-rates. Induced air flotation, IAF and dissolved air flotation, DAF, have been used extensively in the removal of stable oily emulsions or fine particles suspensions (Strickland, 1980; Bennett, 1988; Van Ham et al., 1983). IAF utilizes bubbles sizing between (600–2000 lm), turbulent hydrodynamic conditions and has low retention times, normally <5 min. Conversely, DAF employs micro-bubbles (30–100 lm), and quiescent regimes. However, when retention times are higher (20–60 min) this process is inefficient when treating effluents having high volumes and high flow-rate. A new basis for the separation of oilin-water emulsions, based on the concept of carrier flotation, has been reported (Rubio and Santander, 1997). Here, the carrier solids (coal or coal beneficiation residues) adsorbs or absorbs the oil extensively and flotation is used to separate off the loaded oily adsorbent. The addition of polymeric flocculants may, sometimes, assist the particles settling but the efficiency highly decreases when dispersions are diluted or when particles are in the range of ultrafines or colloidal range (Rubio, 2003). The recovered water in most cases carries suspended particles affecting the clarity and quality. On the other hand, flotation, in those cases, is a more reliable technique for the removal of diluted (<4% solids content) suspensions and oily emulsions from wastewater (Da Rosa et al., 2002). Yet, the classical dissolved air flotation (DAF) is still the most common process removing fine colloidal dispersions and oily emulsions, mainly in refinery wastewaters (Kiuru, 2001; Rubio et al., 2002b). In DAF, a stream of treated wastewater (recycle) is saturated with air at elevated pressures up to 5 atm (40–70 psig). Bubbles are formed by a reduction in pressure of the water pre-saturated forced to flow trough needle valves or special orifices, and clouds of bubbles, 30–70 lm in diameter, are produced just down-stream of the constriction (Rodrigues and Rubio, 2003). More, recently, DAF has been employed to remove suspended solids from neutralized AMD waters (acid mining drainage water) (Menezes et al., 2004) and to remove ions from copper concentrates filtered water (Rubio, 2003; Rubio et al., 2002a).

However, future technologies will have to deal with highly loaded (high solids by weight) process wastewaters, exiting mining and metallurgical industries, and DAF might not meet legislation standards and reuse efficiently the water, due to the low carrying power of the tiny bubbles and the low hold-up. For this reason, DAF may be considered a slow process with high residence time (minutes) and requiring high foot print space. Various publications (Rubio et al., 2002a; Voronin and Dibrov, 1999; Matis, 1995; Mavros and Matis, 1992; Parekh and Miller, 1999), reviewed fundamentals and general features of flotation (usually accompanied by flocculation) for environmental applications. All publications show the great potential of conventional and upcoming novel separation concepts and devices. This article constitutes an advance within this line of research and development.

2. The FF (flocculation–flotation) process: development and main features The flocculation–flotation system (FF, Rubio et al., 2003) is composed of a turbulent ‘‘flocculator’’ to generate aerated polymeric flocs coupled with solid/liquid, solid/liquid/liquid2 or liquid/liquid2 separation devices (columns, tanks, centrifuges). Here, the basic concept is that of a reactor (zigzag or static mixer types) (Fig. 1) of flocculator (Fig. 2) and a floc flotation separator (Fig. 3). The resulting flocs are rapidly formed inside the flocculator, are very light because of the trapped air (see below). Yet, these ‘‘special’’ flocs are generated only in the presence of high molecular weight polymers, bubbles (from the injected air), high shearing forces (caused by the zigzag kind of flow) and a high head loss. Process efficiency was found, in all cases, to be a function of the trilogy, head loss, type (and concentration) of flocculants and air flow rate (Da Rosa et al., 2002; Rubio, 2003; Rubio et al., 2003). In the flotation tank separator the floc float, within seconds, as large units (some millimeters in diameter) having very low densities. The exceeding air abandons the flotation device by the top through a special water seal (avoiding flow turbulence). Conversely, in conventional flocculation, the polymeric floc (non-aerated) are commonly formed after polymer diffusion and adsorption at the solid particle/ water interface under high stirring (agitation) stage, followed by flocs build-up and growth at slow mixing stage. An advanced ASH (air sparger hydrocyclone) type of flotation, which appears to work similarly to FF has been reported in applications to remove oil, grease, BOD, etc. BAF, or bubble accelerated flotation system, uses the contactor-separation concept with very low detention times in the contactor (Owen et al., 1999).

J.J. da Rosa, J. Rubio / Minerals Engineering 18 (2005) 701–707




Pressure gauge Out



Flocculator MS-10

Pressure gauge Compressed Air

Fig. 1. The FF-flocculation–flotation system. Lay out of the aerated floc generation system. MS10 is for the flocculator with 10 zigzags units. The outlet connects with the floc flotation separation unit (cylinder type, see below).

Fig. 2. FF system: different flocculators employed.

3. Applications

Overflow Feed Coaxial Cylinder

3.1. Oil separation from oil-in-water emulsions (synthetic effluent) FF results for the separation of difficult-to-separate oil from emulsified oil dispersions (Table 1), using the

Centrifuge cell Table 1 Oil separation from oil-in-water emulsions by FF (2 m3 h 1). Experimental conditions


Underflow Fig. 3. The aerated floc flotation separator.



Oil concentration, Co Oil mean droplet diameter (volumetric), d(4, 3) Temperature, T NaCl, concentration Nalco, Cationic polymer

695 mg L 10 lm 20 C 60 g L 1 2 mg L 1



J.J. da Rosa, J. Rubio / Minerals Engineering 18 (2005) 701–707

Fig. 4. The FF pilot system used in the removal of oils from oil-in-water emulsions (2 m3 h 1).

Results in Figs. 6 and 7 show that, despite the high degree of emulsification, the oil separation of the aerated floc (Fig. 7) was very rapid and almost complete. The kinetics was very rapid, within seconds, yielding hydraulic loadings of more than 130 m h 1 (m3/m2/h).


Final oil concentration. mg/L


CONDITION Qefluent : 50.0 L/min [Flocculant]: 2 mg/L

80 70

ME-1" ME-3/4"


MS-10 MS-20


3.3. Flocculation–flotation of suspended (dispersed) solids


40 30 20 10 0 0.0












Head loss in the flocculator. kgf/cm 2

Fig. 5. Oil in water emulsion (synthetic) separation by the FF process: effect of head loss and type of flocculator.

equipment described in Fig. 4 showed excellent results only after reaching a minimum head loss in the saturator, i.e. independently on the flocculator design (Fig. 5).

All aqueous suspensions (2% by weight) were prepared by dispersing the solids in water using a high speed stirring. After flocculation–flotation, FF% values were calculated from the dry weights of dispersions, before and after the separation: % FF = (1 0 0)(Do-Df/Do) where Do is the degree of dispersion (‘‘dispersibility’’) calculated from the feed dispersed solids content (by weight) before FF (Do = initial dispersion degree) and Df is the dispersion degree after FF, using the same cationic polymer (Nalco) as in the oils. Particles were ground to 100% less than 37 lm and the solids studied were:

3.2. Oil separation from oil-in-water emulsions (refinery effluent) Studies were conducted with a typical petroleum refinery effluent using the same FF equipment used before (Table 2).

Table 2 Main parameters used in the effluent separation by FF Parameters


Oil Turbidity TSS-total suspended solids OM-organic matter Oil mean droplet diameter (volumetric), d(4,3)

77–115 mg/L 55–67 NTU 43–51 mg/L 480–515 mg O2/L 12 lm

Fig. 6. Oil in water emulsion (petroleum refinery effluent) separation by the FF process, using two different polymer concentrations (Nalco, Cationic polymer).

J.J. da Rosa, J. Rubio / Minerals Engineering 18 (2005) 701–707



Removal, %

90 80 70 60 50 0


40 -1

Surfactant, mgL

Fig. 9. Flocculation–flotation of colloidal Fe(OH)3 floc (using a cationic polyacrylamide) as a function of lauryl sulphate concentration. Removal of the ferric ion content, before and after FF separation. Fig. 7. The aerated oil floc formed with a cationic polyacrylamide in the FF process.

• Bentonite (smectite) from Brasgel-Brazil; • Quartz, pure sample from South Brazil; • Coal beneficiation tailings, corresponds to a ground jigging tailings sample (South Brazil); • Hematite, pure sample from North Brazil (Minas Gerais).

pensions but flocs were readily broken apart (not stable at all!). However, in the presence of sodium lauryl sulphate, these flocs were resistant to shearing in the FF equipment and they were able to separate off. It is believed that the role of the surfactant was to avoid the floc rupture by reducing the air/liquid surface tension.

4. Mechanisms involved Fig. 8 shows comparative results obtained in each case. With the exception of bentonite, a difficult to flocculate solids suspensions, the FF separations of all other solid models was efficient, almost complete and proceeded at a very rapid rate. 3.4. Flocculation–flotation of colloidal dispersions Fig. 9 shows FF results obtained with ferric hydroxide [Fe (OH)3], containing 50 mg L 1 Fe hydrolyzed at pH 6. The ferric colloidal suspensions were flocculated with the same cationic polyacrylamide studied with the solids sus100

FF %

80 60 40 20 0








[Nalco 8589], mgL

Fig. 8. Flocculation–flotation values for various solids suspensions. s = Hematite; d = Quartz; m = Bentonite; h = Coal beneficiation tailing.

It is believed that in the FF process occurs a ‘‘pneumatic’’, on line flocculation whereby the polymer diffusion and adsorption are rapidly ensured by the shearing forces given by the bubbles (minutes ones) and by the head loss in the flocculator, minimum 0.5 to1.0 kgf/cm2 (Fig. 2). More, the turbulent mixing of the flow is characterized as a plug flow regime where all particles have the same residence time (slow dispersion), ideal for flocculation, (Bratby and Marais, 1977). Mechanisms involved are not well elucidated but at least one of the following phenomena may be operating: 1. Small bubble formation and their rapid occlusion (entrapment) within flocs; Nucleation of bubbles at floc/water interfaces and bubbles entrainment. Qualitative measurements of the bubble size generated in the FF system, using the technique developed by Rodrigues and Rubio (2003), yielded values of the order of 100 lm bubbles diameter (Sauter). These values considered very fines (microbubbles) and of the order of those generated in DAF units, are formed at head losses (measured by the pressure differences before and after the flocculator) of the order of 3 atmospheres and at 4 Lmin 1 for the feed rate and 7 Lmin 1 for the air flow rate. Because the uprising rates of the aerated flocs (more than 130 m h 1) are faster than the rising bubbles sizes alone (30–40 m h 1), it is believed that air is entrapped (occluded) inside the flocs, highly decreasing the aggregate density as a function of the incorporated air volume.


J.J. da Rosa, J. Rubio / Minerals Engineering 18 (2005) 701–707

More, because some air must be dissolved in water following the flow pressure inside the flocculator, these microbubbles behave as in DAF. Thus, a very important feature concerns with the mechanisms of bubble/particle (aggregates) interactions other than the common adhesion through hydrophobic forces (Rubio et al., 2002a). Apart from particles-bubbles collisions and adhesion, part of the dissolved air in water, which does not convert into bubbles, remains in solution and ‘‘nucleate’’ at the particle surface (Solari and Gochin, 1993).This mechanism is independent on surface hydrophobicity and allows flotation of hydrophilic particles. In addition, entrainment by the rising bubbles may be also operating. 2. Polymer uncoiling or ‘‘salting out’’ at the aqueous/ air interface. The aerated flocs looked like filamentous, elongated and ‘‘sticky’’ (plastic like) units. This characteristic may be the result of an uncoiling phenomenon of the polymer chains at the solid (liquid1)/liquid2 (water)/air interface (high, because of the small bubbles). Thus, at the flocs/water/air interface, a fraction of the adsorbed polymer may become ordered (small entropy) with a high local ‘‘activity’’. As a result, the polymer appears to suffer a sort of ‘‘salting out’’ or insolubility. This precipitation may explain the shape (form) of the flocs and their high resistance to turbulence and high shearing forces in the flocculation contactor.

5. Conclusions A new in-line flocculation–flotation unit, named FF-Flocculation–Flotation, was developed for the generation and separation of aerated polymeric flocs. The studies showed a high process efficiency for the separation of oil from oil emulsions, suspended solids and colloidal suspensions [Fe(OH)3] with the employment of a cationic, high molecular weight, polymer. The polymer flocs obtained were well structured, big and with elongated (string like) format. The FF appears to have some advantages, namely an adequate turbulence, low area required, absence of mobile parts, simple design, and low mechanical and electrical energy required. Process efficiency depended on a minimum head loss (0.5 e 1.0 kgf/cm2) and not on flocculator design, on polymer concentration and on the air feed flow rate. Finally, the uprising rates of the aerated flocs reached to more than 130 m h 1, values higher than those found in conventional sedimentation tanks or DAF-dissolved air flotation units.

Acknowledgments Jorge Rubio thanks to the students and colleagues from LTMUniversidade Federal do Rio Grande do

Sul, for the friendship, and to all Institutions supporting research in Brazil (FAPERGS, CAPES, CNPq, UFRGS). Authors thank also to the graduate and undergraduate students for their assistance in the hard experimental work.

References Bennett, G.F., 1988. The removal of oil from wastewater by flotation: A review, CRC, vol. 18 (3), pp. 189–253. Bratby, J., Marais, G.V.R., 1977. Flotation. In: Purchas, D.B. (Ed.), Solid/Liquid Separation Equipment Scale-Up. Upland Press, pp. 155–168. Capps, R.W., Metelli, G.N., Bradford, M.L., 1993. Reduce oil and grease content in wastewater. Hydrocarbon Processing. (June), 102–110. Da Rosa, J., Santander, M., Souza, M.L., Rubio, J., 2002. Flotacio´n como proceso de remocio´n de contaminantes: Nuevas te´cnicas y equipos. Minerales (Chile) 57 (243), 29–38. De Oliveira, G.R.C., 1995. Estudo de varia´veis operacionais e interfaciais na flotac¸a˜o de o´leo por ga´s dissolvido., M.Sc. Thesis (in portuguese), Universidade Federal de Rio de Janeiro, pp. 120. Gopalratnam, V.C., Bennett, G.F., Peters, R., 1988. The simultaneous removal of oil and havy metals from industrial wastewater by joint precipitation and air flotation. Environmental Progress 7 (2), 84– 92. Gu, Chiang, 1999. A novel flotation column for oily water cleanup. Separation and Purification Technology 16, 193–203. Kiuru, H.J., 2001. Development of dissolved air flotation technology from the 1st generation to the newest or 3rd one (very thick microbubbles) with high flow-rates (DAF in turbulent flow conditions). Water Science and Technology 8, 1–8. Matis, K.A. (Ed.), 1995. Flotation Science and Engineering. Marcel Dekker, New York, p. 558. Mavros, P., Matis, K.A. (Eds.), 1992. Innovations in Flotation Technology. Kluwer Academic Publishers, Dordrecht. Menezes, C.T.B., Leal Filho, L.S., Santo, E.L., Rubio, J., da Rosa, J.J., Galatto, S.L., Izidoro, G., 2004. Tratamento de drenagem a´cida de mina: experieˆncia da carbonı´fera Metropolitana. XX Encontro Nacional de Tratamento de Mine´rios e Metalurgia Extrativa, Floriano´polis-Brazil, vol. 2, pp. 599–607 (in Portuguese). Owen, J.J., Morse, D.E., Morse, W.O., Jovine, R., 1999. New Developments in Flotation Equipment for Water Treatment Systems. In: Parekh, B.K., Miller, J.D. (Eds.), Advances in Flotation Technology. Society for Mining. Metallurgy, and Exploration, Inc. (SME), Littleton, CO, pp. 381–389. Parekh, B.K., Miller, J.D., (Eds), 1999. Advances in Flotation Technology. SME. Rodrigues, R.T., Rubio, J., 2003. New basis for measuring the size distribution of bubbles. Minerals Engineering 16 (8), 757–765. Rubio, J., Santander, M., 1997. Oil/water emulsions separation by induced air carrier flotation. In: 5th Southern Hemisphere Meetings on Mineral Technology, Buenos Aires, May 6–9, Proceedings, pp. 125–128. Rubio, J., 2003. Unconventional flocculation and flotation. In: Ralston, J., Miller, J., Rubio, J., Flotation and Flocculation: From Fundamentals to Applications, Proceedings from Strategic Conference and Workshop, Hawaii, 2002, pp. 17–32. Rubio, J., Souza, M.L., Smith, R.W., 2002a. Overview of flotation as a wastewater treatment technique. Minerals Engineering 15, 139– 155. Rubio, J., Tessele, F., Porcile, P.A., Marinkovic, E., 2002b. Flotacio´n como proceso de remocio´n de contaminantes: Avances y aplicac-

J.J. da Rosa, J. Rubio / Minerals Engineering 18 (2005) 701–707 iones en la flotacio´n por aire disuelto. Minerales (Chile) 57 (243), 21–28, In Spanish. Rubio, J., Rosa, J.J., Beal, R., 2003. Tratamento e recirculac¸a˜o de a´guas de lavagem de veı´culos, Brazilian Patent, INPI: 0006390. Solari, J.A., Gochin, R.J., 1993. Fundamental Aspects of Microbubbles Flotation. In: Ralston, J., Laskowski, J.S. (Eds.), Colloid Chemistry in Mineral Processing. Cap. 13, pp. 395–418. Strickland, W.T., 1980. Laboratory results of cleaning produced water by gas flotation. Society of Petroleum Engineering, Journal 20, 175–190.


Toyoda, M., Moriya, K., Aizawa, J., Konno, H., Inagaki, M., 1999. Environmental Technology for Oil Spills. In: Hupka, Miller (Eds.). Technical University of Gdansk, Gdansk, pp. 139–142. Van Ham, N.J.M., Behie, L.A., Svrcek, W.Y., 1983. The effect of air distribution on the induced air flotation of fine oil in water emulsions. Canadian Journal of Chemical Engineering 61 (4), 541– 547. Voronin, N.N., Dibrov, I.A., 1999. Classification of flotation processes in wastewater decontamination. Journal of Environmental Engineering, 469.