Separation and Purification Technology 186 (2017) 326–332
Contents lists available at ScienceDirect
Separation and Purification Technology
journal homepage: www.elsevier.com/locate/seppur
Separation of emulsified crude oil in saline water by dissolved air flotation with micro and nanobubbles ⇑ R. Etchepare 1,2, H. Oliveira 1, A. Azevedo, J. Rubio
Laboratório de Tecnologia Mineral e Ambiental, Departamento de Engenharia de Minas, PPGE3M, Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves, 9500, Setor 6 (Centro de Tecnologia), Prédio 43819, Bairro Agronomia 91501-970 Porto Alegre, RS, Brazil3 article info abstract
Article history: This work investigates the separation of emulsified crude oil in saline water (30 g L 1 NaCl) with Received 21 April 2017 microbubbles (MBs, D32 = 30–40 mm) and nanobubbles (NBs, D32 = 150–350 nm). Bubbles were generated Received in revised form 2 June 2017 simultaneously by the depressurization of air-saturated water through a flow constrictor (needle valve). Accepted 3 June 2017 The emulsified oil, after gravity separation of the ‘‘free” oil, was flocculated with a cationic polyacry- lamide (Dismulgan) at pH 7 and removed by: i. Flotation with MBs and NBs; ii. ‘‘Floatation” by NBs; and iii. Flotation with MBs and NBs following floc conditioning by NBs. The best oil removal (>99% effi- Keywords: ciency) was obtained at 5 bar and 5 mg L 1 of Dismulgan, reducing the oil content (feed concentra- Nanobubbles tion = 334–484 mg L 1) in the treated water to <1 mg L 1. Furthermore, the use of low saturation Flotation 1 Separation of oil in water emulsions pressure (Psat = 3.5 bar), resulted in a treated water with oil concentrations lower than 29 mg L (EPA Produced water treatment standards for offshore discharge). Best results were obtained at a low energy for bubble formation, fol- lowed by efficient precipitation and nucleation, at a fairly low air/feed emulsified oil interfacial tension (55 mN m 1). The flotation was very fast and followed a first-order model, with a flotation kinetic con- 1 stant of 1.3 and 1.8 min for Psat of 3.5 and 5 bar, respectively. The injection of isolated NBs (3 108 NBs mL 1), in a conditioning stage after flocculation (with 1 and 3 mg L 1 Dismulgan) increased the hydrophobicity of the aggregates, improved the adhesion between bubbles and oily flocs and the overall efficiency of the flotation process from 73 to 84%, and from 92 to 95%, respectively. ‘‘Floatation” (simply flocs rising with isolated NBs) resulted in oil removal efficiencies of 75 and 90% with and without NaCl (30g L 1). It is believed that the NBs entrap and adhere inside the flocculated oil dro- plets, forming aerated oily flocs, which subsequently assist the MBs in the flotation process. This finding appears to have potential in improving oil separation by flotation. Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction depends on a number of factors, such as the ratio of bubble / dro- plet size, their numeric concentration, salinity, oil viscosity, fluid Flotation is extensively employed in the removal of emulsified velocity and turbulence. Among these factors, the gas dispersion oils from aqueous dispersions, after destabilization by flocculation. parameters are considered the most important in oil-water separa- The size distribution of bubbles and oil droplets plays a key role in tion [3]. flotation efficiency. Small bubbles are preferred because of their Produced water is wastewater generated in petroleum large surface areas, which are shown to be very useful in the aggre- prospecting that is composed of a mixture of water from wells gation of droplets [1,2]. Conversely, larger bubbles tend to rise and process water [4–6]. Flotation is usually applied in its treat- rapidly, which results in lower collision efficiency. The adhesion ment at offshore platforms [7], and the effluent should meet the between bubble and oil droplet is another key factor for flotation emission limit (EPA Oil and Gas Effluent Guidelines [8])of efficiency. The formation of stable bubble-droplets aggregates 29 mg L 1 total oil concentration average for monthly and 42 mg L 1 for daily maximum for offshore disposal. Recent studies have found that in dissolved air flotation (DAF) ⇑ Corresponding author. microbubbles (MBs) and nanobubbles (NBs) are formed [9,10]. E-mail address: [email protected] (J. Rubio). 1 Both authors contributed equally. The generation and applications of NBs are an emerging and fast- 2 Current address: Departamento de Hidráulica e Saneamento, Universidade growing research area because of their physical, chemical and Federal do Paraná, 81531-980 Curitiba, PR, Brazil. physicochemical properties. 3 www.ufrgs.br/ltm. http://dx.doi.org/10.1016/j.seppur.2017.06.007 1383-5866/Ó 2017 Elsevier B.V. All rights reserved. R. Etchepare et al. / Separation and Purification Technology 186 (2017) 326–332 327
Reports and news about flotation with NBs to remove pollu- Table 1 tants from waters and wastewaters are scarce. Recent studies have Crude oil physicochemical and interfacial properties. demonstrated the potential of these fine bubbles in solid-liquid Parameters Value separation and mineral treatment [11–15]. Recently, the removal Density, 25 °C(gmL 1) 0.88 of amine and sulfate precipitates from aqueous solutions using iso- API Grade (°API) 23 lated NBs (F-NBs) and DAF assisted by NBs were studied [11,16]. Viscosity, 25 °C (cP) 42 1 The DAF assisted by NBs consists in the injection of an aqueous dis- Superficial tension (oil/water), 25 °C (mN m )28 persion of NBs in a conditioning stage (with or without flocculants) promoting aggregation and hydrophobization of the particles to be removed. It is believed that NBs may play an important role in the aggre- Fig. 1 shows a schematic representation of the bench system for gation and stabilization of these aggregates (flocs) by adhering to generation and treatment of oily emulsion by flocculation-flotation and/or entraining inside (entrapment) the flocs, generating capil- with MBs and NBs. lary bridges between oil droplets, and improving their hydropho- bicity and the probability of adhesion with MBs [11,17]. The aim 2.2.1. Flotation with MBs and NBs of this study was to evaluate the role of NBs on various multi- The bubbles formed by depressurization and hydrodynamic bubble flotation configurations: MBs and NBs, DAF assisted by cavitation of the flow of air saturated water through the needle NBs, and ‘‘floatation” (flotation without lifting power) with isolated valve were injected directly into the glass column (3) (without sep- NBs (F-NBs), on the removal of emulsified oil after destabilization / aration of MBs and NBs). To perform the tests, a volume of 800 mL flocculation at the bench level. of the oily emulsion was transferred to the glass column (3). Floc- culation was performed using rapid mixing stages (1 min, G = 2400 s 1), followed by slow mixing (5 min, G = 30 s 1) by the 2. Materials and methods use of the mechanical agitator.
The effect of the saturation pressure (Psat) on the oil removal 2.1. Synthetic produced water was observed between 2.5 and 6 bar by injecting compressed air and adjusting the relief valve in the saturator vessel over 30 min. ° Ultrapure (DI) water at room temperature (22 C ± 1) with a Flotation was performed with 200 mL of bubble suspension into l 1 1 conductivity of 3 Scm , a surface tension of 72.5 ± 0.1 mN m the flotation cell, corresponding to a recycle ratio of 25%. and pH 5.5 was used to prepare the synthetic produced water. DI Aliquots of 100 mL were collected for oil concentration analysis water was obtained by ultrapurification of tap water with a reverse after 1 h of separation of the free oil phase, and from the clarified osmosis cartridge and modules of ion-exchange resins and acti- liquid (treated water) after 5 min of flotation. All experimental vated carbon. assays were performed in triplicate and the results were analyzed The petroleum (crude oil) used in this study was supplied by a in terms of oil removal efficiency (%), according to (Eq. (1)). local oil refinery (REFAP, Petrobras, southern Brazil). This oil was ½ðC FÞ 100 characterized in terms of its physicochemical and interfacial prop- Efficiencyð%Þ¼ 100 - f ð1Þ erties (Table 1). C0 The brine solution for synthetic produced water was made by where, C is the final oil concentration; F is the dilution factor (1.2) 1 Ò f dissolving 30 g L of NaCl (99.5%, Vetec , Brazil) into 1 L of DI corresponding to the volume (200 mL) of saturated water injected; water. Then, 1.6 g of crude oil was slowly dripped into the brine C0 is the initial oil concentration. solution for emulsification with an Ultra Turrax mixer (IKA, The air/oil ratio (Eq. (2)) was calculated using the theoretical 24.000 rpm, 10 min). The emulsion was left to stand in an acrylic calculation of air saturation in water (Henry’s Law) and the volume column for 1 h for the separation of the free oil phase to obtain a of precipitated air per liter of water for different values of Psat cal- well stabilized oil-in-water emulsion. Solutions of 0.1 M NaOH Ò culated by Rodrigues and Rubio [18] using a saturation efficiency of (Vetec , Brazil) were prepared for pH adjustment in the destabi- 95% in the saturator vessel. lization/flocculation stage. ð Þ : The destabilization and flocculation of the emulsion was con- air vair F2 0 95 Ò Ratio ¼ ð2Þ ducted using Dismulgan V3377 (Clariant , Rio Grande do Sul, Bra- oil ðCoil F1Þ zil) (1 to 10 mg L 1), a flocculation polymer (polyacrylamide) 1 widely used on offshore petroleum platforms. where Coil is the initial oil concentration (mg L ); F1 is the correc- tion factor corresponding to the volume of liquid (oily emulsion) used (800 mL); V is the volume of precipitated air (mL L 1); F2 2.2. Flocculation-flotation studies for oil/water separation air is the correction factor corresponding to the volume of saturated water used (200 mL). The experiments were performed using the system shown in The effect of Psat on flotation kinetics was evaluated at a P of Fig. 1, which consisted of: i. saturator vessel made of acrylic (1) sat 3.5 and 5 bar. Treated water samples were collected at different for bubble generation (2.5 L, h = 400 mm, diameter = 110 mm, flotation times (0–300 s) for measuring residual oil concentration. equipped with a manometer and a needle valve); ii. a glass column The results were expressed in terms of the percentage of oil (2) for the separation of MBs (2 L, h = 250 mm, diameter = 100 mm) removal (Eq. (1)) and flotation kinetic parameter data was adjusted using the technique described by Calgaroto et al. [9,17]; and iii. a to the first-order flotation model [19]. The flotation kinetic con- glass column (3) for the flocculation and flotation stages (2.5 L, stant k was obtained using (Eq. (3)). h = 330 mm, diameter = 100 mm) and a mechanical stirrer Fisa- Ò tom brand (model 713D). R1 ln ¼ kt ð3Þ The glass column (2) had one input receiving the depressurized R1-R flow (MBs and NBs) from the saturator vessel and one output con- nected to the glass column (3), to inject the isolated NBs. With The García-Zuñiga kinetic model of flotation is given by (Eq. (4)). flotation by MBs and NBs together, the glass column (2) was not ¼ 1ð -ktÞðÞ employed. R R 1-e 4 328 R. Etchepare et al. / Separation and Purification Technology 186 (2017) 326–332
Fig. 1. Oily emulsion generation and treatment systems by flocculation-flotation with air MBs and NBs. (1) Saturator vessel; (2) Glass column for MBs separation and NBs isolation; (3) Glass column for flocculation and flotation stages and mechanical stirrer (FisatomÒ 713D); (4) Ultra Turrax emulsifier for oil in water mixture. where, R is the calculated oil removal (experimental) (%); R1 is the 2.4. Oil concentration analysis removal of oil under stationary conditions (theoretical, at t = 300 s) (%); k is the kinetic constant (s 1); t is the flotation time (s). The oil concentration in the emulsions and in the treated efflu- ent was determined with an oil analyzer (HoribaÒ, model OCMA- 350, Japan). Measurements were based on the energy absorption 2.2.2. Flotation assisted by NBs in the infrared spectrum at the wavelength range of 3.5–3.6 lm. The conditioning of the oily flocs with a NBs suspension ahead Thus, the amount of energy absorbed in this range is assumed to of flotation was evaluated. Two different concentrations of Dismul- be directly proportional to the oil concentration of the sample. gan (1 and 3 mg L 1) were employed. After flocculation, 800 mL of For the extraction of the oil contained in the sample, the solvent Ò NBs suspension were added to 800 mL of oily flocs suspension for poly-trichlorofluoroethylene (S-316, Horiba ) was used. the conditioning of the flocs for 5 min at slow mixing (G = 30 s 1). Finally, flotation was done by depressurizing 400 mL of air satu- 3. Results and discussion rated water in Psat = 5 bar. Blank experiments were conducted by adding pure deionized water instead of NBs suspension in the con- 3.1. Flotation with MBs and NBs ditioning stage. All experimental arrays were performed in tripli- cate and the results of oil removal efficiency were obtained using Fig. 2 shows the effect of the Dismulgan concentration on emul- (Eq. (1)). sified oil removal efficiency by flotation with MBs and NBs. The results showed that oil removal increases as a function of the con- centration of Dismulgan, and the best treatment conditions was 2.2.3. Flotation with isolated NBs (F-NBs) obtained with 5 mg L 1. At this reagent concentration, the removal For the F-NBs, the MBs separation procedure [9,10] was done in efficiency was almost complete (>99%), yielding a residual oil con- the glass column (2) and 200 mL of NBs suspension was placed into centration of 1 mg L 1. the flotation column. The F-NBs was evaluated with and without The role of coagulant-flocculating agents in the destabilization 1 the addition of NaCl (30 g L ) in the oily emulsion. Blank experi- of oil-in-water emulsions to promote the coalescence of oil dro- ments were conducted with pure deionized water instead of NBs plets is well discussed in the literature and the mechanisms suspension for flotation. All results corresponded to triplicate tests involved are based on van der Waals and hydrophobic forces and the oil removal efficiencies were obtained using (Eq. (1)). [20–24]. Fig. 3 shows microphotographs obtained from the oily emulsion (after separation of the free oil phase), the flocs formed with Dismulgan and the treated water after flotation. 2.3. Microscopy imaging Oil droplets with a diameter greater than 20 lm are considered amenable for flotation [25], and those finer (<10 lm) may not Optical microphotographs were taken of the different stages of become coalesced while flocculation due to Brownian motion oily emulsion separation by flocculation-flotation. An optical [26]. Consequently, this process is less efficient when the residence microscope OlympusÒ, model BX41 was employed with an objec- time is low. The emulsions were generated with a technique devel- tive magnification of 1000X coupled to a high-performance digital oped in our laboratory, by Santander [27]. Thus, we utilized an microscope camera, Olympus DP73 (17.28 megapixel resolution). ultra Turrax emulsificator (13.000 rpm) which yielded oil droplets Samples of 100 mL were collected for analysis as follows: i. Oily with a mean Sauter diameter (D32) of 15 lm, and a size distribu- emulsion, after free phase separation; ii. Oily flocs after 5 min of tion between 10 and 17 lm. Herein, oil droplets with diameter flocculation with Dismulgan; iii. Treated water after 5 min of flota- smaller than 10 lm were used, after free oil phase separation tion. One drop of each sample was used for visualization under the (Fig. 3). Thus, the destabilization of the tiny emulsified oil droplets microscope. with 5 mg L 1 of Dismulgan and the application of MBs and NBs R. Etchepare et al. / Separation and Purification Technology 186 (2017) 326–332 329
140 100 With Dismulgan 100
1 Without Dismulgan - 120 80 100 80 Final oil concentration 80 60 Oil removal efficiency 60 60 40 40 40
Oil concentration, mg L 20 20 Oil removal efficiency, % 20 Oil removal efficiency, % 0 0 0246810 0 2.5 3.5 5 6 Dismulgan concentration, mg L-1 Saturation pressure, bar Fig. 2. Flocculation-flotation for oil removal with MBs and NBs as a function of Dismulgan concentration. Conditions: initial [oil] = 337–484 mg L 1; [NaCl] Fig. 4. Oil removal efficiency by flocculation-flotation with MBs and NBs as a =30gL 1; pH 7; Saturation pressure = 5 bar; Saturation time = 30 min; Flotation function of the saturation pressure and in the presence and absence of the flocculant 1 1 time = 5 min. The error bars correspond to the standard deviation of triplicate tests. (Dismulgan). Conditions: initial [oil] = 334–450 mg L ; [NaCl] = 30 g L ; [Dismulgan] concentration = 5 mg L 1; pH 7; Saturation time = 30 min; Recycle rate = 25%; Flotation time = 5 min. The error bars correspond to the standard deviation of the triplicate tests. allowed good collision and adhesion of these bubbles to the flocs, reducing the oil content to a final concentration of 1 mg L 1 in the treated water. Table 2 The effect of Psat on oil removal efficiency is shown in Fig. 4. Oil removal by flocculation-flotation with MBs and NBs as a function of saturation The best results were obtained with saturation pressures of 5 pressure. Conditions: initial [oil] = 334–450 mg L 1; [NaCl] = 30 g L 1; [Dismulgan] 1 and 6 bar, with removal efficiencies >99% and a mean residual oil concentration = 5 mg L ; pH 7; Saturation time = 30 min; Recycle rate = 25%; Flota- tion time = 5 min. Mean values (standard deviation) of quadruplicate tests. concentration of less than 2 mg L 1. It was also observed that the non-flocculated oil droplets did not float properly, leading to an Psat, Feed oil Residual oil Oil removal 1 unsatisfactory oil removal efficiency (<70% and a residual concen- bar concentration, concentration, mg L efficiency, % mg L 1 tration >100 mg L 1). More, in terms of residual oil concentration and practical 2.5 378 (25) 70 (10) 82 (3) 3.5 391 (25) 28 (4) 93 (1) aspects, the results showed that with a Psat of 3.5 bar the removal 5 395 (29) 2 (1.2) 99.6 (0.3) efficiency was about 93%. As shown in Table 2 and Fig. 5, the resid- 6 402 (39) 1.4 (0.5) 99.7 (0.1) ual oil concentration obtained was below the EPA Oil and Gas Efflu- ent Guidelines [8] for monthly average emission limit for offshore 1 waters (29 mg L ), proving the potential of the flotation process kinetic constant of 1.3 and 1.8 min 1 for Psat of 3.5 and 5 bar, with MBs and NBs with low saturation pressures (meaning less respectively. Table 3 summarizes the experimental and calculated energy consumption). oil removal data and corresponding correlation coefficients. Fig. 6 shows the flotation kinetics at saturation pressures of 3.5 and 5 bar. The best results were obtained at higher saturation pres- sures. This fact may be explained in terms of the higher air / oil 3.2. Flotation assisted by NBs ratio (higher collision efficiency), high lifting power of the MBs and the higher concentration of MBs with a narrow size distribu- Fig. 7 and Table 4 show the results of flotation assisted by NBs. tion obtained with 5 bar (0.06 mL mg 1) compared to 3.5 bar It was observed that, in comparison to the flotation process, the (0.04 mL mg 1). The flotation with both saturation pressures was inclusion of the floc conditioning step with NBs promoted an relatively fast, with stabilization of the oil removal efficiency after improvement in the overall oil removal. The NBs appear to interact 3 min. This flotation followed a first-order model with a flotation with the (hydrophobic) oil droplets, aggregating them and forming
Fig. 3. Microphotographs of the different stages of the treatment of oily emulsion by flocculation-flotation: (a) oily emulsion after free oil phase separation; (b) oily flocs; (c) treated water after 5 min of flotation. Conditions: initial [oil] = 390 mg L 1; [NaCl] = 30 g L 1, pH 7; [Dismulgan] = 5 mg L 1; Psat = 5 bar; Saturation time = 30 min; Recycle rate = 25%; Flotation time = 5 min. 330 R. Etchepare et al. / Separation and Purification Technology 186 (2017) 326–332
1 100 - 70 Residual oil concentration Flotation 60 95 Emission limit NBs+Flotation 50 90 40 85 30 80 75 20 70 10 Oil removal efficiency, % 65
Residaul oil concentration, mg L 0 23456 60 Saturation pressure, bar 13 Dismulgan concentration, mg L-1 Fig. 5. Residual oil concentration after flocculation-flotation with MBs and NBs as a function of saturation pressure. Conditions: initial [oil] = 334–450 mg L 1; [NaCl] Fig. 7. Efficiency of oil removal by flocculation-flotation assisted by NBs. Condi- 1 1 =30gL 1; [Dismulgan] = 5 mg L 1; pH 7; Saturation time = 30 min; Recycle tions: initial [oil] = 375–410 mg L ; [NaCl] = 30 g L ; pH 7; Saturation pres- rate = 25%; Flotation time = 5 min; Emission limit = 29 mg L 1. The error bars sure = 2.5 bar; Saturation time = 30 min; Recycle rate of flotation = 25%; Flotation correspond to the standard deviation of the triplicate tests. time = 5 min. The error bars correspond to the standard deviation of the triplicate tests.
100 the repulsive DLVO forces in this liquid film are overlaid by waves of capillary forces that deform the surface of this film. These mech- 80 anisms may explain the results where the NBs facilitated the adhe- sion between bubbles and oily flocs, increasing the overall 60 efficiency of the flotation process.
40 Psat = 3.5 bar 3.3. Floatation with isolated NBs (F-NBs)
Psat = 5 bar Fig. 8 shows the results of the floatation conducted with iso-
Oil removal efficiency, % 20 lated NBs (F-NBs). The residual oil concentrations in the treated water were 34 mg L 1 (without NaCl) and 80 mg L 1 (with NaCl 0 30 g L 1), and oil removal efficiency reached an equilibrium (90% 0 50 100 150 200 250 300 and 75%, respectively), after 10 min of flotation. The results indi- Flotation time, s cated that the NBs contribute to the rising of these aggregates. The NBs are strongly attracted to the oily flocs presumably by Fig. 6. Flocculation-flotation kinetics of oily flocs. Conditions: initial [oil] = 335– 480 mg L 1; [NaCl] = 30 g L 1; [Dismulgan] = 5 mg L 1; pH 7; Saturation time = 30 - hydrophobic forces (interfacial phenomena) and, once adhered, min; Recycle rate = 25%. these minute bubbles are occluded and/or entrapped inside the oily flocs. As a result, the relative density of these flocs decrease allowing the separation from water by gravity. aerated precipitates, a phenomena already reported in literature Yet, these results indicate a slightly lower oil removal efficiency [11,17,28–30]. of floatation with isolated NBs with NaCl 30 g L 1 rather than a Mishchuk [31] carried out a theoretical analysis on the attrac- solution without salt. These results are probably due to the lower tive forces in a particulate system in the presence of NBs and con- concentration of NBs obtained under these conditions (7.1 107 - cluded that these ultra-fine bubbles adhered to the particles, NBs mL 1, Psat = 2.5 bar) when compared to NBs formed in DI reducing the local density of the solid-liquid interface. They also water (3.3 x 108 NBs mL 1; Psat = 2.5 bar), which validates and modify the absolute value of van der Waals forces, improving the reinforces the role of NBs in the separation of these oily aggregates. degree of aggregation by increasing the attraction between these The F-NBs has already been observed in previous studies of this particles. Moreover, the study by Stöckelhuber et al. [32] demon- research group, in the flotation of amine precipitate [11] and ferric strated that NBs cause the rupture of the liquid film between a hydroxide [33]. However, this seems to be the first report in the lit- bubble and a hydrophobic surface without the introduction of a erature of flocculation/NBs floatation for the treatment of emulsi- hydrophobic force mechanism. These authors have found that fied oil.
Table 3 Flocculation-flotation kinetics of oil with MBs and NBs. Experimental data on oil removal and correlation coefficients were calculated using the first order flotation kinetic model. Conditions: k = 1.3 min 1 (Psat = 3.5 bar) and k = 1.8 min 1 (Psat = 5 bar); pH 7; Dismulgan = 5 mg L 1; Saturation time = 30 min; Recycle rate = 25%.
Flotation time, s Psat = 3.5 bar Psat = 5 bar Experimental R,% R1,% Correlation coefficient Experimental R,% R1,% Correlation coefficient 0 0 0.0 1 0 0.0 1 10 53 18 2.97 53 25 2.08 20 69 32 2.14 70 44 1.57 40 71 53 1.33 76 69 1.10 80 91 76 1.19 93 90 1.03 300 96 93 1 99 99 1 R. Etchepare et al. / Separation and Purification Technology 186 (2017) 326–332 331
Table 4 Oil removal by flocculation-flotation assisted by NBs. Conditions: [NaCl] = 30 g L 1; pH 7; Saturation pressure = 2.5 bar; Saturation time = 30 min; Recycle rate = 25%; Flotation time = 5 min. Mean values (standard deviation) of triplicate experiments.
Parameters Dismulgan = 1 mg L 1 Dismulgan = 3 mg L 1 Flotation assisted by NBs Flotation Flotation assisted by NBs Flotation [oil] Initial 398 (8) 380 (9) 413 (4) 389 (32) [oil] Final 64 (5) 104 (7) 22 (2) 32 (7)
100 Department of Energy National Energy Technology Laboratory Under 350 Contract W-31-109-Eng-38, 2004. [6] I.T. Gabardo, E.B. Platte, A.S. Araujo, F.H. Pulgatti, Evaluation of produced water 300 80
-1 from Brazilian offshore platforms, in: K. Lee, J. Neff (Eds.), Prod, Water Environ. 250 Risks Adv. Mitig. Technol, Springer, New York Dordrecht Heidelberg London, 60 2011, pp. 89–113. With NaCl 30 g/L 200 [7] A. Fakhru’l-Razi, A. Pendashteh, L.C. Abdullah, D.R.A. Biak, S.S. Madaeni, Z.Z. Without NaCl Abidin, Review of technologies for oil and gas produced water treatment, J. 150 40 Efficiency (NaCl 30 g/L) Hazard. Mater. 170 (2009) 530–551. doi:10.1016/j.jhazmat.2009.05.044. 100 Efficiency (without NaCl) [8] EPA, EPA Effluent Guideline for Oil and Gas Extraction (CFR 40.435), United 20 States, 1979. https://www.epa.gov/eg/oil-and-gas-extraction-effluent-
Oil concentration, mg.L 50
Oil removal efficiency, % guidelines. 0 0 [9] S. Calgaroto, K.Q. Wilberg, J. Rubio, On the nanobubbles interfacial properties 0 5 10 15 25 and future applications in flotation, Miner. Eng. 60 (2014) 33–40, http://dx.doi. org/10.1016/j.mineng.2014.02.002. Flotation time, min [10] A. Azevedo, R. Etchepare, S. Calgaroto, J. Rubio, Aqueous dispersions of
1 nanobubbles: generation, properties and features, Miner. Eng. 94 (2016) 29– Fig. 8. Oil separation kinetics with isolated NBs, with and without NaCl (30 g L ). 37, http://dx.doi.org/10.1016/j.mineng.2016.05.001. 1 1 Conditions: [NaCl] = 30 g L ; [Dismulgan] = 5 mg L ; pH 7; Saturation pres- [11] S. Calgaroto, A. Azevedo, J. Rubio, Separation of amine-insoluble species by sure = 2.5 bar; Saturation time = 30 min; Recycle rate = 25%. The initial oil corre- flotation with nano and microbubbles, Miner. Eng. 89 (2016) 24–29, http://dx. sponds to the oil concentration after the phase separation step. The error bars doi.org/10.1016/j.mineng.2016.01.006. correspond to the standard deviation of triplicate tests. [12] M. Fan, D. Tao, R. Honaker, Z. Luo, Nanobubble generation and its application in froth flotation (part I): nanobubble generation and its effects on properties of microbubble and millimeter scale bubble solutions, Min. Sci. Technol. 20 (2010) 1–19, http://dx.doi.org/10.1016/S1674-5264(09)60154-X. 4. Conclusion [13] M. Fan, D. Tao, R. Honaker, Z. Luo, Nanobubble generation and its applications in froth flotation (part III): specially designed laboratory scale column flotation Flocculation with 5 mg L 1 of Dismulgan at pH 7, followed by of phosphate, Min. Sci. Technol. 20 (2010) 317–338, http://dx.doi.org/10.1016/ S1674-5264(09)60205-2. flotation (Psat of 5 bar) with MBs and NBs showed an oil removal [14] M. Fan, D. Tao, R. Honaker, Z. Luo, Nanobubble generation and its applications efficiency higher than 99%, yielding a residual oil content of in froth flotation (part II): fundamental study and theoretical analysis, Min. Sci. 1mgL 1. The use of a Psat of 3.5 bar gave a treated water with a Technol. 20 (2010) 159–177, http://dx.doi.org/10.1016/S1674-5264(09) residual oil concentration lower than the emission limit 60179-4. [15] M. Fan, D. Tao, R. Honaker, Z. Luo, Nanobubble generation and its applications 1 (29 mg L ) of oily produced water on offshore oil platforms. The in froth flotation (part IV): mechanical cells and specially designed column flotation followed a first-order model with a kinetic constant of flotation of coal, Min. Sci. Technol. 20 (2010) 641–671, http://dx.doi.org/ 1.3 and 1.8 min 1 at Psat of 3.5 and 5 bar, respectively. A NBs con- 10.1016/S1674-5264(09)60259-3. 1 [16] J. Amaral Filho, A. Azevedo, R. Etchepare, J. Rubio, Removal of sulfate ditioning stage after the flocculation with 1 mg L of Dismulgan ions by dissolved air flotation (DAF) following precipitation and flocculation, resulted in an increase in flotation efficiency from 73 to 84%. Int. J. Miner. Process. 149 (2016) 1–8, http://dx.doi.org/10.1016/j. Floatation of the oily flocs with isolated NBs resulted in oil removal minpro.2016.01.012. [17] S. Calgaroto, A. Azevedo, J. Rubio, Flotation of quartz particles assisted by efficiencies of up to 90%. The results clearly demonstrated the high nanobubbles, Int. J. Miner. Process. 137 (2015) 64–70, http://dx.doi.org/ potential and the role of NBs in the overall oil removal efficiency in 10.1016/j.minpro.2015.02.010. the treatment of these wastewaters. [18] R.T. Rodrigues, J. Rubio, New basis for measuring the size distribution of bubbles, Miner. Eng. 16 (2003) 757–765, http://dx.doi.org/10.1016/S0892- 6875(03)00181-X. Acknowledgements [19] G. Zuñiga, The efficiency obtained by flotation is an exponential function of time, Bol. Min. 47 (1935) 83–86. [20] K. Bensadok, M. Belkacem, G. Nezzal, Treatment of cutting oil/water emulsion The authors would like to thank all the Brazilian Institutes that by coupling coagulation and dissolved air flotation, Desalination. 206 (2007) supported this study, namely, CNPq, CAPES, FAPERGS, and UFRGS. 440–448, http://dx.doi.org/10.1016/j.desal.2006.02.070. We would like to especial thank Alex Rodrigues, Prof Adriana Pohl- [21] F. El-Gohary, A. Tawfik, U. Mahmoud, Comparative study between chemical coagulation/precipitation (C/P) versus coagulation/dissolved air flotation mann and M.Sc. Selma Calgaroto from the Chemistry Department (C/DAF) for pre-treatment of personal care products (PCPs) wastewater, at our university and to all students of LTM for their assistance. Desalination. 252 (2010) 106–112, http://dx.doi.org/10.1016/j. desal.2009.10.016. [22] M. Karhu, V. Kuokkanen, T. Kuokkanen, J. Rämö, Bench scale electrocoagulation References studies of bio oil-in-water and synthetic oil-in-water emulsions, Sep. Purif. Technol. 96 (2012) 296–305, http://dx.doi.org/10.1016/j.seppur.2012.06.003. [1] M. Santander, R.T. Rodrigues, J. Rubio, Modified jet flotation in oil (petroleum) [23] B. Tansel, B. Pascual, Removal of emulsified fuel oils from brackish and pond emulsion/water separations, Colloids Surf. A Physicochem. Eng. Asp. 375 water by dissolved air flotation with and without polyelectrolyte use: Pilot- (2011) 237–244, http://dx.doi.org/10.1016/j.colsurfa.2010.12.027. scale investigation for estuarine and near shore applications (2011), http://dx. [2] J. Saththasivam, K. Loganathan, S. Sarp, An overview of oil–water separation doi.org/10.1016/j.chemosphere.2011.07.006. using gas flotation systems, Chemosphere. 144 (2016) 671–680, http://dx.doi. [24] A. Zouboulis, A. Avranas, Treatment of oil-in-water emulsions by coagulation org/10.1016/j.chemosphere.2015.08.087. and dissolved-air flotation, Colloids Surfaces A Physicochem. Eng. Asp. 172 [3] C.H. Rawlins, Flotation of fine oil droplets in petroleum production ciruits., in: (2000) 153–161, http://dx.doi.org/10.1016/S0927-7757(00)00561-6. D. Malhotra, P. Taylor, E. Spiller, M. Lavier (Eds.), Recent Adv. Miner. Process. [25] R. Moosai, R.A. Dawe, Gas attachment of oil droplets for gas flotation for oily Plant Des., Society for Mining, Metallurgy, & Exploration, 2009: pp. 232–246. wastewater cleanup, Sep. Purif. Technol. 33 (2003) 303–314, http://dx.doi.org/ [4] NSC, The Prevention of Pollution from Offshore Installations, in: Fifth Int. Conf. 10.1016/S1383-5866(03)00091-1. Prot. North Sea, Bergen, Norway, 2002: pp. 29–31. [26] J.J.A. Atarah, The use of Flotation Technology in Produced Water Treatment in [5] J.A. Veil, M.G. Puder, D. Elcock, R.J.J. Redweik, A White Paper Describing the Oil & Gas Industry, University of Stavanger, Norway, 2011. Produced Water from Production of Crude Oil, Natural Gas, and Coal Bed [27] M. Santander, Separação de óleos emulsificados por flotação não convencional, Methane. Prepared by: Argonne National Laboratory. Prepared for: U.S. Masters Dissertation, UFRGS, 1998. (in Portuguese) 332 R. Etchepare et al. / Separation and Purification Technology 186 (2017) 326–332
[28] M.A. Hampton, A.V. Nguyen, Accumulation of dissolved gases at hydrophobic [31] N. Mishchuk, The role of hydrophobicity and dissolved gases in non- surfaces in water and sodium chloride solutions: implications for coal equilibrium surface phenomena, Colloids Surf. A Physicochem. Eng. Asp. 267 flotation, Miner. Eng. 22 (2009) 786–792, http://dx.doi.org/10.1016/j. (2005) 139–152, http://dx.doi.org/10.1016/j.colsurfa.2005.06.052. mineng.2009.02.006. [32] Klaus Werner Stöckelhuber, Boryan Radoev, and Andreas Wenger, H.J. Schulze, [29] P. Attard, Nanobubbles and the hydrophobic attraction, Adv. Colloid Interface Rupture of Wetting Films Caused by Nanobubbles, (2003). doi:10.1021/ Sci. 104 (2003) 75–91, http://dx.doi.org/10.1016/S0001-8686(03)00037-X. LA0354887. [30] H. Schubert, Nanobubbles, hydrophobic effect, heterocoagulation and [33] R. Etchepare, A. Azevedo, S. Calgaroto, J. Rubio, Removal of ferric hydroxide by hydrodynamics in flotation, Int. J. Miner. Process. 78 (2005) 11–21, http://dx. flotation with micro and nanobubbles, Sep. Purif. Technol. 184 (2017) 347– doi.org/10.1016/j.minpro.2005.07.002. 353, http://dx.doi.org/10.1016/j.seppur.2017.05.014.