Microsoft Word - 48dallavecchia.docx CHEMICAL ENGINEERING TRANSACTIONS VOL. 49, 2016 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Guest Editors: Enrico Bardone, Marco Bravi, Tajalli Keshavarz Copyright © 2016, AIDIC Servizi S.r.l., ISBN 978-88-95608-40-2; ISSN 2283-9216 Power Requirements for Complete Suspension and Aeration in an Unbaffled Bioslurry Reactor Alessandro Tamburini*, Andrea Cipollina, Francesca Scargiali, Giorgio Micale, Alberto Brucato Dipartimento di Ingegneria Chimica, Gestionale, Informatica, Meccanica - Università degli Studi di Palermo, Viale delle Scienze Edificio 6 - 90128 Palermo alessandro.tamburini@unipa.it Remediation of contaminated soils is spreading as a matter of crucial importance nowadays. Bioremediation via bioslurry reactors of sites polluted by recalcitrant pollutants has been proved to be a valuable option, although optimization is needed to reduce process costs. Free-surface unbaffled stirred tanks (with central air vortex) have been recently proposed as a promising alternative to the more common systems provided with baffles. In a bioslurry reactor solid-liquid interfacial area, oxygen supply, solid loading per reactor unit volume should be maximized, and, at the same time, operation costs have to be kept low. In this regard, the minimum impeller speeds for complete suspension Njs (suspension of all solid particles) and aeration Nca (air vortex ingested by the turbine and dispersed as bubbles in the system) represents a reasonable compromise between process yield and power requirements. To this purpose, a flat bottomed unbaffled tank with diameter T=0.19 m was investigated. The tank was filled with water up to a height H=T. It was stirred by a radial six- bladed Rushton turbines (RT) with diameter D=T/3 and H=T/3. Mono-dispersed particles with diameter dp=250-300μm and density ρ≈2500 kg/m 3 were employed. Solid loadings B% ranging from 2.5% (weight of solid/ weight of liquid) up to the very high 160% w/w were tested. The visual Zwietering criterion along with the aid of a digital camera was employed to evaluate Njs values. An acoustic criterion was adopted to assess Nca. A static frictionless granite turntable was employed to measure the impeller torque at Njs and Nca and to assess the relevant specific power requirements εjs and εca. Results show that the dependence of Njs and Nca on B% is much lower at low solids loading (B<30%), while a larger dependence was found at larger B% values (B>30%). The relevant specific powers per unit mass of solids (i.e. εjs and εca) were found to exhibit a minimum, at B≈20% for εjs and B≈60% for εca. On overall, data collected suggest that operating a radially stirred unbaffled bioslurry reactor loaded with a concentration B≈30% could be the best compromise to minimize the costs for achieving complete suspension and aeration conditions. 1. Introduction A number of industrial sites have been dramatically polluted during the last century thus leading the interest towards novel and efficient remediation technologies to increase. The technologies making use of aerobic bacteria are commonly considered as particularly suited for soils contaminated by organic compounds such as hydrocarbons, pesticides, pharmaceuticals, etc. (Eweis et al., 1998; Rodríguez-Rodríguez et al., 2012). Unfortunately, the treatment times may be dramatically high when recalcitrant pollutants as polycyclic aromatic hydrocarbons (PAHs) are present. In this regard, bioslurry reactors are a bioremediation technology, which allows the concentration of this kind of pollutants to be reduced to very low values in reasonable times (Zappi et al., 1996; Lewis, 1993). Such systems are in fact provided with stirrers which, by agitating the slurry, increase transfer coefficients thus enhancing pollutants abatement rates (Cassidy et al., 2000). Bioslurry reactors have been seldom employed so far because of the high operation costs. In this regard, the optimization of the fluid dynamics in the tank may allow the costs to sensibly reduce and the technology to consequently spread. In bioslurry reactors for bioremediation, high solid-liquid ratios are employed and insufflation devices require frequent maintenance since the pores could significantly wear or even be clogged. For such kind of processes, free-surface unbaffled stirred tanks may represent a valuable alternative DOI: 10.3303/CET1649076 Please cite this article as: Tamburini A., Cipollina A., Scargiali F., Micale G., Brucato A., 2016, Power requirements for complete suspension and aeration in an unbaffled bioslurry reactor, Chemical Engineering Transactions, 49, 451-456 DOI: 10.3303/CET1649076 451 (Tamburini et al., 2012a). They are characterized by the presence of a central air vortex (vortexing unbaffled tanks). At low impeller speeds N, the vortex does not reach the stirrer (sub-critical conditions) and the oxygen supply occurs only through the vortex surface (Scargiali et al., 2013). The relevant volumetric mass transfer coefficient kLa was found to be much higher (more than one order of magnitude) than the case of a not agitated liquid system with free surface (Scargiali et al., 2014a) and this may be anyway sufficient for some processes (Scargiali et al., 2014b). As N increases, the vortex moves downwards in the central part of the tank. At a particular value of N named Nca, air is ingested and dispersed throughout the tank, thus leading kLa to be largely enhanced. At N ≥ Nca (super-critical conditions) the tank is at the same time agitated and aerated without the air sparger maintenance drawbacks typically occurring in the case of highly dense solid-liquid systems as bioslurry reactors. Moreover, suspension of the soil particles in the tank is a matter of crucial importance to improve the contact with the solid phase and consequently enhance the relevant mass transfer coefficients. In this regard, unbaffled systems have been proved to be less energy demanding than the traditional systems provided with baffles (Brucato et al., 2010; Wang et al., 2012). On overall, in a bioslurry reactor solid-liquid interfacial area, oxygen supply, solid loading per reactor unit volume should be maximized, and, at the same time, operation costs have to be kept low. The aim of the present work is to find a good compromise which could represent an optimum among all these different aspects. The minimum impeller speed for complete suspension (suspension of all solid particles) Njs is well- known to be a good compromise between particle interfacial area maximization and agitation costs minimization (Davoody et al., 2015; Kasat and Pandit, 2005) also in biotechnological processes (Ibrahim and Nienow, 2004; Rafiq et al., 2013). Similarly, the minimum impeller speed for complete aeration (air ingested by the turbine and dispersed as bubbles in the system) Nca represents a reasonable compromise between oxygen supply maximization and agitation costs reduction. Therefore, in the present work Njs and Nca were assessed along with their corresponding power requirements per unit mass of solids εjs and εca. By combining the above quantities, it would be possible to recognize the “optimal” soil loading guaranteeing at the same time lower power requirements for both complete suspension and aeration. 2. Experimental A flat-bottomed unbaffled tank with diameter T=0.19m was investigated. The tank was filled of water up to a height H=T. It was radially stirred by a standard six-bladed Rushton turbine (RT) with diameter D=T/3 and clearance C=T/3. Mono-dispersed particles with diameter dp=250-300μm and density ρ≈2500 kg/m 3 were employed. Bioslurry reactors are usually operated with high solid loadings in order to maximize the soil amount treated per unit volume of the reactor. Thus, solid concentrations B% ranging from 2.5% (weight of solid/ weight of liquid) up to the very high 160% w/w were tested in the present work. The tank was also uncovered and no gas-sparger was provided. As a matter of fact, a bioslurry system operating under aerobic conditions requires the oxygen consumed by the biomass to be suitably replaced (Leunen et al., 2002; Rodríguez-Rodríguez et al., 2012). In vortexing unbaffled tanks this may well occur through the central vortex gas-liquid inter-phase and through air bubbles surface, for systems operated at agitation speeds sufficient for air entrapping. 2.1 Njs assessment The minimum turbine speed insuring the suspension of all particles (Njs) was assessed by the well-known “one second criterion” (Zwietering, 1958). A camera was placed underneath vessel bottom in order to collect a number of images (about 20) at each impeller speed. Camera exposure time was set to one second in accordance with Zwietering’s criterion, so that motionless particles appeared to be well defined, while moving particle were blurred in the pictures. Njs was defined as the minimum impeller speed at which no well-defined particles were observable in all pictures (Tamburini et al., 2015). 2.2 Nca assessment Nca is the minimum agitation speed at which the vortex reaches the impeller turbine plane and blades; this causes the air to be ingested and distributed throughout the whole vessel by the turbine rotation thus leading the system to become three-phasic (complete aeration). For the purpose of the present investigation a simple methodology was employed for the assessment of Nca. It was named “hearing criterion” and is based on the noise produced by the rotating turbine while it is ingesting air. When the vortex depth reaches the turbine and the air is discharged by its blades, a clear hollow noise can be heard: the minimum speed at which this noise is heard is defined as Nca. Some tests were repeated three times, also by different operators, and a surprising reproducibility was found: a maximum discrepancy of about 5% was found. 452 2.3 Power requirement assessment Power measurements were performed by assessing the torque transmitted by the impeller to the tank with the apparatus described by Tamburini et al. (2014a). It is a “static-frictionless” turntable consisting of a granite dish able to rotate around its central axis on a granite table. This arrangement practically cancelled static friction between the surfaces, yet allowing dynamic friction to dump torque oscillations. Notably, power measurements were also performed in a corresponding tank provided with baffles for comparison purposes. 3. Results and Discussion First results are shown in Figure 1 and concern the dependence of the minimal impeller speeds for complete suspension Njs (Figure 1a) and aeration Nca (Figure 1b) on the solid loading. As it can be seen in Figure 1a, Njs values assessed in the vortexing unbaffled vessel are compared with those assessed in a corresponding baffled tank via the well-known Zwietering correlation. The values in the vortexing unbaffled tank are much lower than those pertaining the baffled tank. Values relevant to the baffled configuration are reported up to B=20%, since Zwietering’s correlation may be unreliable at large solid loadings (Tamburini et al., 2012b). Within this range (where comparison is possible), the dependence on of Njs on B% is higher in the baffled system (i.e. Njs ∝ B0.13) than in the unbaffled one (i.e. Njs ∝ B0.03). At larger solid loadings (B>20%, where only unbaffled data are reported), particle-particle interactions became prominent and rheology of the suspension allegedly changes thus leading to a much stronger dependence of Njs on B% (i.e. Njs ∝ B0.54). Surprisingly, a similar behaviour was observed also for the case of the complete aeration speed (Figure 1b). More precisely, Nca seems to be independent of solid concentration at low solid loading (B≤30%), while for B>30%, solid concentration strongly affects the complete aeration speed Nca (i.e. Nca ∝ B0.95 for B>50%). This may be the consequence of a certain effect of solid particle concentration increase on vortex features, but additional data are needed to find an answer to this issue. In this regard, local data provided by CFD simulations might be a viable solution. Figure 1: (a) dependence of Njs on B%; (b) dependence of Nca on B%. Complete suspension and aeration speed represent valuable information, although corresponding power requirements are more suitable to perform comparison among different geometrical configurations. Therefore, in Figure 2a, the minimum power requirements needed to suspend all particles (i.e. power measured at N=Njs) is reported as a function of solid loading. Also, the values experimentally assessed are compared with those relevant to a baffled tank stirred by a A310 stirrer which represents the most efficient configuration available for baffled tanks. The Pjs values relevant to this specific baffled configuration are assessed by firstly calculating the Njs values by means of the Zwietering’s correlation whose geometrical value S is inferred from the data by Wong et al. (1987). Once the Njs values have been obtained, Pjs values can be calculated from the power number expression (Np=Pjs/(ρsuspNjs 3D5)). For this A310-baffled configuration, Np was estimated equal 0.36 from measurements performed at B=0%. As shown in Figure 2a, the Pjs values measured in the vortexing unbaffled configuration are much lower than those relevant to the best option available for baffled vessels, thus suggesting a significant economic convenience in adopting a configuration unprovided with baffles. On the other hand, a similar dependence of Pjs on B% was found for the two configurations. Also in this case, the comparison only concerns values of B% lower than 20% since unreliable predictions may be provided by the Zwietering correlation at larger B% (Oldshue and Sharma, 1992). 100 1000 10000 1 10 100 1000 N js [r pm ] B[%] Vortexing Unbaffled tank Baffled tank_Zwietering correlation 100 1000 10000 1 10 100 1000 N ca [r p m ] B[%] Vortexing Unbaffled tank (a) (b) 453 The different power requirements between the baffled and unbaffled configuration is probably due to the different suspension mechanism which leads to different Njs values (the lower N, the lower P) but also to the different particle distribution in the two tanks. In baffled tanks solids are much better distributed throughout the tank (Tamburini et al., 2013a), although at Njs higher local concentrations may be located in the stirrer region. Conversely, in unbaffled tanks radially stirred, distribution is poor and particles are preferentially concentrated in some zones of the vessel which have been found far from the stirrer (Tamburini et al., 2013b) thanks to the higher centrifugal forces. Thus, the different distribution probably adds, to other geometrical and fluid-dynamic effects, another effect linked to the different apparent density seen by the impeller. As concerns the P vs B% trend, as already observed in Figure 1a for Njs, Pjs shows a lower dependence on B% at lower solid loadings (i.e. Pjs ∝ B 0.39 for B≤20%), while at larger particle concentration a slight increase in the value of B% results into a much larger power consumption increase (i.e. Pjs ∝ B 1.59 for B>20%). A similar behaviour can be observed also in Figure 2b for the case of the power required for complete aeration conditions, Pca: for B≤30%, Pca exhibits a dependence on B% lower (i.e. Pca ∝ B 0.28 for B≤30%) than that pertaining the cases at B>30% (i.e. Pca ∝ B1.65 for B≤20%). Notably, as a difference from Nca, Pca shows some dependence on B% also at low solid loadings: as the solid loading increases within the range 0-30%, the vortex reaches the impeller blades at very close agitation speeds, independently of the particle concentration, while the apparent density encountered by the impeller increases as B% increases. Thus, the increase of Pca with B% within the range 0-30% is allegedly due only to apparent density effects rather than vortex features variations. Figure 2: (a) dependence of Pjs on B%; (b) dependence of Pca on B%. In bioslurry reactors it is important to minimize the remediation cost of an unit mass of contaminated soils. In this regard, power consumption per unit mass of solids is another important parameter that should be assessed. It can be easily estimated as the ratio between the power requirements and the solid loadings (Bong et al., 2015). With reference to our case, the specific power consumption for the just attained suspension of all particles (εjs) and the specific power consumption for the just attained aeration of the tank (εca) are parameters of great importance. Their estimation allow to recognize which is the operating condition, in terms of solid loading, guaranteeing the largest cost savings. This can be easily inferred from Figure 3, which shows the dependence of εjs (Figure 3a) and εca (Figure 3b) on B%. As reported in Figure 3a, εjs exhibits a decreasing-increasing dependence on B%, thus resulting into a minimum at about B=20%. This value of B%, named Bjs, should be regarded as the “optimal solid loading condition for complete suspension” at which the tank should be operated to minimize operation costs for complete suspension. It is worth noting that operating the tank at different conditions may result into dramatic increase of the specific power consumption. Just as an example, operating at B=2.5% results into more than tripling the specific power consumption and significantly reduce the process performance. Analogous consideration can be done for the case of εca: as shown in Figure 3b, also the εca – B% trend is non-monotonic and exhibits a minimum which in this case occurs at B=60%. This value of B%, named Bca, should be regarded as the “optimal solid loading condition for complete aeration” which should be adopted to minimize operation costs for the complete aeration of the system. 0.1 1 10 100 1 10 100 1000 P j s [W ] B[%] Vortexing Unbaffled Tank A310 D=T/3 Baffled_Wong et al. correlation 0.1 1 10 100 1 10 100 1000 P c a [W ] B[%] Vortexing Unbaffled tank (a) (b) 454 Figure 3: (a) dependence of εjs on B%; (b) dependence of εca on B%. Since Bjs and Bca are different, the choice of the best operating condition cannot be performed without referring to the specific process under consideration. For instance, different processes (e.g. including different bacteria and substrates) may require different oxygen demand: when it is not high, the oxygen transfer via vortex surface might be sufficient and no complete aeration is required. Similarly, the burst of bubbles at the air–liquid interface may in some cases result into a damage for some cells (Chisti, 2000). Also, it is well known that operating some processes at impeller speeds slightly lower than Njs (Tamburini et al., 2011) may easily result into power savings which more than counterbalance the yield reduction due to the lower interfacial area (Tamburini et al., 2014b). In case both aeration and suspension are needed, operating the system at B=30% might represent a good compromise between the two aspects. 4. Conclusions In the present work a free surface unbaffled tank stirred by a standard 6-bladed Rushton turbine was investigated. The possibility of adopting this apparatus as a bioslurry reactor was assessed by studying solid- liquid suspensions at different concentrations: in particular different solid loadings B% ranging from 2.5% w/w (weight of solid/ weight of liquid) to 160% were tested. The experimental work concerned the assessment of the minimum impeller speed for complete suspension Njs and the minimum impeller speed for complete aeration Nca, conditions being both recommended for such systems. Moreover, power requirements corresponding to these two conditions were also measured in order to find the optimal concentration guaranteeing the lowest energetic costs for achieving both phenomena. Results show that the dependence of Njs on solid concentration B% is somehow is low and similar to that found by Zwietering (1958) for the case of the baffled vessels up to B≈25%. Conversely, a much larger dependence was found at higher B%. As concerns the identification of the most convenient concentration value for complete suspension requirements, εjs trends showed a minimum at B≈20%.The value of Nca showed a negligible dependence on B% at low solids loading (B<30%), while a larger dependence was found at larger B% values (B>30%). The most convenient concentration for the complete aeration was found to be at B≈60%. Therefore, details on the specific process under consideration are needed in order to recognize which is the controlling phenomenon between suspension and aeration and choose the relevant optimal concentration. When these details are not available or when both phenomena are equally important, it can be concluded that operating a bioslurry reactor loaded with a concentration B≈30% could be the best compromise to minimize the costs for achieving complete suspension and aeration conditions. Reference Bong, E.Y., Eshtiaghi, N., Wu, J., Parthasarathy, R., 2015. Optimum solids concentration for solids suspension and solid-liquid mass transfer in agitated vessels, Chem. Eng. Res. Des. 100, 148-156. Brucato A., Cipollina A., Micale G., Scargiali F., Tamburini A., 2010. Particle suspension in top-covered unbaffled tanks, Chem. Eng. Sci. 65, 3001-3008. Cassidy D.P., Efendiev S., White D.M., 2000. A comparison of CSTR and SBR bioslurry reactor performance. Water Research 34 (18), 4333-4342. Chisti Y., 2000. Animal-cell damage in sparged bioreactors. Trends in Biotechnology 18, 420-432. 0 1 10 100 1 10 100 1000 ε j s [W /k g] B[%] Vortexing Unbaffled tank 0.1 1 10 100 1 10 100 1000ε c a [W /k g] B[%] Vortexing Unbaffled tank (a) (b) 455 Davoody M., Abdul Raman A.A.B., Parthasarathy R., 2015. Maximizing Impeller Power Efficiency in Gas- Solid-Liquid Stirred Vessels through Process Intensification. Ind. Eng. Chem. Res. 54, 11915-11928. Eweis J.B., Ergas S., Chang P.D.Y., Schroeder D., 1998, Bioremediation Principles, McGraw-Hill, New York, United States. Ibrahim S., Nienow A.W., 2004. Suspension of microcarriers for cell culture with axial flow impellers. Chemical Engineering Research and Design 82 (9 SPEC. ISS.), 1082-1088. Kasat G.R., Pandit A.B., 2005. Review on mixing characteristics in solid-liquid and solid-liquid-gas reactor vessels. Canadian Journal of Chemical Engineering 83, 618-643. Leunen L.A., Buggs V.H., Eestep M.E., Enriquez R.C., Leonerd J.W., Blaylock M.J., Huang, J.W., Haggblom, M.M., 2002. Bioremediation of polyaromatic hydrocarboncontaminated sediments in aerated bioslurry reactors. Bioremediation Journal, 6, 125-141. Lewis R.F., 1993. SITE demonstration of slurry phase biodegradation of PAH contaminated soil. Journal of the Air and Waste Management Association 43, 503-508. Oldshue, J.Y., Sharma, R.N.,1992. The effect of off-bottom distance of an impeller for the ‘just suspended speed’, Njs. AIChE Symp. Ser. 88, 72–78. Rafiq, Q.A., Brosnan, K.M., Coopman, K., Nienow, A.W., Hewitt, C.J., 2013. Culture of human mesenchymal stem cells on microcarriers in a 5 l stirred-tank bioreactor, Biotechnology Letters 35 , 1233-1245. Rodríguez-Rodríguez, C.E., Barón, E., Gago-Ferrero, P., Jelić, A., Llorca, M., Farré, M., Díaz-Cruz, M.S., Eljarrat, E., Petrović, M., Caminal, G., Barceló, D., Vicent, T., 2012. Removal of pharmaceuticals, polybrominated flame retardants and UV-filters from sludge by the fungus Trametes versicolor in bioslurry reactor, Journal of Hazardous Material 233-234, 235-243. Sardeshpande M.V., Sagi A.R., Juvekar V.A., Ranade V.V., 2009. Solid suspension and liquid phase mixing in solid–liquid stirred tanks, Ind. Eng. Chem. Res. 48, 9713–9722. Scargiali F., Busciglio A., Grisafi F., Tamburini A., Micale G., Brucato A., 2013. Power consumption in uncovered unbaffled stirred tanks: Influence of the viscosity and flow regime. Ind. Eng. Chem. Res. 52, 14998-15005. Scargiali F., Busciglio A., Grisafi F., Micale, G., Tamburini, A., Brucato A., 2014a. Oxygen transfer performances of unbaffled bio-reactors with various aspect ratios. Chemical Engineering Transactions 38, 1-6. Scargiali F., Busciglio A., Grisafi F., Brucato A., 2014b. Mass transfer and hydrodynamic characteristics of unbaffled stirred bio-reactors: influence of impeller design. Biochemical Engineering Journal 82, 41-47 Tamburini A., Cipollina A., Micale G., Brucato A., Ciofalo M., 2011. CFD Simulations of Dense Solid-Liquid Suspensions in Baffled Stirred Tanks: predictions of suspension curves. Chem. Eng. J. 178, 324-341. Tamburini A., Cipollina A., Micale G., Brucato A., 2012a. Measurements of Njs and power requirements in unbaffled bioslurry reactors. Chemical Engineering Transactions 27, 343-348. Tamburini A., Cipollina A., Micale G., Brucato A., Ciofalo M., 2012b. CFD simulations of dense solid-liquid suspensions in baffled stirred tanks: Prediction of the minimum impeller speed for complete suspension. Chem. Eng. J. 193-194, 234-255. Tamburini A., Cipollina A., Micale G., Brucato A., Ciofalo M., 2013a. CFD simulations of dense solid-liquid suspensions in baffled stirred tanks: Prediction of solid particle distribution. Chem. Eng. J. 223, 875-890. Tamburini A., Cipollina A., Micale G., Brucato A., 2013b. Particle distribution in dilute solid liquid unbaffled tanks via a novel laser sheet and image analysis based technique. Chem. Eng. Sci. 87, 341-358. Tamburini A., Brucato A., Busciglio A., Cipollina A., Grisafi F., Micale G., Scargiali F., Vella G., 2014a, Solid−Liquid suspensions in top-covered unbaffled vessels: influence of particle size, liquid viscosity, impeller size, and clearance. Ind. Eng. Chem. Res. 53, 9587-9599. Tamburini A., Cipollina A., Micale G., Brucato A., Ciofalo M., 2014b. Influence of drag and turbulence modelling on CFD predictions of solid liquid suspensions in stirred vessels. Chemical Engineering Research and Design 92, 1045-1063. Tamburini A., Cipollina A., Grisafi F., Scargiali F., Micale G., Brucato A, 2015. Comparison of agitators performance for particle suspension in top-covered unbaffled vessels. Chemical Engineering Transactions, 43, 1585-1590. Wang S., Parthasarathy R., Bong E.Y., Wu J., Slatter P., 2012. Suspension of ultrahigh concentration solids in an agitated vessel. AIChE Journal, 58, 1291-1298. Wong C.W., Wang J.P., Huang S.T., 1987, Investigations of fluid dynamics in mechanically stirred aerated slurry reactors. Can. Jour. Chem. Eng., 65, 412-419. Zappi M.E., Rogers B.A., Teeter C.L., Gunnison D., Bajpai R., 1996. Bioslurry treatment of a soil contaminated with low concentrations of total petroleum hydrocarbons. Journal of Hazardous Materials 46 1-12. Zwietering T.N., 1958. Suspending of solid particles in liquids by agitators, Chemical Engineering Science, 8, 244-253. 456