Project Topics
Get Complete Project Material File(s) Now! »
Homogenous and heterogeneous regime
Bubble flow regime in bubble columns is characterized by the gas-liquid distribution on the column section. The bubble flow regimes are well described by Zahradník et al. (1997) and Mena et al. (2005). At low gas velocities and with gas diffusers with small orifices the homogeneous flow regime takes place: the interaction between bubbles is small (bubble coalescence and break-up are insignificant), the bubble size distribution is narrow, and the gas hold-up is uniformly distributed in the column section (HoR in Figure I.3 (a)). At higher gas flow rates or with diffusers with large orifices the heterogeneous flow regime occurs: bubble coalescence takes place and consequently larger bubbles appear in the dispersion. The bubble size distribution is wider and the gas hold-up profile in the column section is parabolic with a maximum value at the centre (HeR in Figure I.3 (a)). A transitional regime occurs between the homogenous and heterogeneous regimes. It is characterized by the beginning and complete development of liquid circulation patterns in the bubble bed. Affected by the liquid circulation, bubble rise faster and the gas hold-up stops increasing with the superficial gas velocity and declines. Subsequently when the heterogeneous regime is established at higher gas flow rates, the liquid circulation patterns are reduced and the gas hold-up increases again with the superficial gas velocity though with a less pronounced slope. The typical evolution of the gas hold-up with the superficial gas velocity is shown in Figure I.3 (b).
Factors affecting the bubble regime
The study of Zahradník et al. (1997) summarized some parameters affecting the formation and stability of bubbling regimes. Design parameters such as distributor type and geometry; reactor geometry as well as system properties such as the electrolytes content and viscosity were studied. The main results are described in the following paragraphs.
Diameter orifice and reactor geometry
To study the distributor influence on bubble regime, 8 different perforated plates (orifices diameters between 10mm and 100µm) were used in an air-water system. It was observed that for orifice diameters bigger than 1.6mm, the homogeneous bubbling regime could not be generated. For orifice diameters smaller than 1mm the flow regime and the maximum gas hold-up presented an increased stability. The results have also shown that for all gas diffusers the fully heterogeneous bubbling regime is developed at a superficial gas velocity higher than 0.125m s-1. Concerning the influence of the reactor geometry, it was observed that decreasing the ratio of the liquid height to the column diameter (H/D) significantly increases the gas hold-up for all distributors, thus accentuating the homogeneous bubbling regime. On the contrary, at high gas flow rates in the heterogeneous regime, the gas hold-up was independent on the ratio (H/D).
Dissolved electrolytes and liquid viscosity
In aqueous solutions of electrolytes (NaCl, KCl, MgSO4, KI) Zahradník et al. (1997) observed that the heterogeneous regime was reached at a higher superficial gas velocities compared to the air- water system. The authors attributed this result to the hindrance of coalescence due to the salt concentration (from 0.036 up to 0.38 mol L-1). On the other hand, the influence of the liquid viscosity was evaluated using saccharose solutions. It was observed that at a given gas flow rate, an increment in the fluid viscosity related to an increase in the saccharose concentration, led to a gas hold-up reduction due to an enhancement of bubble coalescence. The cause-effect link between fluid viscosity and bubble coalescence is explained by Stewart (1995) as follows: an increase in the liquid viscosity reduces the turbulence in the wake of individual rising bubbles and as a consequence neighbour bubbles are drawn into other bubble’s wake thus facilitating bubble collision and eventually coalescence.
Design and operation parameters affecting oxygen transfer in clean water
Several research works have evaluated how design and operating parameters influence the oxygen transfer in clean water. The main results (principally obtained in full scale stirred reactors) are summarized in the next paragraphs.
Gas diffusers density and distribution
The gas diffuser density refers to the fraction of the reactor cross sectional area that is covered with membrane diffusers.
Impact on : Oxygen transfer increases for higher gas diffuser densities (Wagner and Pöpel, 1998) and well distributed diffusers on the reactor surface (ASCE, 1992).
Mechanisms: The rising bubble swarm induces an uprising liquid circulation flow associated to drag and buoyancy forces. If the density of gas diffusers is low or gas diffusers are unevenly distributed on the reactor surface, the liquid goes down preferentially in the non-aerated surfaces consequently generating vertical liquid recirculation flows, called ‘spiral flows’ that end up reducing the gas hold-up and oxygen transfer (Gillot, 1997; Capela, 1999; Fayolle, 2006).
Gas diffuser type
Impact on : The oxygen transfer depends on the installed diffuser type: perforated plate, sintered glass porous plate; perforated flexible membrane (Bouaifi et al., 2001).
Mechanisms: The gas diffuser type determines the bubble size and bubble regime and consequently the gas-hold-up, the interfacial area ( ) and also the liquid-side transfer coefficient ( ).
Submergence of diffusers
Impact on : The increase of the liquid height above the gas diffusers leads to a reduction of oxygen transfer (Pöpel and Wagner, 1996; Gillot et al., 2005).
Mechanisms: For a given gas flow rate, an increase in the liquid height promotes the development of vertical liquid circulation patterns which lead to a reduction of the gas hold-up.
Superficial gas velocity
Impact on : The superficial gas velocity is positively correlated with the oxygen transfer coefficient (Bouaifi et al., 2001; Gillot et al., 2005).
Mechanisms: Primarily an increase of the gas hold-up and consequently of the interfacial area (Bouaifi et al., 2001). The oxygen transfer is higher despite an increase of the bubble size and a reduction of the liquid-side transfer coefficient (Colombet et al., 2011). It must be underlined that the oxygen transfer efficiency (the ratio of transferred oxygen to supplied oxygen) is reduced with an increase in the superficial gas velocity. The reason for this reduction is that increasing the superficial gas velocity induces large-scale liquid circulations (spiral flows) that result in lower gas hold-up (Gillot, 1997; Capela, 1999; Fayolle, 2006).
Liquid circulation velocity
Some wastewater treatment reactors have a geometrical configuration in which a horizontal liquid circulation is induced by means of slow speed submerged impellers (loop reactors). The resulting horizontal liquid velocity has an effect on oxygen transfer.
Impact on : The application of a horizontal velocity induces an increase in oxygen transfer (Déronzier et al., 1996; Gillot et al., 2000; Gillot and Héduit, 2000).
Mechanisms: The liquid circulation velocity neutralizes the ‘spiral flows’ (Czarnota and Hahn, 1995), leading to an extended gas hold-up and consequently an expanded interfacial area. According, to Fayolle (2006) the liquid circulation reduces slightly the bubble diameter (-10%) but is not enough to explain the whole augmentation of the interfacial area.
Oxygen transfer in activated sludge
Activated sludge is a complex fluid composed of biological aggregates or flocs and interstitial water. For classical activated sludge and MBR sludge, the water content can range from 98.5 to 99.8 w%. Biological aggregates are a heterogeneous mixture of microbial colonies, colloids, organic fibers and particles, inorganic compounds embedded in a network of extracellular polymeric substances (EPS) of biological origin such as proteins, humic acids, polysaccharides, DNA and nucleic acids (Sheng et al., 2008; Wilen, 2008). Interstitial water is composed by soluble organic matter, soluble N and P species, soluble EPS, surfactants and salts. The physicochemical properties of the continuous and dispersed phases define the size of the biological aggregates. Also, because biological flocs are actually shear sensitive, the hydrodynamic conditions associated to the reactor configuration determine the flocculation-breakup equilibrium and consequently the floc size (Spicer et al., 1998; Biggs and Lant, 2000; Bouyer et al., 2001; Wilén et al., 2003; Jin and Lant, 2004; Bouyer et al., 2005; Coufort et al., 2008).
The diverse components of activated sludge and the associated properties have an impact on oxygen transfer. Under the same design and operating conditions, the oxygen transfer capacity is lower in activated sludge than in clean water. This oxygen transfer depletion is characterised by the alpha factor ( ) defined as: = ′ I.35 with ′ oxygen transfer coefficient in polluted water (h-1) oxygen transfer coefficient clean water (h-1).
The magnitude of this reduction is conditioned on how the oxygen transfer characteristic parameters ( , and ) are impacted by the various components and properties of activated sludge.
Dissolved oxygen saturation concentration in activated sludge (Cs’)
The dissolved oxygen saturation concentration in activated sludge ( ′) is considered to be slightly lower compared to the value in clean water ( ). Under the same operating conditions, this reduction is characterized by means of a coefficient noted as follows: ′ = I.36.
The value of the coefficient ranges between 0.95 and 0.99 depending on the nature of the wastewater influent. A value of =0.99 is classically used for domestic activated sludge (ASCE, 1996).
The similarity of the values in clean water and activated sludge has been recently confirmed by Jimenez (2013) who measured the dissolved oxygen saturation concentration of activated sludge interstitial water and did not observed any variation compared to clean water, within the margin of experimental error (±2%).
Table of contents :
CHAPTER I. LITERATURE REVIEW
I.1 Activated sludge (AS) process
I.1.1 Aeration systems
I.2 Principles of gas-liquid mass transfer
I.2.1 The volumetric oxygen transfer coefficient (kLa)
I.2.2 The liquid-side mass transfer coefficient (kL)
I.2.3 Specific interfacial area (a)
I.2.4 Gas-liquid dispersions
I.2.4.1 Gas and liquid velocities
I.2.4.2 Homogenous and heterogeneous regime
I.2.4.3 Bubble size
I.2.4.4 Bubble terminal velocity
I.2.5 Design and operation parameters affecting oxygen transfer in clean water
I.2.5.1 Gas diffusers density and distribution
I.2.5.2 Gas diffuser type
I.2.5.3 Submergence of diffusers
I.2.5.4 Superficial gas velocity
I.2.5.5 Liquid circulation velocity
I.2.6 Oxygen transfer in activated sludge
I.2.6.1 Dissolved oxygen saturation concentration in activated sludge (Cs’)
I.2.6.2 Activated sludge properties affecting the oxygen transfer coefficient (kLa’)
I.3 Rheology principles
I.3.1 Laminar shear flow
I.3.2 Rheological measurements
I.3.2.1 Rheometers
I.3.3 Rheological behaviours
I.3.3.1 Non-Newtonian fluids
I.3.4 Activated sludge rheology
I.3.4.1 Activated sludge thixotropy
I.3.4.2 Activated sludge rheology modelling
I.3.4.3 Sensitivity of activated sludge rheology to measurement conditions
I.3.4.4 Activated sludge properties affecting the rheological behaviour
I.3.5 Shear rate in aerated bioreactors
I.4 Conclusions on the literature review and work positioning
CHAPTER II. MATERIALS AND METHODS
II.1 Oxygen transfer in clean water and with activated sludge
II.1.1 Experimental setup: Bubble column and aeration system
II.1.2 Measurements of oxygen transfer coefficient in clean water (kLa)
II.1.2.1 Reoxygenation method – Principles
II.1.2.2 Measurement protocol
II.1.3 Measurements of oxygen transfer coefficient in activated sludge (kLa’)
II.1.3.1 Off-Gas method – Principles
II.1.3.2 Experimental setup for oxygen transfer measurements with activated sludge
II.1.3.3 Off-gas measurement protocol
II.1.3.4 Validation of the off-gas method with the reoxygenation method in clean water
II.1.4 Hydrodynamic characterization of the bubble column
II.1.4.1 Measurements of the overall gas hold-up (εG)
II.2 Activated sludge rheological measurements
II.2.1 Construction of a tubular rheometer
II.2.1.1 Rheometer specifications
II.2.1.2 Principles of a rheological measurement with a tubular rheometer
II.2.1.3 Design of the tubular rheometer
II.2.1.4 Description of the constructed tubular rheometer
II.2.2 Rheological behaviour of activated sludges from different plants
II.2.3 Activated sludge rheology and oxygen transfer measurements on site
II.2.4 Temperature effect on the activated sludge rheological behaviour
II.3 Physicochemical characterisation of activated sludge
II.4 Statistical analysis
CHAPTER III. DEVELOPMENT OF A METHODOLOGY TO CHARACTERISE THE RHEOLOGICAL BEHAVIOUR OF ACTIVATED SLUDGE
III.1 Rheometer measurement uncertainty
III.1.1 Theoretical measurement uncertainty
III.1.2 Experimental error with tap water
III.2 Rheological measurements with activated sludge
III.2.1 Setting up a rheological measurement with activated sludge
III.2.1.1 Sample volume
III.2.1.2 Stirring speed in the feeding reservoir
III.2.1.3 Sample storage
III.2.2 Applying the Rabinowitsch-Mooney correction
III.3 Comparing the flow curves obtained with the tubes of different diameter .
III.4 Temperature effect on the rheological behaviour of activated sludge
CHAPTER IV. RHEOLOGICAL BEHAVIOUR OF ACTIVATED SLUDGE FROM DIFFERENT ORIGINS
IV.1 Experimental conditions
IV.2 Characterisation of the rheological behaviour
IV.3 Impact of activated sludge physicochemical properties on rheological behaviour
IV.3.1 Correlation between physicochemical characteristics of interstitial liquid and apparent viscosity
IV.3.2 Correlation between physicochemical characteristics of particulate phase and the sludge apparent viscosity
IV.3.2.1 Impact of mixed liquor suspended solid concentration on apparent viscosity
IV.3.2.2 Impact of other physicochemical characteristics of the particulate phase on apparent viscosity: Introducing the floc structure as an impacting parameter
IV.4 Modelling the rheological behaviour of activated sludge
IV.4.1 Evaluation of existing models
IV.4.1.1 Correlation between rheological parameters and MLSS concentration
IV.4.1.2 Modelling experimental rheograms with MLSS concentration
IV.4.1.3 Comparison of the developed model with other studies
IV.5 Conclusions
CHAPTER V. MEASUREMENT AND INTERPRETATION OF OXYGEN TRANSFER PARAMETERS IN A BUBBLE COLUMN LOCATED ON SITE
V.1 Experimental conditions
V.2 Preliminary measurements in clean water at different temperatures
V.2.1 Temperature effect on the overall gas hold-up
V.2.1.1 Experimental results
V.2.1.2 Influence of temperature on gas hold-up: potential mechanisms
V.2.1.3 Conclusion on the effect of temperature on the overall gas hold-up
V.2.2 Overall gas hold-up temperature correction
V.3 Impact of diffuser type of on oxygen transfer in clean water
V.3.1 Oxygen transfer volumetric coefficients
V.3.2 Characteristics of the gas/liquid dispersion
V.3.2.1 Overall gas hold-up
V.3.2.2 Bubble size, interfacial area and liquid-side volumetric coefficient
V.3.3 Transfer number in clear water
V.3.4 Conclusions
V.4 Oxygen transfer in activated sludge
V.4.1 Oxygen transfer coefficients (FB and CB diffusers)
V.4.2 Characteristics of the gas/liquid dispersion: overall Gas hold-up (FB and CB diffusers) ..
V.4.3 Impact of sludge properties on oxygen transfer parameters
V.4.3.1 Statistical analysis of sludge properties on oxygen transfer
V.4.4 Alpha factor
V.4.4.1 Impact of operating conditions on alpha factor
V.4.4.2 MLSS concentration: A key parameter for alpha factor modelling?
V.5 Conclusions
CHAPTER VI. CONTRIBUTION OF RHEOLOGICAL MEASUREMENTS TO INTERPRET GAS/LIQUID OXYGEN TRANSFER
VI.1 Apparent viscosity for the conditions prevailing in the bubble column
VI.1.1 Rheological models
VI.1.2 Estimation of the average shear rate in the bubble column for the operating conditions
VI.1.3 Impact of the activated sludge apparent viscosity on the oxygen transfer coefficient ……
VI.1.4 Modelling the oxygen transfer coefficient in AS considering its non-Newtonian behaviour
VI.1.4.1 Model development
VI.1.4.2 Transfer number and oxygen transfer coefficient models for fine and coarse bubble aeration
VI.1.5 Interpreting oxygen transfer results with the help of the apparent viscosity
VI.1.6 Alpha factor
VI.2 Conclusions
CONCLUSIONS AND PERSPECTIVES
BIBLIOGRAPHY
