Variation of bubble size distribution in a protein foam fractionation column measured using a capillary probe with photoelectric sensors

Bubble size is used to characterize not only bubble-specific interfacial area but also bubble coalescence in a foam column. The bubble size distributions were obtained in a continuous foam fractionation process for concentrating ovalbumin using a developed photoelectric probe. When the continuous process reached steady state, the bubble size distribution pattern remained stable. Bubble size distribution data above (+1 cm) or below (-1 cm) the bulk liquid-foam interface showed symmetry along the diameter of the column (14 cm ID). The bubble size distribution was affected by the column wall. The nearly constant protein concentration distribution across the column cross-section indicated that the bubble flow distribution approached a flat profile across the column. A log-normal bubble distribution pattern best fit the weighted range of bubbles in the column at column lengths above and below the liquid-foam interface. These observations may prove to be useful in understanding the mechanisms underlying the foam fractionation of proteins.     

Effect of bubble size on foam fractionation of ovalbumin

The bubble size distribution and void fraction (ɛ _g ) (at two bulk liquid pool positions below the bulk liquid-foam interface and one lower foam phase position) in a continuous foam fractionation column containing ovalbumin were obtained using a photoelectric capillary probe. The bubble size and ɛ _g data were gathered for different operating conditions (including the changes in the superficial gas velocity and feed flow rate) at a feed solution of pH 6.5 and used to calculate the specific area, a, of the bubbles. Thus, local enrichment (ER _l ), values of ovalbumin could be estimated and compared with directly obtained experimental results. The ER _l results were also correlated with the bubble size and ɛ _g to understand better the concentration mechanisms of foam fractionation. The high ER _l in the lower foam phase was largely attributable to the abrupt increase in ɛ _g (from 0.25 to 0.75), or the a (from about 12 to 25 cm²/cm³) from the bulk liquid to the foam phase. These changes correspond with enhanced gravity drainage. With an increase in the superficial gas velocity, the bubble size increased and the a decreased in both the bulk liquid and lower foam phases, resulting in a decrease in the local experimentally determined enrichments at high superficial gas velocities. At intermediate feed flow rates, the bubble size reached the maximum. The ɛ _g and a, on the other hand, were the largest for the largest feed flow rate. The ER _l in the lower foam phase was maximized at the lowest feed flow rate. It follows, therefore, that a alone is not sufficient to determine the magnitude of the ER _l in the foam phase.     

Modeling a protein foam fractionation process

A simple staged model for the protein foam fractionation process is proposed in this article. This simplified model does not detail the complex foam structure and gas-liquid hydrodynamics in the foam phase but, rather, is built on the conventional theoretical stage concept considering upward bubbles with entrained liquid and downward liquid (drainage) as counter-current flows. To simulate the protein concentration distribution in the liquid along the column by the model, the bubble size and liquid hold-up with respect to the position must be known, as well as the adsorption isotherm of the protein being considered. The model is evaluated for one stage by data from the semibatch foam fractionation of egg albumin and data from the continuous foam fractionation of bovine serum albumin. The effect of two significant variables (superficial gas velocity and feed protein concentration) on enrichment is well predicted by the model, especially for continuous operation and semibatch operation when initial concentration is high.     

Effect of a natural contaminant on foam fractionation of bromelain

Foam fractionation is a simple, inexpensive method for separating and purifying proteins. Typically, a dilute bromelain solution with a pH ranging from 2.0 to 7.0 foams very well when bubbles are introduced into a foam fractionation column. It was observed, however, that the dilute enzyme solution only foamed between approximately pH 2.0 and 3.0 when the inner wall of the fractionation column was coated with a natural contaminant (okra residue). We studied the separation ratio and the protein mass recovery to explore the effect of a natural antifoaming agent on the foam fractionation of a dilute bromelain solution. The control variables used in this process were the initial bulk solution pH, which ranged from 2.0 to 7.0, and the superficial air velocity, which varied between 1.7 and 6.2 cm/s.     

Effect of denaturation by preheating on the foam fractionation behavior of ovalbumin

Ovalbumin is a globular protein. When it is denatured, it can produce molecular species with different conformational states, each of which has different adsorption properties at a gas-liquid interface. Such changes in adsorption can then affect the foaming behaviors of ovalbumin. Results of semi-batch foam fractionation of both native and denatured ovalbumin aqueous solutions are reported in this paper, along with possible relationships between denaturation and foam fractionation outcomes, such as the enrichment ratio and mass recovery. Bubble size and foam stability are determined in the experiments to show the effect of denaturation on these measured parameters in this system. The relationships between the bubble size, void fraction, and ovalbumin enrichment are also reported to reflect the effect of the presence of denatured species.     

Batch foam recovery of sporamin from sweet potato

The major sweet potato root protein, sporamin (which comprises about 80–90% of the total protein mass in the sweet potato) easily foams in a bubble/foam-fractionation column using air as the carrier gas. Control of that foam fractionation process is readily achieved by adjusting two variables: bulk solution pH and gas superficial velocity. Varying these parameters has an important role in the recovery of sporamin in the foam. Changes in the pH of the bulk solution can control the partitioning of sporamin in the foam phase from that in the bulk phase. A change in pH will also affect the amount of foam generated. The pH varied between 2.0 and 10.0 and the air superficial velocities (V₀) ranged between 1.5 and 4.3 cm/s. It was observed in these ranges that, as the pH increased, the total foamate volume decreased, but the foamate protein (mainly sporamin) concentration increased. On the other hand, the total foamate volume increased significantly as the air superficial velocity increased, but the foamate concentration decreased slightly. The minimum residual protein concentration occurred at pH 3.0 and V_o = 1.5 cm/s. On the other hand, the maximum protein mass recovery occurred at pH 3.0 and at V_o = 4.3 cm/s.     

Effect of environmental humidity on static foam stability

The quality of foaming products (such as beer and shampoo) and the performance of industrial processes that harness foam (such as the froth flotation of minerals or the foam fractionation of proteins) depend upon foam stability. In this study, experiments are performed to study the effect of environmental humidity on the collapse of static foams. The dependency of the rate at which a foam collapses upon humidity is demonstrated, and we propose a hypothesis for bubble bursting due to Marangoni instability induced by nonuniform evaporation to help explain the dependency. This hypothesis is supported by direct experimental observations of the bursting process of isolated bubbles by high speed video recording and the thinning of isolated foam films under different values of humidity and temperature by microinterferometric methods.     

A proposed mechanism for detergent-assisted foam fractionation of lysozyme and cellulase restored with β-cyclodextrin

Foam fractionation by itself cannot effectively concentrate hydrophilic proteins such as lysozyme and cellulase. However, the addition of a detergent to a protein solution can increase the foam volume, and thus, the performance of the foam fractionation process. In this article, we propose a possible protein concentration mechanism of this detergent-assisted foam fractionation: A detergent binds to an oppositely charged protein, followed by the detergent-protein complex being adsorbed onto a bubble during aeration. The formation of this complex is inferred by a decrease in surface tension of the detergent-protein solution. The surface tension of a solution with the complex is lower than the surface tension of a protein or a detergent solution alone. The detergent can then be stripped from the adsorbed protein, such as cellulase, by an artificial chaperone such as beta-cyclodextrin. Stripping the detergent from the protein allows the protein to return to its original conformation and to potentially retain all of its original activity following the foam fractionation process. Low-cost alternatives to beta-cyclodextrin such as corn dextrin were tested experimentally to restore the protein activity through detergent stripping, but without success.     

Adsorption of Water-Soluble Proteins onto Bubbles in Continuous Foam Separation

The mechanism of water-soluble protein enrichment in continuous foam separation was studied. The liquid flow rate and the protein concentration in the foam phase were measured at various heights from the interface between the bulk liquid and foam layer, and the intrinsic values at the interface were estimated by the extrapolation method to determine the accurate adsorption density on the bubble surface. Ovalbumin (OA) and hemoglobin (HB) were used as the soluble proteins. The solution pH values were varied from 3.5 to 6.0 for OA and from 6.0 to 8.0 for HB. The experimental isotherms for OA and HB were compared to the Langmuir isotherm, and the two adsorption parameters of the equilibrium constant, K, and the saturated density, gamma, at each pH were determined. Both gamma values obtained for OA and HB showed maxima at their isoelectric point (pH 4.6 for OA and pH 6.8 for HB). Assuming that OA and HB molecules are spherical in shape and are adsorbed on the bubble surface in a close-packed structure at saturation, the calculated diameters for OA and HB molecules were quite similar to the literature values. The variation in gamma for both OA and HB is discussed qualitatively in relation to the net charge of the protein molecule. Copyright 2000 Academic Press.     

Removing proteins from an aerated yeast fermentation by pulsing carbon dioxide

Salting-out is a common technique used for precipitating proteins and other materials from fermentation and tissue culture processes. It leaves a salt residue in the system. Foam fractionation can also be used to remove proteins by protein precipitation from a dilute solution. In doing so, there is usually a trade-off between enrichment and recovery. An increase in the airflow rate will increase the recovery, but only at the expense of the enrichment. A new method for increasing the recovery in foam fractionations and in yeast fermentations is to add a burst of CO₂ to the process and then restore the air. This CO₂ acts like a temporary salt, but it does not leave behind a residue. The recovery increases as a result of the joint use of these gases, perhaps by more than 10-fold, without sacrificing the enrichment. Chicken egg albumin in a foam fractionation column can serve as a simple, experimental model for the proposed recovery process in lieu of the fermentation process.     

Effect of protein denaturation on void fraction in foam separation column

Foam fractionation is a cost-effective process that uses air to extract protein from a liquid (in this case “crude” dilute egg-albumin solution). This article deals with how the void fraction (fraction of air in the aerated solution) of foam is affected by heat denaturation of the protein. A 2-mm glass tube was used to sample the foam-liquid interface fluid in a 35-mm-diameter column in order to detect small changes in void fraction and foam production, which are not easily detected directly from the bulk foam. The main control variablein this study was the protein solution preheating time. As the preheating time increased, the initial void fraction in the column decreased. The initial void fraction of the undenatured solution ranged from about 0.73 to 0.80, and the void fraction for significant preheating times of 5 min ranged from approx 0.68 to 0.72. Furthermore, the period of foam production increased from 5 to 7 min for undenatured proteins in solution to as long 15 min for 5-min preheated solutions. Side-port sampling through a small capillary tube has the potential to be used as a rapid and inexpensive way to determine the level of protein denaturation by directly determining the void fraction and then estimating the effect of denaturation from a protein denaturation calibration curve of the void fraction.     

Degeneration of β-glucosidase activity in a foam fractionation process

Foam fractionation is a promising technique for concentrating proteins because of its simplicity and low operating cost. One such protein that can be foamed is the enzyme cellulase. The use of inexpensively purified cellulase may be a key step in the economical production of ethanol from biomass. We conducted foam fractionation experiments at total reflux using the cellulase component β-glucosidase to study how continuous shear affects β-glucosidase in a foam such as a fermentation or foam fractionation process. The experiments were conducted at pH 2.4, 5.4, and 11.6 and airflow rates of 3, 6, 15, 20, and 32 cc/min to determine how β-glucosidase activity changes in time at these different conditions. This is apparently a novel and simple way of testing for changes in enzyme activity within a protein foam. The activity did not degenerate during 5 min of reflux at pH 5.4 at an airflow rate of 10 cc/ min. It was established that at 10 min of refluxing, the β-glucosidase denatured more as the flow rate increased. At pH 2.4 and a flow rate of 10 cc/min, the activity remained constant for at least 15 min.     

Sonic wave separation of invertase from a dilute solution to generated droplets

It has previously been shown that a droplet fractionation process, simulated by shaking a separatory funnel containing a dilute protein solution, can generate droplets richer in protein than present in the original dilute solution. In this article, we describe an alternative method that can increase the amount of protein transferred to the droplets. The new metho uses ultrasonic waves, enhanced by a bubble gas stream to create the droplets. The amount of protein in these droplets increases by about 50%. In this method, the top layer of the dilute protein solution (of the solution-air interface) becomes enriched in protein when air is bubbled into the solution. This concentrating procedure is called bubble fractionation. Once the protein has passed through the initial buildup, this enriched protein layer is transferred into droplets with the aid of a vacuum above the solution at the same time that ultrasonic waves are introduced. The droplets are then carried over to a condenser and coalesced. We found that this new method provides an easier way to remove the protein-enriched top layer of the dilute solution and generates more droplet within a shorter period than the separatory funnel droplet generation method. The added air creates the bubbles and carries the droplets, and the vacuum helps remove the effluent airstream from the condenser. The maximum partition coefficient, the ratio of the protein concentration in the droplets to that in the residual solution (approx 8.5), occurred at pH 5.0.     

Batch foam fractionation of kudzu (Pueraria lobata) vine retting solution

The aqueous protein solution from kudzu(Pueraria lobata) vine retting broth, without the addition of other surfactants, was foam-fractionated in a vertical tubular column with multiple sampling ports. Time-varying trajectories of the total protein levels were determined to describe the protein behavior at six positions along the 1-m column. The lowest two trajectories of this batch process represented a loss of proteins from the bulk liquid and tended to merge and decay together in time; the other trajectories displayed a gain in proteins in the foam phase. These upper column port protein concentration trajectories generally increased in time up to 45 min, followed by a decrease, reflecting the removal of proteins from the column ports. The foam became dryer as it passed up the column to the top port. The protein concentration was about 5–8×higher in the top port foam than in the initial bulk solution, mainly as a result of liquid drainage from the foam along the column axis. This concentration increase in the collected foam was dependent on the initial pH of the bulk solution. The mol-wt profile of the proteins in the concentrated foam effluent was determined by one-dimensional gel electrophoresis. An analysis of the gel electropherograms indicated that the most abundant proteins could be cellulases and pectinases.     

Mass transfer of volatile organic carbons through aqueous foams

A model is developed to study diffusive mass transfer of hydrocarbon vapor through a flexible foam blanket. The model accounts for the diffusion of hydrocarbon vapor through gas-phase and liquid lamellae, the combined gravity and capillary drainage from the plateau border, the thinning of foam lamellae caused by the forces of capillary suction, London-van der Waals attraction, and electrostatic double-layer repulsion, and foam collapse. Uniform bubble size is assumed, and hence, interbubble gas diffusion arising out of variation in bubble sizes alone is not incorporated into the model. A high-stability aqueous foam formulation that remains stable in the presence of oil (hexane) at foam-oil contact was developed using surfactants, stabilizers, and viscosifiers. Emission of hexane vapor through the foam was measured. The model predicts that the initially taller foam columns collapse faster. Their mass-transfer resistance is higher before the onset of collapse but not very different from that of the shorter foam columns at long times. If the solubility and diffusivity of the hexane gas in the foam liquid are unaffected, the foams with higher viscosities persist longer and provide greater diffusive mass-transfer resistance. Foam bubble size does not significantly impact the mass-transfer resistance of the foam column before the onset of foam collapse. However, the foams with smaller bubbles collapse earlier, and their ability to act as a mass-transfer barrier to the diffusing hydrocarbon vapor diminishes rapidly. The experimental results compared reasonably with the model for varying initial foam heights and bubble sizes.     

Affinity foam fractionation of Trichoderma cellulase

Cellulase could not be selectively collected from fermentation broth by simple foam fractionation, because of the presence of other more surface-active compounds. A new approach of affinity foam fractionation was investigated for improvement. A hardwood hydrolysate (containing cellulose oligomers, substrates to cellulase) and two substrate analogs, i.e., carboxymethyl cellulose (CMC) and xylan hydrolysate, were added before the foaming process. The substrates and substrate analogs were indeed found to bind the cellulase selectively and form more hydrophobic complexes that partition more readily onto bubble surfaces. In this study, the effects of the type and concentration of substrate/analog as well as the presence of cells at different growth stages were examined. The foam fractionation properties evaluated included foaming speed, foam stability, foamate volume, and enrichment of filter paper unit (FPU) and individual cellulase components (i.e., endoglucanases, exoglucanases, and β-glucosidases). Depending on the broth and substrate/analog employed, the foamate FPU could be more than fourfold higher than the starting broth FPU. Addition of substrate/analog also deterred the enrichment of other extracellular proteins, resulting in the desired cellulase purification in the foamate. The value of E/P (enzyme activity-FPU/g/L of proteins) in the foamate reached as high as 18, from a lactose-based fermentation broth with original E/P of 5.6. Among cellulase components, exoglucanases were enriched the most and β-glucosidases the least. The study with CMC of different molecular weights (MW) and degrees of substitution (DS) indicated that the CMC with low DS and high MW performed better in cellulase foam fractionation.     

Foam/Froth Flotation

Abstract

In this second part of the Review on foam and froth flotation, many other aspects of the flotation process are examined, referring mainly to the separation of particulate matter. The effect of particle size is analyzed, stressing the effect of the finer fractions; the role of bubble size is also examined. Among the flotation techniques investigated are those capable of treating fines, such as electrolytic flotation, dissolved-air flotation and column flotation. Operation and design are also discussed. The subject of mineral processing is finally addressed, since it is the main application of the process.     

Long-Term Cultured Hairy Roots of Chicory—A Rich Source of Hydroxycinnamates and 8-Deoxylactucin Glucoside

Foam fractionation is a promising method for separation and concentration of biochemicals. It is simple, easily scalable, inexpensive, and environment friendly. Foam fractionation thus represents an alternative to the traditional methods used for immunoglobulin enrichment. However, little, if any, literature exists documenting the utilization of foam fractionation in the enrichment of immunoglobulins. Milk were utilized as an immunoglobulin source to serve as examples of a real system in this study. The investigation examined the effects of varying five different process parameters: the initial pH value, the initial concentration of immunoglobulin, the nitrogen flow rate, the column height, and the foaming time. Experimental results demonstrated that immunoglobulin could effectively be enriched from milk by foam fractionation. The maximum enrichment ratio with pretreatment (using pH 4.6 precipitation) was 6.30 along with a more than 92 % recovery for IgG and an enrichment ratio of 5.1 with 85 % recovery for IgM.     

Marangoni effects in aqueous polypropylene glycol foams

The foam behavior of three polypropylene glycols covering the molecular weight range between 192 and 725 g/mol has been examined. Static and dynamic surface tension data, as well as bubble size distribution and retention time in the foam, were incorporated into a simple model of foam stability. The latter clearly indicates that surface tension differences between the plateau border and lamellar region adjacent to the bubble surface are the dominant factor in controlling foamability, causing liquid flow in the direction opposite to liquid drainage, a process termed the Marangoni effect.     

Simple 2D-HPLC using a monolithic silica column for peptide separation

Separation of peptides by fast and simple two-dimensional (2D)-HPLC was studied using a monolithic silica column as a second-dimension (2nd-D) column. Every fraction from the first column, 5 cm long (2.1 mm ID) packed with polymer-based cation exchange beads, was subjected to separation in the 2nd-D using an octadecylsilylated (C18) monolithic sillica column (4.6 mm ID, 2.5 cm). A capillary-type monolithic silica C18column (0.1 mm ID, 10 cm) was also employed as a 2nd-D column with split flow/injection. Effluentof the first dimension (1st-D) was directly loaded into an injector loop of 2nd-D HPLC. UV and MS detection were successfully carried out at high linear velocity of mobile phase at 2nd-D using flow splitting for the 4.6 mm ID 2nd-D column, or with directconnection of the capillary column to the MS interface. Two-minute fractionation inthe 1st-D, 118-second loading, and 2-second injection by the 2nd-D injector, allowed one minute for gradient separation in the 2nd-D, resulting in a maximum peak capacity of about 700 within 40 min. The use of a capillary column in solvent consumption and better MS detectability compared to a larger-sized column. This kind of fast and simple 2D-HPLC utilizing monolithic silica columns will be useful for the separation of complex mixtures in a short time.