- Email: [email protected]
Fermentation of Pullulan Using an Oscillatory Baffled Fermenter
0263–8762/05/$30.00+0.00 # 2005 Institution of Chemical Engineers Trans IChemE, Part A, June 2005 Chemical Engineering Research and Design, 83(A6): 640–645
www.icheme.org/journals doi: 10.1025/cherd.04355
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER H. K. GAIDHANI1, B. MC NEIL2 and X. NI1 1
Centre for Oscillatory Baffled Reactor Application (COBRA), School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh, UK 2 Strathclyde Fermentation Centre, Department of Bioscience, The University of Strathclyde, Glasgow, UK
T
he oscillatory baffled reactor offers enhanced and uniform mixing at very low shear rates compared to conventional mixing devices. This is advantageous for biochemical and biomedical applications where shear sensitive cultures are involved. The work in this paper explores the way in which Aureobasidium pullulans, which produces the versatile biopolymer pullulan, behaves in the low and uniform shear environment of an oscillatory baffled fermenter (OBF), and compares its growth rate and pattern to those in traditional fermenters. A series of pullulan fermentation experiments were performed in the OBF with a particular emphasis placed on the influence of aeration on pullulan and biomass production. The results show that at the optimal volumetric airflow rate of 1 vvm (volume of air per volume of liquid per min) the performance of the OBF is significantly better than the traditional stirred tank fermenter. Keywords: oscillatory baffled fermenter; fermentation; pullulan; biopolymer; mass transfer.
INTRODUCTION
industrial and medicinal applications (Childers et al., 1991; Deshpande et al., 1992). A. pullulans has a number of different morphological forms, with filamentous and yeast-like cellular (or unicellular) structures being the two main types, as shown in Figure 1. There have been however contradictory literature on which type is responsible for pullulan synthesis; some reported for the former (Ono et al., 1977; Bermenjo et al., 1981; Simon et al., 1993), while the others, the majority, for the latter (Catley, 1971; Heald and Kristiansen, 1985; McNeil et al., 1989; Madi, 1995). Pullulan is synthesised intracellularly but relatively little is understood about the mechanism of the bioprocess. A. pullulans cells produce pullulan by receiving nutrients and oxygen from the medium and deposit it on the periphery of the cell as shown in Figure 1 and then release it into the medium. Although the factors affecting pullulan intracellular synthesis and subsequent release are still unclear, the mechanism of other microbial systems, e.g., exopolysaccharide synthesis, can generally be used to describe the pullulan synthesis (Sutherland, 1977). As first noted by Catley (1971) and subsequent confirmed in studies by McNeil and Harvey (1993) and Leathers (2003) synthesis of pullulan commences typically in the late exponential phase and continues into the stationary phase. Thus, it is important to minimize the ‘non-productive’ process phase, the culture exponential growth phase, by maximization of growth rate, in order to reach the stage of pullulan synthesis as quickly as possible. This implies a bioreactor with effective mixing and
Considerable interest has been shown in the production and use of microbial biopolymers, as these are generated from renewable natural sources, and are often biodegradable, and generally non-toxic. Pullulan is one of such biopolymers and is a viscous, water-soluble neutral exopolysaccharide secreted by the polymorphic fungus Aureobasidium pullulans (A. pullulans). Pullulan itself consists of either linear chain of glucopyranose units with regular alteration of two a(1 – 4) and one a(1 –6) linkages or a linear polymer of maltotriose units connected by a(1– 6) linkages (Boa and LeDuy, 1984). The regular introduction of a(1 –6) linkages in pullulan interrupts what would otherwise be linear amylose chain. This difference is thought to impart structural flexibility and enhance solubility, resulting in distinct film- and fibre-forming characteristics that allow pullulan to mimic synthetic polymers derived from petroleum (Wallenfels et al., 1961; Bouveng et al., 1962; Sowa et al., 1963). In the form of a thin film, it is biodegradable, transparent, oilresistant and impermeable to oxygen. Pullulan can also be used as a material for coating and packaging, as a sizing agent for paper, as a starch replacer in low-calories food formulations, in cosmetic emulsions and in other Correspondence to: Professor X. Ni, Centre for Oscillatory Baffled Reactor Application (COBRA), School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK. E-mail: [email protected]
640
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER
641
times higher mass transfer coefficients were reported in an oscillatory baffled column in comparison with a bubble column (Hewgill et al., 1993) and 75% higher with a stirred tank involving yeast culture (Ni et al., 1995). These are contributed by the enhanced gas hold-up due to prolonged residence times of bubbles retained in each baffled cavity and more uniform size distributions of bubbles (Oliveira and Ni, 2001, 2004). The encouraging scientific results in applications of OBR in gas – liquid systems prompted this study investigating the fermentation of pullulan using an oscillatory baffled fermenter. MATERIALS AND METHODS Microorganism
Figure 1. A. pullulans cells and the mechanism of pullulan formation.
oxygen transfer. However, both these processes become increasingly difficult as the viscosity of the process fluid increases from mid-exponential phase onwards. Late exponential phase culture broths, e.g., A. pullulans, become highly viscous and exhibit marked non-Newtonian behaviour (pseudo-plasticity), which limits mass (oxygen, nutrient), momentum and heat transfer (Rho et al., 1988; McNeil and Harvey, 1993), leading to stagnant zones and impeller flooding in the standard stirred tank reactor (STR) type used (Rau et al., 1992). It is clear that process control in such poorly mixed systems will be suboptimal, and since mixing in conventional fermenters profoundly influences pullulan yield, productivity and quality (essentially molecular weight distribution) (Bouveng et al., 1962; Catley, 1971; Bermenjo et al., 1981; Boa and LeDuy, 1984), a fermentation system offering good bulk mixing of viscous fluids is desirable. Additionally, since extended (fed) batch processing has been proposed as a route to higher pullulan productivity in the STR (Catley, 1971), the ability to monitor and control such fermentation systems implies the requirement for good bulk mixing. Fermenters utilizing STR technology have been the workhorse of fermentation and indeed bioprocessing industries worldwide. Such fermenters, although generally considered well-characterized types, are acknowledged to have a number of drawbacks and limitations (Popovic and Robinson, 1993; Leib et al., 2001; Rossi, 2001), including poor bulk mixing and the tendency of gas (air) channelling, leading to inadequate gas dispersion for viscous systems. Generally, these drawbacks worsen as the scale of operation increases, and the process fluid viscosity rises. Thus, there is a pressing need to develop alternatives to the conventional STR for processing viscous fermentation products, such as biopolymers, like pullulan. We have now developed a robust fermenter using oscillatory baffled reactor (OBR) technology capable of providing much more uniform global mixing at significantly lower levels of shear strain rate for a similar power input to traditional STRs (Harrison and Mackley, 1992; Ni et al., 2000). The fluids mixing in the OBR is achieved by the generation and cessation, as well as the interaction of eddies, within each baffled cavity, leading to similar orders of magnitude in axial and radial velocities. Six
The A. pullulans (IMI 145194) was supplied as a lyophilized culture. The culture was resuscitated by addition of sterile diluent and then plated out onto potato dextrose agar (Oxoid Ltd, Basingstoke, UK) plates, and incubated for 5 days at 308C. Inocula were produced by adding 0.005 dm3 of sterile medium consisting of sucrose, 30.0 (g l); (NH4)2SO4, 0.6; KH2PO4, 5.0; MgSO47H2O, 0.2; NaCl, 1.0; yeast extract (Oxoid Ltd), 0.4, and by scraping culture from agar plate to generate a suspension. This suspension was aseptically transferred to 0.125 dm3 sterile medium in a 0.5 dm3 Erlenmeyer flask which was subsequently incubated in an orbital shaker at 308C, 200 rpm for 24 h. The inoculum’s volume in fermenter culture was 5% v/v. Experimental Methods The oscillatory baffled fermenter consists of a cylindrical stainless steel column of 100 mm in internal diameter and 420 mm in height, giving a total liquid capacity of 3.3 dm3 with a working volume of 2.5 dm3 (75% of the total capacity). A set of three orifice baffles were used in this study, and these baffles were made of 3 mm thick stainless steel plate and designed to fit closely to the wall of the fermenter. The baffles were equally spaced at 150 mm apart, and supported by two 6.35 mm diameter stainless steel rods. The orifice diameter was 46 mm, creating a baffle free area of 22%. The baffles were oscillated by the means of a motor and a frequency inverter, which allows frequencies of 0.2 to 10 Hz. Oscillation amplitudes of 5 to 30 mm (centre-to-peak) were generated by adjusting the preset distance between the linkage and the fly arm, as shown in Figure 2. The framework supporting the column was made of mild steel, minimizing vibration during operation. The pH in the OBF was maintained at 4.5 throughout the experiments by automatically adding 1 mol l NaOH or 10% (v/v) H2SO4, depending on the pH of the medium. An Anglicon Solo series 2 pH controller was used for controlling and recording the pH data. The temperature of the fermenter was controlled at 288C using a temperatureindicating controller with the sensing element located inside the fermenter. Air was fed at the base of the vessel after passing through a sterile filter. The gas flow rate was measured and controlled via a mass flow meter. The dissolved oxygen (DO) concentration was measured by a Vernier DO probe. For the fermentation experiments
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
642
GAIDHANI et al. Polysaccharide estimation Polysaccharide (i.e., pullulan) estimation was obtained based on dry weight measurements, i.e., two volumes of absolute alcohol were added to cell free filtrate produced by GFC/filtration. This led to precipitation of the polysaccharide, the mass of which was estimated by filtration through GF/C filters as above. The data of both the total biomass and the polysaccharide were determined at the temperature of the fermentation. RESULT AND DISCUSSIONS Optimal Aeration Rate
Figure 2. Set up of the oscillatory baffled fermenter for fermentation of pullulan from A. pullulans.
reported in this paper, the oscillation amplitude was fixed at 20 mm (centre-to-peak), while varying the frequency. The duration of each fermentation experiment is typically 72 h. Comparative data in the STR are derived from Madi’s work (Madi, 1995; Madi et al., 1996) using an identical A. pullulans strain and medium compositions.
Analytical Methods Biomass estimation In order to measure the profile of biomass during the fermentation, 5 ml samples were taken at regular intervals of 12 h via the sampling port. The filamentous mycelial biomass was estimated by dilution of the samples with two volumes of distilled water and then by filtration through pre-weighed nylon mesh filter (45 mm mesh size, Henry Simon Ltd, Stockport, UK). The trapped biomass was washed carefully with two volumes of distilled water, and dried in a thermostat dryer, then weighed. The filtrate, i.e., the yeast-like unicellular cells, from the above process was filtered through pre-weighed GF/C (Whatman Ltd, Maidenhead) filters, and then washed, dried, cooled and weighed as above. The total biomass is thus the sum of the dry weight of both the filamentous type mycelial cells and yeast-like unicellular cells. In addition, the percentage of the unicellular cell over the total biomass in the culture medium can be calculated, and is used in this work to assess the characteristics of the unicellular cell with respective to pullulan synthesis.
Aeration is essential in the fermentation of pullulan, as the aerobic A. pullulans cells need oxygen to survive and grow. In establishing the optimal aeration rate in the OBF, seven fermentation experiments were carried out at a fixed oscillation condition (xo ¼ 20 mm, f ¼ 2 Hz) and for four aeration rates of 0.5, 1, 1.5 and 2 vvm. The term of vvm stands for the volume of air per unit volume of medium per minute. Figure 3 shows the total biomass (g dm23), the polysaccharide (g dm23) and the percentage of the unicellular cell against the aeration rate. The total biomass and the polysaccharide (pullulan) share the same y-axis on the left and the percentage of the unicellular cell has its own on the right. We examine each parameter in turn. The total biomass (diamonds in Figure 3) initially decreased to a minimum of 6.68 g dm23 with the increase of the aeration rate from 0.5 to 1.0 vvm, and the reason for this is not entirely clear. The total biomass then increased with further increases in the aeration rate to 1.5 vvm and 2 vvm respectively. At these higher aeration rates, oxygen supply is improved, which has impacted on the total biomass, as shown in Figure 3. The trend for the polysaccharide (the pullulan synthesis) in Figure 3 is opposite to that for the total biomass, where the concentration of polysaccharide (squares in Figure 3) reached a maximum of 16.17 g dm23 as the aeration rate increased from 0.5 to 1.0 vvm and then decreased for the rest of the aeration rates. Once again, what caused the increase in the polysaccharide is not known. There are however a number of explanations for the decrease,
Figure 3. Effect of aeration rate on the total biomass, polysaccharide and the percentage of yeast-like cells produced in the OBF (xo ¼ 20 mm, f ¼ 2 Hz) for fermentation of pullulan by A. pullulans.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER for example, (1) the cells consumed more sucrose in the medium at the increased airflow rates via increased respiration, which would lead to a reduction in the sucrose available for pullulan synthesis, hence the decrease in the pullulan with the increase of the aeration rates (Sutherland, 1977, 1982; Lee et al., 2001); (2) oxygen limitation is a stimulatory factor for polysaccharide production (Rau et al., 1992), this may explain why pullulan is the highest at 1 vvm in comparison to that at 1.5 and 2 vvm. From Figure 3, it can also be observed that pullulan production and the percentage of yeast-like cells follow the same pattern, with the highest polysaccharide yield coinciding with the highest percentage of yeast-like cells. This suggests that the yeast-like cell favours the pullulan production in the OBF. These results are in line with some reports in the literature (Wecker and Onken, 1991; Gibbs and Seviour, 1996), where the highest yield of pullulan at a combination of a moderate shear rate and moderate dissolved oxygen condition is associated with a substantially unicellular culture morphology. The results so far indicate that 1 vvm is the optimal aeration rate for the production of pullulan in the OBF, and from here onwards, the optimal aeration rate of 1 vvm has been used in the rest of fermentation experiments. Growth Profile Production of pullulan in traditional STR fermenters generally goes through three phases: exponential growth, stationary or pullulan production phase and finally the decline phase (Sutherland, 1979). These phases are clearly presented in the profile of the dissolved oxygen, in terms of the percentage of saturation, during the fermentation of pullulan in the OBF, and Figure 4 shows such a profile measured at the optimal airflow rate of 1 vvm and a fixed oscillation condition (xo ¼ 20 mm, f ¼ 2 Hz). Also in Figure 4 there are marked phase boundaries for the three phases. We see an initial increase in the dissolved oxygen up to 8 h due to the low cell concentration, and then a sharp decrease from there onwards in the exponential growth phase, because of the fast oxygen consumption by the cells. The dissolved oxygen then experienced a period of rise in Phase B as the pullulan synthesis increased.
Figure 4. Dissolved oxygen profile in the OBF (air flow rate ¼ 1 vvm, xo ¼ 20 mm, f ¼ 2 Hz) for fermentation of pullulan by A. pullulans.
643
Finally the dissolved oxygen leveled-off towards the end of the fermentation, where the demands for oxygen fell as cellular respiration fell. Corresponding to the same phase descriptions in Figure 4, the time profiles of total biomass and polysaccharide in the OBF are shown in Figure 5 at the same operating conditions as above. Note that the two parameters have the same unit of grams per litre, hence share the same y-axis. It can be seen that the total biomass (diamonds in Figure 5) and the pullulan (triangles) in the medium initially followed each other closely in the first stage, i.e., the exponential growth phase (Phase A in Figure 5), the growth of pullulan in the medium continued to increase, while the total biomass levelled-off in the stationary phase. After about 54 h the carbon source in the medium has almost been consumed (Gaidhani, 2004), the growth rate of pullulan reached a plateau, while the biomass decreased a bit, allowing the dissolved oxygen concentration to stabilise at 60 –70% saturation (Figure 4). The highest concentration of polysaccharide obtained in the culture was 16.17 g dm23. The growth pattern of the culture suggests that pullulan elaboration commenced when cells were in the exponential growth phase and continued throughout the entire stationary phase. This indicates that although the cells reached the stationary phase of growth, pullulan production did not cease after cessation of growth (Catley, 1971).
Comparison with STR Figure 6 compares the growth data of pullulan in the OBF with that in two stirred tank fermenters. The results of the latter were extracted from the works of Madi (Madi, 1995), who used the same strain of A. pullulans and identical medium compositions. The details of the physical and operational parameters used in both types of fermenters are given in Table 1. It can be seen from Figure 6a that the OBF gives a much higher rate of growth as well as higher yield of polysaccharide as compared to the stirred tank fermenters, for example, 11.3 and 12.1 g dm23 pullulan was produced in the stirred
Figure 5. Time profiles of total biomass and polysaccharide in OBF (air flow rate ¼ 1 vvm, xo ¼ 20 mm, f ¼ 2 Hz) for fermentation of pullulan by A. pullulans.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
644
GAIDHANI et al.
Figure 6. Comparison of polysaccharide and total biomass produced in the OBF with respect to STR fermenters (operational conditions are listed in Table 1).
tank fermenters 1 and 2 after 96 and 144 h of fermentation respectively, and this was achieved in the OBF at about 37– 39 h [Figure 6(a)]. By examining the time when the stationary phase of growth was reached in both types of fermenters, the results show that it took about 52 h in the OBF compared to about 96 and 128 h in the STF, respectively. As the cell reaches the stationary phase early, more sucrose and more carbon source are available for pullulan synthesis, and in turn the higher pullulan production (Sutherland, 1979). At the optimum aeration rate in the OBF the fermentation process, though broadly similar to that in the STR studies, has one crucial difference: growth and pullulan synthesis commence at very similar times. Since most studies indicate that these two process (growth and pullulan synthesis) are alternative ‘fates’ for the C source (Bermenjo et al., 1981; Boa and LeDuy, 1984), the earlier commencement of synthesis of pullulan in the OBF may also contribute to (1) the increased pullulan synthesis and (2) the decreased biomass. For the production of the total biomass in the OBF [triangles in Figure 6(b)], it took only 36 h to reach to the stationary phase comparing to 72 h and 120 h in STF-1 and STF-2 respectively [Figure 6(b)]. The shorter the process time, the lower the running costs in fermenter operations. The results in Figure 6 indicate that the
production of pullulan in OBF achieves higher pullulan concentrations, much more rapidly than a comparable STR process, and that the biomass is predominately in the unicellular form. This latter aspect is particularly important for process related reasons: for most purposes cells have to be separated from the pullulan during recovery of pure pullulan, yeast cells are unpigmented, whereas mycelial cells are often strongly pigmented with greenish black melanins (Leathers, 2003) increasing the complexity and cost of pullulan synthesis. CONCLUSION The work in this paper has examined the way in which A. pullulans, which produces the versatile biopolymer of pullulan, behaved in the low and even shear environment of an oscillatory baffled fermenter. A series of pullulan fermentation experiments were performed and 1 vvm was identified as the optimal aeration rate for production of pullulan in the OBF. Using the optimal aeration rate, the growth rate and growth pattern of the total biomass and polysaccharide in the OBF are compared to those in traditional fermenters, and the results show that the performance of the OBF is significantly better than the traditional stirred tank fermenter. REFERENCES
Table 1. Operational parameters for OBF and STF. Parameters
OBF
STF-1 (Madi, 1995)
SFT-2 (Madi, 1995)
Operating volume (litre) Internal diameter of fermenter (m) Oscillation frequency (Hz) and amplitude (mm) or rotating speed of impeller (rpm) pH of medium Initial sucrose concentration (kg m23) Volumetric air flow rate (vvm) Temperature (8C)
2.5 0.1
2 0.1
10 0.195
2 Hz 20 mm
300 rpm
300 rpm
4.5 30 1 28
4.5 30 1 28
4.5 30 1 28
Bermenjo, J.M., Dominguez, J.B., Goni, F.M. and Uruburu, F., 1981, Influence of pH on the transition from yeast-like cells to chlamydospores in A. pullulans, Antonie van Leeuwenhoek, 47: 385–392. Boa, J.M. and LeDuy, A., 1984, Peat hydrolysate medium optimisation of pullulan production, Applied Enviornmental Microbiology, 48: 26–30. Bouveng, H.O., Kiessling, H., Lindberg, B. and McKay, J., 1962, Polysaccharides elaborated by pullularia pullunanas. Part 1: the neutral glucans sunthesised from sucrose solutions, Acta Chemistry Scandinavian, 16: 615–622. Catley, B.J., 1971, Role of pH and nitrogen limitation in the elaboration of the extracellular polysaccharide pullulan by pullularia pullulans, Applied Microbiology, 22: 641–649. Childers, R.F., Oren, P.L. and Seidle, W.M.K., 1991, Biopolymers (ChemTec Publisher, USA). Deshpande, M.S., Rale, V.B. and Lynch, J.M., 1992, Aureobasidium pullulans in applied microbiology: a status report, Enzyme and Microbial Technology, 14: 514–527.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER Gaidhani, H.K., 2004, An experimental investigation of pullulan fermentation in a batch oscillatory baffled fermenter, PhD thesis, Heriot-Watt University, Edinburgh, Scotland. Gibbs, P.A. and Seviour, R.J., 1996, Does the agitation rate and/or oxygen saturation influence exopolysaccharide production by Aureobasidium pullulans in batch culture?, Applied Microbiol Biotechnology, 46: 503– 510. Harrison, S.T.L. and Mackley, M.R., 1992, A pulsatile flow bioreactor, Chemical Engineering Science, 47(2): 490 –493. Heald, P.J. and Kristiansen, B., 1985, Synthesis of polysaccharide by yeast-like forms of Aureobasidium pullulans, Biotechnology and Bioengineering, 27: 1516–1519. Hewgill, M.R., Mackley, M.R., Pandit, A.B. and Pannu, S.S., 1993, Enhancement of gas–liquid mass transfer using oscillatory flow in a baffled tube, Chemical Engineering Science, 48(4): 799– 809. Leathers, T.D., 2003, Biotechnological production and applications of pullulan, Applied Microbiol Biotechnology, 62: 468– 473. Lee, J.H., Kim, J.H., Zhu, I.H., Zhan, X.B., Lee, J.W., Shin, D.H. and Kim, S.K., 2001, Optimisation of conditions for the production of pullulan and high molecular weight pullulan by Aureobasidium pullulans, Biotechnology Letters, 23: 817–820. Leib, T.M., Pereira, C.J. and Villadsen, J., 2001, Bioreactors: a chemical engineering perspective, Chemical Engineering Science, 56: 5485– 5497. Madi, N.S., 1995, Exoplysaccharide elaboration by the polymorphic fungus Aureobasidium pullulans, PhD thesis, Strathclyde University, Glasgow, Scotland. Madi, N.S., McNeil, B. and Harvey, L.M., 1996, Influence of culture pH and aeration on ethanol production and pullulan molecular weight in the fungus Aureobasidium pullulans, Journal of Chemical Technology and Biotechnology, 66: 343–350. McNeil, B. and Harvey, L.M., 1993, Viscous fermentation products, Critical Reviews in Biotechnology, 13: 275 –304. McNeil, B., Kristiansen, B. and Seviour, R.J., 1989, Polysaacharide production and morphology of Aureobasidium pullulans in continuous culture, Biotechnology and Bioengineering, 33: 1210–1212. Ni, X., Cosgrove, J.A., Arnott, A.D., Greated, C.A. and Cumming, R.H., 2000, On the measurement of strain rate in an oscillatory baffled column using particle image velocimetry, Chemical Engineering Science, 55: 3195–3208. Ni, X., Gao, S., Cumming, R.H. and Pritchard, D.W., 1995, A comparative study of mass transfer in yeast for a batch pulsed baffled bioreactor and a stirred tank fermenter, Chemical Engineering Science, 50(13): 2127–2136.
645
Oliveira, M.S.N. and Ni, X., 2001, Gas hold-up and bubble diameters in a gassed oscillatory baffled column, Chemical Engineering Science, 56(21–22): 6143–6148. Oliveira, M.S.N. and Ni, X., 2004, Effect of hydrodynamics on mass transfer in a gas–liquid oscillatory baffled column, The Chemical Engineering Journal, 99: 59–68. Ono, K., Yauda, N. and Ueda, S., 1977, The effect of pH on pullulan elaboration by Aureobasidium pullulans, Agriculture Biological Chemistry, 41: 2113–2118. Popovic, M.K. and Robinson, C.W., 1993, Mixing characteristics of external-loop airlifts: non-Newtonian systems, Chemical Engineering Science, 48: 1405– 1413. Rau, U., Gura, E., Olszewski, E. and Wanger, F., 1992, Enhanced glucan formation of filamentous fungi by effective mixing, oxygen limitation and fed-batch processing, Industrial Microbiology, 9: 19–26. Rho, D., Mulchandani, A., Luong, J. and LeDuy, A., 1988, Oxygen requirements in pullulan fermentation, Applied Microbiol Biotechnology, 28: 361 –366. Rossi, G., 2001, The design of bioreactors, Hydrometallurgy, 59: 217–231. Simon, L., Caye-Vague, C. and Bouchonneau, M., 1993, Relation between pullulans production, morphological state and growth conditions in Aureobasidium pullulans, Journal of General Microbial, 139: 979– 985. Sowa, W., Blackwood, A.C. and Adams, G.A., 1963, Neutral extracellular glucan of pullularia pullulans (de Bery) berkhout, Canadian Journal of Chemistry, 41: 2314– 2319. Sutherland, I.W., 1977, Surface Carbohydrate of Prokaryotic Cell (Academic Press, London, UK). Sutherland, I.W., 1979, Microbial Polysaccharides and Polysaccharases (Academic Press, London, UK). Sutherland, I.W., 1982, Biosynthesis of microbial exopolysaccharides, Advanced Microbiol Physiology, 23: 79–150. Wallenfels, K., Bender, H., Keilich, G. and Bechtler, G., 1961, On pullulan, the glucan of the slime coat of pullularia pullulans, Angew Chemistry, 73: 245– 246. Wecker, A. and Onken, U., 1991, Influence of dissolved oxygen concentration and shear rate on the production of pullulan by Aureobasidium pullulans, Biotechnology Letters, 13: 155–160. This paper was presented at the 7th World Congress of Chemical Engineering held in Glasgow, UK, 10–14 July 2005. The manuscript was received 15 December 2004 and accepted for publication after revision 4 April 2005.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
www.icheme.org/journals doi: 10.1025/cherd.04355
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER H. K. GAIDHANI1, B. MC NEIL2 and X. NI1 1
Centre for Oscillatory Baffled Reactor Application (COBRA), School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh, UK 2 Strathclyde Fermentation Centre, Department of Bioscience, The University of Strathclyde, Glasgow, UK
T
he oscillatory baffled reactor offers enhanced and uniform mixing at very low shear rates compared to conventional mixing devices. This is advantageous for biochemical and biomedical applications where shear sensitive cultures are involved. The work in this paper explores the way in which Aureobasidium pullulans, which produces the versatile biopolymer pullulan, behaves in the low and uniform shear environment of an oscillatory baffled fermenter (OBF), and compares its growth rate and pattern to those in traditional fermenters. A series of pullulan fermentation experiments were performed in the OBF with a particular emphasis placed on the influence of aeration on pullulan and biomass production. The results show that at the optimal volumetric airflow rate of 1 vvm (volume of air per volume of liquid per min) the performance of the OBF is significantly better than the traditional stirred tank fermenter. Keywords: oscillatory baffled fermenter; fermentation; pullulan; biopolymer; mass transfer.
INTRODUCTION
industrial and medicinal applications (Childers et al., 1991; Deshpande et al., 1992). A. pullulans has a number of different morphological forms, with filamentous and yeast-like cellular (or unicellular) structures being the two main types, as shown in Figure 1. There have been however contradictory literature on which type is responsible for pullulan synthesis; some reported for the former (Ono et al., 1977; Bermenjo et al., 1981; Simon et al., 1993), while the others, the majority, for the latter (Catley, 1971; Heald and Kristiansen, 1985; McNeil et al., 1989; Madi, 1995). Pullulan is synthesised intracellularly but relatively little is understood about the mechanism of the bioprocess. A. pullulans cells produce pullulan by receiving nutrients and oxygen from the medium and deposit it on the periphery of the cell as shown in Figure 1 and then release it into the medium. Although the factors affecting pullulan intracellular synthesis and subsequent release are still unclear, the mechanism of other microbial systems, e.g., exopolysaccharide synthesis, can generally be used to describe the pullulan synthesis (Sutherland, 1977). As first noted by Catley (1971) and subsequent confirmed in studies by McNeil and Harvey (1993) and Leathers (2003) synthesis of pullulan commences typically in the late exponential phase and continues into the stationary phase. Thus, it is important to minimize the ‘non-productive’ process phase, the culture exponential growth phase, by maximization of growth rate, in order to reach the stage of pullulan synthesis as quickly as possible. This implies a bioreactor with effective mixing and
Considerable interest has been shown in the production and use of microbial biopolymers, as these are generated from renewable natural sources, and are often biodegradable, and generally non-toxic. Pullulan is one of such biopolymers and is a viscous, water-soluble neutral exopolysaccharide secreted by the polymorphic fungus Aureobasidium pullulans (A. pullulans). Pullulan itself consists of either linear chain of glucopyranose units with regular alteration of two a(1 – 4) and one a(1 –6) linkages or a linear polymer of maltotriose units connected by a(1– 6) linkages (Boa and LeDuy, 1984). The regular introduction of a(1 –6) linkages in pullulan interrupts what would otherwise be linear amylose chain. This difference is thought to impart structural flexibility and enhance solubility, resulting in distinct film- and fibre-forming characteristics that allow pullulan to mimic synthetic polymers derived from petroleum (Wallenfels et al., 1961; Bouveng et al., 1962; Sowa et al., 1963). In the form of a thin film, it is biodegradable, transparent, oilresistant and impermeable to oxygen. Pullulan can also be used as a material for coating and packaging, as a sizing agent for paper, as a starch replacer in low-calories food formulations, in cosmetic emulsions and in other Correspondence to: Professor X. Ni, Centre for Oscillatory Baffled Reactor Application (COBRA), School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK. E-mail: [email protected]
640
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER
641
times higher mass transfer coefficients were reported in an oscillatory baffled column in comparison with a bubble column (Hewgill et al., 1993) and 75% higher with a stirred tank involving yeast culture (Ni et al., 1995). These are contributed by the enhanced gas hold-up due to prolonged residence times of bubbles retained in each baffled cavity and more uniform size distributions of bubbles (Oliveira and Ni, 2001, 2004). The encouraging scientific results in applications of OBR in gas – liquid systems prompted this study investigating the fermentation of pullulan using an oscillatory baffled fermenter. MATERIALS AND METHODS Microorganism
Figure 1. A. pullulans cells and the mechanism of pullulan formation.
oxygen transfer. However, both these processes become increasingly difficult as the viscosity of the process fluid increases from mid-exponential phase onwards. Late exponential phase culture broths, e.g., A. pullulans, become highly viscous and exhibit marked non-Newtonian behaviour (pseudo-plasticity), which limits mass (oxygen, nutrient), momentum and heat transfer (Rho et al., 1988; McNeil and Harvey, 1993), leading to stagnant zones and impeller flooding in the standard stirred tank reactor (STR) type used (Rau et al., 1992). It is clear that process control in such poorly mixed systems will be suboptimal, and since mixing in conventional fermenters profoundly influences pullulan yield, productivity and quality (essentially molecular weight distribution) (Bouveng et al., 1962; Catley, 1971; Bermenjo et al., 1981; Boa and LeDuy, 1984), a fermentation system offering good bulk mixing of viscous fluids is desirable. Additionally, since extended (fed) batch processing has been proposed as a route to higher pullulan productivity in the STR (Catley, 1971), the ability to monitor and control such fermentation systems implies the requirement for good bulk mixing. Fermenters utilizing STR technology have been the workhorse of fermentation and indeed bioprocessing industries worldwide. Such fermenters, although generally considered well-characterized types, are acknowledged to have a number of drawbacks and limitations (Popovic and Robinson, 1993; Leib et al., 2001; Rossi, 2001), including poor bulk mixing and the tendency of gas (air) channelling, leading to inadequate gas dispersion for viscous systems. Generally, these drawbacks worsen as the scale of operation increases, and the process fluid viscosity rises. Thus, there is a pressing need to develop alternatives to the conventional STR for processing viscous fermentation products, such as biopolymers, like pullulan. We have now developed a robust fermenter using oscillatory baffled reactor (OBR) technology capable of providing much more uniform global mixing at significantly lower levels of shear strain rate for a similar power input to traditional STRs (Harrison and Mackley, 1992; Ni et al., 2000). The fluids mixing in the OBR is achieved by the generation and cessation, as well as the interaction of eddies, within each baffled cavity, leading to similar orders of magnitude in axial and radial velocities. Six
The A. pullulans (IMI 145194) was supplied as a lyophilized culture. The culture was resuscitated by addition of sterile diluent and then plated out onto potato dextrose agar (Oxoid Ltd, Basingstoke, UK) plates, and incubated for 5 days at 308C. Inocula were produced by adding 0.005 dm3 of sterile medium consisting of sucrose, 30.0 (g l); (NH4)2SO4, 0.6; KH2PO4, 5.0; MgSO47H2O, 0.2; NaCl, 1.0; yeast extract (Oxoid Ltd), 0.4, and by scraping culture from agar plate to generate a suspension. This suspension was aseptically transferred to 0.125 dm3 sterile medium in a 0.5 dm3 Erlenmeyer flask which was subsequently incubated in an orbital shaker at 308C, 200 rpm for 24 h. The inoculum’s volume in fermenter culture was 5% v/v. Experimental Methods The oscillatory baffled fermenter consists of a cylindrical stainless steel column of 100 mm in internal diameter and 420 mm in height, giving a total liquid capacity of 3.3 dm3 with a working volume of 2.5 dm3 (75% of the total capacity). A set of three orifice baffles were used in this study, and these baffles were made of 3 mm thick stainless steel plate and designed to fit closely to the wall of the fermenter. The baffles were equally spaced at 150 mm apart, and supported by two 6.35 mm diameter stainless steel rods. The orifice diameter was 46 mm, creating a baffle free area of 22%. The baffles were oscillated by the means of a motor and a frequency inverter, which allows frequencies of 0.2 to 10 Hz. Oscillation amplitudes of 5 to 30 mm (centre-to-peak) were generated by adjusting the preset distance between the linkage and the fly arm, as shown in Figure 2. The framework supporting the column was made of mild steel, minimizing vibration during operation. The pH in the OBF was maintained at 4.5 throughout the experiments by automatically adding 1 mol l NaOH or 10% (v/v) H2SO4, depending on the pH of the medium. An Anglicon Solo series 2 pH controller was used for controlling and recording the pH data. The temperature of the fermenter was controlled at 288C using a temperatureindicating controller with the sensing element located inside the fermenter. Air was fed at the base of the vessel after passing through a sterile filter. The gas flow rate was measured and controlled via a mass flow meter. The dissolved oxygen (DO) concentration was measured by a Vernier DO probe. For the fermentation experiments
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
642
GAIDHANI et al. Polysaccharide estimation Polysaccharide (i.e., pullulan) estimation was obtained based on dry weight measurements, i.e., two volumes of absolute alcohol were added to cell free filtrate produced by GFC/filtration. This led to precipitation of the polysaccharide, the mass of which was estimated by filtration through GF/C filters as above. The data of both the total biomass and the polysaccharide were determined at the temperature of the fermentation. RESULT AND DISCUSSIONS Optimal Aeration Rate
Figure 2. Set up of the oscillatory baffled fermenter for fermentation of pullulan from A. pullulans.
reported in this paper, the oscillation amplitude was fixed at 20 mm (centre-to-peak), while varying the frequency. The duration of each fermentation experiment is typically 72 h. Comparative data in the STR are derived from Madi’s work (Madi, 1995; Madi et al., 1996) using an identical A. pullulans strain and medium compositions.
Analytical Methods Biomass estimation In order to measure the profile of biomass during the fermentation, 5 ml samples were taken at regular intervals of 12 h via the sampling port. The filamentous mycelial biomass was estimated by dilution of the samples with two volumes of distilled water and then by filtration through pre-weighed nylon mesh filter (45 mm mesh size, Henry Simon Ltd, Stockport, UK). The trapped biomass was washed carefully with two volumes of distilled water, and dried in a thermostat dryer, then weighed. The filtrate, i.e., the yeast-like unicellular cells, from the above process was filtered through pre-weighed GF/C (Whatman Ltd, Maidenhead) filters, and then washed, dried, cooled and weighed as above. The total biomass is thus the sum of the dry weight of both the filamentous type mycelial cells and yeast-like unicellular cells. In addition, the percentage of the unicellular cell over the total biomass in the culture medium can be calculated, and is used in this work to assess the characteristics of the unicellular cell with respective to pullulan synthesis.
Aeration is essential in the fermentation of pullulan, as the aerobic A. pullulans cells need oxygen to survive and grow. In establishing the optimal aeration rate in the OBF, seven fermentation experiments were carried out at a fixed oscillation condition (xo ¼ 20 mm, f ¼ 2 Hz) and for four aeration rates of 0.5, 1, 1.5 and 2 vvm. The term of vvm stands for the volume of air per unit volume of medium per minute. Figure 3 shows the total biomass (g dm23), the polysaccharide (g dm23) and the percentage of the unicellular cell against the aeration rate. The total biomass and the polysaccharide (pullulan) share the same y-axis on the left and the percentage of the unicellular cell has its own on the right. We examine each parameter in turn. The total biomass (diamonds in Figure 3) initially decreased to a minimum of 6.68 g dm23 with the increase of the aeration rate from 0.5 to 1.0 vvm, and the reason for this is not entirely clear. The total biomass then increased with further increases in the aeration rate to 1.5 vvm and 2 vvm respectively. At these higher aeration rates, oxygen supply is improved, which has impacted on the total biomass, as shown in Figure 3. The trend for the polysaccharide (the pullulan synthesis) in Figure 3 is opposite to that for the total biomass, where the concentration of polysaccharide (squares in Figure 3) reached a maximum of 16.17 g dm23 as the aeration rate increased from 0.5 to 1.0 vvm and then decreased for the rest of the aeration rates. Once again, what caused the increase in the polysaccharide is not known. There are however a number of explanations for the decrease,
Figure 3. Effect of aeration rate on the total biomass, polysaccharide and the percentage of yeast-like cells produced in the OBF (xo ¼ 20 mm, f ¼ 2 Hz) for fermentation of pullulan by A. pullulans.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER for example, (1) the cells consumed more sucrose in the medium at the increased airflow rates via increased respiration, which would lead to a reduction in the sucrose available for pullulan synthesis, hence the decrease in the pullulan with the increase of the aeration rates (Sutherland, 1977, 1982; Lee et al., 2001); (2) oxygen limitation is a stimulatory factor for polysaccharide production (Rau et al., 1992), this may explain why pullulan is the highest at 1 vvm in comparison to that at 1.5 and 2 vvm. From Figure 3, it can also be observed that pullulan production and the percentage of yeast-like cells follow the same pattern, with the highest polysaccharide yield coinciding with the highest percentage of yeast-like cells. This suggests that the yeast-like cell favours the pullulan production in the OBF. These results are in line with some reports in the literature (Wecker and Onken, 1991; Gibbs and Seviour, 1996), where the highest yield of pullulan at a combination of a moderate shear rate and moderate dissolved oxygen condition is associated with a substantially unicellular culture morphology. The results so far indicate that 1 vvm is the optimal aeration rate for the production of pullulan in the OBF, and from here onwards, the optimal aeration rate of 1 vvm has been used in the rest of fermentation experiments. Growth Profile Production of pullulan in traditional STR fermenters generally goes through three phases: exponential growth, stationary or pullulan production phase and finally the decline phase (Sutherland, 1979). These phases are clearly presented in the profile of the dissolved oxygen, in terms of the percentage of saturation, during the fermentation of pullulan in the OBF, and Figure 4 shows such a profile measured at the optimal airflow rate of 1 vvm and a fixed oscillation condition (xo ¼ 20 mm, f ¼ 2 Hz). Also in Figure 4 there are marked phase boundaries for the three phases. We see an initial increase in the dissolved oxygen up to 8 h due to the low cell concentration, and then a sharp decrease from there onwards in the exponential growth phase, because of the fast oxygen consumption by the cells. The dissolved oxygen then experienced a period of rise in Phase B as the pullulan synthesis increased.
Figure 4. Dissolved oxygen profile in the OBF (air flow rate ¼ 1 vvm, xo ¼ 20 mm, f ¼ 2 Hz) for fermentation of pullulan by A. pullulans.
643
Finally the dissolved oxygen leveled-off towards the end of the fermentation, where the demands for oxygen fell as cellular respiration fell. Corresponding to the same phase descriptions in Figure 4, the time profiles of total biomass and polysaccharide in the OBF are shown in Figure 5 at the same operating conditions as above. Note that the two parameters have the same unit of grams per litre, hence share the same y-axis. It can be seen that the total biomass (diamonds in Figure 5) and the pullulan (triangles) in the medium initially followed each other closely in the first stage, i.e., the exponential growth phase (Phase A in Figure 5), the growth of pullulan in the medium continued to increase, while the total biomass levelled-off in the stationary phase. After about 54 h the carbon source in the medium has almost been consumed (Gaidhani, 2004), the growth rate of pullulan reached a plateau, while the biomass decreased a bit, allowing the dissolved oxygen concentration to stabilise at 60 –70% saturation (Figure 4). The highest concentration of polysaccharide obtained in the culture was 16.17 g dm23. The growth pattern of the culture suggests that pullulan elaboration commenced when cells were in the exponential growth phase and continued throughout the entire stationary phase. This indicates that although the cells reached the stationary phase of growth, pullulan production did not cease after cessation of growth (Catley, 1971).
Comparison with STR Figure 6 compares the growth data of pullulan in the OBF with that in two stirred tank fermenters. The results of the latter were extracted from the works of Madi (Madi, 1995), who used the same strain of A. pullulans and identical medium compositions. The details of the physical and operational parameters used in both types of fermenters are given in Table 1. It can be seen from Figure 6a that the OBF gives a much higher rate of growth as well as higher yield of polysaccharide as compared to the stirred tank fermenters, for example, 11.3 and 12.1 g dm23 pullulan was produced in the stirred
Figure 5. Time profiles of total biomass and polysaccharide in OBF (air flow rate ¼ 1 vvm, xo ¼ 20 mm, f ¼ 2 Hz) for fermentation of pullulan by A. pullulans.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
644
GAIDHANI et al.
Figure 6. Comparison of polysaccharide and total biomass produced in the OBF with respect to STR fermenters (operational conditions are listed in Table 1).
tank fermenters 1 and 2 after 96 and 144 h of fermentation respectively, and this was achieved in the OBF at about 37– 39 h [Figure 6(a)]. By examining the time when the stationary phase of growth was reached in both types of fermenters, the results show that it took about 52 h in the OBF compared to about 96 and 128 h in the STF, respectively. As the cell reaches the stationary phase early, more sucrose and more carbon source are available for pullulan synthesis, and in turn the higher pullulan production (Sutherland, 1979). At the optimum aeration rate in the OBF the fermentation process, though broadly similar to that in the STR studies, has one crucial difference: growth and pullulan synthesis commence at very similar times. Since most studies indicate that these two process (growth and pullulan synthesis) are alternative ‘fates’ for the C source (Bermenjo et al., 1981; Boa and LeDuy, 1984), the earlier commencement of synthesis of pullulan in the OBF may also contribute to (1) the increased pullulan synthesis and (2) the decreased biomass. For the production of the total biomass in the OBF [triangles in Figure 6(b)], it took only 36 h to reach to the stationary phase comparing to 72 h and 120 h in STF-1 and STF-2 respectively [Figure 6(b)]. The shorter the process time, the lower the running costs in fermenter operations. The results in Figure 6 indicate that the
production of pullulan in OBF achieves higher pullulan concentrations, much more rapidly than a comparable STR process, and that the biomass is predominately in the unicellular form. This latter aspect is particularly important for process related reasons: for most purposes cells have to be separated from the pullulan during recovery of pure pullulan, yeast cells are unpigmented, whereas mycelial cells are often strongly pigmented with greenish black melanins (Leathers, 2003) increasing the complexity and cost of pullulan synthesis. CONCLUSION The work in this paper has examined the way in which A. pullulans, which produces the versatile biopolymer of pullulan, behaved in the low and even shear environment of an oscillatory baffled fermenter. A series of pullulan fermentation experiments were performed and 1 vvm was identified as the optimal aeration rate for production of pullulan in the OBF. Using the optimal aeration rate, the growth rate and growth pattern of the total biomass and polysaccharide in the OBF are compared to those in traditional fermenters, and the results show that the performance of the OBF is significantly better than the traditional stirred tank fermenter. REFERENCES
Table 1. Operational parameters for OBF and STF. Parameters
OBF
STF-1 (Madi, 1995)
SFT-2 (Madi, 1995)
Operating volume (litre) Internal diameter of fermenter (m) Oscillation frequency (Hz) and amplitude (mm) or rotating speed of impeller (rpm) pH of medium Initial sucrose concentration (kg m23) Volumetric air flow rate (vvm) Temperature (8C)
2.5 0.1
2 0.1
10 0.195
2 Hz 20 mm
300 rpm
300 rpm
4.5 30 1 28
4.5 30 1 28
4.5 30 1 28
Bermenjo, J.M., Dominguez, J.B., Goni, F.M. and Uruburu, F., 1981, Influence of pH on the transition from yeast-like cells to chlamydospores in A. pullulans, Antonie van Leeuwenhoek, 47: 385–392. Boa, J.M. and LeDuy, A., 1984, Peat hydrolysate medium optimisation of pullulan production, Applied Enviornmental Microbiology, 48: 26–30. Bouveng, H.O., Kiessling, H., Lindberg, B. and McKay, J., 1962, Polysaccharides elaborated by pullularia pullunanas. Part 1: the neutral glucans sunthesised from sucrose solutions, Acta Chemistry Scandinavian, 16: 615–622. Catley, B.J., 1971, Role of pH and nitrogen limitation in the elaboration of the extracellular polysaccharide pullulan by pullularia pullulans, Applied Microbiology, 22: 641–649. Childers, R.F., Oren, P.L. and Seidle, W.M.K., 1991, Biopolymers (ChemTec Publisher, USA). Deshpande, M.S., Rale, V.B. and Lynch, J.M., 1992, Aureobasidium pullulans in applied microbiology: a status report, Enzyme and Microbial Technology, 14: 514–527.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645
FERMENTATION OF PULLULAN USING AN OSCILLATORY BAFFLED FERMENTER Gaidhani, H.K., 2004, An experimental investigation of pullulan fermentation in a batch oscillatory baffled fermenter, PhD thesis, Heriot-Watt University, Edinburgh, Scotland. Gibbs, P.A. and Seviour, R.J., 1996, Does the agitation rate and/or oxygen saturation influence exopolysaccharide production by Aureobasidium pullulans in batch culture?, Applied Microbiol Biotechnology, 46: 503– 510. Harrison, S.T.L. and Mackley, M.R., 1992, A pulsatile flow bioreactor, Chemical Engineering Science, 47(2): 490 –493. Heald, P.J. and Kristiansen, B., 1985, Synthesis of polysaccharide by yeast-like forms of Aureobasidium pullulans, Biotechnology and Bioengineering, 27: 1516–1519. Hewgill, M.R., Mackley, M.R., Pandit, A.B. and Pannu, S.S., 1993, Enhancement of gas–liquid mass transfer using oscillatory flow in a baffled tube, Chemical Engineering Science, 48(4): 799– 809. Leathers, T.D., 2003, Biotechnological production and applications of pullulan, Applied Microbiol Biotechnology, 62: 468– 473. Lee, J.H., Kim, J.H., Zhu, I.H., Zhan, X.B., Lee, J.W., Shin, D.H. and Kim, S.K., 2001, Optimisation of conditions for the production of pullulan and high molecular weight pullulan by Aureobasidium pullulans, Biotechnology Letters, 23: 817–820. Leib, T.M., Pereira, C.J. and Villadsen, J., 2001, Bioreactors: a chemical engineering perspective, Chemical Engineering Science, 56: 5485– 5497. Madi, N.S., 1995, Exoplysaccharide elaboration by the polymorphic fungus Aureobasidium pullulans, PhD thesis, Strathclyde University, Glasgow, Scotland. Madi, N.S., McNeil, B. and Harvey, L.M., 1996, Influence of culture pH and aeration on ethanol production and pullulan molecular weight in the fungus Aureobasidium pullulans, Journal of Chemical Technology and Biotechnology, 66: 343–350. McNeil, B. and Harvey, L.M., 1993, Viscous fermentation products, Critical Reviews in Biotechnology, 13: 275 –304. McNeil, B., Kristiansen, B. and Seviour, R.J., 1989, Polysaacharide production and morphology of Aureobasidium pullulans in continuous culture, Biotechnology and Bioengineering, 33: 1210–1212. Ni, X., Cosgrove, J.A., Arnott, A.D., Greated, C.A. and Cumming, R.H., 2000, On the measurement of strain rate in an oscillatory baffled column using particle image velocimetry, Chemical Engineering Science, 55: 3195–3208. Ni, X., Gao, S., Cumming, R.H. and Pritchard, D.W., 1995, A comparative study of mass transfer in yeast for a batch pulsed baffled bioreactor and a stirred tank fermenter, Chemical Engineering Science, 50(13): 2127–2136.
645
Oliveira, M.S.N. and Ni, X., 2001, Gas hold-up and bubble diameters in a gassed oscillatory baffled column, Chemical Engineering Science, 56(21–22): 6143–6148. Oliveira, M.S.N. and Ni, X., 2004, Effect of hydrodynamics on mass transfer in a gas–liquid oscillatory baffled column, The Chemical Engineering Journal, 99: 59–68. Ono, K., Yauda, N. and Ueda, S., 1977, The effect of pH on pullulan elaboration by Aureobasidium pullulans, Agriculture Biological Chemistry, 41: 2113–2118. Popovic, M.K. and Robinson, C.W., 1993, Mixing characteristics of external-loop airlifts: non-Newtonian systems, Chemical Engineering Science, 48: 1405– 1413. Rau, U., Gura, E., Olszewski, E. and Wanger, F., 1992, Enhanced glucan formation of filamentous fungi by effective mixing, oxygen limitation and fed-batch processing, Industrial Microbiology, 9: 19–26. Rho, D., Mulchandani, A., Luong, J. and LeDuy, A., 1988, Oxygen requirements in pullulan fermentation, Applied Microbiol Biotechnology, 28: 361 –366. Rossi, G., 2001, The design of bioreactors, Hydrometallurgy, 59: 217–231. Simon, L., Caye-Vague, C. and Bouchonneau, M., 1993, Relation between pullulans production, morphological state and growth conditions in Aureobasidium pullulans, Journal of General Microbial, 139: 979– 985. Sowa, W., Blackwood, A.C. and Adams, G.A., 1963, Neutral extracellular glucan of pullularia pullulans (de Bery) berkhout, Canadian Journal of Chemistry, 41: 2314– 2319. Sutherland, I.W., 1977, Surface Carbohydrate of Prokaryotic Cell (Academic Press, London, UK). Sutherland, I.W., 1979, Microbial Polysaccharides and Polysaccharases (Academic Press, London, UK). Sutherland, I.W., 1982, Biosynthesis of microbial exopolysaccharides, Advanced Microbiol Physiology, 23: 79–150. Wallenfels, K., Bender, H., Keilich, G. and Bechtler, G., 1961, On pullulan, the glucan of the slime coat of pullularia pullulans, Angew Chemistry, 73: 245– 246. Wecker, A. and Onken, U., 1991, Influence of dissolved oxygen concentration and shear rate on the production of pullulan by Aureobasidium pullulans, Biotechnology Letters, 13: 155–160. This paper was presented at the 7th World Congress of Chemical Engineering held in Glasgow, UK, 10–14 July 2005. The manuscript was received 15 December 2004 and accepted for publication after revision 4 April 2005.
Trans IChemE, Part A, Chemical Engineering Research and Design, 2005, 83(A6): 640–645