Continuous ethanol production using induced yeast aggregates

Summary The induction of yeast cell aggregates in a column reactor was initiated by packing yeast cell paste of Saccharomyces uvarum into the column, and then YMP broth was fed into the column from the bottom at a linear flow rate of 2.5 cm/h. Thereafter, yeast cells aggregated in the column within 48 h without a supply of oxygen. When this yeast aggregate column reactor was used for continuous ethanol production, a final ethanol concentration of 10.8% (w/v) was obtained from 23% (w/v) of glucose in a YMP broth with a dilution rate of 0.05 h^-1, and 4.9% (w/v) was obtained from 10% (w/v) of glucose with a dilution rate of 0.6 h^-1. The theoretical yield was above 97% in both cases. The ethanol production rates were 13 g¹ h^-1 l^-1 and 90 g¹ h^-1 l^-1 for producing 10.8% (w/v) and 4.9% (w/v) of ethanol respectively. This column reactor was maintained at a steady state for more than one month.     

Construction of a flocculating yeast for fructose production from inulin

Construction of flocculating yeast lacking for fructose utilisation was realised by integration of the FLO1 flocculation gene in the ribosomal DNA of an hexokinase deficient (hxk1, hxk2) Saccharomyces cerevisiae strain (ATCC36859). Simultaneous production of ethanol and fructose was obtained from glucose/fructose mixtures or from hydrolysed Jerusalem artichoke extracts using the transformed yeast in batch fermentations and in a continuous reactor with internal biomass recycle. This allowed the production of 5 g ethanol/L and 48 g sugars/L containing up to 99 % fructose from diluted hydrolysed Jerusalem artichoke extracts containing 60 g sugars/L. © Rapid Science Ltd. 1998     

Acetone-butanol-ethanol (ABE) fermentation in an immobilized cell trickle bed reactor

Acetone-butanol-ethanol (ABE) fermentation was successfully carried out in an immobilized cell trickle bed reactor. The reactor was composed of two serial columns packed with Clostridium acetobutylicum ATCC 824 entrapped on the surface of natural sponge segments at a cell loading in the range of 2.03-5.56 g dry cells/g sponge. The average cell loading was 3.58 g dry cells/g sponge. Batch experiments indicated that a critical pH above 4.2 is necessary for the initiation of cell growth. One of the media used during continuous experiments consisted of a salt mixture alone and the other a nutrient medium containing a salt mixture with yeast extract and peptone. Effluent pH was controlled by supplying various fractions of the two different types of media. A nutrient medium fraction above 0.6 was crucial for successful fermentation in a trickle bed reactor. The nutrient medium fraction is the ratio of the volume of the nutrient medium to the total volume of nutrient plus salt medium. Supplying nutrient medium to both columns continuously was an effective way to meet both pH and nutrient requirement. A 257-mL reactor could ferment 45 g/L glucose from an initial concentration of 60 g/L glucose at a rate of 70 mL/h. Butanol, acetone, and ethanol concentrations were 8.82, 5.22, and 1.45 g/L, respectively, with a butanol and total solvent yield of 19.4 and 34.1 wt %. Solvent productivity in an immobilized cell trickle bed reactor was 4.2 g/L h, which was 10 times higher than that obtained in a batch fermentation using free cells and 2.76 times higher than that of an immobilized CSTR. If the nutrient medium fraction was below 0.6 and the pH was below 4.2, the system degenerated. Oxygen also contributed to the system degeneration. Upon degeneration, glucose consumption and solvent yield decreased to 30.9 g/L and 23.0 wt %, respectively. The yield of total liquid product (40.0 wt %) and butanol selectivity (60.0 wt %) remained almost constant. Once the cells were degenerated, they could not be recovered.     

Very high gravity ethanol fermentation by flocculating yeast under redox potential-controlled conditions

ABSTRACT: BACKGROUND: Very high gravity (VHG) fermentation using medium in excess of 250 g/L sugars for more than 15 % (v) ethanol can save energy consumption, not only for ethanol distillation, but also for distillage treatment; however, stuck fermentation with prolonged fermentation time and more sugars unfermented is the biggest challenge. Controlling redox potential (ORP) during VHG fermentation benefits biomass accumulation and improvement of yeast cell viability that is affected by osmotic pressure and ethanol inhibition, enhancing ethanol productivity and yield, the most important techno-economic aspect of fuel ethanol production. RESULTS: Batch fermentation was performed under different ORP conditions using the flocculating yeast and media containing glucose of 201 [PLUS-MINUS SIGN] 3.1, 252 [PLUS-MINUS SIGN] 2.9 and 298 [PLUS-MINUS SIGN] 3.8 g/L. Compared with ethanol fermentation by non-flocculating yeast, different ORP profiles were observed with the flocculating yeast due to the morphological change associated with the flocculation of yeast cells. When ORP was controlled at [MINUS SIGN]100 mV, ethanol fermentation with the high gravity (HG) media containing glucose of 201 [PLUS-MINUS SIGN] 3.1 and 252 [PLUS-MINUS SIGN] 2.9 g/L was completed at 32 and 56 h, respectively, producing 93.0 [PLUS-MINUS SIGN] 1.3 and 120.0 [PLUS-MINUS SIGN] 1.8 g/L ethanol, correspondingly. In contrast, there were 24.0 [PLUS-MINUS SIGN] 0.4 and 17.0 [PLUS-MINUS SIGN] 0.3 g/L glucose remained unfermented without ORP control. As high as 131.0 [PLUS-MINUS SIGN] 1.8 g/L ethanol was produced at 72 h when ORP was controlled at [MINUS SIGN]150 mV for the VHG fermentation with medium containing 298 [PLUS-MINUS SIGN] 3.8 g/L glucose, since yeast cell viability was improved more significantly. CONCLUSIONS: No lag phase was observed during ethanol fermentation with the flocculating yeast, and the implementation of ORP control improved ethanol productivity and yield. When ORP was controlled at [MINUS SIGN]150 mV, more reducing power was available for yeast cells to survive, which in turn improved their viability and VHG ethanol fermentation performance. On the other hand, controlling ORP at [MINUS SIGN]100 mV stimulated yeast growth and enhanced ethanol production under the HG conditions. Moreover, the ORP profile detected during ethanol fermentation with the flocculating yeast was less fluctuated, indicating that yeast flocculation could attenuate the ORP fluctuation observed during ethanol fermentation with non-flocculating yeast.     

New concepts for rapid yeast settling. II. pH switching with an inert powder

A new technique is outlined for the rapid settling of yeast cells in fermentation media. The technique involved the addition of dense, inert particles (nickel powder) to a yeast suspension (Saccharomyces cerevisiae) at pH 4.5 and a rapid change of pH to 8.0-9.0. When the pH was changed large flocs formed immediately and settled rapidly, leaving a clear supernatant. On returning the pH to 4.5 the flocs were destroyed. This technique gave larger flocs and higher settling rates than the constant pH method, and much lower nickel/yeast ratios were required. Good flocculation also occurred in a fermentation medium. The technique was used to recycle yeast cells to a semicontinuous ethanol fermentation. Application of the technique to this and similar systems is discussed. The factors affecting yeast/inert powder flocculation are also discussed and a model is proposed to explain the observed experimental behavior for flocculation with a rapid change in pH.     

Identification and functional study of a new FLO10-derivative gene from the industrial flocculating yeast SPSC01

Yeast flocculation is an important property for the brewing industry as well as for ethanol fermentation to facilitate biomass recovery by sedimentation from the fermentation broth, which is cost-effective. In this study, a new flocculating gene FLO10 (spsc) of 4,221 bp homologous to FLO10 was identified in the industrial flocculating yeast SPSC01. Sequence analysis indicated that the N- and C-terminus of the deduced protein of this new FLO gene are 99 % identical to that of FLO10, but more intragenic repeats are included. The study on the function of FLO10 (spsc) by its integrative expression in the non-flocculating industrial yeast indicated severe inhibition in the flocculation of the transformant by mannose and maltose, moderate inhibition by sucrose and glucose and no inhibition by xylose and galactose, and thus the NewFlo type was established. Meanwhile, the flocculation of the transformant was stable when the temperature was below 50 °C and the pH was in the range of 4.0-6.0. Furthermore, the medium containing 250 g/l glucose was completely fermented within 48 h by the transformant, with about 110 g/l ethanol and 5.5 g(DCW)/l biomass produced, and no significant difference in ethanol fermentation performance was observed compared to its wide-type strain. Therefore, the FLO gene and corresponding transformation strategy provide a platform for engineering yeast strains with the flocculation phenotype to facilitate biomass recovery.     

Ethanol fermentation in a continuous tower fermentor

A mutant of Saccharomyces cerevisiae, which forms large, multicellular flocs in liquid culture, rapidly fermented media containing high concentrations of glucose (100-180 g/L) in a continuous nonaerated tower fermentor at 30 degrees C. The fermentor operated continuously for seven months. Batch and tower fermentor data were fitted to a kinetic model incorporating linear ethanol inhibition and Monod dependence on glucose. Conversion, ethanol yield, and ethanol productivity were related to the apparent fermentation time for initial glucose concentrations of 130 and 180 g/L. Productivities of 8-12 g ethanol/L h were achieved through the yeast bed giving conversions exceeding 90% of the theoretical yield.     

Continuous fermentation of feed streams containing D-glucose and D-xylose in a two-stage process utilizing immobilized Saccharomyces cerevisiae and Pachysolen tannophilus

In the U.S., forest and crop residues contain enough glucose and xylose to supply 10 times the country's usage of ethanol and ethylene, but an efficient fermentation scheme is lacking,(1,2,3) To develop a strategy for process design, specific ethanol productivities and yields of Pachysolen tannophilus NRRL Y-2460 and Saccharomyces cerevisiae NRRL Y-2235 were compared. Batch cultures and continuous stirred reactors (CSTR) loaded with immobilized cells were fed glucose and xylose. As expected from previous reports, Y-2235 fermented glucose but not xylose. Y-2460 consumed both sugars but fermented glucose inefficiently relative to Y-2235, and it suffered a diauxic lag lasting 10-20 h when given a sugar mixture. Immobilized Y-2235 exhibited increasing productivity but constant yield with in creasing glucose concentration. In contrast, Y-2460 exhibited an optimum productivity at 30-40 g/L xylose and a declining yield with increasing xylose concentration. Immobilized Y-2235 tolerated more than 100 g/L ethanol while the productivity and yield of Y-2460 fell by 80 and 58%, respectively, as ethanol reached 50 g/L. A 38.8-g/L ethanol stream could be produced as 103 g/L xylose was continuously fed to Y-2460. If it was blended with a 274 g/L glucose stream to give a composite of 23.7 g/L ethanol and 107 g/L glucose, Y-2235 could en rich the ethanol to 75 g/L. Taken together these results suggest use of a two-stage continuous reactor for pro cessing xylose and glucose from lignocellulose. An immobilized Y-2460 CSTR (or cascade) would convert the hemicellulose hydrolyzate. Then downstream, an immobilized Y-2235 plug flow reactor would enrich the hemicellulose-derived ethanol to more than 70 g/L upon addition of cellulose hydrolyzate.     

Carbohydrate carbon sources induce loss of flocculation of an ale-brewing yeast strain

AIMS: To identify the nutrients that can trigger the loss of flocculation under growth conditions in an ale-brewing strain, Saccharomyces cerevisiae NCYC 1195. METHODS AND RESULTS: Flocculation was evaluated using the method of Soares, E.V. and Vroman, A. [Journal of Applied Microbiology (2003) 95, 325]. Yeast growth with metabolizable carbon sources (glucose, fructose, galactose, maltose or sucrose) at 2% (w/v), induced the loss of flocculation in yeast that had previously been allowed to flocculate. The yeast remained flocculent when transferred to a medium containing the required nutrients for yeast growth and a sole nonmetabolizable carbon source (lactose). Transfer of flocculent yeast into a growth medium with ethanol (4% v/v), as the sole carbon source did not induce the loss of flocculation. Even the addition of glucose (2% w/v) or glucose and antimycin A (0.1 mg l(-1)) to this culture did not bring about loss of flocculation. Cycloheximide addition (15 mg l(-1)) to glucose-growing cells stopped flocculation loss. CONCLUSIONS: Carbohydrates were the nutrients responsible for stimulating the loss of flocculation in flocculent yeast cells transferred to growing conditions. The glucose-induced loss of flocculation required de novo protein synthesis. Ethanol prevented glucose-induced loss of flocculation. This protective effect of ethanol was independent of the respiratory function of the yeast. SIGNIFICANCE AND IMPACT OF THE STUDY: This work contributes to the elucidation of the role of nutrients in the control of the flocculation cycle in NewFlo phenotype yeast strains.     

自絮凝颗粒酵母乙醇连续发酵耦合酵母回用工艺的研究

模拟现有酒精发酵行业普遍采用的多级发酵罐串联系统,建立了一套由三级串联操作的搅拌式发酵罐和两个沉降罐组成的反应器系统,以脱胚脱皮玉米粉双酶法制备的糖化液为发酵底物,培养基初始还原糖浓度为220g/L,添加(NH4)2HPO41.5g/L和KH2PO42.5g/L,以0.057h-1的恒定稀释速率流加,将自沉降浓缩后的酵母乳先后经活化和不活化两种方式处理并循环至第一级发酵罐,系统在两种操作条件下分别达到了拟稳态。实验结果表明活化处理对改善发酵工艺技术指标方面发挥了显著的作用,发酵终点乙醇浓度达到101g/L,还原糖和残总糖分别在3.2和7.7g/L左右,发酵系统的设备生产强度指标为5.77g/(L.h),与无酵母回用的搅拌式反应器系统中自絮凝颗粒酵母乙醇发酵工艺相比,提高了70%。     

Observed quasi-steady kinetics of yeast cell growth and ethanol formation under very high gravity fermentation condition

Using a generalSaccharomyces cerevisiae as a model strain, continuous ethanol fermentation was carried out in a stirred tank bioreactor with a working volume of 1,500 mL. Three different gravity media containing glucose of 120, 200 and 280 g/L, respectively, supplemented with 5 g/L yeast extract and 3 g/L peptone, were fed into the fermentor at different dilution rates. Although complete steady states developed for low gravity medium containing 120 g/L glucose, quasi-steady states and oscillations of the fermented parameters, including residual glucose, ethanol and biomass were observed when high gravity medium containing 200 g/L glucose and very high gravity medium containing 280 g/L glucose were fed at the designated dilution rate of 0.027 h⁻¹. The observed quasi-steady states that incorporated these steady states, quasi-steady states and oscillations were proposed as these oscillations were of relatively short periods of time and their averages fluctuated up and down almost symmetrically. The continuous kinetic models that combined both the substrate and product inhibitions were developed and correlated for these observed quasi-steady states.     

Production of ethanol by adsorbed yeast cells

Summary Various ion exchange resins were tested for their ability to adsorb cells of Saccharomyces cerivisiae with the ultimate intention of developing a packed bed immobilized cell reactor for the continuous production of ethanol. The resins varied greatly in their ability to adsorb cells - the least effective resins retained less than 1 mg S. cerivisiae cells (dry weight)/g of resin (dry weight), and the most effective, 130–140 mg cells/g of resin. A column reactor packed with adsorbed yeast cells was operated continuously for over 200 hours using a 12% (w/v) glucose medium at dilution rates of 1.1 h^-1 and 1.44 h^-1 (based on void volume). High ethanol productivities of 53.1 and 62.0 g ethanol/l-h were obtained.     

Physical and hydrodynamic properties of flocs produced during biological hydrogen production

Dense flocs readily form in continuous culture bioreactors used for hydrogen production, but the fractal and hydrodynamic properties of these flocs have not been previously analyzed. We therefore examined the size distribution, fractal dimension, and hydrodynamic properties of flocs formed in a continuous flow, well-mixed reactor treating synthetic wastewater at a fixed condition of a 4.5 h hydraulic detention time (23 degrees C, pH 5.5). The reactor was operated for a total of 3 months at three different organic loading rates (27, 53, and 80 g-COD/L-d) with influent glucose concentrations of 5, 10, and 15 g-COD/L. At all three loading rates the removal of glucose was nearly complete (98.6-99.4%) and biomass was produced in proportion to the organic loading rate (0.86 +/- 0.11, 2.40 +/- 0.26, and 4.59 +/- 1.55 g/L of MLVSS in the reactor). Overall conversion efficiencies of glucose to hydrogen, evaluated on the basis of a maximum of 4 mol-H2/mol-glucose, increased with organic loading rates in the order 17.7%, 23.1%, and 25.6%. The gas contained 56.1 +/- 4.9% hydrogen, with the balance as carbon dioxide. No methane gas was detected. Under these conditions, flocs were produced with mean sizes that increased with organic loading, in the order 0.12 cm (5 g-COD/L), 0.35 cm (10 g-COD/L), and 0.58 cm (15 g-COD/L). As the average floc size increased, the flocs became on average denser and less fractal, with fractal dimensions increasing from 2.11 +/- 0.17 to 2.48 +/- 0.13. Floc porosities ranged from 0.75-0.96, and resulted in aggregate densities that allowed little intra-aggregate flow through the floc. As a result, average settling velocities were not appreciably larger than those predicted by Stokes' law for spherical, impermeable flocs. Our results demonstrate that dense, relatively impermeable flocs are produced in biohydrogen reactors that have settling properties in reasonable agreement with Stokes' law.     

Inulin enrichment by fermentation in a flocculating yeast reactor

Summary Simultaneous production of ethanol and fructose enriched syrups was obtained from Jerusalem artichoke extract using a Saccharomyces diastaticus flocculating yeast in a continuous gas-lift reactor with internal biomass recycle. This allowed the production of 42 g/L of ethanol and 70 g/L of inulin containing up to 92% fructose (fructose/glucose ratio of 11). These results can be compared to the batch and chemostat fermentations which gave a higher ethanol concentration but a lower fructose enrichment. Mass transfert limitations can explain both the productivity decrease and the selectivity improvement in the gas-lift reactor.     

Ethanol production by Kluyveromyces lactis immobilized cells in copolymer carriers produced by radiation polymerization

The conditions for batch and continuous production of ethanol, using immobilized growing yeast cells of Kluyveromyces lactis, have been optimized. Yeast cells have been immobilized in hydrogel copolymer carriers composed of polyvinyl alcohol (PVA) with various hydrophilic monomers, using radiation copolymerization technique. Yeast cells were immobilized through adhesion and multiplication of yeast cells themselves. The ethanol production of immobilized growing yeast cells with these hydrogel carriers was related to the monomer composition of the copolymers and the optimum monomer composition was hydroxyethyl methacrylate (HEMA). In this case by using batch fermentation, the superior ethanol production was 32.9 g L(-1) which was about 4 times higher than that of cells in free system. The relation between the activity of immobilized yeast cells and the water content of the copolymer carriers was also discussed. Immobilized growing yeast cells in PVA: HEMA (7%: 10%, w/w) hydrogel copolymer carrier, were used in a packed-bed column reactor for the continuous production of ethanol from lactose at different levels of concentrations (50, 100 and 150) g L(-1). For all lactose feed concentrations, an increase in dilution rates from 0.1 h(-1) to 0.3 h(-1) lowered ethanol concentration in fermented broth, but the volumetric ethanol productivity and volumetric lactose uptake rate were improved. The fermentation efficiency was lowered with the increase in dilution rate and also at higher lactose concentration in feed medium and a maximum of 70.2% was obtained at the lowest lactose concentration 50 g L(-1).     

Effect of oxygen on ethanol fermentation in packed-bed tapered-column reactor

Summary In ethanol production with immobilized yeast a major problem is the provision of nutrients to these highly concentrated cells. O₂ being one of the nutrients of utmost importance to yeast cells, was fed into a column packed with beads with a cell loading of more than 40 g/l. Since addition of large volume of air or O₂ to a cylindrical column reactor would aggravate the problems of pressure build up and channelling caused by the evolving CO₂ gas, a tapered-column reactor and pulsed flow of oxygen gas was used. The supplement of O₂ gas to the tapered column increased the productivity from 21.1 g ethanol x (l gel x h)^-1 to 26.7 g x (l gel x h)^-1, when the ethanol concentration at the outlet was about 80 g/l. The yield coefficient of ethanol was also increased from 0.41 g ethanol/g glucose to 0.43 after O₂ supplement was started. The effects of frequency and duration of O₂ supplement were also determined.     

自絮凝酵母高浓度重复批次乙醇发酵

利用发酵性能优良的自絮凝酵母Saccharomyces cerevisiaeflo,研究开发了重复批次高浓度乙醇发酵系统,以节省下游加工过程的能耗。在终点乙醇浓度达到120g/L左右的条件下,发酵系统的乙醇生产强度达到8.2g/(L·h)。然而实验中发现,随着发酵批次的增多,自絮凝酵母沉降性能逐渐下降,从发酵液中沉降分离所需时间相应延长,导致发酵液中高浓度乙醇对酵母的毒害作用加剧,影响其发酵活性和发酵系统运行的稳定性,发酵装置运行11个批次后无法继续运行。实验结果表明,絮凝能力下降导致的酵母絮凝颗粒尺度减小是其沉降性能下降的主要原因。进一步研究发现,酵母的絮凝能力通过再培养可以恢复。在此基础上对发酵系统操作进行改进,每批发酵结束后可控采出一定比例菌体,调节系统的酵母细胞密度和乙醇生产强度以刺激酵母增殖,保持其絮凝能力。在达到相同发酵终点乙醇浓度条件下,虽然发酵系统的乙醇生产强度降低到4.0g/(L·h),但运行10d后絮凝颗粒酵母尺度趋于稳定,继续运行14d,未发现絮凝颗粒酵母尺度继续下降的现象,系统可以稳定运行。     

Optimization of medium formulation and seed conditions for expression of mature PsaA (pneumococcal surface adhesin A) in Escherichia coli using a sequential experimental design strategy and response surface methodology

PsaA, a candidate antigen for a vaccine against pneumonia, is well-conserved in all Streptococcus pneumoniae serotypes. A sequence of two-level experimental designs was used to evaluate medium composition and seed conditions to optimize the expression of soluble mature PsaA in E. coli. A face-centered central composite design was first used to evaluate the effects of yeast extract (5 and 23.6?g/L), tryptone (0 and 10?g/L), and glucose (1 and 10?g/L), with replicate experiments at the central point (14.3?g/L yeast extract, 5?g/L tryptone, 5.5?g/L glucose). Next, a central composite design was used to analyze the influence of NaCl concentration (0, 5, and 10?g/L) compared with potassium salts (9.4?g/L K(2)HPO(4)/2.2?g/L KH(2)PO(4)), and seed growth (7 and 16?h). Tryptone had no significant effect and was removed from the medium. Yeast extract and glucose were optimized at their intermediate concentrations, resulting in an animal-derived material-free culture medium containing 15?g/L yeast extract, 8?g/L glucose, 50?μg/mL kanamycin, and 0.4% glycerol, yielding 1?g/L rPsaA after 16?h induction at 25°C in shake flasks at 200?rpm. All the seed age and salt conditions produced similar yields, indicating that no variation had a statistically significant effect on expression. Instead of growing the seed culture for 16?h (until saturation), the process can be conducted with 7?h seed growth until the exponential phase. These results enhanced the process productivity and reduced costs, with 5?g/L NaCl being used rather than potassium salts.     

Must deacidification with an induced flocculant yeast strain of Schizosaccharomyces pombe

The use of flocculant cells of the yeast strain Schizosaccharomyces pombe for the deacidification of grape musts in continuous culture was developed. An external loop reactor was used to induce flocculation. The flocs obtained were stable in the pH range 3.0–6.0 and in the presence of several sugars. Some inhibition was observed for high (above 6.0) and low (below 3.0) pH values. Once induced, flocculation could no longer be completely inhibited. Vinho Verde, a typical Portuguese wine, has a relatively low ethanol content and a high acid concentration. The external loop reactor loaded with the flocculant cells was used to deacidify a synthetic medium with sugar and malic acid concentrations similar to the ones found in Vinho Verde grape must. A desirable malic acid decrease with moderate glucose consumption was obtained at a dilution rate of 0.7 H^–1. Improved results were obtained when the synthetic medium was replaced by Vinho Verde grape must. Correspondence to: M. Mota     

Lactic acid fermentation in a recycle batch reactor using immobilized Lactobacillus casei

Lactic acid production by recycle batch fermentation using immobilized cells of Lactobacillus casei subsp. rhamnosus was studied. The culture medium was composed of whey treated with an endoprotease, and supplemented with 2.5 g/L of yeast extract and 0.18 mM Mn(2+) ions. The fermentation set-up comprised of a column packed with polyethyleneimine-coated foam glass particles, Pora-bact A, and connected with recirculation to a stirred tank reactor vessel for pH control. The immobilization of L. casei was performed simply by circulating the culture medium inoculated with the organism over the beads. At this stage, a long lag period preceded the cell growth and lactic acid production. Subsequently, for recycle batch fermentations using the immobilized cells, the reducing sugar concentration of the medium was increased to 100 g/L by addition of glucose. The lactic acid production started immediately after onset of fermentation and the average reactor productivity during repeated cycles was about 4.3 to 4.6 g/L . h, with complete substrate utilization and more than 90% product yield. Sugar consumption and lactate yield were maintained at the same level with increase in medium volume up to at least 10 times that of the immobilized biocatalyst. The liberation of significant amounts of cells into the medium limited the number of fermentation cycles possible in a recycle batch mode. Use of lower yeast extract concentration reduced the amount of suspended biomass without significant change in productivity, thereby also increasing the number of fermentation cycles, and even maintained the D-lactate amount at low levels. The product was recovered from the clarified and decolorized broth by ion-exchange adsorption. (c) 1997 John Wiley & Sons, Inc. Biotechnol Bioeng 55:841-853, 1997.