SciELO - Scientific Electronic Library Online

vol.20 issue2Removal of SO2 with particles of dolomite limestone powder in a binary fluidized bed reactor with bubbling fluidization author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Brazilian Journal of Chemical Engineering

Print version ISSN 0104-6632On-line version ISSN 1678-4383

Braz. J. Chem. Eng. vol.20 no.2 São Paulo Apr./June 2003 

The influence of oxygen supply on the production of acetaldehyde by Zymomonas mobilis



M.F.MastroeniI, *; P.V.GurgelII; M.M.SilveiraIII; I.M.De MancilhaI; R.JonasII, IV

IDepartamento de Tecnologia de Alimentos, Universidade Federal de Viçosa, 36570-000, Viçosa - MG, Brazil
IUniversidade da Região de Joinville, UNIVILLE, Faculdade de Farmácia, Campus Universitário S/N, Cx. P. 1361, CEP 89.201-972, Phone: 55 47 461-9091, Fax: 55 47 473-0811, Bairro Bom Retiro, Joinville - SC, Brazil, E-mail:
IICentro de Desenvolvimento Biotecnológico, Rodovia SC 301, Km 0, 89239-970, Joinville - SC, Brazil
IIIUniversidade de Caxias do Sul, Instituto de Biotecnologia, Cx. P.1352, 95001-970, Caxias do Sul - RS, Brazil
IVGesellschaft für Biotechnologische Forschung mbH, Mascheroder , Weg 1, D-38124 Braunschweig, Germany




The influence of oxygen supply rate on the growth and the production of both ethanol and acetaldehyde by the aerotolerant fermentative bacterium Z. mobilis is discussed in this work. The results showed similar values of cell mass yield (0.043 g/g) for the five different levels of initial volumetric oxygen transfer coefficient (KLa) studied. The maximum specific growth rate (µx,m.) under anaerobic conditions was higher than those found in aerated runs. Anaerobic cultivation resulted in the best ethanol yield (0.38 g/g). For initial KLa values of 62, 94, and 118 h-1, ethanol yields were 0.10, 0.12, and 0.09 g/g, respectively, whereas for KLa of 30 h-1, an intermediate value (0.24 g/g) was achieved. Under anaerobiosis, no acetaldehyde was produced. With initial KLa values of 62, 94, and 118 h-1, acetaldehyde yields were similar (0.12 to 0.17 g/g), and for KLa of 30 h-1 only 0.07 gram of acetaldehyde was formed per gram of glucose. Although increasing values for the maximal specific acetaldehyde formation rate were calculated as KLa was increased, our results showed that the presence of an excess of dissolved oxygen throughout fermentation is enough to provide appropriate conditions for the production of acetaldehyde by Z. mobilis.

Keywords: Zymomonas mobilis, acetaldehyde production, oxygen supply rate.




Zymomonas mobilis is a preferentially anaerobic microorganism, wich is also able to grow aerobically (Viikari, 1986), that ferments glucose, fructose, or sucrose as carbon sources (Viikari, 1988). These carbohydrates are metabolized via the same biochemical route, the Entner-Doudoroff pathway. This pathway produces only one mol of ATP per mol of hexose fermented, which requires a high level of activity, making this an important basis for the industrial application of this bacterium (Doelle and Doelle, 1989).

When grown under aerobic conditions, Z. mobilis needs to remove the dissolved oxygen from the medium (Bringer et al., 1984). In Z. mobilis, the activity of this defense reaction is NADH and NADPH dependent. Since alcohol dehydrogenase (ADH) is a NADH-dependent enzyme, the conversion of acetaldehyde to ethanol is coupled with NAD+ formation (Pankova et al., 1988). The NADH oxidase enzyme also uses NADH as a substrate, and due to the competition between the two enzymes, acetaldehyde will accumulate in the medium (Ishikawa et al., 1990).

Acetaldehyde is a flammable liquid with a characteristic pungent odor, miscible with water and alcohol (Budavari, 1996). Acetaldehyde is used in different areas and is an important compound in the production of plastics and synthetic rubber, acetic acid, acetic anhydride, ethyl acetate, butanol, and pyridines. In the food industry this product is used as a flavor additive, contributing to the freshness, fruitiness, and/or nuttiness of many food systems (Wecker and Zall, 1987; Zall and Wecker, 1990). The freshness effect can be extended to dry flavors and instantized foods by using dry acetaldehyde systems (Byrne and Sherman, 1984).

Although some work on the behavior of Z. mobilis under aerobic conditions has been published (Bringer et al., 1984; Tanaka et al., 1990; Ishikawa and Tanaka, 1992), there was no study focusing on the production of acetaldehyde in a fermentor under different oxygen supply conditions. In this work, the production of acetaldehyde and also ethanol was followed during the growth of Z. mobilis in batch mode under different initial oxygen supply conditions.




Zymomonas mobilis CP1 was provided by the Instituto Oswaldo G. de Lima of the Universidade Federal de Pernambuco (Recife, Brazil). The microorganism was maintained in liquid medium containing (g/L) sucrose 40.0, peptone 20.0, yeast extract 10.0, KH2PO4 1.0, (NH4)2SO4 1.0, and MgSO4.7H2O 0.5 at pH 5.5 and stored at 4ºC. Pre-inoculum was grown in 25 mL culture tubes with exhaust filters in 5 mL of a medium of the same composition, but containing glucose instead of sucrose. The cells were incubated for 16 hours at 30ºC and 150 rpm. The inoculum, containing 1% pre-inoculum, was grown under the same conditions as those used for pre-inoculum preparation in a 250 mL flask with 100 mL of the medium previously described by Tanaka et al. (1990) of the following composition (g/L): glucose 40.0, peptone 2.5, yeast extract 5.0, KH2PO4 1.0, (NH4)2SO4 1.0, and MgSO4.7H2O 0.5, at pH 5.5).

Fermentation Assays

Fermentation experiments were carried out in batch mode at 30ºC with the same medium as that used for inoculum production in a BIOSTAT B (B.BRAUN BIOTECH, Germany) bioreactor with a 5 L vessel containing 3 L of medium. The pH was automatically controlled at 5.5 by adding 1.0 M KOH. Impeller speed and aeration rate varied according to the value of initial volumetric oxygen transfer coefficient (KLa) desired (Table 1). The fermentor was equipped with three stirrers with six flat blades.



Volatile compounds generated during the process were captured by using a trapping system (Figure 1) similar to that used by Ishikawa and Tanaka (1992). Three flasks were connected in series; the first two contained 3 L of 20.0 g/L NaHSO3 at a pH lower than 2.0 and the third flask contained 200 mL of deionized water. The exhaust air from the fermentor was bubbled into the solutions. All flasks were maintained in an ice bath.



Analytical Methods

Samples were taken from the fermentor and flasks every two hours for quantification of cell mass, glucose, ethanol, and acetaldehyde.

Cell growth was determined by measuring the optical density of cell suspensions at 560 nm (SHIMADZU model UV-160A, Japan). These measurements gave a linear relationship with dry cell mass concentration. Glucose concentration was determined by using an enzymatic method (GOD-POD) (Bergmeyer, 1974). Acetaldehyde and ethanol concentrations in samples taken from the bioreactor were determined by HPLC (MERCK-HITACHI), using an ERC7515A refraction index detector and an ORH-801 column (300 X 6.5 mm) at 70ºC. H2SO4 0.01 N was used as eluent at 0.35 mL/min.

Samples obtained from the trapping flasks were processed using GC (HEWLETT PACKARD) with a flame ionization detector at 250ºC, using hydrogen (60 mL/min) and synthetic air (430 mL/min) as flame gases. The injector was kept at 250ºC, and a PORAPAK Q column (1.0 m X 0.53 mm) at 110ºC was used with a mixture of helium (15 mL/min) and nitrogen (25 mL/min) as carrier. Acetone (0.5 g/L) was used as internal standard.

In bioreactor runs, the pH was determined with a potentiometric electrode. Dissolved oxygen concentration was monitored by using a polarographic probe and the relative measurements (% of saturation in O2) were converted to concentration (mmol/L) by the method described by Schumpe and Quicker (1982). The initial volumetric oxygen transfer coefficients (KLa) were determined by the gassing out method (Rainer, 1990). All fermentation assays were done in duplicate.



The effect of different oxygen supply levels on the production of acetaldehyde and ethanol and on cell growth was studied in batch cultures. Since oxygen transfer rate is dependent on several factors such as the composition and rheology of the medium, as well as the shape of each particular bioreactor, aeration and agitation could not be used to define the operational condition of each run. Thus, the initial KLa was chosen as a reproducible parameter to characterize the oxygen supply rates in the runs carried out.

The effect of oxygen supply on the growth of Z. mobilis was assessed by comparing the specific growth rate profiles (µx) (Figure 2), the maximal specific growth rates (µx,m) (Table 2), and the cell mass production (Figure 3) at the different KLa levels. The results presented in Table 2 show that the maximal specific growth rate (0.62 h-1) during the anaerobic cultivation of Z. mobilis is significantly higher than that observed in the aerated cultures (0.47 to 0.51 h-1), in accordance with the results reported by Viikari (1986). The similar values of µx,m under aeration suggest that Z. mobilis growth is not dependent on oxygen supply rate. According to O'Brien and Morris (1971), the presence of oxygen caused a decrease in µx,m due to its extremely effective action as an acceptor of electrons, impairing the reducing power required for biosynthesis and other cell functions.




In the cultivation of Z. mobilis under anaerobic conditions, a fast decrease in the specific growth rate (µx) is normally observed following the exponential phase, as shown in Figure 2. During the aerated runs, that µx trend was also observed, but with a slow decrease in µx values from approximately 5 hours after the beginning of the fermentation up to the end. Another feature that can be observed in Figure 2 is the similarity of the specific growth rate profiles for the aerated runs. Inhibition of cell growth by acetaldehyde was not expected, since the concentration of this product in fermentation media has never been higher than 0.7 g/L (data not shown), considerably lower than the inhibitory concentration of 1.2 g/L reported by Tanaka et al. (1990). Therefore, the lower values of µx in aerated experiments can not be related to this product, but only to the presence of oxygen in the medium.

Data presented in Figure 3 show cell growth as a function of fermentation time at the different levels of initial KLa studied. It can be observed that the final cell concentrations at the end of each fermentation are similar for all treatments. In all cases, the cell yields remained around 0.043 g of cells per gram of glucose consumed. These results agree with those from Bringer et al. (1984), who reported that Z. mobilis is able to grow in the presence of oxygen without altering the cellular yield. Under anaerobic conditions, however, the cells grow much faster, resulting in a short process time.



The effect of oxygen supply on the kinetics of acetaldehyde and ethanol formation at different levels of KLa is represented in Figure 4, and the yields achieved for each product are shown in Table 3. As expected, no acetaldehyde was formed under anaerobic conditions. With initial KLa levels of 62 to 118 h-1, similar profiles for acetaldehyde production were observed with no significant difference was observed in yields (0.12 to 0.17 g/g). At KLa of 30 h-1, the yield (0.07 g/g) was significantly lower than that observed in the other aerated runs. With respect to the production of ethanol, from the results shown in Table 3, one can see that under anaerobiosis a yield of 0.38 g/g was achieved. At an initial KLa value of 30 h-1, an intermediate yield of 0.24 g/g was calculated. This value is significantly lower than that obtained in the anaerobic run, but higher than those found at levels of KLa of 62, 94, and 118 h-1 (0.10, 0.12, and 0.09 g/g, respectively). These results can be explained by the lowest oxygen supply conditions in the experiment with an initial KLa of 30 h-1, in which the cells were in an oxygen-limited medium from 8 to 14 hours (Figure 6), more favorable conditions for the action of the enzyme alcohol dehydrogenase, which resulted in increasing ethanol formation and reducing acetaldehyde accumulation.




In Figure 5, the variation in the specific acetaldehyde production rate (µP) in each run with time of process is depicted. As can be seen, as the oxygen supply rate was increased, increasing values for the maximum µp were achieved. That behavior is possibly related to the concentration of oxygen in the medium in each case (Figure 6). With increasing levels of dissolved O2, the action of alcohol dehydrogenase would be retarded, thereby improving the specific formation of acetaldehyde. In the sequence, decreasing values for µp were noticed. The results of this work do not provide a complete explanation for that fact. The reduced values of µp were probably influenced by the decreasing concentrations of dissolved oxygen. On the other hand, as mentioned previously, reduced values of the specific growth rate (µx) are observed over time in this fermentation even under conditions favorable to the microorganism. Thus, since the fermentation pathway is coupled to cell growth in Z. mobilis, µp could also be affected by µx.

In spite of the differences observed between the curves relating µP with time, from our results it can bee seen that similar final acetaldehyde yields were obtained for initial KLa values over 62 h-1. That apparently contradictory behavior could be due to the lower concentrations of dissolved oxygen noticed in the second half of these runs, when most of the acetaldehyde was formed. Therefore, in contrast with results reported in other studies (Ishikawa and Tanaka, 1992; Zall and Wecker, 1990), higher supplies of oxygen do not necessarily result in increasing acetaldehyde formation and relatively good yields of this product can be achieved by maintaining a constant excess of dissolved oxygen in the medium. Under these conditions, NADH oxidase would be activated by oxygen and NADH would be limited for ADH to convert acetaldehyde to ethanol.



In this work, we have confirmed that accumulation of acetaldehyde during the fermentation of glucose by Z. mobilis is affected by the oxygen supply rate. Nevertheless, similar product yields (0.12 to 0.17 g acetaldehyde/g glucose) are found when high concentrations of dissolved oxygen are kept in the medium.

Although, for awhile, this process can not be considered for the economical fermentative production of acetaldehyde, further studies focusing on biochemical and microbiological aspects of the cultivation of Z. mobilis under aeration could change this situation. However, Z. mobilis cultivation in the presence of oxygen could not be analyzed by application of conventional theory to facultative anaerobes. Although this bacterium is able to grow in oxygenated media, it is preferentially anaerobic. Therefore, concepts such as critical dissolved oxygen concentration and the relationship between oxygen consumption rate (QO2) and specific growth rate (µx) should be restudied and eventually redefined for this particular case.



We are grateful to the Conselho Nacional de Desenvolvimento Científico e Tecnológico/CNPq (Brazil) for the financial support and the Centro de Desenvolvimento Biotecnológico (Brazil) for the use of its laboratories.





Bergmeyer, H.U., Methods of Enzymatic Analysis. 2nd ed., Academic Press, New York, 1205-14 (1974).        [ Links ]

Bringer, S., Finn, R.K. and Sahm, M., Effect of Oxygen on the Metabolism of Zymomonas mobilis, Arch Microbiol, 139, 376-81 (1984).        [ Links ]

Budavari, S., The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals, 12th ed., Whitehouse Station: Merck (1996).        [ Links ]

Byrne, B. and Sherman, G., Stability of Dry Acetaldehyde Systems, Food Technol, 38, 57-61 (1984).        [ Links ]

Doelle, H.W. and Doelle, M.B., Zymomonas Ethanol Technology – Present State and Development, Aust J. Biotechnol, 3, 218-22 (1989).        [ Links ]

Ishikawa, H., Nobayashi, H. and Tanaka, H., Mechanism of Fermentation Performance of Zymomonas mobilis under Oxygen Supply in Batch Culture, J Ferment Bioeng, 70, 34-40 (1990).        [ Links ]

Ishikawa, H. and Tanaka, H., Effect of Ventilation on the Production of Acetaldehyde by Zymomonas mobilis, J Ferment Bioeng, 73, 297-302 (1992).        [ Links ]

O'Brien, R.W. and Morris, J.G., Oxygen and the Growth and Metabolism of Clostridium acetobutylicum, J Gen Microbiol, 68, 307 (1971).        [ Links ]

Pankova, L.M., Shvinka J.E. and Beker M.J., Regulation of Intracellular H+ Balance in Zymomonas mobilis 113 during the Shift from Anaerobic to Aerobic Conditions, Appl Microbiol Biotechnol, 28, 583-8 (1988).        [ Links ]

Rainer, B.W., Determination Methods of the Volumetric Oxygen Transfer Coefficient KLa in Bioreactors, Chem Biochem Eng, 4, 185-96 (1990).        [ Links ]

Schumpe, A. and Quicker, G., Gas Solubilities in Microbial Culture Media, Adv. Biochem. Eng., 24, 1-38 (1982).        [ Links ]

Tanaka, H., Ishikawa, H., Osuga, K. and Takagi, Y., Fermentative Ability of Zymomonas mobilis under Various Oxygen Supply Conditions in Batch Culture, J Ferment Bioeng 4, 234-9 (1990).        [ Links ]

Viikari, L., By-product Formation in Ethanol Fermentation by Zymomonas mobilis, Technical Research Center of Finland, 27, 3-29 (1986).        [ Links ]

Viikari, L., Carbohydrate Metabolism in Zymomonas mobilis, Crit Rev Biotechnol, 7, 237-61 (1988).        [ Links ]

Wecker, M.S.A. and Zall, R.R., Fermentation Strategies: Acetaldehyde or Ethanol?, Process Biochem, 10, 135-8 (1987).        [ Links ]

Zall, R.R. and Wecker, M.S.A., United States Patent 4 900 670 (1990).        [ Links ]



Received: January 3, 2001
Accepted: September 19, 2002



* To whom correspondence should be addressed

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License