Bioethanol production performance in a packed bed solid-state fermenter: evaluation of operational factors and intermittent aeration strategies

Abstract

A mixture of carob pods and wheat bran as filler was evaluated as substrates for bioethanol production using Zymomonas mobilis in a 0.5 L-glass packed bed column fermenter. The experiments were initially performed under the optimum conditions defined previously in a flask-scale fermentation system. Under these conditions, removal of the generated heat and CO₂ from the solid medium was more challenging than in the flask-scale system. In addition, the traditional aeration method was not applicable for removing the heat and CO₂ since bioethanol production should be conducted under anaerobic conditions. The packed bed column system was modified, and an intermittent aeration method was used to remove the entrapped CO₂ from the solid bed, thereby enabling the anaerobic condition to be maintained. The effect of operating temperature and carob particle size on the amount and rate of ethanol produced was also studied. We determined that the optimum operating conditions for this packed bed column fermentation system were 28 °C, a carob particle size of 1 mm, and 0.1 L min⁻¹ intermittent aeration for 15 min each hour during the second 15 h of the fermentation process. Under these conditions 60.9 g ethanol per kilogram dry mixture of carob pod and wheat bran was produced.

Introduction

Increasing world population, extended industrialization, improvement of life quality worldwide, and increasing oil prices have enhanced the need for new sources of energy as alternatives to traditional fossil fuels (Balat and Balat 2009; Nigam and Singh 2011). One of the alternatives, bioethanol, is a renewable and environmentally friendly fuel which can improve air quality, lower greenhouse gas emissions, and promote domestic rural economies (Börjesson 2009; Nigam and Singh 2011).

However, cheap feedstocks as substrates and new cost-effective production systems are needed before bioethanol can become a profitable alternative to the traditional fuels on an industrial scale. One of these cost-effective technologies may be solid-state fermentation (SSF), in which microorganisms grow on the surface of solid substrates in the absence or near absence of free water (Singhania et al. 2009). In a SSF system, the sugar extraction process is eliminated and less wastewater is produced, resulting in lower distillation and purification costs (Pandey et al. 2008; Singhania et al. 2009). However, the low heat conductivity of the solid medium and the heterogeneity of the process are two major drawbacks of SSF processes (Pandey et al. 2008; Singhania et al. 2009), especially under the anaerobic condition, where forced aeration cannot be used for eliminating the heat and CO₂ from the solid medium.

Many feedstocks have been used for bioethanol production in SSF bioreactors, such as grape and sugar beet pomace (Amin and Khalaf-Allah 1992; Rodriguez et al. 2010), sweet sorghum (Yu et al. 2008; Kwon et al. 2011; Li et al. 2013), mahula flowers (Mohanty et al. 2009), cassava flour (Ogbonna and Okoli 2010), and carob pods (Roukas 1994b; Mazaheri et al. 2012). Among these, Ceratonia siliqua, known as carob, is a promising substrate for bioethanol production due to its high content of fermentable sugars (about 50 % w/w). Carob is an evergreen shrub or tree native to the Mediterranean area, Southwest Asia, and many areas of North America where its drought resistance enables it to grow under dry conditions (Sánchez et al. 2010). Carob contains fructose (102–115 g kg⁻¹), glucose (33.0–36.8 g kg⁻¹), and sucrose (299–384 g kg⁻¹) (Biner et al. 2007), which can be easily consumed by ethanol-producing microorganisms. Carob extract has been used for bioethanol production with Saccharomyces cerevisiae in liquid fermentation systems, with maximum yields of 0.37 (Turhan et al. 2010), 0.47 (Sánchez et al. 2010) and 0.21 g g⁻¹ of initial sugars (Roukas 1994a) reported in the different studies. In addition, an analysis of the process design and economics of bioethanol production from carob extract has also shown that the estimated ethanol production cost and discounted cash flow rate of return are 0.55 € L⁻¹ and 7 %, respectively (Sánchez-Segado et al. 2012).

In our previous studies, we used the Gram-negative bacterium Zymomonas mobilis as a promising alternative to S. cerevisiae for bioethanol production from carob pods in both liquid (Vaheed et al. 2011) and solid-state fermentation systems (Mazaheri et al. 2012; Saharkhiz et al. 2013). In these liquid and solid-state fermentation systems using carob with Z. mobilis, a maximum of 0.34 and 0.3 g ethanol g⁻¹ of initial sugars was produced in flask-scale experiments, respectively (Vaheed et al. 2011; Mazaheri et al. 2012). In the SSF experiments of Mazaheri et al. (2012), wheat bran was mixed with carob particles to provide a support for bacterial growth [high concentration of sugars in carob may inhibit bacterial growth on carob particles; in the presence of a filler (e.g.,wheat bran), the bacteria use the surface of the filler for growth]. These authors also observed that in a SSF system using carob pods, no pretreatment, hydrolysis, or sugar extraction processes were required, which can strongly affect the final cost of ethanol production (Mazaheri et al. 2012).

Zymomonas mobilis is a facultative anaerobe bacterium that metabolizes glucose, fructose, and sucrose, the sugars present in carob pods, through the Entner–Doudoroff pathway. Zymomonas mobilis has fast growth rate and high ethanol production rate (He et al. 2009). High concentrations of oxygen have been reported to negatively affect cell and ethanol yield during Z. mobilis fermentation (Yang et al. 2009). Therefore, low air flow rate or intermittent aeration in liquid and solid state fermentations may be an effective strategy to improve bioethanol production.

In this study, after the successful results of bioethanol production from carob pods using Z. mobilis in flask-scale SSF process, we applied the process to a packed bed column. The latter is inherently associated with a number of challenging problems, such as the removal of the generated heat and CO₂ that accumulates during the fermentation process. Because Zymomonas mobilis produces bioethanol under semi-anaerobic conditions, the traditional aeration method (from the beginning to the end of the fermentation process) for removing the heat and CO₂ is not efficacious. Here, we investigated the effect of different factors and also a number of intermittent aeration strategies on bioethanol production from the mixture of carob pods and wheat bran in the packed bed solid state column.

Materials and methods

Microorganism and substrate

Zymomonas mobilis (PTCC 1718) was purchased from the Persian Type Culture Collection. The freeze-dried strain was activated and cultured at 30 °C and 120 rpm for 17 h in a medium containing 10 g peptone from meat (Merck, Darmstadt, Germany), 10 g yeast extract (Merck), and 20 g glucose (Merck) per liter of distilled water.

Carob fruits were obtained from a Cypriot local market and stored at 4 °C. The carob fruits were first pulverized and the seeds removed; the pod particles were then dried in an oven at 70 °C to a constant weight and subsequently passed through sieves with mesh sizes of 0.25, 0.5, 1, and 1.5 mm, respectively.

Wheat bran was purchased from a local market in Tehran, Iran. Wheat bran particles were oven-dried in the same process as described for the carob particles and then passed through sieves with mesh sizes between 1–1.7 mm.

SSF in the packed bed column

The fermentations were performed in a jacketed glass column (height 28 cm, internal diameter 5 cm). The temperature of the column bed was adjusted by water circulation through the jacket around the column bed using a water circulator (Julabo F-32; Sigma-Aldrich, St. Louis, MO). A sieve glass plate was also fixed to the bottom of the column as a sparger to distribute the air during aeration.

The optimum conditions determined in our previous study on SSF in flasks (Mazaheri et al. 2012) were used in our initial experiments on SSF in the packed bed column. The solid substrate, which consisted of 50 g of a carob/wheat bran mixture (60 % w/w carob particles of the appropriate size, 40 % w/w wheat bran), was moistened with peptone solution (0.7 % w/w) to achieve a moisture content of 80 % (w/w). After sterilization by autoclaving at 121 °C for 15 min, the medium was inoculated with Z. mobilis at a concentration of 6.74 × 10⁸ Z. mobilis cells g⁻¹ carob. The temperature of the solid bed during the fermentation was monitored by a sensor and adjusted when necessary by adjusting the temperature of the water in the jacket around the column. During aeration periods, aeration was carried out by an air pump, and the air flow rate was controlled by a digital mass flow controller (MASS-STREAM®; Bronkhorst Hi-Tech, Ruurlo, the Netherlands). The outlet air stream from the column was connected to a gas analyzer (model 4100; Servomex, Woburn, MA) for monitoring the amount of CO₂ produced during the SSF process.

We first studied the effects of circulating water temperature (from 26 to 32 °C) and carob particle size (0.25–1.5 mm) on bioethanol production in the column. We then investigated the efficacy of four different intermittent aeration strategies to overcome the accumulation of CO₂ in the bed and increase the yield of bioethanol production.

Analytical methods

For evaluating the ethanol and sugar content of the solid bed, we harvested the fermented substrate from the column at the end of the fermentation process. The solid samples were mixed with an appropriate amount of distilled water and shaken at 200 rpm on an orbital platform shaker (Unimax 2010; Heidolph Instruments Labortechnik, Schwabach, Germany) for 30 min to leach the contents. The leachate was then centrifuged (B. Braun Biotech Int, Sartorius BBI Systems GmbH, Goettingen, Germany) at 4,000 rpm for 10 min and the supernatant collected; the sediment was then washed again with distilled water and recentrifuged and the supernatant collected. For analysis, the two supernatants were combined and a 10-ml sample was distilled at atmospheric pressure. The ethanol content of the liquid extract was determined using the Caputi method (Caputi and Ueda 1968). The total sugar content of the samples was quantified by hydrolysis with 1 M HCl at pH 1, 80–85 °C, for 30 min, followed by neutralization with 1 M NaOH. The 3,5-dinitrosalicylic acid method was used to determine the total sugar content based on the glucose calibration curve (Miller 1959).

Results and discussion

The experiments in the packed bed column were carried out under the optimum conditions determined in previous study using a flask-scale fermentation system. In this previous study (Mazaheri et al. 2012), a maximum ethanol production of 63 g ethanol kg⁻¹ dry mixture of carob pod/wheat bran was achieved under the optimal conditions, which consisted of a mixture of carob and wheat bran as substrate, fermentation at 31 °C, initial moisture content of 80 % (w/w), carob particle size of 1 mm, peptone concentration of 0.7 % (w/w), initial cell concentration of 6.74 × 10⁸ cells g⁻¹ carob, and a fermentation time of 43 h. In contrast, in our initial experiments using the pack bed column, ethanol production was 35.7 g kg^-1 dry mixture of carob pod/wheat bran under the optimum conditions for the flask-scale system, a fermentation time of 40 h, and the anaerobic condition. This lower ethanol production could have been due to the accumulation of generated heat and CO₂ during the fermentation in the solid bed. With the aim to increase ethanol production, we then investigated the effect of various parameters on the performance of the packed bed column.

Effect of circulating water temperature

Ethanol production in the packed bed column was studied at four different temperatures: 26, 28, 30, and 32 °C. The results are shown in Fig. 1. The maximum amount of 42 g ethanol kg⁻¹ dry mixture of carob pod/wheat bran was produced at 28 °C. The results revealed that the amount of heat and CO₂ accumulating in the packed bed column was higher than that in the flask-scale system, indicating that it was necessary to remove the heat by the circulating water system in the jacket around the column.

Fig. 1

Effect of circulating water temperature on bioethanol production in the solid-state fermentation system using a packed bed column

In order to study the effect of circulating water temperature on the heat accumulation and increases in bed temperature, the fermentations were carried out at different circulating water temperatures. Figure 2 shows the temperature monitored at the bottom of the bed for circulating water temperatures of 28 and 30 °C. This graph shows that after 10 h, the temperature of the solid bed increased as a result of enhanced bacterial metabolic activity and subsequent heat generation. The temperature decrease during stationary phase is also related to lower metabolic activity of bacteria in the system and depletion of nutrients. The same trend was observed by Figueroa-Montero et al. (2011) and Kwon et al. (2011). The most interesting finding, as shown in Fig. 2, is that the temperature in the bed not exceed the optimum temperature of bacterial activity; for example, at a circulating water temperature of 30 °C, the bed temperature reached 31.7 °C, which is close to the optimum temperature of the bacteria (based on the results obtained in flasks). Consequently, we conclude that the accumulation of generated heat and subsequent increased temperature of the solid bed were not the main factors for the decreased ethanol production (compared to flasks) in the packed bed column and suggest that the entrapped CO₂ in the moistened solid bed, which can inhibit bacterial metabolism, may be the main reason for the lower ethanol production yield. Since at a circulating water temperature of 30 °C the microorganism is near its optimum temperature, the rate of bacterial growth and CO₂ production is higher than that at 28 °C and, consequently, the inhibition effect of CO₂ on ethanol production would be higher, resulting in lower ethanol production.

Fig. 2

Solid bed temperature variation during fermentation at a circulating water temperature of 28 °C (broken line) and 30 °C (solid line)

A gas analyzer was also connected to the air outlet stream of the column to measure CO₂ production. In this condition (without aeration), CO₂ was not detected by the gas analyzer, confirming that the produced CO₂ was entrapped in the moistened solid bed and unable to leave the column.

Effect of continuous aeration

The results obtained in the previous section reveal that the entrapment of CO₂ in the solid bed may inhibit ethanol production in the packed bed column. Forced air convection is the most common method used to remove CO₂ from SSF systems (Figueroa-Montero et al. 2011; Kwon et al. 2011). However, ethanol fermentation by Z. mobilis requires anaerobic conditions, and forced air convection may adversely affect the ethanol production process. One alternative is to blow inert gases into the solid bed, but this procedure may not be economically feasible, especially in large-scale bioreactors. Therefore, we tested different aeration strategies for CO₂ removal while concomitantly maintaining the anaerobic condition to some extent. We studied the effect of continuous aeration at a low flow rate for a limited period of time from the beginning of the process and we evaluated the effect of intermittent aeration on ethanol production in the packed bed column.

The experiments were conducted in the packed bed column under the optimum conditions described for the flask-scale study at a circulating water temperature of 28 °C. Aeration was performed for 5, 10, 15, 20, and 40 h, respectively, from the beginning of the fermentation process at flow rate of 100 mL min⁻¹ , following which time that the process was followed without aeration. Ethanol production under these different aerations conditions is shown in Fig. 3. Continuous aeration during the early stages of the fermentation was found to slightly increase ethanol production. Based on these results, we suggest that aeration for up to 10 h from the beginning of the fermentation process could facilitate ethanol production by removing the CO₂ from the solid bed; however, after 10 h the aerobic condition of the packed bed column and evaporation of ethanol adversely affect ethanol production in the solid bed.

Fig. 3

Effect of different lengths of continuous aeration time (flow rate 100 mL min⁻¹) on ethanol production in the solid packed bed fermentation system

Effect of carob particle size

The experiments on the effect of carob particle size were performed in the packed bed column under the optimum conditions mentioned above, but with continuous aeration for 10 h from the beginning of the process. Carob particle sizes of 0.25, 0.5, 1, and 1.5 mm were tested separately. Figure 4 shows that a carob particle size of 1 mm resulted in the highest ethanol production in the packed bed column, which is the same result obtained in the flask-scale study. We suggest carob particles of <1 mm are not suitable as fermentation substrate due to the adhesion properties of the carob particles. In contrast, however, the reduced surface area/volume ratio of the larger particles provides a smaller surface for bacterial growth.

Fig. 4

Effect of carob particle size on ethanol production in the solid packed bed fermentation system

Effect of intermittent aeration

Different intermittent aeration strategies were tested for their efficacy to remove the entrapped CO₂ from the solid bed and simultaneously maintain the anaerobic condition for a higher production of ethanol. In this experiment, intermittent aeration strategy consisted of providing a low flow rate of air (100 mL min⁻¹) to the fermentation system for a short period of time (15 min), followed by a period of no air flow (i.e., fermentation under anaerobic conditions); this cycle of air/no air was repeated for different lengths of time during the fermentation process. For example, in aeration program no. 3, aeration was provided for 15 min per hour during the exponential growth phase (about 15 h). The aeration strategies and the associated ethanol production are reported in Table 1. We found that 15 min aeration per hour during the whole process (program No. 1) produced better results in terms of CO₂ removal and maintenance of the anaerobic condition than the anaerobic process and continuous aeration. In program No. 2, intermittent aeration was provided during the first 15 h of the fermentation process, which contains the lag phase. Since metabolic activity is low in the lag phase, CO₂ generation is not challenging in this period and, consequently, intermittent aeration in this period had a limited effect on ethanol production. However, in the exponential phase, when the metabolic activity of cells is at its maximum, CO₂ production will increase, possibly inhibiting ethanol production. We found that intermittent aeration in this phase (program No. 3) achieved an ethanol production of 60.9 g kg⁻¹ dry mixture of carob pod and wheat bran, which is close to that obtained under flask-scale fermentation conditions. Under this condition, 70 % of the initial sugar was converted to ethanol, with yield of 40 %. In the stationary phase (program No. 4), similar to the lag phase, intermittent aeration has a limited effect on ethanol production due to the lower metabolic activity and CO₂ production in this phase. Moreover, aeration in this phase may cause evaporation of the ethanol produced in the solid medium.

Table 1 The intermittent aeration strategies applied in our study and the associated ethanol production

The removal of entrapped CO₂ from the solid bed was tested by using a gas analyzer. Figure 5 shows the CO₂ percentage in the outlet of the packed bed column. Before the aeration (before t = 15 h), the gas analyzer was unable to detect CO₂ in the outlet stream, which shows that CO₂ was entrapped in the solid bed and could not leave the column without forced air convection (Fig. 5a). After aeration (after t = 15 h), CO₂ could be removed from the bed by the forced air convection and was detected by the gas analyzer in the outlet stream. Figure 5b. also shows that only during the aeration periods could CO₂ be detected by the gas analyzer.

Fig. 5

CO₂ concentration in the outlet of the packed bed column while intermittent aeration strategy No. 3. a The overall process, b enlarged diagram between 17 and 19 h

Conclusions

We evaluated the performance of a packed bed column for bioethanol production from carob pods. Our results show that the accumulation of metabolic CO₂ in the packed solid bed column can inhibit bacterial activity and ethanol production in the bed. Since the process is anaerobic and the use of inert gases may not be economically advantageous, we tested various intermittent aeration strategies and identified a strategy for improving the ethanol production of the packed bed column. We also found that carob particles of <1 mm had a negative effect on bioethanol production due to the adhesion of the carob particles.