Solid bioreactor

from Wikipedia, the free encyclopedia

Solid bioreactors (English Solid State Fermentation-Bioreactors ; SFB ) are bioreactors that for the culture of microorganisms and for industrial production of enzymes , drugs and foods can be used.

Problems with the design of the solid bioreactors result from the limitations in the oxygen supply and the ventilation of excess CO 2 , which lead to intolerable temperature gradients , drying out and irreversible damage to the culture.

By adding suitable fillers , the tolerable layer thickness and thus the space-time yield of the reactor can be increased.

A general distinction is made between stationary and unsteady bioreactors. The stationary reactors include Petri dishes , Fernbach flasks, wooden incubators , the Covered Pan bioreactor and column SSF bioreactors (English Solid State Fermentation , SSF, solid fermentation). The unsteady reactors include the rotary drum SFB, the stirred SFB and the tumbling SFB.

history

While solid fermentation has been neglected in western countries since its late discovery at the beginning of the 20th century, it experienced a steady upward trend in the Far Eastern countries after World War II towards fully automated industrial processes with an estimated market volume of $ 8-10 billion /Year. The SSF is not only of interest for classic East Asian applications, such as the production of enzymes, foodstuffs and condiments, but also for

The classic method for mass culture of molds by SSF has been for millennia, especially in Far Eastern countries, the stationary culture in perforated trays, which are filled a few centimeters high with the inoculated substrate .

In the East Asian countries, especially Japan, the SSF was scaled up to the fully automated industrial scale with capacities of more than 100 t of inoculated substrate per SSF bioreactor (SFB). With a substrate capacity that is five to seven times larger, this corresponds to the performance of a submerged bioreactor with a working volume between 700 and 1000 m³. The recognition of significant economic and procedural advantages of SSF compared to submerged fermentation (STF) led, especially in the USA since the 1970s, to a revival of research activities in the field of SSF.

However, the breakthrough to large-scale use of this comparatively simpler and more cost-effective process depends on the availability of automated SSF bioreactors on an industrial scale, that is, with capacities above 10 t of inoculated substrate.

The following is an overview of the SSF bioreactors used to date. Depending on whether the inoculated substrate is moved during the fermentation or not, they are divided into stationary and transient processes.

Stationary SSF bioreactors

In the stationary processes of the SSF, the substrate is either not at all or is only circulated once or twice in the middle of the fermentation.

Petri dishes

Petri dish with culture

Petri dishes with 20–30 g of moist substrate in a thin layer are the first step in the cultivation of mold on solid substrates. The inoculation takes place from an agar plate by transferring sterile substrate to the sporulated culture . The plate is shaken to distribute the spores and aerial mycelium evenly on the surface of the substrate particles. The Petri dish is incubated under suitable conditions in incubators until sporulation. The densely grown, compact cake is dried and ground and serves as an inoculum (the amount of cells with which a fermenter is inoculated) for cultivation in larger, flat dishes with around 200 g of inoculated substrate. All manipulations are carried out under aseptic conditions. This method is simple, safe, and inexpensive.

Fernbach flask

Fernbach flasks are used exclusively for SSF on a small laboratory scale. For example, 70 g of wheat bran are soaked and sealed with cotton stoppers or caps. The flasks are autoclaved . The inoculum is mixed with the substrate by vigorous shaking, and the flasks are then incubated in a stationary manner. Petri dishes and Fernbach flasks are only suitable for the first steps in the laboratory due to the lack of a means for ventilation, humidification, removal of excess CO 2 and the resulting low substrate capacities. Both methods are relatively labor-intensive and unsuitable for scale-up .

Wooden incubators

In order to increase the substrate capacity, wooden incubators with a volume of 27 L for holding 4.5 kg of wheat bran with a layer thickness of 25 cm were developed in the USA since around 1938 and during the Second World War to produce aflatoxin production on the intended large scale. The bioreactor had devices for ventilation with reversal of the direction of flow, temperature and pressure display and temperature-controlled air humidification.

The results obtained with these bioreactors were worse than those obtained with the flat layer. The causes were uneven aeration, ventilation, humidification and temperature gradients of 30 to 50 ° C due to preferred flow channels within the pack. The results were viewed as the basis for further developments. Attempts to scale-up to the semi-industrial scale with substrate capacities of up to 1 t gave even more moderate results. After all, the growth was reproducible at a much lower level than in a thin layer. Attempts to increase the layer thickness while reducing the diameter at the same time failed completely. The results with increased layer thicknesses were so clearly negative, the causes of the failures so obvious that, looking back, as early as the 1940s and even more later, further attempts to increase the capacity rate by increasing the layer thickness would have been superfluous.

Covered Pan Bioreactor ( Pot Method )

In the early 1940s, large amounts of inoculum were required for the SSF in rotary drum bioreactors. As an alternative to the wooden incubators, closed, flat pans measuring 61 × 61 × 10 cm (total volume approx. 37 l) have been developed. The air supply took place in the center of the lid, the exhaust air outlets in the four corners originated from the depth of the substrate package . This should ensure a forced flow through the substrate. The capacity was approximately 4 kg of inoculated substrate. The layer thickness could not exceed 1 cm, otherwise oxygen limitation would take place. The capacity rate (w / v) was less than 10%. This method was in rotating rotary drum bioreactors ( Rotating Drum Bioreactors ) superior growth, enzyme production and sporulation were far more intense. The facility was less expensive to operate than the rotary drum bioreactors. In addition, the mycelium was not subjected to constant shear stress during growth, as in the rotary drum bioreactors, which led to a strong retardation of growth with significantly longer fermentation times and significantly reduced enzyme yields.

A smaller version consisted of a round aluminum - saucepan (diameter 15 cm and height 11,3 cm, total volume of approximately 2 l) whose bottom perforated was. The capacity was around 750 g of moist wheat bran (capacity rate 38%, w / v). The bioreactor had devices for ventilation, reversing the direction of flow and temperature control. Moistening of the substrate, sterilization, inoculation with Aspergillus oryzae (0.6–1%, w / w), incubation (30 ° C) etc. were carried out in situ. Ventilation, ventilation of excess CO 2 , humidification and temperature control were carried out with humidified, temperature-controlled air. Comparisons of this type with rotary drum bioreactors revealed a number of advantages for the stationary process: less space required, more uniform growth, better reproducible and higher enzyme yields, no devices for circulating the fermenting substrate. Here, too, the negative effect of shear stress in rotary drum bioreactors on growth, yield and process duration was emphasized. A disadvantage of the Covered Pan Bioreactors, however, was that this method was not capable of being scaled up due to uneven ventilation when the scale was increased, it was too labor-intensive and therefore not suitable for industrial SSF.

SSF in flat trays (Shallow Layer Tray Method)

The traditional Koji process for the production of enzymes on soybeans in Japan uses flat trays (layer thickness up to approx. 4 cm) made of wood, aluminum or stainless steel with a perforated base (holes approx. 3 mm in diameter at intervals of approx. 2.5 cm), in which the substrate inoculated with Aspergillus oryzae is incubated for about 3 days / 30 ° C. The trays are stacked on top of one another in air-conditioned rooms with clear gaps of around 10 cm. The layer thickness of the inoculated substrate must not exceed 4 cm in cultures with a fermentation time of more than 48 h.

In the case of rapidly growing cultures (15–24 h), the maximum reasonable layer thickness is 2–3 cm. With higher layer thicknesses, the outer (top and bottom) 1-2 cm grow perfectly, while within them the temperatures reach such high values ​​(up to about 57 ° C) that the mycelium is irreversibly inactivated and autolysed after a small initial growth. Laboratory and pilot-scale trays are usually handled manually. For weight reasons, they therefore have dimensions of around 50 × 60 × 2-4 cm³ (6-12 l volume) with a substrate capacity of 4-8 kg of inoculated substrate, depending on the layer thickness. With steel sheets (18/10), the total weight after filling is between 8–12 kg. Trays for commercial production have dimensions of about 150 × 500 × 2-4 cm³ (150-300 l volume) with a capacity of up to about 200 kg of inoculated substrate. These trays are automatically filled, emptied, cleaned and sterilized. After inoculation, the trays, inclined preferably 10-15 ° to improve ventilation, are stacked in stacks with vertical clearances of down to 2 cm depending on the ventilation and are incubated under defined conditions (temperature, humidity, ventilation) . Regardless of the desired product, there is usually no need to monitor other factors. The main advantages of the tablet method are its simplicity, immunity to interference, excellent results and reproducibility. The fully automatic Japanese systems, some of which have been in operation since the 1960s, have capacities of around 30 to several hundred tons of substrate per day with continuous process management.

Modern SSF bioreactor on a pilot scale

A modern SSF bioreactor for the incubation of flat trays has a variable substrate capacity depending on the number of incubated trays from <1 to 240 kg. The trays contain the inoculated substrate in a thin (2-4 cm) layer. They are inclined at around 10 ° to improve ventilation, in racks with clear vertical distances between the trays of around 2 cm. Defined temperature-controlled and humidified air flows in a directed forced flow through a side opening in the bottom area of ​​the reactor interior over the top and bottom of the trays. In the headspace of the bioreactor, the air flow is sucked in by a fan and fed to the floor area in a loop on the rear of the bioreactor in a downward forced flow. It passes large heat exchangers and freshly conditioned air is supplied.

The maximum capacity rate is 25-30% (w / v = 70%, volume fraction ). Because of the variable substrate capacity of this SSF bioreactor, the results obtained with, for example, 200 g substrate can be easily transferred in one step to its full capacity of 240 kg. Excellent results are obtained with comparatively significantly shorter fermentation times. The price / benefit ratio (Eur / kg fermentation capacity) falls, depending on the equipment, from smaller to around a quarter of the costs for bioreactors with larger capacity. 

Automated systems (conveyor bioreactor)

Automatic systems were developed for the production scale decades ago in the USA. They include all steps of a continuous SSF: soaking the substrate, sterilization , cooling, inoculating, distributing on the trays, stacking the trays, incubation, downstream processing , cleaning and sterilizing the cleaned trays. Nothing significant has changed in this principle to this day.

Industrial plants

Industrial, automated systems have been in operation in Japanese companies for the production of enzymes, foodstuffs and condiments since the 1960s. Continuous cookers, automatic inoculators, mixers, rectangular (5.4 × 60 × 0.6 m³) tubs with perforated walls and bottom (total volume approx. 190 m³) with capacities of up to 100 t of inoculated substrate / tub and batch, set up in closed Rooms with automatic forced ventilation, temperature control and devices for the occasional circulation of the fermenting substrate. The inoculated mass is spread onto the perforated trays and conditioned air is circulated through the substrate. After 48 to 72 hours, the finished koji (Aspergillus oryzae) is harvested and the enzymes extracted. Since the circulation to break up the aggregates leads to growth retardation of several hours, this process is limited to once or a maximum of twice during the fermentation period. The delay can be estimated at 8–12 hours.

Aspergillus oryzae produces enzymes within a relatively broad, predominantly suboptimal physiological spectrum (26–42 ° C in the package, 35–52% {w / w} substrate moisture , 10–100% saturation of oxygen requirements , CO 2 levels of up to 21% in the atmosphere). In this respect, enzymes in SSF with Aspergillus oryzae can also be produced in a relatively high layer (if necessary, suitable spacers are mixed in) with a correspondingly suitable process management with good yield.

In contrast, the optima, in particular the optimal degree of saturation of the oxygen demand, for the production of other metabolites are much narrower. This limits the layer thicknesses to avoid gradients more than in the case of enzyme production.

SSF in high layer (high heap method)

Various efforts have been made to increase film thickness in order to improve the capacity rate. This was done up to a layer thickness of 1.5 m. In one of the first SSF processes with a large layer thickness (high heap), the moist, sterilized material was mixed with porous fillers such as rice husks, inoculated and placed in a container with a perforated bottom through which conditioned air was forced.

For the production of citric acid with Aspergillus niger , diluted molasses was mixed with rice or wheat bran , heated to 90-95 ° C for 30 minutes, cooled to below 40 ° C and, after inoculation, mixed with pre-sterilized sawdust or rice husks to create a porous structure . Moisture-saturated, temperature-controlled (30 ° C.) air was forced through the packing. During the lag phase, the substrate was heated to 33 ° C. with water passed through cooling coils. The yields of citric acid were higher in SSF than in surface culture or in submerged fermentation (STF).

A laboratory SSF bioreactor with a square floor plan (30 × 30 × 25 cm³, 22.5 l total volume), perforated base and cover for supply and exhaust air, with a substrate capacity of around 9 kg (capacity rate 40%, w / v) gave satisfactory results, although not comparable with the tablet method. Conditioned compressed air was required to ensure the fresh air, moisture, ventilation of CO 2 and oxidation heat required during the SSF . As growth progressed, the interparticle spaces clogged with mycelium until growth came to a standstill despite the increase in the pre-pressure. In spite of the moderate results compared to the results of the tablet method, a scale-up of this type was attempted with a further increase in the layer thickness. However, the results turned out to be a complete failure. There were various reasons for this: The main reason was that the pressure loss (> 1 bar) with progressive growth, in particular due to the relatively long diffusion paths, could not be compensated despite the increase in pressure, which meant that ventilation, humidification and ventilation could not be maintained within an adequate range . The fermenting substrate then overheated and dried out in a short time, and the mycelium was irreversibly damaged. Satisfactory growth was only found in an area spatially limited to the immediate vicinity of the air inlet opening.

These clear failures proved that without the aid of porous spacers an increase in the layer thickness only led to satisfactory results within very narrow limits. This was repeatedly confirmed involuntarily in the later decades. The fixation on increasing layer heights, perhaps in analogy to the successful results of submerged fermentation, did not deter the scientists of the time.

And so a cylindrical, upright SSF bioreactor was created, consisting of three concentric, ring-shaped compartments with perforated dividing walls. In the center there was a perforated air inlet tube, surrounded on the outside by the compartment for receiving the inoculated substrate. The direction of flow ran horizontally through the fermenting substrate in a relatively thin layer (approx. 9 cm). The exhaust gas left the SSF bio-reactor through the peripheral, annular space. The height of the area filled with substrate was 90 cm.

Although this type of SSF bioreactor is generally discussed in connection with the high heap process , it is obvious that this is more of a modification of the tray process with forced flow by increasing the pressure.

The scientists had basically recognized the futility of their earlier attempts with the high heap principle and turned back to the SSF in a thin layer ( shallow layer ) with this variant . The capacity of the bioreactor was 24 kg of inoculated substrate. The device could be easily filled and emptied through removable plates on the top and bottom. However, cleaning the extensive perforated surfaces proved to be very labor intensive. Added to this were the considerable temperature gradients (around 10 ° C) of the fermenting pack between the inlet and outlet side. These were just within the tolerable range for growth and enzyme formation by Aspergillus oryzae.

From the point of view of today's optimized fermentation process, however, such gradients appear unacceptable. Even then, the results obtained with this SSF bioreactor could not compete with those obtained in a flat layer (tray method). In SSF bioreactors of the tray method, temperature gradients of 3–4 ° C are achieved under full load, which are within the various product-related optima. In any case, Underkofler concluded in 1947 from the mediocre results of his high-heap attempts that it would be foolish to try to increase the scale with this principle.

Inclined high heap SSF bioreactor

The inclined high-heap SSF bioreactor was inclined at an angle of 45 ° from the vertical position in order, as Underkofler et al. argued to make filling and emptying easier. These types of reactors were much larger than the aforementioned vertically oriented type. However, a scale-up to an industrial scale also failed with this type.

Column SSF bioreactor

A column SSF bioreactor consists of a glass or plastic column with connections for ventilation. The temperature is controlled via a double jacket with heated water, provided the column is not in a temperature-controlled room such as an incubator or similar. With such bioreactors, diameter 10 cm, height 150 cm, total volume about 12 l, with a substrate capacity of 8 kg (humidity 22%, w / w), attempts were made in the USA at the beginning of the 1940s to mass- produce aflatoxins by fermenting wheat , Oat husks , oat groats and maize with Aspergillus flavus or Aspergillus parasiticus. The humidity was checked by means of humidified (80-85% relative humidity) air. The fermentation lasted 10 days at 33 ° C. The aflatoxin levels were more than 1 g / kg fermented substrate. These results agree with those of Hesseltine insofar as during most of the SSF there must have been an acute oxygen deficiency with a considerable excess of CO 2 and in some places overheating in the fermenting pack. In contrast to the results of Hesseltine, who worked in vigorously shaken flasks without aeration, the maximum production of aflatoxins was achieved in static culture. This example shows the poor comparability and transferability of the results obtained with different SSF culture methods.

A battery consisting of 24 small column SSF bioreactors was used by Raimbault and Alazard. Column SSF bioreactors are only suitable for laboratory scale because of their disproportionately long diffusion distances and the associated deficiencies in aeration, humidification and ventilation. The results are not transferable. In the case of column SSF bioreactors in particular, only a slight pressure drop across the column can often be observed despite considerable gradients in the ventilation, ventilation and temperature control. This is due to the formation of preferred flow channels within the packing. In these channels the rapidly flowing air is heated and dries out the areas surrounding the flow channel. This prevents the canal from overgrowing. In contrast, the immediately adjacent areas that have grown over with mycelium are inadequately ventilated and ventilated. Most of the packing in the column is therefore far outside of optimal fermentation conditions . The problems with the columnar SSF bioreactors are exacerbated the same as discussed for the high heap type .

Plastic bags

Plastic bags with 2.8 or 40 kg of wheat straw as substrate were fermented under suitable conditions. About the same process was propagated six years later by Hesseltine for the production of tempeh , an Indonesian staple food , on soybeans. Tempeh can be characterized as a vegetarian schnitzel with antioxidant properties, positive effects on the biosynthesis of hemoglobin , increased levels of vitamin B2 , vitamin B6 , vitamin B12 , pantothenic acid , niacin and tocopherol . It is classified as a health food in the American diet . In accordance with Hesseltine's recommendation to simplify the SSF process, the moist, sterilized and, after cooling, inoculated soybeans are filled into perforated plastic bags, flattened, shrink-wrapped and incubated at 30–32 ° C in normal incubators. The temperature inside the pack is around 37 ° C. As a result of the lack of oxygen and the relatively high humidity inside the pack (saturation with condensed oxidation water), the growth of Rhizopus oligosporus or other closely related species is greatly delayed with a duration of 48-72 hours compared to a culture in flat trays (relative humidity of the atmosphere about 95%, of the fermenting substrate 48–52% {w / w}, duration 15–18 h). The lack of oxygen and the relatively high humidity of the substrate also promote the growth of facultative anaerobic bacteria - in addition to the Klebsiella pneumoniae or Citrobacter freundii undesired enterobacteria  , which are required for the production of vitamin B12 .

In any case of defects in industrial hygiene occurs in the nutrient-rich substrate because of the short doubling times (about 30 min), the undesired bacteria by selective enrichment in severe contamination (3 * 10 8 microorganisms / g of product) with coliforms , other fecal bacteria and Pseudomonas , such as showed itself in the natural food sector and in the administration of health food in clinics to end-stage cancer patients because of its suspected anticarcinogenic ingredients.

These examples show that even the relatively easy-to-operate SSFs require a certain amount of training and a sense of responsibility. As the above explanations show, there is a high risk with the plastic bag method because of the process-related, selective accumulation of undesirable, possibly human pathogenic microorganisms. Such risks are not to be feared when cultivating in flat trays.

Conclusions

In the previous sections, inpatient SFBs were discussed. It was shown that layer heights of the fermenting substrate above approx. 4 cm lead to limitations in oxygen supply, ventilation of excess CO 2 , intolerable temperature gradients, drying out and, in the worst case, irreversible damage to the culture. Numerous attempts to increase the layer thickness of the fermenting substrate, perhaps analogous to the successful results in submerged fermentation (STF), to over 1.50 m, have led to poor results over the past five decades. Underkofler, one of the great old pioneers of the SSF in the USA, came to the conclusion after extensive investigations that the attempt at an industrial SSF in a high, static layer was foolish. He hoped for a way out of the gradient dilemma of high fermenting substrate layers through their regular circulation.

Unsteady systems

Despite numerous publications on unfavorable results in solid fermentations (SSF) with stirred, shaken or rotating SSF bioreactors, especially when trying to transfer results from the laboratory to the pilot or production scale, there have been isolated reports up to the last few years about successfully run SSF with moving substrate. It is therefore the object of this section to critically evaluate the apparently contradicting results of the last seven decades or so.

A circulation of the fermenting substrate is mentioned by various authors. In the case of greater layer thicknesses, they reported a more even distribution of the inoculum, more even growth than would otherwise have been achievable because of the extreme gradient formation in the higher layer. Aggregates were broken open, gas and heat exchanges were improved. The speed of the circulation was proportional to the oxygen consumption. However, the reported fermentation times were significantly longer and the growth significantly weaker than with stationary SFB in a flat layer.

Other authors, on the other hand, reported that the effect of the substrate upheaval was in no way promoting growth. Underkofler et al. experienced difficulties with insufficient growth when scaling up their rotary drum bioreactors from the 20 L scale onwards. The cause was a destruction of the sensitive mycelium by the shear forces occurring during circulation; the larger the rotary drum bioreactor, the greater the difficulties. Extracted sugar beet pulp , impregnated with sucrose for conversion to citrate by Aspergillus niger, gave unsatisfactory results when the SSF took place in slowly rotating rotary drum bioreactors.

In contrast, Takamine achieved relatively good results with his Rhizopus fermentations in rotating drums. Takamine admitted that the mycelium on the surface of the substrate particles was damaged by shear forces and explained his satisfactory results by the fact that the mycelium growing on the concave areas of the substrate particles was protected from shear forces. The devastating effect of shear stresses on mycelial growth has been confirmed by further work. 

It should be mentioned at this point that the humidity of the substrate for SSF in rotating or stirred SFB must be set to significantly lower values ​​than for stationary ones in order to avoid clumping of the substrate during the circulation.

Kunz & Stefan used a small (1.4 kg capacity) SFB whose carefully tumbling, three-dimensional movement, according to the inventors in SSF with Monascus purpureus, largely avoids clumping of the moist substrate. SSF with Monascus purpureus on peeled rice are a special case insofar as this fungus only provides acceptable pigment yields with a comparatively low substrate moisture content (25–28%, w / w). Kunz & Stefan had around 30–35% (w / w) in their "SWING" -SFB. Under these conditions, none of the unsteady SFBs tested by the author (rotary drum SFBs with or without baffles, stirred SFBs with flat, rounded or helical stirring elements, the SFBs had 4-16 kg substrate capacity) did not clump the substrate, which appeared dry on the outside. It remained free-flowing during the entire fermentation period. The peculiarity associated with the "SWING" -SFB of a lump-free substrate turnover would still have to be confirmed in cultures with, for example, Rhizopus or Aspergillus, which, if the fermentation times are not to exceed any acceptable level, at significantly higher substrate moisture levels (45-60%, w / w) must be operated. With a larger "SWING" -SFB (200 kg substrate volume), practical SSF experience has so far been lacking.

However, similar to Underkofler's work, it can be expected that with the increasing intensity of the shear forces due to the pressure of the loading substrate pack, difficulties similar to those described for the Rotary Drum - SFB, which have made it impossible to scale up to production scale, will arise.

A comparison of stationary SFB (tray method) with the transient systems shows that the fermentation time in the transient SFB is much longer. Apart from the fact that the investment costs for unsteady SFB (approx. DM 5000–3000 / kg substrate capacity) are considerably higher than for stationary ones (DM 2000–300 / kg substrate capacity - the value decreases with increasing capacity), the delay caused by circulation is the From an economic point of view, SSF can only be justified if the higher process costs are offset by higher throughput, higher yield and quality.

From the information in the literature and from our own experimental experience, however, the opposite must be concluded. One exception should be mentioned, however: the industrial production of enzymes by SSF in Japan. There, 100 t substrate batches are briefly circulated after about half of the fermentation period. The delay caused by this (in industrial strains from around 24 in shallow layer to> 140 hours) seems to be justifiable in view of the broad physiological range in which, regardless of the strong gradient formation, acceptable enzyme production takes place.

Methods of substrate turnover in SSF

When difficulties arose in keeping the temperature, humidity, ventilation and removal of CO 2 constant when the layer thickness of the fermenting substrate was increased, substrate circulation appeared to be the method of choice to eliminate these problems, as in the STF. The benefits associated with substrate upheaval should be:

  • even distribution of the inoculum
  • more even growth on the substrate particles
  • Prevents the formation of aggregates
  • improved ventilation of the individual substrate particles
  • Facilitation of gas exchange
  • Improvement of the heat dissipation
  • Avoidance of local gradients
  • Guarantee of homogeneous conditions during the entire fermentation period

According to these ideas, the rate of circulation (similar to the STF) seemed to be in direct correlation with the oxygen demand of the culture. These authors found that the results in the moving SFBs were worse than those in the stationary SFBs (tray method). The different concepts of substrate turnover during SSF are discussed below.

Rotary Drum SFB

construction

The construction of a rotating drum SFB - this type is mainly of historical interest - corresponds to that of a rotating hollow cylinder , mounted on rollers, which also drive the hollow cylinder. As an alternative to this, a drive based on a clockwork (with low gear ratio) was used via a central shaft. The speed of rotation was usually 1-5 / min, but in other cases 6-16 / min. The containers were 19 L glass balloons or iron kettles, mortar buckets, barrels with a capacity of around 215 L, concrete mixers with capacities of 70 and 114 L, and pressure vessels with capacities of 0.1, 1, 5, 20, and 70 kg up to industrial scale . The Rotary Drum SFB were provided with connections for ventilation. The air inlet tube reached almost to the bottom of the drum. In another embodiment it branched several times to whorls , the ends being designed as nozzles. The ventilation was carried out by applying negative pressure or compressed air. The fresh air was passed through sulfuric acid for sterilization and then through sterile distilled water for humidification . The preparation of the substrate (soaking, steaming, inoculation, incubation and drying) took place in situ .

Bullying Rotary Drum SFBs were found to be superior to their unreinforced counterparts. For series tests, the interior of the drum was divided axially into three to four separate compartments, each with baffles. In the small Rotary Drum SFB, growth was rapid and steady. Difficulties arose with the scale-up because the sensitive mycelium was crushed during early growth by the shear forces of the particles sliding against each other. This problem increased with the size of the Rotary Drum SFB. In addition, other factors such as:

  • Temperature control
  • Microbial contamination
  • Clumping of the substrate
  • Inhibition of growth due to mechanical attrition of the mycelium

than cannot be solved with Rotary Drum SFB.

Laboratory scale

Rotary Drum SFB gave satisfactory results (the authors emphasized several times that these could not keep up with those in stationary SFB {tray method}) as long as the load in a 20 l tank did not exceed 1–2 kg. The reasonably usable substrate capacity was therefore less than 10% (w / v). Air was slowly introduced. The fermenting substrate was tempered as far as possible in a range around 30 ° C. Under these conditions the temperature inside the fermenting substrate reached about 37 ° C. If the outside temperature reached 35 ° C, the temperature inside the substrate rose to over 42 ° C, which led to irreversible damage to the culture.

During the germination of the inoculum, i.e. 12-15 hours after inoculation, the drum was rotated periodically (every two hours for 15-20 min each time). After germination, the drum was rotated continuously for 40-45 hours (1 / min). Before the start of sporulation, the substrate was removed, dried, ground and used as an enzyme mixture for the hydrolysis of corn starch .

Comparative studies on the production of Koji (mushroom enzyme mixtures) for the production of soy sauce (Shoyu) with the Rotary Drum SFB and the stationary tray method were to the disadvantage of the Rotary Drum SFB. Although (due to the lack of amylases , proteases etc.) the analytical and sensory values ​​of the Shoyus produced with the "Rotary Drum Koji" did not come close to those produced with the tray method, the authors diplomatically assessed the result, which was disappointing for them, as "satisfactory" . The SFB circulated by rolling (4 / min) was a 20 l glass vessel, inclined 20 °. The maximum useful payload was 1.8 kg of a mixture of soybeans and wheat flour (humidity 43–49%, w / w). This corresponded to a substrate capacity rate of only 9% (w / v). The inoculum ( Aspergillus oryzae ) was 1010 spores / 1.8 kg (approx. 1%, w / w). The fermentation temperature was 30 ° C. Humidified air (0.05 vvm) was blown in. Even with this small batch size, it was not possible to keep the temperature in the fermenting substrate below 39-42 ° C. during the strongest growth (40 hours after inoculation). This problem worsened with a larger load of substrate. The fermentation time was 72–74 hours (compared to 24–28 hours in modern SFBs with the stationary tray method).

The attempts to scale-up the Rotary Drum SFB failed. The reason was that the mycelium was massively damaged by the shear forces. As a way out of this dilemma, Underkofler turned again to the inpatient SSF in high shifts. Another disadvantage of the SSF in Rotary Drum SFB was the low enzyme content. This disadvantage has also been attributed to the destruction of the growing mycelium by shear forces [18, 19]. In the opinion of other authors, however, the advantage of the higher substrate capacity rate (kg substrate / L volume, w / v) outweighed the advantage compared with the stationary tray method at that time.

Stirred SFB

In particular, on the laboratory and pilot scale, there is a great effort to record a large number of measured variables with probes as continuously as possible in order to better understand and control a fermentation process. Connection and cabling of probes etc. cause considerable difficulties with Rotary Drum SFB if ​​the installation cannot be carried out in the area of ​​the rotation axis. Instatic SFBs that do not have this disadvantage are those with installed agitators or SFBs that perform tumbling movements. In view of a possibly necessary scale-up to the production scale (>> 1 t / batch), an SFB that executes tumbling movements is less of a consideration.

At the beginning of the eighties, SFBs with installed agitators (substrate capacity 4 or 20 kg) had connections for all possible operations. The substrate was circulated continuously or in a programmable, discontinuous manner by self-wiping agitators with a variable speed of rotation (0.5-30 / min and reversible direction of rotation). All process steps (soaking, draining the excess soaking water, sterilizing up to 138 ° C with pressurized steam, inoculating, tempering, ventilation, humidification, sampling, if necessary drying of the fermented product, etc.) could be carried out in situ. The substrate capacity rate was 32% (w / v) or approximately 80% ( volume fraction ). This type of SFB delivered results that came close to those obtained with the tablet method. However, the SSF lasted significantly longer than in the stationary SFB. The reasons for this lay in:

  • Because of the greater thickness of the layer, it is occasionally necessary to stir the substrate to break up aggregates, usually for a few minutes during half-time
  • lower humidity (25-35%, w / w) to which the substrate had to be adjusted in order to avoid clumping of the fermenting substrate during circulation

A disadvantage of this type of SFB, however, was the relatively high costs, around DM 4000 / kg for a laboratory or calculated DM 1500-2000 / kg for a small production SFB. Since this concept was several times more expensive than a stationary SFB using the tray method, especially on a production scale, it was stopped. Regardless of this, these CRCs are still in industrial R&D activity after more than 25 years.

"Taumel" SFB

Another unsteady SFB type with three-dimensional tumbling movement (system with internal gyration = SWING), which, as the inventors emphasize, does not have the disadvantages mentioned for the rotary drum SFB, consists of a transparent plastic hollow cylinder with connections for loading and unloading Ventilation as well as for external temperature control. The device was originally designed and developed as a mixer, but as expected it was equally suitable for mixing free-flowing, fermenting substrate. The sterilized and inoculated substrate is placed in the vessel. The maximum capacity rate of this laboratory SFB (2 L total volume) is 28% (w / v). This corresponds to a filling of around 70% (volume fraction). Understandably, the device cannot be sterilized in situ. The container is sterilized by wetting it with 70% ethanol (Kunz {1997}, personal communication).

The inventors of the SWING mixer emphasize the outstanding feature of this mixer: the lump-free mixing of the rice inoculated with Monascus purpureus (approx. 30% moisture, w / w) during the SSF on a 1 to 1.4 kg scale. The practical test of the SWING concept in SSF, especially with substrate moisture levels of significantly more than 30% (w / w), is therefore pending. Although a 200 L mixer is allegedly available, there is a lack of relevant studies with SSF processes. The high investment costs, around DM 5000 / kg substrate (Doman (1997), personal communication) appear, analogous to the agitated SFB dealt with in the previous chapter, as a serious disadvantage that will probably stand in the way of economic operation on a production scale.

Industrial-scale Rotary Drum SFB

Large Rotary Drum SFBs have been operating in Japanese companies for the production of miso since the late 1950s . This can be seen as an indication that the difficulties mentioned in the previous chapters do not exist in Japanese companies or do so to a lesser extent. Miso production takes place semi-continuously and is largely automated. Rotary Drum SFB's are mainly used for the production of rice and barley koji (substrate-enzyme mixture according to SSF). Cooked and inoculated rice is filled into the large drum, in which tempered and humidified air circulates. The drum is rotated occasionally to avoid agglomeration of the fermenting substrate. The rotary drums have a diameter of around 3 m with an estimated capacity between 10 and 20 t, maybe more. They are battery-wise in large rooms. Undoubtedly, in these processes, the careful adjustment of the substrate moisture to relatively low values ​​to prevent the substrate particles from sticking to one another plays a decisive role.

Interpretation criteria

Various concepts for the layout and design of solid bioreactors are known. One of the types most frequently discussed in the specialist literature is the so-called "rotary drum reactor". This consists of a hollow cylinder lying on its side, in the interior of which microorganisms are to be cultivated on solid substrates. Since the substrate in this type of bioreactor is in a relatively thick layer, it must be circulated continuously or at regular intervals during fermentation in order to maintain adequate ventilation (dissipation of the heat of oxidation, removal of excess carbon dioxide, adequate supply of oxygen and moisture). A significant disadvantage associated with this, however, is that the fungal mycelium located on the surface of the particulate substrate is damaged by the shear forces that occur during circulation . As a result, the metabolic activity of the culture in question is greatly reduced, or it even comes to a standstill before it can gradually recover again after several hours without agitation. The rotary drum reactors have disappeared from the market because of this disadvantage.

Attempts were therefore made to cultivate dormant, static cultures with a large layer thickness (up to 1.50 m). However, this has the disadvantage that considerable diffusion resistances occur because of the high layer thickness . These can only be overcome by blowing in air or other suitable gas mixtures under pressure. In addition, increases with the progress of mycelial growth of the diffusion resistance, as a result of Zuwachsens the interparticle spaces so that before reaching the maximum possible growth of the process due to overheating, lack of ventilation and carbon dioxide lack collapses of oxygen supply and therefore must be stopped prematurely. In another variant, the SSF is carried out in a relatively thin layer on flat trays. This avoids the disadvantages described for the other two types, but there is another disadvantage; Because of the spaced-apart substrate distributed over many flat trays, the payload in relation to the total volume of the bioreactor drops considerably. Another disadvantage of the bioreactors previously used for culture in flat trays is that the heat of oxidation is compensated for by a corresponding intensity of the flow through the reactor space with humidified, sterile air. With increasing growth this leads to the necessary air exchange rates of up to more than 100 air changes / h in the reaction space, as is usual in practice only slightly lower temperature of the supply air. However, this creates overflow velocities at which the mycelium located on the surface grows phenotypically in the manner of weather spruce with significantly reduced metabolic activity. However, this is undesirable in the interests of optimal process management. Another disadvantage of the bioreactors previously used for culture in flat trays is that, particularly in larger systems under full load and intensive growth, considerable amounts of odor-intensive and germ-contaminated exhaust air have to be disposed of. Another disadvantage of the bioreactors used for culture in flat trays is that the steam for humidifying the blown air or the gas mixture is provided in thermal steam generators. This requires considerable efforts to cool the hot steam down to fermentation temperatures (25-35 ° C). When cooling down, however, a large part of the water vapor is lost again through condensation. This results in a relatively low degree of efficiency and the resulting oversized system for humidifying the air. Because of these disadvantages, as already mentioned, the humidified air to be blown is therefore generally cooled to a temperature which is only insignificantly below the desired internal reactor temperature. This results in a relatively low level of efficiency in the necessary dissipation of heat of oxidation.

An SSF bioreactor that avoids the disadvantages listed above is described below. It is characterized in that:

  • trays inclined at an angle of 10-15 ° are arranged in the interior
  • the culture is supplied with a defined gas atmosphere by a flow- or pressure-regulated gas supply device
  • the humidification of the tempered gas volume flow to a relative humidity between 40 and 99% is carried out by a device for generating cold water mist
  • Inside a device for air circulation is arranged, and the air or the gas mixture in a forced flow through baffles and with a low flow velocity from the lower side of the tray on the upper and lower side of the same sweeps to its higher side
  • to dissipate the heat of oxidation on the baffles, heat exchangers are provided, the temperature of which is 2-3 ° C below the desired fermentation temperature
  • A pressureless hot steam supply or temperature control to 180 ° C is provided for the sterilization of the SFB
  • A drain with a heat-sterilizable trap is installed to drain off condensate at the lowest point of the interior of the reactor
  • to discharge the exhaust air above the SFB there is a heat-sterilizable device for discharging, sterilizing and, if necessary, deodorizing the odorous and germ-laden exhaust air

The advantages for the process flow achieved by the aforementioned criteria are explained below:

For example, gas exchange and growth of an Aspergillus niger culture are significantly accelerated when the trays are inclined by 15 ° compared to other systems on the market: The growth is optimal after about 24 hours, compared to 48 to 72 hours in other SSF bioreactors. This improves throughput by over 100%. As a result of the improved ventilation, the clear gaps between the trays lying on top of one another can be reduced to 1 to 4 cm compared to the usual 5 to 15 cm, and the payload capacity is increased by 35%.

In cultures with Rhizopus oligosporus, the time to achieve maximum growth is 15-18 hours compared to 48-72 hours in systems with less efficient ventilation, which also have to be mixed at intervals to improve the same. This improves throughput by over 100%.

As a result of the controllable circulation of the gas atmosphere to avoid turbulent flows over the vegetated substrate, the speed of growth, the rate of product formation of secondary metabolites and the yield of biomass or product in cultures with Penicillium chrysogenum can be increased by about 25%.

In cultures with Rhizopus oligosporus, the design of the interior with plate-shaped, flat heat exchangers and baffles to generate a directed forced flow leads to a reduction in the circulating power required for adequate ventilation and the necessary fresh air supply by around 50% in each case compared to other systems.

The metered supply of humidified fresh air to compensate for the oxygen consumed by a culture of Aspergillus oryzae and to ventilate the CO 2 produced results in a reduction in the odor-intensive and germ-contaminated exhaust air volume by around 80% compared to the usual fresh air supply. This results in a significant reduction in costs for the disposal of the exhaust air and a corresponding reduction in emissions .

The metered addition of humidified oxygen to optimize the oxygen supply for the production of secondary metabolites in a culture of Penicillium chrysogenum leads to a yield increase of 45% compared to a pure variation of the fresh air throughput when the oxygen demand of the culture is saturation of around 80%. The growth and the substrate yield coefficient of various molds is promoted in a defined gas atmosphere consisting of N 2 , O 2 and CO 2 by partial pressures of CO 2 between 3 and 5% by around 25% compared to corresponding, optimally aerated cultures.

The feeding in of cold water mist results in an energy saving of 70% compared to the usual feeding in of cooled down hot steam. As a result of the reduced heat input described in the previous example, the payload with a culture of Aspergillus oryzae can be increased by a factor of 2.5, i.e. H. by 250% compared to other SSF bioreactors.

The formation of extensive, flat, plate-shaped heat exchanger surfaces instead of simple baffles to generate a directed air flow enables efficient cooling within narrow temperature limits of +1 ° C when the maximum temperature is exceeded, in which the temperature of the heat exchanger surfaces is only about 2 ° C below the desired fermentation temperature. In this way, undesired drying out of the gas atmosphere is largely avoided. As comparisons of cultures with Aspergillus , Penicillium , Rhizopus , Mucor , Sporotrichum , Thermoascus , Monascus etc. with those in conventional SSF bioreactors showed, growth or product formation occurs at maximum capacity (240 kg fermenting substrate) due to the improved maintenance of temperature and Humidity within the narrow, optimal range (+1 ° C or +1% relative humidity) increased by 20 to 30%. As a result of the more efficient humidification of the atmosphere compared to thermal vaporization systems in other SSF bioreactors, the humidification capacity can be reduced by around 60% and the relevant installation costs by around the same percentage.

As a result of maintaining the relative humidity, a very critical variable in this special case, within relatively narrow limits (95 + 1%), the fermentation time of Monascus purpureus to produce the purple pigment is shortened to six days, compared to 9 to 21 days in others Systems.

Cultures with Sporotrichum thermophile and Thermoascus aurantiacus at 51 ° C in SSF to generate cellulases result in a significantly higher productivity per volume of medium with higher temperature stability of the enzymes than in STF with the same strains. The fermentation time in SSF until sporulation is 24 hours for Sporotrichum thermophile and 28–30 hours for Thermoascus aurantiacus.

Investigations to simulate the microbially caused and economically not insignificant self-heating of the hay were carried out up to a temperature of 95 ° C. In these experiments, the temperature of the bioreactor was adjusted to that in the loosely packed hay in order to minimize the heat losses from the hay pack to the environment. Humidity and air supply were controlled. In this way, the maximum growth of thermophilic microorganisms ( Bacillus stearothermophilus and Thermoactinomyces vulgaris ) and the start of the so-called pyrogenic phase (from approx. 95 ° C) were reached after 2-3 days. This means that the SSF bioreactor described allows a simulation of the microbial processes during the self-heating of hay that is closer to the conditions in barns (2-3 days) than the devices specially designed for this purpose (7 to 13 days, lack of defined humidity and defined supply of air or oxygen). Such investigations are not possible with other SSF bioreactors due to their limited working areas.

The lignin degradation of milled, pelleted wood flour, which is loaded with the usual mineral nutrients, by Phanaerochaete chrysosporium is 70% after 6–7 days, compared with periods between 30 and 60 days in other SSFs until a comparable degree of degradation is reached. It should be noted that in this case the preparation of the substrate also has a significant influence on the shortening of the fermentation time.

Future aspects

At the end of his experiments with the Rotary Drum - SFB without having achieved anything, Underkofler returned to the SSF in a thin layer: The SFB with moving substrate could not keep up with the stationary SSF (tray method) in terms of product quality. The western experience with transient SFB in the university area has remained limited to the laboratory scale (20 kg substrate). In some western companies, however, systems run with 10 to 100 times the capacity.

In Japan, however, industrial plants with capacities of 10 to 100 t of inoculated substrate / batch have been in operation for the production of enzyme mixtures for decades. These systems have devices for briefly circulating the fermenting substrate after about half the fermentation period. Soaking, draining , sterilizing and inoculating the substrate are done outside of the SFB. The processes take place under semi-sterile conditions. The fermentation times specified for Koji of around 50 hours are significantly higher than those in stationary SFBs (tray method, 24–28 hours), the enzyme yields significantly lower. As the production of enzymes by Aspergillus oryzae takes place within a wide physiological range (temperature 20 to 41 ° C, humidity 25 to 60%, w / w), ventilation, the gradients occurring in the large rotary drums are in the sense of a still satisfactory enzyme production tolerable. However, because of the considerable gradients, such SFBs appear to be unsuitable for the large-scale production of secondary metabolites , the production optima of which are much narrower. Therefore, if the SSF is to be used to search for previously unknown metabolites or metabolites that can be prepared in low yield, SFBs that are designed according to the tray method seem to be most suitable.

Interesting research programs for the future as well as promising applications of the SSF could be: 

Research and Development

  • Development of mathematical models for the reliable prediction and optimization of SSF processes
  • Investigations into the transfer of substances, the dynamics of growth and the absorption of swollen substrates by mold
  • In-depth study of the key factors for SSF
  • genetic engineering optimization of industrially valuable strains
  • Research into the uses of yeast and bacteria
  • Testing with regard to previously unused areas of industrial microbiology
  • Testing of mixed cultures except for enzyme production
  • Development of a fast, simple and reliable method for the determination of the microbial biomass in SSF
  • Physiological differentiation of airborne mycelium and mycelium penetrating into the substrate with regard to their different ability to produce secondary metabolites
  • Determination of the factors for the metabolic synchronization of air and penetrating mycelium

Commercial Aspects

  • Development of inexpensive automated systems for the pretreatment of the substrate
  • Production of individual enzymes instead of complex enzyme mixtures, as was previously the case
  • Modification of food (appearance, texture, taste, nutritional value) by SSF or microbial enzymes
  • Improvement of microorganisms used in the production of food
  • Production of secondary metabolites, mycotoxins , novel antibiotics , other active ingredients
  • Production of fungal spores
  • Transformation of organic compounds
  • Production of flavors, fragrances
  • Production of microbial agents against plant pathogens
  • Protein enrichment of carbohydrate-rich agricultural raw materials / waste for use as animal feed
  • Removal of persistent compounds from the soil

Individual evidence

  1. ↑ a b c d e f Hesseltine, CW (1977). Process Biochem. 12 No. 6, 24-27.
  2. ↑ a b c d J. Takamine. Ind. Eng. Chem. 6, 1914, 824-828.
  3. ↑ a b c Lindenfelser, LA, Ciegler, A. (1975). Appl. Microbiol. 29, 323-327.
  4. ↑ a b c Knapp, JS, Howell, JA (1980). In: Topics in Enzyme and Fermentation Biotechnology, A. Wiseman, ed., Vol. 4, pp. 85-143, Ellis Horwood Ltd., Chichester.
  5. ↑ a b Cahn, FJ (1935). Ind. Eng. Chem. 27, 201-204.
  6. ↑ a b c d e Aidoo, KE, Hendry, R., Wood, BJB (1982). Adv. Appl. Microbiol. 28, 201-237.
  7. ↑ a b c Vollbrecht, D. (1988). Forum microbiol. 9, 386-394.
  8. ↑ a b Kunz, B., Stefan, G. (1992). Bioforum 15 No. 5, 160-163.
  9. ↑ a b c d Doman, M., Haukohl, M. (1991). Chem.-Techn. 20 No. 10, 1-2.
  10. ↑ a b Doman, M. (1997). personal message
  11. ↑ Underkofler, LA, Fulmer, EI, Schoene, L. (1939). Ind. Eng. Chem. 31, 734-738.
  12. ↑ a b c d e f g h i j K.L. Schulze, Appl. Microbiol. 10, 1962, 108-122.
  13. ↑ Hesseltine, CW, Shotwell, OL, Ellis, JJ, Stubblefield, RD (1966). Bacteriol. Rev. 30, 795-805
  14. ↑ Han YW, Anderson, AW (1975). Appl. Microbiol. 30, 930-934.
  15. ↑ LA Under Kofler, RR and SS Barton Rennert: Production of Microbial Enzymes and Their Applications ; In: Appl Microbiol. 1958 May; 6 (3): 212–221, PMC 1057391 (free full text)
  16. ↑ a b Underkofler, LA, Severson, GM, Goering, KJ, Christensen, LM (1947). Cerium. Chem. 24, 1-22
  17. ↑ a b c d e Vollbrecht, D. (1997). Chem. Ing. Techn. 69, 1403-1408.
  18. ↑ Terui, G., Shibasaki, I., Mochizuki, T. (1958). Technology Report, Osaka Univ. 8, 213-219.
  19. ↑ Arima, K. (1964). In: Global Impact of Applied Microbiology, MP Starr, ed., Pp. 277-293, John Wiley & Son, New York.
  20. ↑ a b c Aidoo, KE, Hendry, R., Wood, BJB (1984). J. Food Technol. 19, 389-398
  21. ↑ a b c d e f Fukushima, D. (1979). J Am Oil Chem Soc . 56, 357-362
  22. ↑ a b Lonsane, BK, Ghildyal, NP, Budiatman, S., Ramakrishna, SV (1985). Enzymes microb. Tech-nol., 7 No. 6, 258-265.
  23. ↑ D. Vollbrecht: Adv. Food Sci. 19 No. 3/4, 1997, 87-89.
  24. ↑ a b Underkofler, LA, Severson, GM, Goering, KJ, Christensen, LM (1947). Cereal Chem. 24, 1-22.
  25. ↑ Terui, G., Shibasaki, I., Mochizuki, T. (1957). Hakko Kogaku Zasshi 35, 105-116.
  26. ↑ Lakshminarayana, K., Chaudhary, K., Ethiraj, S., Tauro, P. (1975). Biotechnol. Bioeng. 17, 291-293.
  27. ^ A b c d Silman, RW, Conway, HF, Anderson, RA, Bagley, EB (1979). Biotechnol. Bioeng. 21, 1799-1808.
  28. ↑ a b c Hao, LC, Fulmer, EI, Underkofler, LA (1943). Ind. Eng. Chem. 35, 814-818.
  29. ↑ Vollbrecht, D. (1994). DE 44 06 632.

literature

Commons : Biotechnology  - collection of images, videos and audio files

Trade journals