Ghasem D. Najafpour. Biochemical Engineering and Biotechnology

Подождите немного. Документ загружается.

284 BIOCHEMICAL ENGINEERING AND BIOTECHNOLOGY

used in the present study, obtained from the American Type Culture Collection.

Ferrocyanide-treated molasses [K

Fe(CN)

200 ppm] medium containing 100 g⭈l

⫺1

sugar is

used as the basal fermentation medium. Different cultural conditions such as incubation

temperature (30 ⬚C), initial pH (6.0), air supply (1.0 l

⫺1

min), agitation intensity (200

rpm) and incubation time (about 5 days) will be optimised for enhanced citric acid produc-

tion. A maximum amount of anhydrous citric acid with 100 g⭈l

⫺1

of substrate is in the range

71–88 g⭈l

⫺1

. If the product is removed as it is formed, the yield can be maximised. Final pH

and cell dry weight are about 2.1 and 10 g⭈l

⫺1

, respectively.

A. niger remains the organism of choice for the production of citric acid. In a submerged

fermentor, either purified compressed air or oxygen and agitation are used. Molasses are

a desirable raw material for citric acid fermentation because of their availability and rela-

tively low price.

Incubation temperature plays an important role in the production of citric acid.

Temperatures between 25 and 30 ⬚C are usually employed for culturing of A. niger, but

above 35 ⬚C citric acid formation is inhibited because of the increased production of by-

product acids and the inhibition of culture development. Citric acid production by A. niger

is sensitive to the initial pH of the fermentation medium. The maximum production of cit-

ric acid (6.5%) was obtained at pH 5.4 in the molasses medium. The appropriate pH is

important for the progress and successful termination of fermentation. The citric acid pro-

duced by A. niger is extremely sensitive to trace metals present in the molasses. Trace met-

als such as iron, zinc, copper and manganese present a critical problem in submerged

fermentation. The organisms need major elements such as carbon, nitrogen, phosphorus

and sulphur in addition to various trace elements for growth and citric acid production.

4,5

12.7 ANALYTICAL METHOD

12.7.1 Cell Dry Weight

Cells, spores and mycelia are filtered (0.45␮m), the filter is dried in an oven at 80 ⬚C, and

the cell dry weight is determined according to the procedure explained in the earlier bio-

process module.

TABLE 12.1. Initial media composition for seed culture preparation

Media composition Concentration, mg⭈l

⫺1

4Cl 50

MgSO

⭈7H

O20

FeSO

⭈7H

O10

ZnCl

CuSO

⭈5H

O10

Distilled H

O 1000 ml

Ch012.qxd 10/27/2006 10:54 AM Page 284

PRODUCTION OF CITRIC ACID 285

12.7.2 Carbohydrates

Sugars are estimated with reducing DNS reagent, based on the colorimetric method devel-

oped in an earlier module. The samples are determined by colour developed, which is

detected by a spectrophotometer at a wavelength of 540 nm. In large-scale operations,

mycelia are separated by a rotary drum filter.

12.7.3 Citric Acid

The citric acid obtained from fermentation is removed from the culture by precipitation.

The precipitation is formed by the addition of Ca(OH)

200 g⭈l

⫺1

, at 70 ⬚C. The pH of solu-

tion is adjusted to 7.2. Tri-calcium citrate tetrahydrate is collected by filtration. The tri-

calcium citrate as filter cake is dissolved in H

at 60 ⬚C with 0.1% excess, the solid

retained is CaSO

and the free citric acid is obtained. The free concentration of citric acid

is determined with an enzymatic kit available from Merck. GC/HPLC is recommended for

high accuracy of any research work.

12.8 EXPERIMENTAL RUN

Based on the above discussion we are now ready to start a real experiment. Molasses are

transferred from the sugar industry and kept in a cool room. The ATCC culture order has

arrived and hydrated. Stock culture was prepared and aseptic transfer successfully done.

Prepare spores, fungal conidia/small pellets in Petri dish or cotton-plugged flask, for 48

hours. Harvest the spore in separate flask with media for propagation. Once in the separated

flask the concentration of spores has reached to about 3 million per litre. It is now ready

to be transferred to a 2 l B. Braun biostat fermenter. The minimum volume of harvested

spores in the flask is 300 ml. Media must be prepared based on sufficient carbon source

TABLE 12.2. Experimental run for production of citric acid

Time, DO, Citric acid Cell dry Sugar

hmg⭈l

⫺1

concentration, g⭈l

⫺1

weight, g⭈l

⫺1

concentration, g⭈l

⫺1

0 8 0 0.1 10

12 7 0.4 0.3 9

24 7 0.9 0.5 8

36 6 1.35 0.7 7

48 6 2.1 1 6

60 5 2.9 1.1 4

72 5 3.6 1.2 2

84 4 3.9 1.3 1

96 4 4 1.3 1

108 6 4 1.3 1

120 7 4 1.3 1

Ch012.qxd 10/27/2006 10:54 AM Page 285

286 BIOCHEMICAL ENGINEERING AND BIOTECHNOLOGY

with 10 g⭈l

⫺1

sugar, nitrogen sources and trace metals, as explained before and defined in

Table 12.1. Air or pure oxygen is supplied to ensure availability of oxygen for sufficient

aeration. The experimental data obtained are reported in Table 12.2.

REFERENCES

1. Baily, J.E. and Ollis, D.F., “Biochemical Engineering Fundamentals”, 2nd edn. McGraw-Hill, New York, 1986.

2. Doran, P.M., “Bioprocess Engineering Principles”. Academic Press, New York, 1995.

3. Schmauder, H.P., (ed.) “Methods in Biotechnology”. Taylor & Francis, UK, 1997.

4. Stanbury, P.F. and Whitaker, A., “Principles of Fermentation Technology”. Pergamon Press, Oxford, 1984.

5. Matthews, R.F. and Braddock, R.J., Food Technol. 41, 57 (1987).

Ch012.qxd 10/27/2006 10:54 AM Page 286

CHAPTER 13

Bioprocess Scale-up

13.1 INTRODUCTION

Microbial processes are developed and implemented in many ways. These processes are

operated at different scales, such as bench, pilot and plant scales. For economic reasons, large-

scale bioprocesses should not be disturbed and any modification to the running processes may

not be considered. For research, many small-scale processes are developed to screen and

develop better strains, improve existing culture, use efficient media and handle the process

more efficiently with maximum productivity. The role of biochemical engineers in microbial

processes is well defined. First, they should have clear understanding of the bioprocess and

know the action of biocatalysts to stimulate biochemical reactions. Secondly, they should

maintain optimal growth and the environmental conditions required to obtain maximum pro-

duction yield. The microbial environment involves substrate and product concentrations, media

composition, pH and temperature. There are many physical and chemical parameters involved

in the growth environment once the organisms are propagated in a well-advanced bioreactor

with instrumentation control. The process parameters are the flow dynamics and chemical

dynamics of the biochemical system such as mass transfer, mixing, shear generated by agita-

tion, power input, dilution rate, nutrients and growth stimulants. The chemical para-meters of

the process are well controlled and optimised for growth kinetics; physical factors are based

on process design. The process design, geometric shape and the size of the bioreactor are

affected by dilution rate and substrate consumption rate. The operating parameters are well

controlled. Changing bioreactor size must follow certain rules to meet special criteria. Most

bioprocess designers have attempted to follow only vessel geometric similarities and some

dimensionless numbers. However, only increasing volume and dimensions by keeping some

dimensionless number constant may not satisfy the needs of bioprocess design.

13.2 SCALE-UP PROCEDURE FROM LABORATORY

SCALE TO PLANT SCALE

The usual procedure for scale-up of a fermenter based on concepts is well known. The

following criteria are translated between two scales of operation. They are selected as a pro-

cedure for scale-up.

287

Ch013.qxd 11/21/2006 9:51 AM Page 287

1. Similar Reynolds number or momentum factors.

2. Constant power consumption per unit volume of liquid, P

/V.

3. Constant impeller tip velocity, ND

4. Equal liquid mixing and recirculation times, t

5. Constant volumetric of mass transfer coefficient, K

6. Keep all environmental factors for the microorganism constant.

Based on the practical history of scale-up, most fermentation processes for alcohol and

organic acid production have followed the concepts of geometric similarity and constant

power per unit volume. From the above concept, and as a strong basis for translation of

process criteria, only physical properties of the process were considered in the scale-up cal-

culation. For power consumption in an agitated vessel, there is a fixed relation between

impeller speed, N, and impeller diameter, D

. The constant power per unit volume, for a

mechanical agitated vessel is given by:

(13.2.1)

The power per unit volume is constant. From power consumptions in a bench-scale biore-

actor, the necessary agitation rate is calculated for the scale up ratio, using Equation

(13.2.1). The choice of criterion is dependent on what type of fermentation process has

been studied. The following equation expresses relations for the impeller size and agitation

rate in small and large bioreactors.

(13.2.2)

The agitation rate is proportional to impeller diameter to the power of 2/3.

(13.2.3)

Similarly, that is true for the rotational speed of large tank, which is related to a small tank

with the ratio of impeller diameter of large and small tanks to the power of 2/3.

(13.2.4)

The characteristics of an agitated vessel with impeller spacing are illustrated in Figure 13.1.

rpm rpm



i,2



i,1

ND ND

ii,,



Re

for

288 BIOCHEMICAL ENGINEERING AND BIOTECHNOLOGY

Ch013.qxd 11/21/2006 9:51 AM Page 288

13.2.1 Scale-up for Constant K

The mass transfer coefficient K

a is constant; the general correlation is considered by many

as proportional to the power per unit volume with constant exponent, and gas superficial

velocity to another constant power as shown below:

1,2

(13.2.1.1)

There has to be a relation between k

a with aeration rate and agitation speed, and scale-up

factor has to be determined. To eliminate the effect of viscous forces, the rheology of the

media and broth for a large vessel have to be similar to that of a bench-scale vessel. For

scale-up based on geometric similarity, the constant values a and b are proposed for the

mass-transfer correlation in Table 13.1.

The oxygen transfer rate is calculated based on oxygen concentration gradient, by deter-

mination of the oxygen level in the liquid phase and the equilibrium value.

(13.2.1.2)

At steady-state condition, change of oxygen concentration with time approaches zero. The

value of K

a should be estimated by a correlation developed for various sizes of fermenta-

tion vessel.

(13.2.1.3)

where is oxygen transfer rate and X is biomass concentration.





OTR

AL O

KaC C XQ

()

 a

()

BIOPROCESS SCALE-UP 289

FIG. 13.1. Characteristics of agitated vessel with impeller spacing.

Ch013.qxd 11/21/2006 9:51 AM Page 289

13.2.2 Scale-up Based on Shear Forces

Scale-up calculation is based on constant shear forces, where shear is directly related to

impeller tip velocity. Shear forces are defined as

(13.2.2.1)

where, t is the shear stress, du/dx is the shear rate, and m is the fluid viscosity. Since the shear

is defined as S, which is proportional to ND

, the concept of constant impeller tip velocity

and constant power per unit volume were applied. Now introduce variable S for impeller tip

velocity ND

into Equation (13.2.2), then multiply both sides and divide by impeller diameter,

so that the power equation is reduced to shear forces related to impeller diameter:

(13.2.2.2)

The constant shear concept has been applied for bioreactor scale-up that utilises mycelia,

where the fermentation process is shear sensitive and the broth is affected by shear rate of

impeller tip velocity. For instance, in the production of novobicin, the yield of antibiotic

production is dependent on impeller size and impeller tip velocity.

The impeller is a device where shear forces are transmitted to the fermentation broth.

Since the process is sensitive to high agitation rate, it is necessary that the special bioreac-

tor be scaled up, based on shear forces for maximising product yield such as antibiotics.

This is the main reason for keeping impeller tip speed constant; in practice, it is recom-

mended that this is in the range 0.25–0.5 ms

1

. If the above conditions hold, it is not nec-

essary to follow the geometric similarity constraint; however, it may violate the rule of

power consumption per unit volume. Therefore keeping the impeller at a constant speed

and keeping K

a constant may be required for special case studies for bioreactor design in

antibiotic fermentation.

13.2.3 Scale-up for Constant Mixing Time

The problem associated with poor mixing in a large vessel was identified as low dissolved

oxygen in the aerated vessel. The mixing time has been correlated with turbulent flow. In



i,L

i,S

13/

tm

290 BIOCHEMICAL ENGINEERING AND BIOTECHNOLOGY

TABLE 13.1. Constants in mass transfer correlation

for various fermenter size

Fermentation vessel size, l ab

5 0.95 0.67

500 0.60–0.7 0.67

50,000 0.40–0.5 0.50

Ch013.qxd 11/21/2006 9:51 AM Page 290

a large-scale operation, the corrected function for mixing time incorporates the actual mix-

ing time with impeller speed, tank diameter and liquid height. The function of mixing, f(t)

is defined as constant mixing time:

(13.2.3.1)

where, t

is the mixing time, g is gravity, D

is the tank diameter, and Y the depth of the liq-

uid. Substitute (13.2.2) into (13.2.2.2) as power per unit volume is constant; the resulting

equation is the mixing time:

(13.2.3.2)

That is proportional to impeller diameter to the power of 0.65. Then, solving for mixing time:

(13.2.3.3)

Equation 13.2.3.3 is used for geometric similarity, where t

and t

represent the mixing

time for small and large fermentation vessels. Mixing and mixing time have been major

concerns for several investigators, because large-scale vessels may not have uniform mix-

ing whereas in small vessels mixing is not a problem. The mixing time in large vessels has

been correlated based on dimensional analysis.

In 1953, Rushton proposed a dimensionless number that is used for scale-up calculation.

The dimensionless group is proportional to N

as shown by the following equation:

2,3

(13.2.3.4)

where the Reynolds number for an agitated vessel was defined as:

(13.2.3.5)

Since the process is more complex, the proposed method may not be valid for scale-up cal-

culation. The combination of power and Reynolds number was the next step for correlating

power and fluid-flow dimensionless number, which was to define power number as a func-

tion of the Reynolds number. In fact, the study by Rushton summarised various geometrics

of impellers, as his findings were plotted as dimensionless power input versus impeller



c  kN

()

mL mS



11 18 0 65/.



16/

f( )

mi i t

ttND gDYD= ()

23 16

BIOPROCESS SCALE-UP 291

Ch013.qxd 11/21/2006 9:51 AM Page 291

292 BIOCHEMICAL ENGINEERING AND BIOTECHNOLOGY

Reynolds number; the plot is known as a power graph. The plot is presented in Figure 6.6,

Chapter 6. Equation 13.2.3.6 correlates the power number as a function of Reynolds number.

(13.2.3.6)

In fact, the power number is a dimensionless number that is the ratio of ungassed power to

gassed power in a normal bioreactor.

(13.2.3.7)

where P is ungassed power, in W or hp.

In a mixed agitated vessel with high agitation rate, at the centre of the vessel a vortex

often forms. To prevent a central vortex in tanks less than 3m in diameter, four baffles each

with a baffle width of 15–20cm are necessary. A basic assumption is to select a ratio of liq-

uid height to tank diameter from 2:1 to 6:1.

 2:1–6:1 (13.2.3.8)

It is common for an agitated vessel to have the tank diameter equal to the liquid working

volume.

(13.2.3.9)

Also, select a ratio of tank diameter to baffle width from 10 to 12:

 10–12 (13.2.3.10)

A suitable impeller interspacing (H

) is preferred for good mixing that has to be in distance

of less than 2D

(13.2.3.11)

It is well understood that mixing and mass transfer are affected by agitation and aeration

rates. Typically, in an agitated vessel, the working volume is about 75% of nominal CSTR

volume.

DH D

ii i

2

1





()Re

Ch013.qxd 11/21/2006 9:51 AM Page 292

13.3 BIOREACTOR DESIGN CRITERIA

The Bioreactor is the major equipment used in biochemical processes. It differs totally from

a simple chemical reaction vessel. To control physical operating parameters and microbial

environmental conditions, there are several influential variables:

• Biomass concentration

• Sterile condition

• Effect of agitation and mass transfer

• Heat removal for temperature control

• pH control of the media

• Correct shear conditions

Bioreactors in terms of operation and specific criteria are well defined. The most useful and

applied types of bioreactor are:

• CSTR

• Air lift: air is used for fluid circulation by pressurised air

• Loop reactor: modified air lift, pump transport the air and fluid through the vessel.

• Immobilised system: air and fluid circulate over a film of microorganisms grown on a

solid support.

• Tower fermenter

• Membrane bioreactor

13.3.1 General Cases

If the height of liquid, H

is equal to D

, one set of agitators is enough to obtain sufficient

mixing. For H

equal to 2D

or 3D

, additional sets of agitators separated by distance H

are

required, mounted on the same shaft. Spargers are placed at a distance D

/2 from the bot-

tom of the tank. Power input is greater than 100Wm

3

and the tip impeller speed is ND

greater than 1.5 ms

1

. Also, Froude number is greater than 1.0. H

is impeller inter

spacing.

Foam in the bioreactor is troublesome: it can reduce the oxygen transfer rate (OTR).

Antifoam is use to prevent foam formation. However, excess antifoam may cause growth

inhibition in the course of fermentation. The simplest device is known as a foam breaker,

which is mounted on the stirrer shaft located on the surface of liquid. It is a flat blade.

13.3.2 Bubble Column

For production of baker’s yeast, beer, vinegar, and in wastewater treatment, bubble columns

are used. Generally these columns should be long enough so that bubbles are utilised as

they rise while passing through the column. Column diameters range from 10cm to 7.5 m.

The height of column is three to six times greater than the diameter. The common height

BIOPROCESS SCALE-UP 293

Ch013.qxd 11/21/2006 9:51 AM Page 293