221.full.pdf

Microbioogical

Process Discussion

Calculation of Heat Sterilization Times for Fermentation Media FRED H. DEINDOERFER1 Merck Sharp & Dohme Research Laboratories, Merck & Co., Inc., Rahway,

New

Jersey

Received for publication September 14, 1956

Bacterial spores are by far the most heat resistant forms of microorganisms. The thermal resistance of bacterial spores is inherently different among species and even among strains of the same species. Besides thermal resistance inherent to spores of a particular species, variations can be caused by a number of environmental factors. These can be generalized into two groups: 1) environmental factors affecting sporulation prior to sterilization, and 2) environmental factors affecting spores during the sterilizing heat exposure. The most important factors in nutrient media sterilization fall in the latter category and include factors such as the pH during sterilization and the osmotic nature of the media. The presence of suspended solids also affects sterilization by physically insulating spores from heat exposure. Thermal-Death Relationships of Bacterial Spores Because bacterial spores are the most heat resistant forms of microorganisms, their germinated cells are the most frequent contaminants encountered in industrial fermentations due to improper sterilization. The ensuing discussion will deal, therefore, with the thermal-death relationships of these forms. Relationships between the number of viable spores and exposure time to heat demonstrate a logarithmic rate of spore viability destruction. Two typical survivor curves for spores of Bacillus stearothermophilus strain 1518, an organism often used for sterilization studies in the food industry, are shown in figures 1A and lB. These curves were obtained in buffer solutions and would differ from their respective curves in other media. Spore destruction at the higher temperature is more than 400 times faster than at the lower temperature.

Thermal Resistance of Microorganisms The ability of microorganisms to withstand heat is much greater than that of other forms of life. Thermophiles capable of even tolerating common heat sterilization conditions, such as exposure to steam at 250 F for 20 to 30 min, exist. Their occurrence in nutrient media, however, is rare. Relative resistances of several microorganisms to moist heat are shown in table 1.

TABLE 1. Relative resistances of microorganisms to sterilization by moist heat* Organism

Organism

Escherichia coli ............................. Bacterial spores ............................. Mold spores ................................. Viruses an(d bacteriophages .................

1 Present Address: E. R. Squibb & Sons, New Brunswick, New Jersey.

*

221

Rahn (1945).

~~~~~~Relative Resistance 1

3,000,000 2-10 1-5

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

Sterilization of nutrient media is an operation essential to all industrial fermentation processes requiring pure culture maintenance. AMethods for sterilizing nutrient media include 1) heating, 2) filtration, 3) irradiations, 4) sonic vibration, and 5) exposure to chemical agents. Because heating the medium, usually by steam, is the most reliable and on a large scale the easiest to control, it is the method of choice throughout the fermentation industry. The increasing availability of radioactive isotopes undoubtedly will stimulate further study of the use of gamma radiation for sterilization. However, a great deal of work is needed before commerically feasible gamma ray applications will be used in the fermentation industry. The other methods, although successful on a laboratory scale, present too many operational drawbacks for large-scale use in fermentation processes. How long should a nutrient medium be exposed at high temperatures to achieve sterile conditions? This question arises often and usually is answered in benchscale or pilot plant tests. Improper translation of these test results with large equipment may subject the medium to unnecessary overheating. A method by which minimum exposure time to achieve sterile conditions can be calculated from easily obtainable thermaldeath relationships and the temperature conditions in a heat sterilization process is presented in this paper. The method can be used to correlate sterilization conditions among various sized fermentation vessels. It also permits evaluation of the temperature and retention time relationship in continuous sterilizers. It need not be restricted to fermentation processes, but should be applicable to any process involving a heat sterilization or pasteurization operation.

F. H. DEINDOERFER

222

N1 6=2.3l k N

spores of a particular species will be a function of temperature only. It will behave similarly to other velocity constants in its relation with temperature, quantitatively expressed by the empirical Arrhenius equation as follows: k = Ae RT

where A is a proportionality constant, R is the gas constant, T is the absolute temperature, and ,u is an apparent activation energy for heat destruction of the spores. No theoretical significance need be attached to this equation for this discussion. The conformity of many destruction rates to the Arrhenius equation provides sufficient justification for its generalization here. In logarithmic form, equation 3 may be written as follows:

(2)

where N1 is the number of particular spores at the start of the heat exposure and N2 is the number of the same spores surviving at any time, 0, after exposure has started. N2 in this relationship can never reach zero, but practikally this does not matter. For destruction of spores wAith only one chance in one hundred of failing to destroy all spores, N2 should be set equal to 0.01. Higher degrees of confidence result from smaller values of N2. In a given environment the velocity constant, k, for

(3)

logk= - - /+logA

(4)

For all practical purposes, log A may be treated as a constant.2 Then equation 4 is of the form y = mx + b 2 The significance of the term, log A, is evident from examination of the Eyring rate equation. See Johnson et al. (1954). It includes an entropy term, which in cases of spore destruction is quite large. Occasionally, values of AS along with values of /Aare listed for thermal spore destruction. This completely defines the velocity constant, k, as a function of temperature.

05\

'OS

A)

14

TEMPERATURE 220 F. REDRAWN FROM DATA OF BALL (1943)

B)

104

_

TEMPERATURE 268.7 F. REDRAWN FROM DATA \OF STERN AND PROCTOR (1954)

-I

w

-J 103

\ k .

102

.

00057

SEC.-'I

k - 0.25

,o

SEC.-'

0

z z

*

TIME

MINUTES

-0-

TIME

MINUTES

FIG. 1. Survivor curves for spores of Bacillus stearothermophilus strain 1518 at two different temperatures

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

The curves in figures IA and lB represent first-order reactions, where the rate of spore destruction is directly proportional to the number of surviving spores. This can be described mathematically as follows: dN (1) dO where N is the number of surviving spores in the volume under consideration, 0 is the time of exposure, k is a is the rate of spore destrucvelocity constant, and dN dO tion. By integrating this equation, it can be modified to a form that can be used to calculate the time for sterilization at a particular temperature where the velocity constant, k, is known. Then,

[VOL. 5

STERILIZATION OF NUTRIENT MEDIA

1957] a

plot of log k

for

versus

spores

of

a

T

-

Injection of steam introduces some dilution to the medium, but this is taken into account during batch make-up prior to sterilization. Batch sterilization conditions are often specified as a holding period at a certain temperature. The heat effects involved during the time required to reach the desired sterilizing temperature and the time required in cooling down from this temperature are usually neglected. In large-scale equipment, the rising and falling temperature portions of the heating cycle are much longer than the constant temperature portion. Figure 4 compares actual fermentor heating cycles for various sized vessels. Note that 110 F is the starting temperature for the heating cycle. Batch make-up using warm water can save considerable time in the solution

Batch Sterilization

The most common method of heat sterilizing nutrient media is the batch method. The medium ingredients are charged directly into the fermentor. The fermentor and medium are sterilized by heat transferred across the jacket and/or coil surfaces from condensing steam. Fermentors are usually of the geometric design shown in figure 3. The heat transfer surfaces are marked in figure 3 by heavy lines. The medium is agitated and often steam is injected directly into the medium through the air sparger to speed up the sterilization. 3Various other terms commonly are used to describe thermal behavior of bacterial spores during heat sterilization, especially in the food industry. The terms used in this article are consistent with those used in chemical reaction kinetics. Some other terms and their relationship to the velocity constant, k, are tabulated for reference purposes.

-1

f

I.z 4n

z 0

Fa 3-0

-J

>

1i0

0.00135

Term

Relation

Definition

kT

The ratio of the velocity constant at a particular temperature to the velocity constant at a temperature ten degrees lower. This ratio is often falsely assumed constant over the entire temperature range. It diminishes as temperature is raised

Qio

klTjo

TDT

F

Two synonymous terms representing the time required for 90 per cent destruction of a spore population at a particular temperature, often called the decimal reduction time

2.3 k 2.3 k

D

2 3 log N,

k

N2

as above

Thermal death time, a term attached

required for "complete" destruction of spores in a particular environment. Algebraically, this term is meaningless unless N2 has some finite value to the time

The thermal death time at 250 F

000137

0.00139

0.00141

0.00143

0.00145

RECIPROCAL TEMPERATURE T

0.00147

0.00149

R

~~~~~~~R

FIG. 2. Effect of temperature on the velocity constant for destruction of spores of Bacillus stearothermophiluis strain 1518. TABLE 2. Activation energies and entropies for first order degradation of B-complex vitamins and death of bacterial spores Energy

Vitamin or Bacterial Spore

A

Entropy AS

cal/mole

cal/mole K

Folic acid ............................... 16,800* ........... 21,000* d-Panthothenyl alcohol ....... Cyanocobalamin ............. ............ 23,100* Thiamine hydrochloride .................. 26,000* Bacillus stearothermophilus strain 1518.... 67,700 Putrefactive anaerobe NCA 3679 ..... .... 72,400t ........... 82,l00t Clostridium botulinum .........

-14.1 t 5. it 2.2t

*

Garrett (1956).

t Calculated from data of Garrett (1956). t Levine (1956).

11.4t 105

1234 160t

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

particular species in a given medium should yield a straight line having a slope equal to 23RT* A typical plot is shown in figure 2. Such plots are constructed by determining the velocity constant, k, at at least two temperatures. If there is doubt that the kinetics of the spore destruction are such that the velocity constants are not related to temperature as in the Arrhenius equation, more k values should be determined. The subsequent treatment is not affected by another relationship, as long as it is known. For spores of B. stearothermophilus strain 1518, characterized in figure 2, the activation energy for destruction is 67.7 Kcal/mole. Activation energies and entropies for thermal destruction of other bacterial spores are listed in table 2.3

and

223

F. H. DEINDOERFER

224

of batch ingredients and in heating time. If water at 50 F were used in batch make-up it would have taken an additional 30 min to heat the medium in the 15,000gal fermentor to 110 F. In order to avoid extending the graphs unduly, 110 F was used as the finishing temperature. Actually, temperatures below 200 F have little effect on the sterilization. All the vessels characterized in figure 4 are reasonably geometrically similar. Because of this congruency, the heat transfer area per unit volume decreases as vessel size increases. Also, since the vessels operate at similar power input per unit volume levels, the heat transfer coefficient also decreases as vessel size increases. This

[VOL. 5

explains the different shapes of the heating cycles in figure 4. Coils are added to larger fermentors to provide additional heat transfer surface. Very often, too, the amount of steam directly injected into the medium per unit volume is increased for larger fermentors. Despite this, however, the disadvantages of increased size are not fully compensated. Obviously, the rising and falling portions of the sterilization cycle contribute significantly to fermentor and medium sterilization, especially in larger vessels. A method for determining the contribution of these portions to the sterilization is suggested through the use of the thermal-death relationships discussed earlier.

STEAM INILETS

{=

L

AND COOLING WATER

EXHAUST AIR OUTLET (STEAM VENT DURING STERILIZATION)

OUTLEETS

3 STEAM CONDENSATE OUTLETS AND

3 COOLING WATER INLETS C STERILE AIR INLET ($TEAM DURNIN STERILIZATION)

FIG. 3. Geometric design of large-scale fermentors

o

TimE

MINUTES

*

TIME

MINUTES

FIG. 4. Temperature rising and falling curves during sterilization of medium in various sized fermentors

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

RAW INGREDIENTS WATER ACID OR BASE

225

STERILIZATION OF NUTRIENT MEDIA The method involves a graphical integration of the velocity constant over these portions of the cycle. k is a function of T as shown in figure 2 and described in equation 3. T is a function of time, 0, as described by the heating cycle. Therefore, k is also a function of 0. An average value of k for each portion of the cycle can be obtained by graphically integrating k over the time period involved and dividing the integral by the time represented by the portion, as in equation 5 below. 2

lCv

kavg

=

k dO

(5)

1

02

-

01

It should be kept in mind that the heating cycle rising and falling curves for a vessel will vary with a number of factors. These include the liquid physical properties of the nutrient medium such as density, viscosity, thermal conductivity, and specific heat; extent of fouling of the heat transfer surfaces; amount and enthalpy of steam sparged directly into the medium; medium charge volume; temperature and enthalpy of the heating steam and temperature of the cooling water in the vessel coils and jacket. For a given process and given vessel these variables are maintained reasonably constant, and good replication in heating cycles occurs. The method is illustrated by way of the following hypothetical example. Example 1. A 15,000-gal fermentor containing 12,000 gal of a penicillin production medium is to be sterilized. The medium contains 4 per cent corn steep liquor. Corn steep provides an excellent source of contaminating organisms. Laboratory checks have shown bacterial counts not exceeding 20 X 106 cells per ml in the

.024

A)

8) FALLING PORTION OF CYCLE

RISING PORTION OF CYCLE

-10 I- .020

z

z

0 .016

0 -I

w .012

.008o f

.004

kd

-

fkde

o. 331

2kav0.335 9

0.0112

SEC.-l

~~~~I 0

5

10

15

25

20 0

FIG. 5. Velocity constant variation

over

TIME

kav

i-31 0

0.087 0.0870,0066

I ______

I.

I

0

S

SEC. 1 I

I

10

MINUTES

rising and falling temperature portions

of sterilization

cycle

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

presterilized medium. For how long must the medium be sterilized at 250 F to be sure of sterilization in 999 out of 1000 batches? Solution. For sake of illustration, assume the curve in figure 2 characterizes the most heat resistant bacterial spores in the penicillin medium. Also, let the heating cycle of the 15,000-gal fermentor shown in figure 4 represent the heating cycle of the fermentor under consideration. 220 F is chosen as the minimum lethal temperature. This is an arbitrary choice, of course. Actually lower temperatures are lethal but their relative lethality is small and neglecting them does not introduce any significant error. Since the percentage of the laboratory bacterial count that is contributed by the most heat resistant spores was not determined, assume that the entire count was contributed by these spores. This is equivalent to adding a safety factor in the calculation. Then N1 = (20 X 106 spores/ml) (3.78 X 103 ml/gal)(12 X 103 gal) = 9.07 X 1014 spores The chance for failure has been set at one in a thousand so that N2 = 0.001 spore. The velocity constants at the temperatures along the rising portion of the sterilization cycle in figure 4 are determined from figure 2. These values are plotted versus the time corresponding to the temperatures along the cycle as in figure 5A. The procedure is repeated for the falling portion of the cycle in figure 5B. Note that the time of each of these portions below 220 F is not shown on the figures. Graphical integration of the area under the curves in figures 5A and 5B, and solu-

226

F. H. DEINDOERFER

tion of equation 5, yields values of kavg for each of these portions of the cycle. For 29.5 min of the rising portion, kavg is equal to 0.0112 sec-1 and for 13.2 min of the falling portion, kavg is equal to 0.0066 sec-'. Therefore, for portions of the heating cycle totalling 42.7 min, a kavg of 0.0098 sec-' can be used to calculate the contribution of these portions to the sterilization. Using equation 2, the population remaining if these two periods followed each other can be calculated. X 1014 log-9.07 N2

(0-0098 sec') (60 sec/min) (42.7 min)

TABLE 3. Time at sterilization temperature to achieve the same degree of sterilization in different sized fermentors Fermentor Size

Total of Heating 220 F Above CycleTime

gal 50 150

1,500 15,000

Time at 250 F

m#

mn

28.0 33.7 41.3 51.5

17.5 12.6 11.3 8.8

Thus, only 8.8 min at 250 F are required for the sterilization of the fermentor. For comparison, similar calculations have been made for the other sterilization cycles depicted in figure 4. The results are listed in table 3. As vessel size increases, the respective contributions of the rising and falling portions of the cycle increase, and consequently the time required at so-called sterilization temperature decreases. Heat Effects on Nutrients The time required for batch sterilization is not usually the optimum time of heating when product yield is considered. The thermal effects on nutrient quality of the medium must also be taken into account. Sometimes, productivity is increased by prolonged heating of the medium. In most media, however, the deleterious effects of extensive heating are more apparent than any beneficial effects, and overheating of the medium must be minimized. An easily destroyed nutrient quality might be any one of the B-complex vitamins. Vitamin degradation is known to occur rapidly at high temperatures. Although figures for vitamin destruction in fermentation media are not published, Garrett (1956) has studied vitamin stability at elevated temperatures in liquid preparations. He found activation energies of from 16,800 to 26,000 calories/mole for destruction of several B vitamins. The vitamins investigated by Garrett are listed with their activation energies and entropies in table 2. The activation energies and entropies for the vitamins are much lower than for the three bacterial spores also listed. STERILE

STEAM

RAW INGREDIENTS WATER ACID OR BASE

MEDIUM

MAKE -UP TANK

UNSTERILE

MEDIUM

FIG. 6. Steam-injection type of continuous sterilizer

I

WATER

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

2.3 N2 = 1.14 X 10' spores The N2 just calculated becomes the new N1 to be used in equation 2, this time to calculate the time the fermentor must be held at 250 F to reduce the population to the desired N2 of 0.001. 2.3 1.14 X 103 0 = 0 (0.0265 sec-) (60 sec/min) log 1 X 10-3 0 = 8.8 min

[VOL. 5

227

STERILIZATION OF NUTRIENT MEDIA

1957]

The conditions in a batch sterilization carried out at 250 F cannot always achieve sterilization without impairing nutrient quality of the medium. Higher temperature-shorter time batch sterilizations can be carried out, but fermentation vessels are usually limited to operation at not more than 30 psig (274 F) by design restrictions.

TABLE 4. Operating conditions employed for continuous sterilization of mnedia* Media Used

Suspended Solids

%t70 Riboflavin ........... 1.8 corn steep liquor Cyanocobalamin ..... 4.0 soybean meal and distillers' solubles Acetone, butanol ...1 1.8 ground corn Sodium gluconate 0.4 corn steep liquor Itaconic acid ........0 .2 corn steep liquor Fungal amylase...... 4.0 distillers' solubles and 3.0 ground corn

* Pfeifer and

Vojnovich (1952).

pH

Temra- Time F

min

4.5 4.5

275 325

4 13

6.5 4.5 6.1 5.0

275 275 300 325

3 5 5 13

N1 = (3 X 106 spores/ml) (3.78 X 103 ml/gal)(1 X 101 gal) = 1.13 X 1015 spores

Since one chance in one hundred was chosen as the margin of failure, N2 is equal to 0.01. Knowing the required retention time, equation 2 can be used to solve for the necessary velocity constant. 2.3 300 = 0.131

1.13 X 1015 1 X 10-2

sec-1

The calculated value of k is used to read the required sterilization temperature from figure 2. For k = 0.131 sec- 1 T = 0.1382. Thus, T = 724 R and a sterilization temperature of 264 F is required. To achieve the same sterilization at 250 F, a retention time of 24.8 minutes would be required. Figure 7 illustrates the relationship between time and tempera-

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

Continuous Sterilization A sterilization method becoming increasingly popular and having several advantages over the conventional batch sterilization is continuous sterilization. Continuous sterilization of media for riboflavin fermentations has been reported by Pfeifer et al. (1950) and for penicillin fermentations by Whitmarsh (1954). In one type of continuous sterilization, preheated unsterile medium passes through an injection heater in which steam is introduced. The vigorous and almost instantaneous mixing obtained raises the medium to sterilization temperature immediately. This temperature is maintained for the required amount of time in an insulated retention tube through which the hot medium flows. A diagram of a typical continuous sterilizer is shown in figure 6. The hot medium passes through a heat exchanger where it is cooled to below its flash point as it preheats unsterile medium. Final cooling to process temperature is accomplished in the fermentor. The activation energy of bacterial spore destruction is much higher than the activation energy of simpler chemical reactions. It is, therefore, an advantage to use a high temperature-short exposure time sterilization operation whenever nutrient degradation occurs to the extent that it lowers the process yield. Continuous sterilization not only overcomes unfavorable nutrient destruction, but has a number of operational advantages. Pfeifer and Vojnovich (1952) point out these advantages in an excellent paper on continuous sterilization. Operating conditions employed by these authors for continuous sterilization of several types of media are shown in table 4. In a batch sterilization, probably

only those sterilizations at 275 F could be approached, if at all, in conventional fermentors. The calculation of sterilization conditions for continuous operation is simplified by almost instantaneous rising time to sterilization temperature and a rapid cooling period thereafter. For this type of sterilization, only the velocity constants of the most resistant spores present in the medium and the total initial bacterial concentration need be known to perform the calculation. Usually, in production operations the sterilizer is of a fixed length. Different retention times are achieved by controlled pumping. If temperature conditions are specified, the retention time can be calculated easily from equation 2. If process conditions dictate a certain flow rate, as may be the case in continuous fermentations, any modification in sterilizer operation will have to be made temperature-wise. Consider the following example. Example 2. Process conditions are such that a retention time of 5 min is required in the sterilization of a sodium gluconate production medium containing 0.4 per cent corn steep liquor. Plate counts of 3 X 106 bacteria per ml in the presterilized medium have been determined in the laboratory. At what temperature should the sterilizer be operated to achieve a sterilization having 99 per cent certainty of being successful? One hundred thousand gallons of medium are to be sterilized in this manner. Solution. For sake of illustration, the curve in figure 2 again will be assumed as characterizing the most heat resistant spores in the medium. The initial population again will be assumed as entirely consisting of this species. Then

F. H. DEINDOERFER

228

ture to achieve the above sterilization. Also listed in table 5 are the relative effects of an adverse chemical reaction destroying a hypothetical vitamin during the sterilization, the degradation of which is characterized by an activation energy of 22.6 Kcal/mole, one-third that of the activation energy for bacterial spore destruction. Even for the retention time and temperature used in example 2, the vitamin was almost completely destroyed. Much of the vitamin quality can be retained, however, by higher temperature-shorter retention time sterilizations.

40

2 w

go-2 Il

o2~ 0~~~~~~~~~~~~~

0.00129

0.00131

0.00133

0.00135

0.0057

0.00139

RECIPROCAL TEMPERATURE

000141

0.00143

I

FIG. 7. Time-temperature relationship to achieve the same degree of sterilization in a continuous sterilizer. TABLE, 5. Time-temperature relationship and its effect on vitamin content in a continuous steri,lization Temperature

Time

Relative Retention of Original Vitamin Content*

F

min

%

250 265 280 295 310 325

24.8 4.1 0.72 0.14 0.029 0.0061

0.0 0.0 2.3 28 64 89

* k for vitamin destruction assumed equal to k for spore destruction at 250 F.

Design Method The use of well-known thermal behavior characteristics of bacterial spores and the temperature characteristics of the sterilization cycle in calculating the time for sterilization of fermentation medium has been illustrated. This method offers an approach to the correlation of sterilization conditions among various sized fermentation vessels and between temperature and retention time in continuous sterilizers. The advantages of continuous sterilization have been pointed out in a sterilization where nutrient damage occurs. The steps involved in calculating process conditions for sterilization operations can be summarized as follows: (1) Periodically determine the thermal-death relationship of the most heat resistant bacterial spore in the medium to be sterilized. (2) Determine the initial population concentration of the medium and choose a confidence level for the sterilization. An added safety factor is introduced by assuming that the total population consists entirely of the most heat resistant spores. (3) Calculate the contribution to the sterilization of the rising and falling portions of the sterilization cycle; this step is unnecessary for steam injection type continuous sterilizations. (4) Calculate the time required to hold the medium isothermally at the highest temperature chosen for the sterilization. For continuous sterilizations, the time of exposure is often chosen, and calculations are carried out to find the required temperature. REFERENCES BALL, C. 0. 1943 Short-time pasteurization of milk. Ind. Eng. Chem., 35, 71-84. GARRETT, E. R. 1956 Prediction of stability in pharmaceutical preparations. II. Vitamin stability in liquid multivitamin preparations. J. Am. Pharm. Assn., 45, 171-178. JOHNSON, F. H., EYRING, H., AND PILISSAR, M. J. 1954 The Kinetic Basis of Molecular Biology, p. 220. John Wiley & Sons, New York, N. Y. LEVINE, S. 1956 Determination of the thermal death rate of bacteria. Food Research, 21, 295-301. PFEIFER, V. F., TANNER, F. W., VOJNOVICH, C., AND TRAUFLER, D. H. 1950 Riboflavin production by fermentation with Ashbya gossypii. Ind. Eng. Chem., 42, 1776-1781. PFEIFER, V. F. AND VOJNOVICH, C. 1952 Continuous sterilization of media in biochemical processes. Ind. Eng. Chem., 44, 1940-1946. RAHN, 0. 1945 Physical methods of sterilization of microorganisms. Bacteriol. Reviews, 9, 1-47. STERN, J. A. AND PROCTOR, B. E. 1954 A micro-method and apparatus for the multiple determination of rates of destruction of bacteria and bacterial spores subjected to heat. Food Technol., 8, 139-143. WHITMARSH, J. M. 1954 Continuous sterilization of fermentation media. J. Appl. Bacteriol., 17, 27.

Downloaded from http://aem.asm.org/ on March 7, 2019 by guest

10.2

[VOL. .5