Alcoholic Fermentation of
Molasses RAPID CONTINUOUS FERMENTATION PROCESS H. R. BILFORD, R. E. SCALF, W. H. STARK, AND PAUL J. ICOLACHOV Joseph E. Seagram & Sons, Inc., Louisville, Ky.
Necessary conditions such as temperature, agitation, pH, and cell count have been determined for the development of a continuQUS fermentation process. A single-vessel continuous fermentation system has been developed. With the system and the proper conditions molasses mashes containing 12-13 grams per 100 cc. of sugar can be fermented to completion in a 5-7 hour cycle. Yeast propagation is maintained so that inoculation is required only at the beginning of a run. The principal value of this process is the extensive reduction in equipment requirements.
W H E N THE FAST, COXTIWJOUS PROCESS I S FULLY DEVELOPED, THE MODERN BATCH FERMENTER SHOWN ABOVECAN B E CONVERTED T O A CONTINUOUS UXIT REPLACING A NUMBEROF
BATCH FERMENTERS
of water, and lowering the pH t o 4.0-4.5 by the addition of acid. Although molasses generally contains most of the nutrient substances required for fermentation, ammonium salts, such as phosphates or sulfates, may be added t o the mash to supply nitrogen or phosphates in which the molasses may be deficient. The prepared mash is inoculated with a suitable strain of yeast grown on molasses mash. The inoculum usually represents 4 to 6 per cent of the total volume of the main mash. The temperature of fermentation ranges from 21.1" to 32.2" C., or slightly higher. Most of the industrial alcohol companies modify the mashing and fermenting procedures according t o their experience H E success of any commercial process used in the and plant facilities. Therefore the above described procedure is to be considered, at best, a very general picture of existing manufacture of ethyl alcohol by fermentation depends on its adherence to certain fundamental factors vitally methods. It was intended to develop an alcoholic fermentation procrelated to the activities of the yeast cell. These factors iness which would be continuous and in which sugar concentravolve the use of an optimum concentration of sugar, an tions normally used in the batch methods would be fermented optimum pH, and an optimum temperature; the addition to completion in an appreciably shorter time. The conof supplementary substances to the medium if it is deficient tinuous system was to involve a minimum number of fermenin any essential constituent; the inhibition of bacterial tation vessels. growth; the use of a vigorous strain of yeast with high alcohol There are many reports in the literature describing contolerance; and the maintenance of anaerobic conditions tinuous processes for alcohol fermentation. Most of them during fermentation proper. All of the above factors are are from Russian investigators, but their systems involve dependent on the nature of the raw material to be fermented. many phases, usually three to nine fermentation vessels. Beverage alchohol, except rum and wines, is produced from Also, the time of fermentation ranges from 34 to 48 hours, grains, while industrial alcohol is produced principally from depending on the material fermented and other experimental molasses in this country ( 2 ) . conditions. I n 1941, Alzola patented a continuous process The preparation of molasses, usually blackstrap, for for molasses fermentation, but his system also requires a fermentation requires adjustment of the sugar concentration number of fermentation vessels for continuous operation ( I ) . t o approximately 12 per cent or slightly higher by the addition 1406
T
November, 1942
INDUSTRIAL AND ENGINEERING CHEMISTRY
1407
an outlet for the gas which was passed to scrubbers to remove the alcohol vapors. A third tube was employed Introductory work involved the determination of experito remove samples for analysis during the run and to mental conditions necessary for the rapid fermentation of add ammonia for pH adjustment. The feed medium enpure glucose solutions fortified with yeast extract and amtered the fermenter through a storage flask and connectmonium phosphate. No attempt was made at first to ing tube. The completely fermented material is withdrawn conduct fermentations on a continuous basis. It was by a draw-off tube. desired only to obtain conditions that would completely During the initial phase of operation the sugar was comferment the sugar present in the shortest possible time. The pletely fermented in 4 to 10 hours, depending on the initial experimental conditions obtained and applied to the concell count and other conditions. Continuous operation was tinuous fermentation system appear in Table I. started by feeding fresh (uninoculated) fermentation medium of high sugar concentration a t a constant rate from the storage flask into the fermenter, and the completely ferCONDITIONS FOR RAPIDCONTINUOUS mented material was withdrawn a t the same rate. The rate TABLE I. EXPERIMENTAL of feed and draw-off was dependent on the material being FERMBNTATION fermented, initial cell count, and other conditions. The Single fermenter plus storage tank Equipment entire system was kept free from contamination. It is Reducing sugar concentration, 10-16 grams/lOO Feed cc. depending on material; pH, 4.6-6.0 important for successful commercial operation that a pure Reducing sugar concentration 0.1-1.5 grams/100 Fermentation culture system be constantly maintained. Fermentation Conditions
Supplements Cycle time
cc. depending on material; pH, 4.6-6.0 adjusted during operation with NHtOH; temperature, 32.2' C.; yeast count, 150-350 million cells/cc. ; agitation, mechanical or Cot (1.8-4.8 liters/min. per liter of medium) Depending on fermentation medium used 4-10 hours, depending on above conditions
Fermentation of Glucose-Yeast Water
Initial work on this system involved the continuous fermentation of a medium composed of 10-12 per cent glucose dissolved in 10 per cent yeast water containing 0.1 per cent (NH4)2HP04. The results of an experiment in which the rate for the system under certain experimental conditions was determined are presented in Table IIA. The initial yeast cell count in the fermenter was approximately 150 million per cc. The strain of yeast was Seagram No. 1, a variety of Saccharomyces cerevisiae. To prepare the yeast inoculum, large flasks of medium were inoculated and aerated for 18 hours at 30" C. Upon the completion of the aeration period the cell concentration was determined, the count usually varying from 150 to 200 cells X 106 per cc. The necessary
Under these conditions it was possible to ferment continuously and completely solutions containing 12 grams sugar per 100 cc. in a 4-10 hour cycle. Only a single vessel was required. The reducing sugar concentration of the feed medium ranged from 10 to 15 grams per 100 cc., and the pH of the medium was adjusted to 4.5-5.0. Reducing sugar was determined by the method of Stiles, Peterson, and Fred (3). I n the fermentation vessel during continuous operation, the reducing sugar concentration varied from 0.1 to 1.5 grams per 100 cc., depending on the material being fermented. The pH of the ferSTORACIL FLASK Con menting material was kept a t 4.5-5.0 by SUI.AR S O L U T I O N (APPROX. 12.0G/IOOCCJ adjustment with ammonia every half hour throughout the fermentation period. Besides adjusting the pH, the ammonia is a source of nitrogen for the yeast. The fermentation was maintained at 32.2" C. (90" F.). This temperature was found to give the fastest fermentation without inhibition of the reproductive activity of the cells. The initial cell count ranged from 150-350 million per cc. As would be expected, the fermentation was faster with RUBBER STOPPER the higher cell count. Mechanical agitation or agitation with 1.8 to 4.8 liters of carbon dioxide per minute per liter of medium has been found necessary for rapid fermentation. Agitation accelerates the fermentation by providing ADDITION T u e e FOR SUOAR SOLUTION better diffusion of nutrients and carbohydrate. D R A W OFF A diagram of the laboratory unit employed for the continuous fermentation system is L l W l D LEVEL shown in Figure 1. It consisted of a single fermenter, a wide-mouthed Pyrex jar or flask. A two-fermenter system was initially employed successfully, but operations with one fermenter were easier to control and produced just as successful results. Inoculated fermentation medium was placed inside the jar. ALOXITP BALL Carbon dioxide was introduced through B glass tube and dispersed by one-inch Aloxite spheres. As stated before, mechanical agiFERMENTeR tation could be successfully substituted for UNIT FOR CONTINUOUS FERMENTATION carbon dioxide, Another glass tube provided FIGURE 1. LABORATORY
1408
Vol. 34, No. 11
INDUSTRIAL A N D ENGINEERING CHEMISTRY
CONVENTIONAL ALCOHOL FERMESTATION PLANTS REQUIRE A GREAT MANYFERMENTATION VESSELS
amount was centrifuged and cells resuspended in 3000 cc. of fermentation medium. The inoculated medium was then placed,in the fermentation vessel, and agitation with 4.8 liters of carbon dioxide per minute per liter was begun. The temperature of fermentation was held a t 32.2" C. Sugar and Balling determinations were made every hour, and the pH was determined every half hour. The pH was maintained a t 5.0 during the run by adjustment with 3 N ammonium hydroxide. Under the experimental conditions employed, a period of 5 hours was needed to ferment 98.8 per cent of the sugar in a solution containing 10 grams of sugar per 100 cc. At this point the yeast count had increased slightly from 133 to 182 million cells per cc., and the stationary phase was finished. Continuous operation was then begun a t this point; that is, the feed of fresh medium into the fermenter and the draw-off of the fermented material from the fermenter at equal rates of 5 per cent per hour were started. This rate was maintained for only one hour and the rate was then increased to 10 per cent, for the sugar concentration indicated that the system could accommodate the 5 per cent per hour easily. The 10 per cent rate was held for the next two hours. The sugar concentration in the fermented medium during this period was very low. The cell count remained constant. The rate was therefore increased to 15 per cent per hour, and this rate was held for three hours. However, under the experimental conditions this rate appeared to be slightly too rapid. The sugar concentration slowly increased and, after 3 hours a t this rate, had risen to about 0.5 gram per 100 cc. Simultaneously the cell count dropped to about 160,000,000 per cc., indicating further that the rate was too rapid. The rate was then decreased to 13 per cent; it appears that the system could probably have operated a t 13 per cent per hour, since the amount of sugar and the cell count in the fermenter medium remained constant a t a desirable level with the 13 per cent per hour rate. The results of a similar experiment appear in Table IIB. The rate of throughput for the system was determined with an initial cell count of approximately 350 rather than 150 million per cc. These data indicate that 10 grams sugar per 100 cc. of medium were almost completely fermented in 2.5 to 3 hours. Therefore, after 2.5 hours of fermentation the
system could be put into continuous operation. This is about half the time necessary with a cell count of 150 million per cc. After 2.5 hours about 98.5 per cent of the sugar was fermented. The rate for the first 3 hours was 20 per cent per hour. This rate was satisfactory; the sugar concentration
TABLE 11. RAPIDCONTINUOUS FERMENTATION OF GLUCOSEYEASTWATERMEDIUM Hours 0 21 3 4 5 6 7 8 9 10 11 12 13
0
1
2
2, 3.5
3 N Reducing Yeast Throughput, NHIOH, Sugar, G./ Milliln Total pH Cc. Balling 100 Cc. Cells/Cc. %ol./Hr. A . Initial Yeast Count, Approx. 150,000,000 Cells/Cc. 5.03 44 .. 42 00 4.25 4.50 4.92 5.00 4.85 4.75 4.75 4.65 4.60 4.30
C.
...
i:i
3 3 4 4 7
10.2 9.9 7 6 4.7 2.1 0.6 0.7 0.6 0.6 0.7 1.1 1.1 0.7 0.7
10.28 6.92 8.S4 4.04 1.24 0.11 0.10 0.10 0.10 0.12 0.50 0.41 0.32 0.57
0.6 0.6 0.6 0.6 1.8
0.10 0.10 0.32 0.58 1.33
133
1::
... ...
182 183 196
ii9
...
160
...
{ti;-
Sta-
phase 5 10 10 15 15 15 13 13 13
... 165 B . Initial Yeast Count, Approx. 350,000,000 Cells/Cc. 5.40 ... 10.3 10.40 ... {?& -; 3.30 15 6.6 5.12 350 4.80 5 1.1 0.26 ... E phase 5.20 .5.. 0.9 0.16 36? 20 0.11 350 20 4.80 0.7 ....
2:; 2:; 7":8 :. 5 4j:;:. 6 5 9.5
6 8 5
4.65
5"3
....
.....
...
325 340 310 163
...
20 26 27 34 35 35
Initial Yeast Count, Approx. 350,000,000 CelWCc.; Throughput Rate Constant 0 1 2
6.30 4.00 4.20
3 4 5 6 7 8 0 10
4.80 4.65 4.70 4.70 4.70 4.80 4.75 4 70
... 15
10.6 8.1 4.0
10.52 6.24 2.39
355
10 5 5 3 3 3 3 3 3
1.4 1.4 1.2 1.1 1.0 1.0 1.0 1.0
0.15 0.20 0.25 0.32 0.22 0 37 0.13 0.11
365 350 357 351 327 334 321 331
... *..
{f::
tion-
phase 25 25 25 25 26 25 25 25
November, 1942
INDUSTRIAL AND ENGINEERING CHEMISTRY
TABLE 111. RAPIDCONTINUOUS FERMENTATION OF MOLASSES 3 N
Hours
pH
Nn40n, Cc.
Balling
Reducing Yeast Sugar, G./ Million 100 Go. Cells/Co.
Throughput, % ' Total Vol./Hr.
A . Cuban Blackstrap, No Supplement 0 1
2 3 4 5 6 7 8 9
4.92 4.80 4.75 4.85 4.80 4.70 4.80 4.70 4.80 4.68
5''
2.5 2 3
3
3 3 2
...
18.0 17.3 15.4 12.9 9.8 8.3 8.3 8.7 9.0 9.1
12.02 10.27 8.15 5.60 2.56 1.03 1.14 1.24 1.25 1.31
396
.,.
... ... 545 550
460 417 515
phase 19 25 25 25 25
E . Refined Molasses 4- 75 Mg. (NH4)&04/100 Grams 0 1 2 3 4 5 6 7 8 9 10
5.10 4.80 4.75 4.75 4.56 4.80 4.80 4.60 4.70 4.80 4.83
0 1 2 3 4 5 6 7 8 9 10 10" iia
4.50 4.45 4.50 4.45 4.35 4.50 4.48 4.50 4.40 4.52 4.50 4.50 4.50 4.50
C.
...
3 5 5 8 5 5 7 6 5 5
18.0 17.1 15.4 13.5 11.4 9.3 7.2 6.4 6.5 6.9 7.1
...
13.45 9.87
315
..... 3.95
515
..... 4.11 1.69 1.12
.....
1.51 .....
... ... ... ...
580 580 540 508
phase 15 15 20 20 20 20
Beet Molasses 4- 100 Mg. (NH4)sHP04/100 Grams
... ... ... ... 2
... ...
... 1
... ... ... ... ...
120 a Mechanical agitation.
18.5 15.6 12.9 10.2 7.6 6.8 6.8 7.2 7.4 7.2 7.2 7.2 6.9 7.1
10.15 8.81 6.10 3.77 1.07 0.53
350 400
e....
iA6 ii6
0.85 0.62 0.87 0.76 0.76 0.76 0.82
...
ii6
...
360
3ii
...
phase 15 15 15 15 15 15 15 15 15
in the draw-off was practically negligible, and the high cell count was maintained. The rate was then steadily increased to 26, 27, 34, and 35 per cent per hour. However, when the sugar analyses were made at a later time they indicated that the 26-27 per cent per hour was slightly too high for suitable operation. Table IIC indicates the results of another similar experiment except that the rate of feed and withdrawal was maintained a t 25 per cent per hour throughout the period of continuous operation. It was desired to determine whether the experimental conditions employed and this rate would yield successful results during a long operating period. The results confirm those of Experiment B (Table 11). The medium containing 10 grams sugar per 100 cc. was 98.5 per cent fermented in about 3 hours during the stationary phase. The continuous phase was started at this point. Proof of the feasibility of the 25 per cent per hour rate is found in the following facts: The sugar level in the fermeqted material remained between 0.1 and 0.4 gram per 100 cc. during the entire 7-hour period of continuous operation, and the cell count was maintained a t its high initial figure. Fermentation of Molasses
Since this system appeared to work satisfactorily on partially synthetic medium, it was decided to attempt the continuous fermentation of materials such as molasses, Previous experimental work in these laboratories had involved only standard batch fermentations of various kinds of molasses, but certain facts obtained were applicable to the fast, continuous fermentation work on molasses. The Seagram No. 1 yeast strain was not suitable for molasses fermentation, but the Seagram No. 90 strain, another variety of Saccharomyces cerm'siae Hansen (a Java molasses distillery yeast, American Type Culture No. 4125) was satisfactory.
1409
The three types tested in the continuous fermentation system were Cuban blackstrap, refined, and beet molasses. The types and amounts of supplements necessary for the fermentation of the three kinds of molasses were determined. Cuban blackstrap required no supplement, the refined molasses required the addition of 75 mg. (NH&S04 per 100 grams of molasses, and the beet molasses required 100 mg. of (NH&HPOl per 100 grams. The rate of throughput for the one-fermenter system as determined with Cuban blackstrap molasses is shown in Table IIIA. The initial cell count of approximately 350 million cells per cc. was obtained by aerating large flasks containing 18' Balling molasses a t 28.8' C. for 18 hours. The cells were separated by centrifugation and resuspended in the fermentation medium, which was of the same composition and pH as the aeration medium. The other experimental conditions, such as pH, temperature, and agitation, were the same as for the work previously described. A period of 5 hours was required to ferment 91.4 per cent of the sugar in the medium. This is essentially complete fermentation of this material. The cell count had increased from the initial 350 to 550 million per cc. The continuous phase of the fermentation was then started at a 19 per cent per hour rate of feed and withdrawal. Since this rate was satisfactory, it was increased to 25 per cent. At this point the sugar concentration slowly increased and the cell count began to decrease, indicating too rapid a rate. The optimum rate of throughput for Cuban blackstrap under the above described experimental conditions lies between 19 and 25 per cent per hour. Table IIIB presents data from a similar experiment in which refined molasses was fermented. The addition of 5 grams of malt sprouts per 100 grams of refined molasses in addition to ammonium sulfate was necessary to secure a large cell crop for inoculum. This batch of refined molasses seemed to lack certain necessary growth and fermentation factors for yeast. Other experimental conditions were the same as for the preceding experiment. About 6 hours were required for the system to complete the stationary phase. This may have been due in part to the fact that the refined molasses contained more sugar than did the Cuban blackstrap. Refined molasses (18"Balling) was equivalent to about 13.5 per cent reducing sugar, while 18' Balling Cuban blackstrap was equivalent to approximately 12 per cent reducing sugar. At the completion of the stationary phase 91.6 per cent of the sugar was fermented, and the cell count had increased from the original 315 to 515 million per cc. Rate of feed and withdrawal was started at 15 and later increased to 20 per cent per hour. The latter rate appeared too rapid for the system under the experimental conditions. A rate of throughput between 15 and 20 per cent per hour or a 5-7 hour cycle appears to be satisfactory. A third experiment of this type was run on fortified beet molasses (Table IIIC). On aeration it also provided a good medium for the growth of yeast for inoculum. Conditions were the same as those employed in the experiments with Cuban blackstrap and refined molasses, except that toward the end of the run mechanical agitation was substituted for carbon dioxide agitation in order to determine the effect of this type of agitation on the rate of throughput. About 5 hours were required for the completion of the stationary phase of the fermentation. The sugar was 94.7 per cent fermented a t this time. The cell count had increased from an initial 350 to about 530 million per cc. Rate of feed and withdrawal was begun at 15 per cent per hour. At first it appeared that this rate of throughput would be too rapid, as the sugar increased slightly and the cell count dropped. However, after the system reached an equilibrium point, the 15 per cent per hour rate appeared to
1410
I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY
be satisfactory. Substitution of mechanical agitation for carbon dioxide agitation produced no noticeable effect. Data on experimental work in which alcohol yields were determined on molasses continuously fermented in the system for a period as long as 31 hours have not been reported, but it can be stated that the alcohol yields obtained with the continuous system were comparable to those obtained’ from standard 50-hour batch fermentations Of the same material.
Vol. 34, No. 11
Literature Cited (1) Alzola, Francisco, Mem. 14th conf. a n d , Asoc. tec. uzucar. Cuba, 1940,326-6. ( 2 ) Prescott, 9. c., and ~ u n n C. , G., “Industrial Miorobiology”, 1940. (3) Stiles, H. R., Peterson, W. H., and Fred, E. B., J . Bact., 12, 42835 (1926). PRESENTED before the Division of Sugar Chemistry and Technology a t the 103rd hfeeting of the AMERICAN CHEMICAL SOCIETY, Memphis, Tenn
National Survival through Science HARRY N. HOLMES Oberlin College, Oberlin, Ohio
H E S I speak oi national sur\i\ sl tlirougli science, I refer not only to the vit.11 aid of science in ninniiig this nar, but ilso to it, great service in the difiicult years that follon-. To the peqsimiits 11 ho believe that the Allies will lose and thst tlic rnited States nil1 finally be forced, by economic strangulation, to yield to Hitler’s orders I am coninelled to sav that under such t h r o t“tl i n g HARRYN. HOLMES our chief hoDe of survival as a free nation \vi11 lie in the resourcefulness of our scientists. The profound influence on our civilization of anesthetics and antiseptics, the steam engine, the electric dynamo and motor, the telegraph, telephone, and wireless, the cotton gin, portland cement, the pig iron furnace and steel mill, refrigeration, and the motor car convinc:s every thoughtful person that this is a scientific civilization. To be truly cultured yau must have some understanding of the achievements, the methods, and the possibilities of scientific research. Centuries ago recovery from great disasters such as plague, famine, flood, war, and oppression was slow, fatally slow for some nations. Medical science can now check pestilence in most of its forms, although it did not check the world epidemic of virulent flu in 1918 until millions of lives had been lost. The encouraging fact today is that science learns from every disaster, be it yellow fever, typhus, bubonic plague, an earthquake, a great flood on the Yellow River in China or on our own Mississippi. Typhus fever has killed 200,000,000 people in Europe and Asia during recorded time, and it is again threatening Europe in the war areas. The body louse that carries it is said to have done more than the Russian winter to defeat Napoleon. We are told that Hitler is in great fear of a typhus epidemic in his armies. The American and British armies have quantities of an effective serum. Malaria, carried by a vicious type of mosquito, weakened ancient Greece and Rome, helped bring them t o their fall. Recently you have read that MacArthur’s army in the Bataan
Peninsula had so little quinine that their resistance t o the Japs was tragically weakened by malaria. When the mosquito carrying yellow fever interfered with the digging of the Panama Canal, it was too much. General Gorgas and his medical staff got rid of the mosquito. At this moment we learn of many scattered cases of the terrible black plague in the West, a plague with a high percentage of mortality for which there is as yet no known cure, Fleas, carried by rats, squirrels, and rabbits, transmit the disease. Birds spread Rocky Mountain spotted fever. The tide of medical battle ebbs and flows, but the “men in white” always gain ground. They even make side forays against laziness, once a sin, and prove that much of it is caused by infected teeth, hookworm, malaria, hay fever, a deficiency of vitamins C and B1, and by other understandable troubles. We hesitate to call this a new distinction between virtue and vice. Famines force modern improvements in irrigation and flood control, stimulate the attack on crop diseases, force extension of transportation, and may lead t o planned redistribution of populations and to birth control. ‘(Modern science”, said Vice President Wallace, “when devoted whole-heartedly to the general welfare has in i t potentialities of which we do not dream.” T H E present disaster of a World JT7ar calls upon every resource, material and mental, if we are t o survive as a free people. To say that we were ill prepared in a military way is not enough. Our natural resources, which we viewed with complacency, had led us to prodigal extravagance. Forests disappeared and were seldom replaced. Antiquated farming methods permitted rains to wash away rich topsoil and depleted or ruined great areas by unwise cropping. The plowing of marginal grasslands of the West helped create the Dust Bowl, now being improved by contour tillage, deeper plowing, and alternate strip cultivation. Our most alarming extravagance, it seems, has been in the use of mineral wealth. Since 1900 world consumption of mineral resources has exceeded that of all previous ages. This acceleration cannot continue indefinitely. Recovery of scrap metal must become part of a carefully planned national economy. Substitution of products derived from the soil, such as wood, laminated plywood, and certain plastics, can help in conservation of metals. Power will not be produced indefinitely from coal and petroleum, so we will ultimately rely upon water power, alcohol
