Biological battery studies and related problems

Denver, Colorado, July JO,

l9L.8

Mr .. A. A. Clark Mr• C. H • Criswall tAr. H. L. Hartburg )!r. G. V,. Rienke /

Mr.

c.

E. Hirsch

Ur. H. W. Dahl'ber Gentlamm, Mro A. R. Nees

Mr.

E. Gonyou

Mr. V. V. Hartman Hr. N. J Muscavi tch Hrd

c.

R. Koontz Mr ..

c.

A. Horne

I

send you

herewith copy of a SwedjJlh pa._per on

biological battery studies which Mr o Dahloerg has .ki.n4cy" brought

to

my

attent10110 I believe that those

or

us who are concermd with ths problems of the Sil var dif fu.ser will find · considerable of interest in ito The Sv.iedish Sugar Conip3.IW

apparently has considerable difficulty wi t.h tl:e infection o! regular bat,teries because of tha fact that they return tl'ie pulp water and pulp press water t.o diffusiono This accounts for

tm

low pH of the battery supply water, des:..gna.ted a.s "press wa:rer11 _{in the paper o}

S. J. Osborn SJO:bg

Enc ..

BIOLOGICAL BATT6RY STUDIES AND ~LAT.J:;D Pi10BUMS

By Nils Vien:ian, Grad,_._r.._,_n.,..g_. _ _ _ _ _

An infection in a diffusion battery manifests itself by acid production in the juice as well as, in more serious cases, by gas formation. The purpose of the present study is to shed some light on certain phenomena associated with this

acidification. Direct :rrethods for the determination of the number of bacteria in the juice do not _6ive that information concerning the course of events during the

diffusion process which interests the technician. The acid production of the bacteria, on the other hand, gives an idea of the destructicn of material in an

infected battery, and it is primarily these aspects of the subject that have been investigated.

During the campaign

1944

investigations were carried out at a small number of factories with the priL,ary purpose of ascertaining the activity of

thermophilic bacteria in a diffusion battery. Bacterial activity of this kind was demonstrated by means of pH determinations in the juice taken from different

diffusers in the battery. Simultaneously the temperatures in the different

diffusers were read off the thermometers mounted on the calorisators and in certain cases the quantity of volatile acids in the juice-samples taken.

Fig. 1 shows an example of a seriously infected battery with a pH of

4.6

in the four.th diffuser from the water-end.

The test was performed as follows. First of all the pH of the pressure water was measured pressure for pressure. When the water first measured had reached the juice-end of the battery, samples were taken from all the diffusers at the same time. It was found then and at subse~uent tests that no correlation existed between the pH of the pressure water and that of the juice.

In the tests conducted in

19!..5

certain changes were made in the rethod. The temperature measurements were carried out by means of recording thermometers,

which were mounted on sockets specially fitted for the purpose (Fig. 2). The juice samples were taken direct from the pipe connecting the diffuser with the following calorisator. When juice temperature is mentioned hereafter, it has reference to

the degrees of the juice leaving the diffuser. Thus, temperature determinations and samplings took place at one anci the sarre point in the battery chain. The tedious determination of volatile acids was replaced by a potentiometric titration of the juice-samples at 20°C. By this rr~ans the advantage was also obtained that only entirely fresh samples were tested.

The first problem taken up for treatn~nt during the campaign of

19!..5

was

the determination of the incubation tirre for the Arl3v factory, i.e. the time

required for the bacteria to become acclimatized to the factory environment., Fully one week passed before sugar destruction and increased acidity in the pressure water could be detected by kee~ing-quality tests, and then small mobile

rods in this water were also discernible in the microscope. The Arl~v factory ran the battery at a relatively high temi,erature during the first period, and the

battery curves were in no way remarka.ole. Fig. J gives the first curves, which were taken about four hours after thu start. The top curve shows the temperature

of the juice leaving each diffuser and the pH of the juice at that time, measured at the same point. The bottom curve is the diffusion curve or the refractometric

of sugar per litre, no account having he re been taken, however, of the purity quotient of the juice.

The lowest curve but one is the acidity curve. By potentiometric titration the acidity of the juice-samples was determined and then converted into milli-e4uivalents per litre. With a good battery work this curve ought to run roughJ.y parallel to the diffusion curve, it being here assumed that the "neutral acidi ty11 _{of the beet material diffuses out into the juice at the same rate as sugar}

and other soluble substances. This assumiition is not, of course, fully correct, but it may be used as a starting-point so long as exhaustive investigations have not been made into this matter. In Fig.

4,

which also has reference to the Arl~v

factory, another state of the battery appears. The temperature is lower and the pE creeps under the 6 line. The diffusion curve is the usual one, but the acidity curve has taken on another appearance and tends to bulge upwards,

All the diffusion batteries of the Swedish Sugar Company, except that of the Lidk~ping factory, were tested in the same way as described above. At Arl~v altogether some ten measurements were made, and, as it may be of some interest to follow these battery conditions during the course of the campaign, still another measurerrent is submitted below. Fig.

5

is the last in this series ar.d it also shows a good battery work at the end of the campaign. As a general conclusion it may here be ment~oued that, with the exception of a short period at the end of October, the Arlov factory operated its battery without any noteworthy increases of acidity.

The battery work of the year 1945 in the H~lsingborg factory differed so much from that of year 1944 that it may be of interest to make a comparison.

Fig. 1 visualizes the conditions in 1944, with low temperatures at the water-end and a substantial lowering of the pH, whereas in 1945, Fig. 6, the temperature at the water-end was above 70° and no depression of the pH could be detected. The acidity curve follows the extraction curve, and no gas production was noticed.

An example of a much infected battery is presented in Fig. 7. The infection manifested itself in the first place by an intense gasification,

CO2+ H2, and the battery curves show that this was accompanied by a heavy fall in the pH and an increase in the acidity.

The pH of the pressure water was 2 .9 and the acidity was abnormally high right up to the 9th diffuser. None the less, the raw juice had normal pH and acidity values, a fact that will be explained in a later connection.

Exactly the opposite conditions are exemplified by the battery work shown in Fig.

8.

The temperature curve is already remarkably flat after the first

diffuser. At the water-end it has risen to above 72°c. The acidity curve runs concentric with the extraction curve, and the pH curve is noticeably straight. The battery worked very rapidly and no gasification occurred.

In order to get sorre idea of the correlation between the refractometric solids, natural acidity of the beet and other factors during the diffusion process recourse was had to the experirrental battery of the Central Laboratory, and a number of extraction curves were plotted for Pol., refractometric solids, acidity, total nitrogen and ash (Fig.

9).

If the values for acidity and nitrogen are commuted so that the end-points for diffuser

14,

i.e. the raw juice, coincide with the end-point for the refractometric solids, it will be found that the curves also become practically identical. On the other hand, the ash, determined by Lunden1_s

technique, does not seem to follow exactly the same laws as the sugar, acidity and

nitrogen. At the water-end of the battery more salts are extracted than the diffusion curve indicates, but after the 4th diffuser of the experLnental battery the ash constituents are extracted according to the diffusion curve. This will be seen best if the concentrations are graphed on logarithmic scale, when the

diffusion curve becomes rectilinear (Fig. 10).

To control the electrochemical ash determination as calculated by

Lunden 1 _{s method, potassium determinations were carried out with the flame photometer}

of the soil laboratory. Identically the same kind is obtained in this potassium -extraction curve as in the salt curve, which is rather natural, seeing that

potassium is the metal which occurs most abundantly in the plants.

In the study of the relationship between sugar loss and acid formation in much diluted beet juices, such as pressure water and jnices from the diffusers at the water-end of the battery, the keeping quality tests fB rformed at the factories may afford a link of son~ interest. The material is extremely lacking in

homogeneity, and unfortunately certain analytical results have had to be omitted, being obviously erroneous.

Factory It Arlov Hasslarp Angelholm

~'k" .

o opinge Jordberga Karpalund K11 av 1· inge K11 • b opin~e ro II Garsnas _{It :)} Morbylanga Skivarp telleborg tofta TABIE 1 KEiPING QUALITY T~STS

Equivalents of acid per mol of consumed saccharose

No. of

₄

Tests

_-

Hours 12 2.18 1S 0.27

54

0.35

18 o.86 s s.09

s

1.40 26 1.86

19

o.84 53 o.S3 3S 0.76

7

0.01

16 14.10 8 1.67 8 Hours 2.24

o.44

0.37

0.91 3.10 l.bO 1.67

o.45

0.75 0.78 0.20 15.14 1.17 Fermentation of a saccharose solution with for instance butyric acid bacteria follows the course

C12H22Di1 + H20 ~ 2CH3CH2CH2COOH + 4 CO2+ 4H2

One molecule of saccharose gives two molecules of acids besides gases.

A 1_{r ov,} 11 K" 1 · _{av inge an} d _{possibly Karpalund and urtofta have acid formations} · 11 that indicate this course, but the values given for the other factories differ considerably from the fundamental forr,mla. The extreme values are found at Skiva~p with 0~07 and at Trelleborg with 14. At Trelleborg and Jordberga a sub-stantial quantity of other substances than sugar take part in the acid formation wh_ile at Skivarp and most of the other factories sugar is consumed without any ,''

corresponding ac:idification. The only thing that can be said with certainty is that we are here faced with a very complicated problem.

A small number of keeping-quality tests have been carried out at the Central Laboratory, primarily with a view to studying the effect of the ammonium ion on the fennentation process. 1Ni thout addition of ammonia, 2 .0-2

.5

equivalents of acid i-,er mol of sugar were here obtained but ·with an addition of 1 g of H

₄

NC1 per litre of

3-4

e4uivalents. As a curious fact it may be mentioned that at one of these tests the same value wa.s obtained as at T:rellebo:rg, viz.,

14

equivalents of acid per mol of sugar,. F1_~0t1 the factories pressure water was taken after the campaign for a closer investigation of this correlation, but unfortunately these samples, in spite of being stored in a refrigerator, were rapidly attacked by moulrl.

and became useless for the purpose.

A more detailed discussion of these acidity r.urves will necessitate our assuming that the acidity curve which runs uniformly with the diffusion curve is de}lendent upon two factors, viz., the natural acidity of the beet and the proportion of bases contained in the beet, the acid component obviously being in dominance. In those cases in which the acidity curve deviated from the sha}le of the normal curve and showed higher values for the acidity than what can be expected in normal work, this deviation should imply an acid formation caused by a process of fermenta-tion. That the final acidity will always be about the sane - 0.026 (%Cao or

9 .J milli-equivalents/li tre - ought, superficially viewed, to depend on the fact that at the juice-end of the battery so many alkaline substances pass into the juice that the newly formed acid is neutralized. However, this problem is not so simple, for at the same moment as the concentration of acid in the juice becomes higher than that in the pulp there arises a diffusion in the opposite direction, and the pulp becomes enriched with acid. If the diffusion process is regarded as a counter-current process, it can be outlined as follows:

Pressure water

_{+ - - - -}

_--,)1

_{- - - - - Raw juice}

A B

Pulp & discharge w a t e r - - - -- ---,.Cassettes

The beet slices travel in one direction, losing their sugar and other substances, while on the other hand the juice moves in the opposite direction and is enriched by these substciilces. Fermentative processes With acid and gas formation mainly occur at the water-end of the battery, and the shape of the acidity curves shovs, as already pointed out, that the acidity diminishes at the juice-end. If we now

assurre that the acid-forming fermentation occurs in the diffusers A-B, the acid will naturally accompany tre juice towards the juice-end of the battery but then meets beet with a lower acid concentration, whereupon some of the acid changes its

direction of motion and accompanies the pulp and the battery discharge water out of the battery. Between Band A there should not, reasonably seen, be any notable diffusion of acid, as the concentrations are approximately equal in juice and beet.

To test the accuracy of this thesis recourse was had to the experimental battery, and as pressure water use was made of saline solutions. First, however, the battery was run to a state of inertia with distilled water and juice-samples were taken, whereupon the diffusion contined with saline solution until a new state of equilibrium was obtained (Fig. 11). In the first test the migration of the potassium ion in the battery was studied. Curve I is a diffusion curve for the natural potassium con tent of tre beet expressed as K20, and curve II that for l per cent KCl. From the latter it is seen that the }X)tash content falls rather

rapidly and approaches the nat~ral potash content of the beet in the last diffuser.

Only

ho%

of the potash of the potassium chloride has reached the juice-end.

Curve III represents a test with pressure water containing 0.25% KCl. The test

was carried out a few days later and with another delivery of beets and therefore

the end-point was not exactly the sarre as for curve I, but it is distinctly seen

that tre potash content first drops in order to rise again when it approaches

curve I. The chloride-ion content is here detennined at three points, and at the

7th diffuser was

73%

and at the last

14%

of its original value. The chloride-ion

content of a raw juice is difficult to fix, and these figures must be regarded as

approximate.

Another experiment was made with a 1-per cent ammonium sulphate solution,

curve IV. In this case only one-tenth of the original SOJ entered the raw juice,

Stanek and Vondr1ik havu earlier made similar tests with common salt and

carbamide in another connection,...,:·

Even if the chief reason for these diffusion tests was to prove the changed direction of movement of the acid in the battery caused by fermentation,

trey are also of practical interest. The fact is that there are good grounds for

assuming that those substances, foreign to the beet material, which enter the

system with the pressure water, such as sulphuric acid, ammonium hydrate, formalin,

etc., only partially pass into the raw juice while the bulk of trem vanish by the

same route as they entered, i.e. through the water-end of the bettery.

When the numberical material after the battery investigations of the year

1945

is considered, it is difficult to get a clear view of the part played by the

battery temperature on the bacterial fermentations that occurred in the diffusers.

The addition of formalin caused so great a disturbance of tre fermentation process

that direct indications as to the influence of higher or lower temperatures cannot

be univocally elicited. Judging from everything, it may be expected that a

power-ful fermentation can still arise at round about 80°C. When the samples were

taken, the K~vlinge factory was running wj_ th fermentation in the battery but

simultaneously with a flat temperature curve at about 78°c. None the less, we

ought not to rely entirely upon formalin, since even this disinfectant nay fail to

accomplish its purpose, as was shown by a test at Roma.

During the campaign of

19L5

formalin was used as a disinfectant to a

considerably greater extent than had previously been the case. An account of how

this formalin was utilized may therefore be relevant in this connection. A close

study of the disinfecting records of the factories reveals that formalin was

intro-duced into the operations at different points in the return system as well as

direct into the pressure water. In some cases, among others at Skivarp,

Staffanstorp, sibyholm and Teckomatorp, the formalin was pumped direct into the

battery. At some factories it was also passed into the juice-end of the battery,

and against this method the objection may be urged trat this is rather late, for

by then most of the damage r~s already been done. The r..ajority of the factories

add the main portion of the formalin in the Babrowski filters or the pressure

water tank. Several factories introduce small amounts of formalin in the beet

elevators, cutting machines, pulp pjt,s and at other secondary points. The size and frequency of the individual formalin additions vary from a few litres per addition

to 100 litres a time at one point. There is a considerabln vari~tion in the

litres of formalin added per 1000 tons, as will be seen from Table 2.

Arl~v jjasslarp Angelholm Hlhsingborg II II • Hokopinge Jordberga Karlshamn Karlpalund K11 av inge 1· K6pin~ebro GM.rsnas TABI.E 2

LITR.c;S OF' FORMALIN FER 1000 TONS OF BiETS

ACCORDING TO FACTORY DISINFECTION R.-:;CORDS

58

Lidk~ping

99

lonk~ping

89

rbyl~nga

94

Roma

55

Skivarp

67

Staffanstorp

73

Sabyholm 11 10 Teckomatorp

71

Trelleborg

94

'ortofta

58

41

79

66

122

65

64

93

80

71

77

No correlation between formalin addition and unkown losses has been found

on comparison between the factories.

Various formalin-adding methods of special interest will be described

below.

At Karlshamn the whole dose goes to the pressure water. As a dosing

device a small petrolometer of ordinary type is employed, and the quantity of formalin measured off is allowed to flow into the discharge pipe from the

Babrowski diffusion filter. Formalin is given twice daily, beginning at

8

A.M.

and 8 P .M. £,very seven minutes,

5

litres of formalin are introduced, and this is

repeated nine times~ In this way,

45

litres of formalin are given in the course

of 63 minutes. The time of 63 minutes has been chosen because it coincides as

nearly as possible with a cycle of operations in the battery. The plant worked

very well, and no irritating odour of formalin could be detected on the premises. At Teckomatorp there is an original and practical system of pumping

formalin into the diffusers. Leading to the neck of each diffuser is a brass

tube, which is fed from a main pipe. At the conunencement of each shift,

JO

litres

of formalin are pumped into the battery, distributed over all the diffusers. On

account of the difference in pressure existing between trn water- and juice-ends of the battery it is expected that the larger ~ortion of the disinfectant enters

by the juice-end, where the pressure is lowest. This has been the object desired

at this factory because it is primarily the raw juice _-for which disinfection is

wanted.

At Skivarp, Staffanstorp and S~byholm the formalin dose is forced into

the water-end of the battery by means of test-pressure pumps. The amounts at each

dosage range from

10

to

Lo

litres,

Judging from the figures for the losses, the increased addition of

formalin together With other anti-fermentation measures has'been of benefit.

However, no special method cf introducing formalin and no definite quantity of

formalin can be pointed out as giving the best result. This question is therefore

difficult to judge, but sorre points of view may nevertheless be submitted.

When a fire is being extinguished, the jet of water is directed a~ainst

the battery should be prevented against the flow of juice and not with it. From

tests undertaken at Arl~v at the beginning of the camµaign the conclusion may be

drawn that tha bacteria require a certain time to become acclimatized to the

factory environment. It is undoubtedly not the bacteria introduced with the beets

which are the most dangerous, but those which enter with the pressure water, since the latter micro-organisms are acclimatized to the battery environment. The most

serious gasification and acidification are invariably found at the water-end of the

battery. If counter-current disinfection is to be adopted, which is doubtless the technically correct procedure, this should be effected either in a diffuser at the water-end or in tre pressure water immediately before this enters the battery.

As regards the course of the formalin through the battery the sane views

must be adopted as previously in the case of newly formed acid. Large propcrtio~,

of the formalin in the pressure water diffuse over into the beet, change the

direction of motion and pass out With the pu1.p and the discharge water. What does not remain in the squeezed pulp, re-enters the pressure water.

Earlier investigations by Mr.

s.

M~nsson into the effect of different disinfectants on the bacteria of the pressure water show that there is a lower limit of concentration for the bactericidal power of the formalin. Accordi.ng to M~nsoon, this limit lies at 2-3 ml. of 37 per cent formalin per litre of pressure water. If a formalin addition is to be of real value, it ought to be carried out

so that all the return water in circulation will become practically speaking

sterile. After that a certain time may be assumed to be required for the bacteria

to accumulate again in the pressure water. The length of this period of growth cannot be stated here, but the interval between two immediately successive

dis-infection·s must be fixed separately for each factory. Keeping-quality tests and microscopy can give information as to·when it is time for a fresh addition of formalin,

Large quantities of formalin can doubtless be saved by scavenging and flushing away the pulp that remains lying under the diffusers on the stands and masonry. The shaping of the pulp-elevator pit plays a great role in a hygienic respect. Several factories have reported formalin addition in the pulp pit, and

such a procedure is no doubt a good one, but a better one is to provide place for a water-hose and to cleanse the masonry, girders and machine-parts. Without

fre-quent cleansings of those places where pulp and battery discharge water collect the formalin treatment will be more ex_p3nsive than it need be.

Means of determining how large a sugar loss is caused by these bacterial

fermentations has long been an object of desire. Confronted with those

conclu-sions concerning the diffusion ~rocess tc which more especially the acidity curves

have led, we must for the present abandon any expectation of this. Another

question that for the present must be left where it stands is the species determina •.

tion and study of the specific vital conditions of the thermophilic bacteria.

Hitherto the work has been directed to localizing the foci of infection, to

working out special analytical methods and, on the basis of current opinions on the subject, to quickly evolving technically applicable procedures for the prevention as far as possible of the damage caused by the bacteria.

The analytical nethods used in this work are so simple that they can be applied by all sugar factory chemists. liereafter tre factories ought themselves to be able to plot the battery curves ~resente~ in this paper whenever there is

Much still remains to be investigated in the purely microbiological region before we can attain to technically sterile working methods in our sugar factories.