T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
Feb., 1918
TABLEV-EFFECT OP DIFFERENT BRANDS OF ALLSPICE O N THE GROWTH OF MOLDSAND YEASTS Dilution Brand Brand Brand ORGANISM of spice A B C Alternaria tenuis(?) 1 : 25 0 0 50 0 0
...............
Penicillium glaucum..
............. 1
100
:
25 50
100
Rhieopus nigricans ................. 1 : 25 50 Aspergillus n i g e r . .
................ 1 :
Yeast Foam-culture
from..
.......
Fleischma nn’s compressed-culture from
...........................
Saccharomyces cerevisiae..
..........
Saccharomyces ellipsoideus. Saccharomyces anomalus..
.........
100
25 50 100 1 : 25 50 100
1:
25
50 100 1 : 25 50 100 1 : 25
.......... 1 :
50
100
25 50 100
0 0 0
+ 0
0 0 0 0
+0 +0 0
+0 +00 +0 0 +0 0
+ 0 +00 +0 0 ++0 ++ ++0 ++0 0
-!-
+ ++0
.... ..
++0 ++0 ++ +
.. .. .. .. .. .. ..
.. ..
*.
.. .. .. .. ..
on media containing cinnamon t h a t i t shows no difference a t all between Brands B and C in the dilutions used. Penicillium, on the other hand, is very sensitive and in dilutions up t o I : I O O showed no difference in the brands used for the results tabulated above. It may be t h a t some species of Aspergillus would, by its growth on media containing cinnamon, show ,he comparative preservative value of many brands. It is highly probable t h a t other organisms may be found which will serve the purpose even better than these which I have used. Some observations have been made on the variation in sensitiveness of two other strains of Rhizopus. For the results given in this paper I have used a culture of Rhizopus nigricans which I have designated No. 4 in my cultures. It produces a very vigorous growth on culture media and grows much more readily on media containing some kinds of spice t h a n a plus strain of Rhizopus nigricans. With a dilution of I : 7500 of cinnamic aldehyde, I found when I used Rhizopus nigricans KO. 4 t h a t there was scarcely any retardation of growth. With a minus strain of Rhizopus nigricans growth was somewhat delayed, but later quite vigorous, while with the plus strain there was no evidence of germination of the spores even after incubating the culture for a week. Using a I : 600 dilution of three brands of cloves, I found the plus strain of Rhizopus nigricans somewhat less sensitive t o this spice t h a n the minus strain. It is again evident from the data recorded above t h a t there is considerable difference in sensitiveness of any one organism t o the different spices and also t h a t no one spice has a n equally inhibiting effect on the growth of different organisms. This makes i t difficult t o determine the minimum amount of spice necessary t o preserve any food product. I t is necessary t h a t further data on the effect of spices on other organisms be obtained. T h a t , of the different spices, cinnamon is the most generally effective as a preservative, as I stated earlier,l does not seem t o have been proved. Results of further study with different brands of spice 1
LOC. cit.
123
indicate t h a t cloves may be just as effective as cinnamon. The best grades of these spices certainly exert a very considerable preservative effect, and although the amount used in flavoring a food product may not be sufficient t o preserve i t from spoilage, yet i t may be a large factor in its preservation. CONCLUSIONS
Molds, yeasts and bacteria show a marked variation in sensitiveness to different brands of spices. The amount of growth of such organisms in a given time on media containing spice may be used as a means of determining the relative preservative values of the different brands of the spice. SUM M A R Y
Microorganisms have been used t o determine the preservative value of different brands of spices. Spices of molds, yeasts and bacteria were grown on nutrient agar containing varying amounts of spice. Tabulated results of such a study using five brands of cloves, three of cinnamon, and three of allspice are given. The results show t h a t there is considerable variation in the preservative value of the brands used and t h a t the growth of microorganisms on a spiced medium may be used as a criterion of the preservative value of the brand of the spice. BACTERIOLOGICAL LABORATORY AGRICULTURAL COLLEGE MADISON, WISCONSIN
DISINFECTION WITH FORMALDEHYDE A SUBSTITUTE FOR THE PERMANGANATE-FORMALIN METHOD By C. G. STORM Received December 7 , 1917
The method proposed by H. D. Evans and J. P. Russell in 1 9 0 4 ~for t h e rapid liberation of formaldehyde gas from its water solution, the “formalin” of commerce, in a condition suitable for practical disinfection, has been found by numerous investigators t o be superior t o most of the other known methods of formaldehyde disinfection, as regards simplicity, rapidity, cost and efficiency. This method consists in pouring the formalin quickly upon crystals of potassium permanganate contained in any suitable metallic vessel (for example, a water bucket), the oxidation of a part of the formaldehyde furnishing sufficient heat t o cause rapid evaporation of the remainder of the formaldehyde together with the water. The permanganate method has found very general application and is widely used in this country in practical disinfection work. I t s use has, however, received a serious set-back by reason of the present scarcity of potassium permanganate and the resulting excessive cost of this chemical. Prior t o the war in Europe, potassium permanganate was obtainable in this country a t prices ranging usually from g t o I O cents per lb. It is now obtainable only a t many 1 H. D. Evans and J. P. Russell, “Formaldehyde Disinfection,” 13th Ann. Rept., State Board of Health of Maine, and J . A m . Chem. SOC.,27 (1905), 7 14. See also Daniel Base, “Formaldehyde Disinfection,” J . A m . Chem. SOC.,28 (1906), 964-96.
124
T H E JOURNAL OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY
times its former price, having held a t approximately $4.00 t o $4.25 per lb. for the past six months. This fact has impressed the writer with the desirability of publishing a note regarding an analogous method for generating formaldehyde, devised by him in October, 1911, which i t is believed has never been proposed for practical use, which preliminary tests indicate t o be safer, as rapid, and almost as simple of operation, and which will be much less expensive, owing t o the relatively low price of the material used. The new method depends upon the action between the water solution of formaldehyde and a soluble chlorate, and is apparently analogous t o the permanganate method, in t h a t the oxidation of a part of the formaldehyde furnishes a sudden evolution of heat which serves t o vaporize the remainder of the formaldehyde. It was suggested t o the writer in the course of analysis of a potassium chlorate explosive, On adding formalin t o the water solution of the explosive and heating the mixture, a violent evolution of gas resulted, increasing in intensity even after the tube containing the mixture was removed from the flame of the burner. The gas evolved was largely formaldehyde, apparently liberated from its solution by the heat generated in the oxidation of a, part of the formaldehyde by the chlorate. An examination of the solution remaining after the reaction had subsided, showed the presence of large amounts of chloride which had resulted from reduction of the chlorate. Repeated trials showed t h a t the violent evolution of gas resulted only from concentrated solutions, but t h a t the reduction of the chlorate to chloride, with a corresponding oxidation of formaldehyde, took place even in the case of very dilute solutions of chlorate. It has been demonstrated t h a t under proper conditions this reaction is quantitative, and the results of the study of this quantitative method for determining chlorates will shortly be published. The object of this paper is merely t o call attention t o what it is hoped will be a satisfactory substitute for the permanganate method of disinfection, and t o offer an opportunity for a more complete study of the method, the writer's investigation having necessarily been quite. incomplete because of lack of time and facilities for conducting work of this nature. Potassium permanganate reacts immediately on coming in contact with formalin a t ordinary temperatures, and if the permanganate is finely powdered instead of crystalline, the reaction may be violently explosive in character.l If formalin is poured on crystals of sodium or potassium chlorate, no action results until the mixture is warmed by application of external heat t o about 6 5 O C. This may be considered as a disadvantage, but as a matter of fact the reaction may be started with very little difficulty. The chlorate and formalin are placed together in a suitable metal container, such as a water bucket, of sufficient size t o prevent the reaction mixture from foaming over, and the bucket, properly weighted so i t will not float, placed in a large shallow pan (an ordinary dish pan 1.
1234.
G . B. Frankforter and R. M. West, J . A m . Chem. SOC.,28 (1906),
Vol.
IO,
No. z
will answer the purpose) containing water heated t o about the boiling point. The mixture becomes heated t o the reaction temperature in a few minutes, when bubbles of gas begin t o be evolved, this evolution increasing rapidly until it is so violent t h a t the mixture may foam over t h e top of the bucket. The action is completed in 2 or 3 min., and with the proper proportion of chlorate and formalin the residue remaining in the bucket is practically dry and consists chiefly of chloride together with some unreduced chlorate. Sodium chlorate seems t o give just as satisfactory results as potassium chlorate, and has the distinct advantage of costing less than one-half as much as t h e latter. Potassium chlorate is now quoted a t 5 0 t o 5 5 cents per lb., while sodium chlorate is listed a t 24 t o 2 5 cents. Several investigators have attempted t o determine what proportion of formaldehyde used in t h e permanganate process is liberated as gas and what proportion is oxidized by the reaction. Frankforter and West1 obtained a n evolution of 62 per cent t o 75 per cent of the formaldehyde from formalin by this process in a long series of experiments under laboratory conditions, using glass apparatus and absorbing the evolved gas in water, the strength of the resulting solution being determined. D. Base,2 in experiments with the process on a practical scale, used a specially prepared room of 2 , 0 0 0 cu. f t . capacity, determining the amount of formaldehyde gas in the room by drawing 5 t o I O liter samples through standard K C N solution, adding excess of standard AgNOa solution and titrating t h e excess of the latter with sulfocyanate. Base found t h a t not over 40 per cent of the total amount of formaldehyde used as formalin was evolved in the state of gas in the room. It has been suggested t h a t the reaction between formaldehyde and potassium permanganate is probably as follows: 4Kn/In04 3HCHO H20 = qMn(0H)z 2KzC03 COZ I t is, however, likely t h a t other reactions proceed a t the same time, in which part of the formaldehyde is oxidized to formic acid. Assuming this reaction, however, i t is calculated t h a t with the proportions recommended by Evans and Russell (roo cc. of 4 0 per cent formalin t o 37.5 g. K M n 0 4 ) 5.34 g. of formaldehyde, or about 13.35 per cent of the formalin, would be oxidized by t h e K M n 0 4 . With the proportions recommended by Base (100 cc. of formalin to 5 0 g. KMn04) 7.12 g. H C H O or 17.8 per cent of the formalin would be destroyed. These figures are considerably lower t h a n those found in the investigations mentioned above. I n any event it is apparent t h a t in the permanganate process a considerable part of the formaldehyde used as formalin is destroyed b y oxidation, the reaction supplying t h e heat which causes the rapid volatilization of the remainder. I t may be assumed t h a t the behavior of the chlorate with the formalin is entirely analogous t o t h a t of t h e
+
1 2
+
J . A m . Chem. Sac., 28 (1906), 1234. I b i d . , 28 (1906), 964.
+
+
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
Feb., 1918
permanganate, and t h a t the reaction occurs according t o the equation 2KC103 3HCHO = 2KCl 3H2O 3C02. Experiments have shown t h a t the proportion of 2 5 g . of chlorate t o IOO cc. of formalin is approximately the one giving best results, t h a t is, with these proportions there is practically no liquid left in the residue after the reaction subsides, the formaldehyde being either driven off as gas or oxidized and the water ,evaporated a t the same time. A simple calculation shows t h a t , according t o the above reaction, 25 g. KC103 will theoretically oxidize 9.18 g. HCHO or nearly 23 per cent of the formaldehyde in t h e IOO cc. of formalin, leaving the remaining 7 7 per cent t o be volatilized. It is probable, however , t h a t other reactions occur, such as KC103 3HCHO = KC1 3HCOOH. I n fact, appreciable amounts of formic acid, as well as COa, are evolved by t h e reaction of formalin with either permanganate or chlorate. I n a n attempt t o determine in a simple manner the best proportions of formalin and chlorate, a series of roughly quantitative experiments were made, using varying proportions of t h e two materials. A weighed amount of powdered KC1O3 was treated in a beaker with a weighed amount of the 40 per cent formalin and the beaker immersed in hot water in order t o start the reaction. The residue left in the beaker after the reaction had ceased was dissolved in water and titrated with standard solution of silver nitrate t o determine the amount of chloride present. From the amount of KC1 found, t h e weight of formalin representing t h e formaldehyde destroyed in the reduction of KC108 t o KC1 was then calculated from the reaction 2KC103 3HCHO = 2KCl 3C02 3Hz0. The results of these tests are shown in the following table. I t is t o be noted t h a t if the reaction KC103 3HCHO = KC1 3HCOOH is assumed, the calculated amounts of formaldehyde consumed will be just twice those given in the table.
+
+
+
+
+
125
chlorate was effected and the amount of formalin oxidized much less. A number of qualitative tests were made using formalin and sodium chlorate in proportions varying from 6 : I t o 2 : I, the maximum temperatures reached during the reaction being noted. With the ratios 2 : I , 2 . j : I, and 3 : I, this temperature was 108-109’ C., while with lower proportions of chlorate (4 : I and 6 : I) the temperature was slightly less, 104’ t o I O j o C. I n each case the reaction started a t 60-65’ C., was violent a t about 7 j ’ C., and lasted only about 30 seconds, the maximum temperature being indicated near the end of t h e reaction. The writer hopes t h a t comparisons of the actual disinfecting efficiencies of t h e permanganate and chlorate methods will be made by those who may be interested in the practical side of t h e question and t h a t the chlorate method may be found t o be of some use. O R ~ N A N CDFPARTMENT, E U. S. R. WASHINGTON,
D.
c.
EFFECT OF FERTILIZERS ON HYDROGEN-ION CONCENTRATION IN SOILS‘ By F. W. MORSE Received September 2 9 , 1917
Most of t h e fertilizer plots a t the Massachusetts Agricultural Experiment Station have been continuously treated for more t h a n 25 years, a n d there are marked differences in their crop-producing powers, which in some instances appear t o be due t o chemical or physical changes in the soil and not t o a deficit of the usual constituents of a fertilizer. Among methods of investigating these soils, t h e measurement of the hydrogen-ion concentration in water extracts of t h e soils has given some interesting results. The method of procedure has been as follows: 2 5 grams of air-dry soil were weighed into a n Erlenmeyer flask of 300 cc. capacity, and 250 cc. distilled water were added. The flask was repeatedly shaken during a period of a n hour, and then the mixture was filtered through a dry paper filter. The first portions of the filtrate were usually cloudy and were returned TESTS OF RESIDUEREMAININGAFTER REACTION BETWEEN FORMALIN t o t h e soil flask. When t h e paper became well coated AND POTASSIUM CHLORATB with soil, the filtrate would, as a rule, be clear, with HCHO oxi- Formalin (40%) Calc. KClOa dized (equiv. equiv. t o HCHO KClOa Formalin the exception of some limed samples which would Reduced t o KC1 found) oxidized TEST Used Used Grams Grams Grams persistently retain a slight turbidity from clay. The Grams No. Grams 1.439 3.59 1 7 12 3.917 soil and water were in contact for about 3 hours before 1,402 3.50 2 6 12 3.817 1.329 3.33 3 6 12 3.620 filtration was completed. 1.375 3.44 4 5 12 3.750 1.369 3.42 5 4 12 3.726 The colorimetric method was used for determining 6 3 12 2.990 1.098 2.75 the hydrogen-ion concentration. The range for the 0.738 1.85 7 2 12 2.010 soils was found t o be covered by the indicators methyl I n Tests 5 and 6 a very small amount of liquid red, paranitro phenol and rosolic acid. The standard remained in the residue after the reaction; in Test 7 a n salt mixtures used were Walpole’s2 acetic-acid-sodiumappreciable amount of liquid remained and a de- acetate mixture, Sorensen’s3 mono- and dibasic phostermination of formaldehyde showed 1.104g. HCHO, phates, and Clark and L u ~ s mixture ’~ of monopotassium equal t c ? 2.76 g. of 40 per cent formalin. I n Tests I phosphate and sodium hydroxide. The last named t o 5 , inclusive, where the weight of KClOs was a t least covers practically the same range as Sorensen’s and one-third of the weight of the formalin, the amount of is much more convenient t o prepare. HCHO oxidized was fairly constant, the KC1 found 1 Presented before the Fertilizer Division. a t t h e 5 5 t h Meeting of the indicating t h a t only part of the KC103 had been American Chemical Society, B o s t o n , September 10 t o 13. 1917. 1 Biochem. J . , 1914; J . Chem. Soc.. 1914. reduced. I n Tests 6 and 7 the excess of formalin a Ergebnksse Physiol., 1912. was such t h a t practically complete reduction of the 4 J . B i d . Chem.. 26 (1916). 504.
+
+
+
+
+
