T H E JOCR-VAL OF I-VD L - S T R I A L A S D E i Y G I S E E R I S G CHE:\fISTRY.
5 40 TABLE 1,
EXTRACTION OF BENZOIC ACID
Aug., 1909
A
mitted to other laboratories for collaborative work and for comparison with other proposed methods. Tabulated results of this work are given in U . S. Defit. of Agriculture, Bureau of Chemistry, Bulletin 122, Proceedings of the Association of Oficial Agricultural Chemists for 1908.
This table shows that the process of extraction with ether is complete. The result on sample No. 4 is probably due to incomplete drying,
ECONOMIC REASONS FOR THE REDUCTION OF MILK TO POWDER.'
WITH
ETHER
FROM
WATERSOLUTION. Sample number. 1 2 3 4 Average
TABLE2.
EXPERIMENTS
Amount used. Gram.
Amount extracted. Gram. 0.091 0.1018 0.102 0.114 0.102
0.1
0.1 0.1 0.1
ON THE USE O F WEIGHED FILTER PAPERS FOR FINAL WEIGHING.
By LEWISC. MERRPI.I.. Received May 3, 1909.
The present method of milk supply is wasteful, (Water solution of sodium benzoate + 1 H20 used.) uneconomical and generally unsanitary. It is Amount of Weight of silSodium benzoate recovrrrd. not to be supposed that a civilized community sodium ben- vex benzoate Sample zoate present. obtained. Weight.' in which sanitary science is advancing so rapidly Gram. Gram. Gram. number. Percentage. will continue to tolerate the menace of impure 1 0.100 0.1366 0.1122 112.0 milk, when a satisfactory solution of the whole 2 0.060 0.077 0.0545 90.8 3 0.060 0.0556 92.70 0.0786 milk question is offered. 4 0.030 0.0300 0.0212 70.0 While the production of certified milk offers 5 0.030 0.0326 0.0231 77 . 0 decided advantages over old methods in the matter The use of a tared filter paper for final weighing of cleanliness, i t by no means solves the universal does not give as satisfactory results as when an asbesmilk question, for there is no present prospect of tos gooch is used. a sufficient supply of certified milk to make its TABLE3. &PLICATION O F THE FINALM,ETHODTO XVATER STAXDuse a t all common, and the expense of producing ARDS. i t places it beyond the reach of the modest income (Using a water solution of sodium benzoate + lHzO and an asbestos gooch.) except for the special uses of infant and invalid. Amount of Weight of Sodium benzoate recovered. Contrary to general opinion, the production of sodium bensilver hen----.clean milk is not much more expensive than the Sample zoate present. zoate obtained. Weight.' number. Gram. Gram. Gram. Percentage. production of unsanitary milk, The expense neces1 0.030 0.043 0,0307 102.0 sary in the production of certified milk is charge2 0.030 0.041 0.029 96 .7 3 0.030 0.040 0.0283 94.4 able rather to methods and apparatus used in TABLE 4. APPLICATIONo f METHOD TO KETCHUPS CONTAINING keeping the milk clean from contamination and KNOWN AMOUNTS O F S O D l U M BENZOATE. protecting i t from unfavorable conditions during (Gooch asbestos filter used.) shipment and storage. Such experience as I Sodium ben- Weight of silSodium benzoate recovered, have had in the production of clean milk leads zoate ver benzoate Sample added. obtained. Weight.' me to believe that any careful farmer with a reanumber. Grant. Gram. Gram. Percentage. sonably well equipped dairy can produce milk 1 h-one 0.0000 0.0000 ..... 2 0.006 0.004 ..... containing not more than twenty-five hundred 3 0.0009 0,0006 ..... to four thousand bacteria per cc. in the milk pail. 4 0.0000 0 .0000 ..... 5 0.004 0,0028 ..... The bacteria count of this same milk a t subsequent 6 0.004 0.0025 ..... stages depends entirely upon the conditions under 7 0.0019 0.0014 ..... 8 0.030 0.041 0.0290 97.0 which the milk is kept. My point is that good, 9 0.060 0.090 0.0637 106.0 clean milk, well below the certified milk limit as 10 0.060 0.0783 0.056 93 . O 11 0.060 0.0783 0.056 93 . O to bacteria count, can be produced without much 12 0.060 0 ,0792 0,0560 93.5 additional expense in any dairy, but the expense 13 0.060 0.0834 0.0591 98.5 14 0.100 0.1410 0.1005 100.5 involved in preparing such milk for the market, 15 0.100 0.1410 0.9996 99.96 that is to say the expense of preserving it, is such This method has been used in the routine work as to make the use of certified milk very limited. of analyzing commercial samples of wine, jams, I t is apparent that we need an inexpensive ciders, molasses, sirups and ketchups, in the Divi- method of preserving milk for the market, a method sion of Foods. The method has also been sub1 Read before t h e Syracuse Section of the American Chemical
-----
------
1
Calculated from preceding columns
Society, November 23, 1908.
MERRELL O S REDCCTIOS OF -1IILK T O POWDER. less expensive than is employed in preparing certified milk so that it will keep. The ideal method of milk preservation should prevent decomposition or increase of bacteria count, without the use of preservative substances, without changing the chemical composition of the material or be preserved, and without altering such characteristic qualities as taste, odor and appearance. All these advantages may be secured, as I will presently show you, by a properly conducted system of desiccation. This will a t the same time prevent injury to the milk from estremes of heat and cold and will eliminate the possibility of contamination during shipment and storage. Reduction of bulk and weight are also secured, resulting in considerable saving in freight. Before entering into a further discussion of the advantages and possibilities of milk desiccation, I wish to take up, in a general way, the composition of milk and explain some of the difficulties which have tended to prevent the accomplishment of this extremely desirable result, namely, the reduction of the solids of milk to a powder. I will take the following analysis as typical of a n axyerage milk: Pat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.60 Ifilk s u g a r . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.58 Casein. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Albumen ............................ Ash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solids. . . . . . . . . . . . . . . W a t e r . . .............................
2 .81 0.56 0.75 12.30 87.70
If we remove the moisture from this milk, we will have approximately the following analysis : Casein.. ............................. Ash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
29.27 37.21 22.85 4.55 6.09
Let us consider these different constituents in their relation to desiccation. If the butter-fat of milk is heated for a considerable length of time, especially if in contact with a heated metal surface, the butter-fat loses its finely subdivided globular condition and melts into a grease. For this reason all milk-desiccating processes, which require heating the milk in a container or upon a metal surface, have been most successful with milk from which the cream has been separated. If any desiccated milk is to solve the problem of our milk supply, i t must contain all the cream, in no wise altered from its natural condition. Milk sugar presents few obstacles in the desic-
541
cation of milk, provided the temperature be kept below 2 1 2 ' F. This temperature may be exceeded for short periods of time without caramelizing the milk sugar, but great care must be exercised. Some processes depend upon desiccation in z'acuo to prevent the browning of the milk sugar. That this is not always successful is due to the fact that in order to maintain sufficient temperature within the vacuum to cause boiling and consequent evaporation of the water of the milk, i t is necessary to use a temperature above 2 1 2 O F. in the steam jacket of the vacuum chamber. Hot water has been tried as a source of heat but i t renders the evaporation so slow as to be impracticable. The wall of the vacuum chamber, therefore, is above 2 1 2 ' F. even though the interior atmosphere of the chamber is only 130' to 140' F. In consequence of the temperature of the chamber wall, some of the milk sugar is usually discolored. Lactic acid, while not mentioned in the above analysis, is always a constituent of milk. It results from bacterial action upon the milk sugar, and exists in larger or smaller amounts according to the care with which the milk has been drawn and kept. It is not decomposed by any drying process, but will, under certain conditions, combine with other constituents of the milk, giving undesirable results. The removal of the water from milk concentrates the acid to eight times its strength in the liquid milk. -1 good quality of fresh milk will contain from ten to fifteen hundredths of one per cent. acid. IThen this same milk is reduced to dryness, its solids will contain approximately one per cent. of acid. One of the constituents of milk which is affected by this concentration of the lactic acid is casein, which is precipitated from its normal suspension and becomes insoluble. This effect is more pronounced under the influence of heat. I t is sometimes avoided by neutralizing the acidity of the milk with suitable alkalies. If this is overdone, a soapy taste is imparted to the milk, and even if the neutralizing is accurately performed, the milk tastes salty on account of the formation of lactates. If desiccation is performed a t a comparatively low temperature and very rapidly, the concentration of the acid has very little effect on the casein. The other constituent of milk which is affected by the concentration of the lactic acid is the albumen. Unlike the casein, the lact-albumen is in true solution. It is coagulated by acidity and heat,
542
T H E JOC'RlVAL OF I N D U S T R I A L A N D ENGIIVEERING C H E - W I S T R Y .
and even though the acidity be neutralized, the coagulating effect of the heat cannot be avoided as with casein. This coagulating effect is produced at any temperature above 149' F. and depends upon the length of time of the exposure to heat. It is evident t h a t to avoid coagulating the albumen, either a very low temperature must be employed in desiccation or else t h a t the drying must be accomplished in a space of time decreasing in length as the temperature increases. No process of desiccation which depends upon reduction to dryness by boiling in open air or in vacuo has ever, so far as I a m aware, produced a milk powder containing a completely soluble albumen. The relation of the milk salts to desiccation is not important and I will not discuss i t here. There is still a further difficulty which lies in the way of those who have attempted to reduce milk to dryness. When its water content is reduced below 40 per cent., it acquires a consistency somewhat like library paste. After i t reaches this condition, i t is very hard to dry. It does not reach a condition where it can be powdered until the water content has been reduced to perhaps IO per cent. It differs in this respect from most other food substances that are ordinarily dried. I had occasion some time ago to make some exper ments with starch. The particular sample I was examining became liquid with a water content of 5 5 per cent. When this water content was reduced to j o per cent. b y the addition of more starch, the starch became solid and capable of being powdered. There was no intermediate sticky, pasty condition. I t is this peculiarity of milk more than anything else that prevented the perfection of a satisfactory milk-desiccating process for more than fifty years. When milk is in this condition, i t is very easily injured b y heat. The water cannot be boiled off for bubbles cannot rise through the sticky mass. I n fact, the water is very difficult to remove in any manner. Various methods have been employed for treating milk to remove the moisture while in this sticky state, as for instance stirring i n vacuo, pressing through a shredding machine, and exposing to dry air, rolling into thin sheets either in zlucuo or in the open air upon a heated surface and scraping the dry material off with a knife. All of these methods require so much mechanical manipulation, rubbing and stirring of the pasty milk, that the globular form of the butter-fat is destroyed, and the manufacturers are unable to make a whole
Aug., 1909
milk powder or a cream powder that resembles fresh milk and cream when dissolved, nor are they able to manufacture a powder containing butterfat that will keep. There is no reason why the butter-fat of milk should not keep if its original globular form is preserved. Many seeds and different kinds of meal and flour contain large percentages of f a t and oils, which do not become rancid unless the substance be heated or otherwise treated so as to break down the globular structure of the f a t or oil, when i t becomes rancid. A good illustration of this is the peanut. It contains a large amount of oil, yet does not become rancid while raw. After i t is roasted the oil soon begins to oxidize, and the peanut butter, which is the expressed oil of the peanut, becomes rancid very easily. I said earlier in my paper 'that a properly conducted system of milk desiccation was the ideal method of milk preservation, preventing decomposition or increase of bacteria count, without the use of preservative substances, without changing the chemical decomposition of the milk and without altering its state, odor and appearance. I will now describe such a process, showing you how the difficulties before mentioned may be avoided and a perfect milk powder be produced which offers all the advantages of certified milk a t a much less cost. Fresh whole milk is drawn into a vacuum pan and a portion of its water removed. This condensation is halted while the milk is still in a fluid condition and before any of the milk albumen has been cooked on to the walls of the vacuum chamber. The milk is then drawn from the vacuum pan and sprayed into a current of hot air. The moisture of the milk is instantly absorbed by the air and the particles of milk solids fall like snow. Upon examination, they are found to contain less than 2 per cent. and sometimes not more than one-half of one per cent. of moisture. The hotter the air is the more rapid the drying effect and the less. danger there is of injuring the milk solids by heat. I say this advisedly. Let us assume that each little particle of milk is a sphere, and that evaporation is proceeding upon all portions of its surface. This evaporation not only takes heat from the air, b u t also from the interior of the milk particle, so that the milk solids become cooler and cooler until completely dry. After that, the teniperature does not affect them. As illustrating the fact that thoroughly dried albumen cannot be coagulated by heat, I took
MERRELL O N REDUCTION OF ,MILK T O POWDER. a sample of egg albumen, dried b y this process, and hermetically sealed i t in several cans. These cans were placed in boiling hot water for periods varying from one-half to four hours. Afterward the albumen was examined for solubility which was found not to have been impaired, the albumen in the can subjected to heat the longest showing the same solubility as the original unheated albumen. This method of desiccation does not destroy the globular condition of the butter-fat, i t does not burn the milk sugar nor does it coagulate the albumen of the milk. It is not necessary to neutralize the acidity of the milk, for the moisture is removed so quickly that there is no ,chance for chemical action, and neither the casein nor the albumen are affected in any way by the concentration of the acid. The difficult pasty condition of the milk solids is passed while the milk particle is suspended in the air and not in contact with heated metal. As nearly as I can estimate, one pint of milk presents about two acres of surface when sprayed into the air. The individual dried particles are from one two-thousandth to one ten-thousandth of a n inch in diameter. No bacterial action has been discovered in milk powder containing less than 3 per cent. moisture, and no chemical deterioration takes place. It is, therefore, evident t h a t the milk powder product described above, fulfils my definition of a n ideal preserved milk, for decomposition is prevented merely by dryness and without the use of preservative substances and without changing the chemical composition of the milk. As to the odor, taste and appearance, I will leave you to judge from the samples I will show you. I will say, however, t h a t this whole milk powder is in use in place of fresh milk a t several of the United States soldiers’ homes and military posts as well as in the navy. It has been subjected to the most exhaustive tests by the United States Department of Agriculture, Bureau of Chemistry, and by the Experiment Stations of different states. The Pacific fleet carried a ton of it around the Horn under a n absolute guaranty as to keeping quality and has since reordered largely. A IOC package makes one and a half quarts of milk, a t $0.06 2 / 3 a quart. With these results in mind i t is not too much to assume that the reduction of milk to powder offers a satisfactory solution of the universal milk question. Primarily there are two reasons for the preservation of any perishable food substance: in the first place, that i t may be stored for future use
543
and in the second place so that i t may be transported from place to place economically while retaining its desirable qualities. Both of these reasons apply with special force to milk which is one of the most difficult of food substances to store and transport. Giving our consideration to the first of these questions, there are several important reasons for the preservation of milk by reducing it to powder. Two-thirds of the milk supply is produced in six months from April to September, and the remaining one-third in the six coldest months of the year. The price drops in summer because the supply exceeds the demand and because milk is more difficult to keep in the hot weather. The price rises in winter because the demand exceeds the supply. The surplus :n summer goes into butter and cheese b u t there is no way to augment the winter supply when i t falls short. The storage of whole milk in summer in the form of a powder assures the country of a n adequate supply in winter, The storage of food always tends to steady prices, enabling merchants to do business on smaller margins, thus reducing prices to the community in general. The advantages of storing fresh milk for future use on the sea and in very cold and very hot climates are well understood and need no comment here. The second reason for the preservation of milk, namely, that it may be transported from place to place economically without deterioration is probably even more important. There is great difficulty and expense connected with the transportation of a highly fluid and perishable substance like milk. New York City gathers its supply largely within a radius of three hundred miles though in certain cases milk for domestic consumption is drawn four hundred miles to New York. The daily average of milk required to supply New York City during 1907 was 49,306 cans, a n increase of 1335 cans per day over 1906. The freight rate was $0.23 for the first forty miles, $0.26 for the first one hundred miles, $0.29 for the first one hundred and ninety miles and $0.32 for all beyond this distance. If a person were to ship milk from the outer zone to New York City, i t would cost $0.32 for a fortyquart can containing eighty-five pounds of milk, A discount of I j per cent. is allowed on carload lots of not less than 10,000 quarts, making a net cost per can $0.272. Eighty-five pounds of milk would contain I O 314 pounds of powder. Assuming, therefore, that the solids represent the
544
T H E JOCRNAL OF I.VDUSTRIAL ALYD EZI’GISEER1,VG CHE.I/IISTRY.
Aug., 1909
value of the milk, it would cost $2.53 to ship to in proportion to population may be taken as inNew York, liquid milk equivalent to IOO pounds dicating a continual increase in the consumption of powder. The freight rate from the same zone to of fresh milk. Milkisnot, however, by any means New York on milk powder i.s $0.25 per hundred, as widely used as i t should be. This is partly due in less than carload lots or $0.18 per hundred in to the reluctance of the housewife to purchase an carload lots. As the material is not perishable, ample supply, owing to its perishable nature. It i t is fair to take the carload rate. We can, there- is a well-known fact that convenience increases fore, ship for $0.18 in powder form an amount of use. Street cars are more depended upon when milk which i t would cost us $2.53 to ship in liquid they run every three minutes than if there are form. Besides this, the shipper of liquid milk five-minute intervals, for many people who would runs the risk of having his milk spoil, while the walk rather than wait, will ride if there is a car at shipper of milk powder runs no such risk. The cost hand. The average housekeeper takes a given of-shipping the milk in powder form is only about number of pints or quarts of milk a day. Some 7 per cent. of the cost of shipping the same amount days she needs more than she has, other days she of milk in liquid form. Assuming that the average has more than she needs. Either she is wasting rate per can to New York is $0.25,it cost last year money by buying too much or else the milk dealer $12,326.00per day freight on Eew York supply is losing business. I t is evident that with a of milk. This same milk could have been delivered powdered milk always on hand ready when wanted, in New York in the form of powder for $863.00 the use of milk would be popularized. per day, a saving of $11,463.00 per day, a saving I n England where ice is but little used for preof 94,183,995.00 per year. serving food, the housewife buys only as much of The average cost of New York’s supply of milk is meat or vegetables as can be used during one day. $1.30 per forty-quart can or $0.03$ per quart, Milk peddlers are passing in London every few mindelivered a t New York. This milk costs the New utes and milk is bought only when wanted. Such Yorker about $0.08 a quart, leaving a margin of methods seem strange to us because we are ac$0.043 per quart for cost of delivery and profit customed to depend upon preservation of food by to the dealer. The same amount of milk can be ice. I t is not impossible therefore that our present delivered to the consumer in the form of powder method of supplying milk may seem strange to on a margin of $ 0 . 0 2 1 per quart for cost of delivery people a few years from now, when milk powder and profit to the dealer, showing a saving of $0.024 has become a staple article like flour. per quart over present methods. This would mean One reason for the expensiveness of a house-toa saving of $0.90 a can or ’$44,375.00per day on house delivery of liquid milk is the fact that the New- York’s milk supply, a saving of $16,197,021.00demand for milk varies considerably from day to per year over present methods. I am assuining day, whereas the cow pays no attention to the that the grocer handles this milk powder on reg- whims of the public. The milk dealer must alular margins, namely 20 per cent. for the retailer ways have enough to supply his patrons when the and I O per cent. for the wholesaler. demand is large. At other times he must make At $0.08 per quart New York’s milk supply is butter and Dutch cheese out of his surplus milk worth $j7,589,408.00 per year. I have shown and lose money on every quart of milk utilized in above a saving per year of ~16,197,021.00 on this way. I a m told by a prominent milk dealer dealers’ profit and cost of selling, amounting to that the demand for milk is usually large on Monabout 28 per cent. of what New York City’s in- day. On Tuesday i t is very small. On Wedneshabitants pay for milk. On this basis milk could day i t recovers and on Thursday i t drops away be sold in New York City in powder form a t a n again. On Friday it is large and on Saturday his equivalent of $0.06 per quart and afford an at- demand always exceeds his supply. A permanent tractive margin of profit for all concerned, the product would adjust supply to demand every day saving in freight being sufficient to pay the cost in the week. of reducing the milk to powder, and the manuAnother economic reason for the reduction of facturer’s profit. milk to powder is the fact that there are many There are no reliable data available as to the districts which are eminently suited to milk proconsumption of fresh milk per capita, but the duction where the dairy industry does not thrive constant rate of increase in the number of cows because of their distance from market. The
RA EKELA-VD 0.V C O S D E S S A T I O S PROD L-CTS.
545
stitute. He boiled two molecules of formaldehyde in solution with two molecules of a phenol and one molecule of an oxyacid (tartaric acid) and thus prepared a resinous mass which is purified by washing and can be melted like any ordinary resin and is soluble in alcohol as well as in watery solutions of NaOH. This substance, according to its method of preparation and the amount of free phenol it contains, may be made to melt a t widely variable temperatures ranging from 50' C. to considerably above I O O O C . According to whether it is prepared with pure white phenol or dark commercial cresol, it will vary in hardness or color. I t is always more brittle than shellac and its appearance may vary from that of a colorless transparent vitreous mass to yellow-, brown or black. I t dissolves in alcohol in all proportions and gives a varnish which dries in about the same way as an alcoholic solution of shellac, but the layer it leaves after thorough evaporation is not so elastic nor flexible as that of shellac and is more brittle; this is probably the reason why it does not sand-paper nor polish as well as shellac. Furthermore, this varnish has a decided tendency to oxidation. The alcoholic solution, which can be made colorless, assumes a darker color after a few days and may keep on darkening until it finally becomes deep ruby. In the same way, objects varnished with this solution mill ultimately acquire a decidedly darker color which in time becomes a deep red mahogany stain. ON SOLUBLE, FUSIBLE, RESINOUS CONDENThis resin does not dissolve well in weak alkalies SATION PRODUCTS OF PHENOLS like shellac does, but it is soluble in aqueous NaOH. AND FORMALDEHYDE.' The latter solution oxidizes rapidly in the air and By L H B A E K E L A S D . S c D. may acquire a vivid purplish red color. Carbonic ReceivedMay IS. 1909. gas and any acids reprecipitate the resin from this I n a former paper2 I have described the con- alkaline solution. ditions under which the condensation of phenols The varnish made with the resin obtained from and formaldehyde may produce infusible, insoluble, commercial impure phenol or cresol has a decided homogeneous bodies of great chemical inertness coal tar odor, which clings persistently to all obwhich can be used for many technical purposes. jects coated with it. I stated then and there that, starting from the This resinous material is decidedly different in same raw materials, but operating under slightly chemical and physical properties from the product modified conditions, we may obtain, in other described by me as Solid A, or Initial Condensainstances, resinous substances which are soluble tion Product,l from which it is easily distinguished and fusible and of very different chemical properties. by its great tendency to oxidation and its inAs far as I know, Blumer3 was the first one to ability of being transformed by simple heating publish an attempt to utilize this reaction com- under pressure into Bakelite C. At the time when mercially for the manufacture of a shellac sub- Blumer published his English patent, he seems to have given considerable importance to the chemical 1 Read before the New York Section, May 14, 1909. 2 ' The Synthesis, Constitution and Uses of Bakelite THIS r61e the oxyacid plays in this reaction. However,
location of milk-desiccating plants in such districts would do much not only to promote the interests of remote farming communities, but would largely increase the milk supply of the country. The question of time would be entirely eliminated and the distance from market would not be a factor except for a slight increase in freight rate. I t is well known that the railroads make very close rates on long distance shipments of western products to eastern states and in fact the freight rates from Illinois and Iowa may be lower to Kew York than from some district in northern Vermont or northern New Tork. There is no reason therefore why New I'ork City should not draw its milk supply from the western states as it now largely draws its supply of butter and eggs. I n fact most of the economic reasons for the reduction of milk to powder might be summed up in one phrase, a world market for milk. There is one reason, howeyer, which I wish to dwell upon that is rather sanitary than economic. Fresh milk is a substance upon which no pure food guaranty is possible. I t is drawn to-day and gone to-morrow. This renders adequate milk inspection practically impossible. n'ith milk supplied in the form of a permanent powder, it is entirely possible to create a high standard of quality and purity because the stability of the product would permit of the enforcement of rigid regulations.
"
JOURYAL, 3
1, 149.
Louis Blumer, Eng P a t , June 5 , 1902, No 1'2,880.
1
L o c cal.
