THE HEAT COAGULATION OF MILK H. H. SOMMER and E. B. HART I* oUU(Nc;;h i. (HESiS (From the Department of Agricultural Chemistry, University OF Wisconsin, Madison) Reprinted from THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. XL, No. 1, November, 1919 Reprinted from The Journal of Biological Chemistry, Vol. XL, No. 1, 1919 %> \^'^THE HEAT COAGULATION OF MILK* (J- By H. H. SOMMER and E. B. HART. ^^"^ {From the Department of Agricultural Chemistry, University of Wisconsiii, Madison.) (Received for publication, September 17, 1919.) The coagulation of milk by heat was first observed by Ham- mersten/ who found that it occurred at from 130-150°C. with different samples of milk. Since then the question has been studied very little, and no tenable explanation has ever been given for the difference in the coagulating points of milks from different cows. In recent years a knowledge of the factors which determine this difference has become very desirable, for these same factors undoubtedly determine whether a condensed milk will coagulate when it is sterilized. The coagulation of condensed milk on sterilizing causes serious losses in the milk-condensing industry. In the manufacture of condensed milk, the fresh milk is first pasteurized or "preheated" at from 180-210°F. for from 1 to 20 minutes. The condensing is done under vacuum at 130-160°F. After the desired concentration has been attained, the milk is put into cans, sealed, and then sterilized at 224-240°F. for 20 to 50 minutes. It is during the sterilizing process that the coagulation occurs. Because it occurs so frequently, all the condensed milk is placed into shaking machines to break up any loose coagulum that may have formed. However, frequently the coagulum is so firm that even after shaking the milk remains lumpy. Such a product is rejected by the consuming public, and thus is a loss. Manufacturers have sought to solve the problem by controlling the acidity of the milk. They have set an arbitrary standard such as 0.18 per cent acid (calculated as lactic acid), above which * Published with the permission of the Director of the Wisconsin Agri- cultural Experiment Station. 1 Quoted from Iterature referred to in, Kastle, J. H., Chemistry of milk, Hyg. Lab., Bull. 56, 1909. 137 138 The Heat Coagulation of Milk thej^ reject all milk. This has led to much difficulty, because often, immediately after it is drawn from the cow, milk has a higher titratable acidity than 0.18 per cent. Thus the con- denseries may be rejecting perfectly fresh milk, believing that they are remedying their difficulty in this way, although it has never been demonstrated that titratable acidity is related to the coagulation. The factors involved in the coagulation have never been determined, and no explanation is available on which to base a remedy for this difficulty. To offer an explanation for the difference in the coagulating points of different milk samples the following factors were studied : titratable acidity, hydrogen ion concentration, concentration of the milk, and composition and balance of the milk salts. The Heat Test. The temperature at which the milk was heated was arbitrarily set at ISG^C. At first the heating was done in an autoclave at 50 pounds pressure for 20 minutes, and in that way the milks were dif- ferentiated into coagulating and non-coagulating. With the auto- clave, it took about 10 minutes to get up to the desired pressure j' and, after the milk had been heated, the pressure had to be released gradually to prevent the milk from boiling over. The disad- vantages of this method were such that there were no sharp limits from which to calculate the 20 minute interval, and it was im- possible to determine the relative rates at which the milk samples coagulated. To overcome these disadvantages the milk was placed into small glass tubes, sealed, and then heated in a xylene vapor bath which was constant at 136°C. within 0.5°C. The sealed tubes were clamped in a rack, s6 arranged that it could be tilted to invert the tubes, to see how the milk would flow, and in that way it was possible to determine the exact length of time required for each sample to coagulate. The milk in the sealed tubes was up to 136°C. in less than 1 minute, so the point from which to calculate the time was practically the instant the tubes were inserted into the vapor bath. ( -vv -V «^ -u-^ -,< H. H. Sommer and E. B. Hart 139 Titratdble Acidity. Since condenseries are attempting to remedy the coagulation problem by rejecting milk above 0.18 per cent acid (calculated as TABLE I. Titratable Acidity and Coagulation. May 8, 1919. May 10, 1919. May 16. 1919. Cow No. Titratable acidity. Coagu- lation. Cow No. Titratable acidity. Coagu- lation. Cow No. Titratable acidity. Coagu- lation. Lactic acid. Lactic acid. Lactic acid. per cent min. per cent min. ■per cent min. 1 0.257 20-* 1 0.241 6 31 0.203 4| 2 0.235 3 2 0.231 11 2 0.193 20- 3 0.216 20- 26 0.228 5 1 0.192 20- 4 0.214 20- 31 0.222 ^ 3 0.188 4| 6 0.210 8 27 0.212 6 27 0.188 6| 6 0.207 20- 3 0.212 5§ 6 0.188 20- 7 0.206 6 4 0.211 20- 4 0.188 20- 28 0.201 20- 5 0.205 5i 28 0.186 20- 9 0.200 20- 9 0.197 20- 12 0.184 10 10 0.200 20- 28 0.199 20- 26 0.184 3 11 0.195 4.1 6 0.195 20- 5 0.183 6| 12 0.191 20- 29 0.192 5i 9 0.182 20- 13 0.190 20- 10 0.190 20- 14 0.182 9 14 0.189 6i 13 0.186 9 29 0.182. If 15 0.182 20- 12 0.185 11 11 0.178 3^ 16 0.179 4 11 0.184 5 13 0.175 4 8 0.174 9 8 0.174 20- 10 0.171 20- 17 0.172 20- 14 0.175 6i 30 0.165 20- 18 0.167 20- 15 0.172 20- 15 0.163 20- 19 0.158 12 17 0.166 20- 8 0.163 6| 20 0.157 20- 18 0.160 20- 17 0.156 20- 21 0.156 20- 30 0.162 4 18 0.154 20- 22 0.148 3 7 0.162 2 7 0.148 2 23 0.146 3^ 21 0.157 20- 20 0.143 20- 24 0.143 20- 20 0.147 20- 19 0.135 20- 25 0.120 2 23 0.145 51 22 0.133 2 19 0.144 20- 23 0.130 4 22 O.IM 2 24 0.128 20- 24 0.141 20 21 0.128 20- 25 0.131 H 25 0.102 U "20— = no coagulation in 20 minutes. 140 The Heat Coagulation of Milk lactic acid), it was of interest to know how much variation there was in the tritratable acidity of milk from individuaf cows, and what relation the acidity would bear to the coagulation. To study this, samples were taken from the University herd and titrated immediately, and the heat test in the xylene vapor bath applied as soon as possible. The results given in Table I were obtained. The titratable acidity varies from 0.102 to 0.257 per cent. Out of the 86 samples, 45 are above 0.18 per cent. In fresh milk there is no direct relation between titratable acidity and coagulation, as is evident from Table II. If fresh milk samples were more nearly alike in titratable acidity, then titratable acidity might bear a direct relationship to the heat TABLE II. Summary of Titratable Acidity and Coagulation. Date. No. of samples. No. above 0.18 per cent acid.* No. that are 20+.t No below 0.18 pet- cent acid.t No. that are 20+. May 8 " 10 " 16 26 30 30 15 14 16 5 7 11 11 16 14 6 7 6 Total 86 45 23 41 19 * Per cent above 0.18 per cent acidity, coagulating 20+ = 51.2 per cent. t 20+ means coagulation within 20 minutes. t Per cent below 0.18 per cent acidity, coagulating 20+ = 46.4 per cent. coagulation of commercial milk samples. The acidity would then be a measure of the amount of fermentation that had taken place. Lactic acid fermentation lowers the coagulating point in two ways; (1) it changes the reaction, and (2) it lowers the citric acid content of the milk very rapidly.^ Both of these are factors in lowering the coagulating point, as will be shown later. Since fresh milk samples vary so widely in titratable acidity, it is impossible to measure the extent of acid fermentation in a sample by titration. For this reason it is impossible to use titra- table acidity as a criterion of coagulability. 2 Bosworth, A. W., and Prucha, M. J., Tech. Bull. 14, N. Y. Agric. Exp. Station, 1910. TABLE III. Hydrogen Ion Concentration and Coagulation. Date. Cow. pH Ch Coagulation in 20 min. 1919 Feb. 21 31 6.55 2.82X10-^ + + + * 32 6.83 1. 48X10-' + 13 6.58 2. 63X10-' + + + 24 6.83 1. 48X10-' — " 25 • 13 6.25 5. 62X10-' 32 6.66 2.19X10-' — 31 6.66 2. 19X10-' + + + " 26 13 6.58 2. 63X10-' + + + 32 6.69 2. 04X10-' + + 31 6.70 1. 99X10-' + + 24 6.73 1. 86X10-' + 14 6.70 1. 99X10-' + + + " 27 13 6.44 3. 62X10-' + 32 6.64 2. 29X10-' + + 31 6.64 2. 29X10-' + + + " 27 24 6.70 1. 99X10-' 14 6.44 3. 62X10-' - 33 6.59 2. 56X10-' + + + " 28 13 6.50 3. 16X10-' _ 31 6.94 1. 15X10-' + + + 32 6.92 1. 20X10-' - 24 6.93 1.17X10- + + 14 6.50 3. 16X10-' - 33 6.68 2. 09X10-' + + + Mar. 3 13 6.67 2. 14X10-' 31 6.84 1. 44X10-' - 33 6.69 2. 04X10-' + + " 5 33 6.64 2. 29X10-' + + + 13 6.64 2. 29X10-' - 31 6.93 1. 17X10-' + 32 6.79 1. 62X10-' - 24 6.79 1. 62X10-' - 14 6.58 2. 63X10-' — " 6 24 6.97 1. 07X10-' + 33 6.59 2. 57X10-' + + + " 7 31 6.85 1. 41X10-' + + + " 10 33 6.79 1. 62X10-' + + + * Number of plus signs indicates degree of firmness. 141 142 The Heat Coagulation of Milk Hydrogen Ion Concentration. • Titratable acidity does not give an index to true acidity, or hydrogen ion concentration, so that if there is any relation between acidity and coagulation, it would be most likely to exist between the hydrogen ion concentration and coagulation. To study this possibility the hydrogen ion concentration of fresh milk was determined by means of the gas chain method, and the heat test was applied by means of the autoclave. The results given in Table III were obtained. From a study of the data it becomes evident that the hydrogen ion concentration is not the determining factor in the coagulation. Samples of equal Ch do not always respond alike to the heat test; one may remain liquid, and the other may form a firm coagulum. In a large number of cases samples of high Ch did not coagulate, whereas samples of lower Ch did, the exact reverse of what should happen if true acidity was the cause of the coagulation. We must conclude from this that in fresh milk Ch is not the determining factor in the coagulation. However, it may become a factor, for if we change the reaction of a milk sample by adding small amounts of acids the coagulating point is lowered. Concentration. The concentration of the milk would be expected to influence the coagulating point. This was found to be the case when milk was diluted (Table IV). TABLE IV. Relation, of Coagulation to Concentration. 25 cc. of milk + H2O. Coagulation time. H2O added. cc. min. 0.0 H 10 2 2.0 2f 3.0 14 4.0 35- 5.0 35- 6.0 35- H. H. Sommer and E. B. Hart 143 Not only the concentration of the casein influences the coagu- lating point, but also the concentration of the serum. This was determined by comparing the effect of water dilution to the effect of dilution with milk serum obtained by filtering the milk through Pasteur-Chamberlain filters (Table V). In the dilution with water, where the casein and the serum are both diluted, the effect is greater than where the casein alone is diluted by adding serum; therefore, the concentration of the serum is also a factor influencing the coagulating point. Concentration of casein and of serum may in part explain the, difference in the coagulating points of different milk samples. TABLE V. Relation of Coagulation to Concentration of Serum. 25 cc. of milk + serum. Coagulation time. Serum added. cc. min. 0.0 n 0.1 n 0.2 2 25 cc. of milk + H2O. Coagulation time. H2O added. cc. mm. 0.0 n 0.1 2i 0.2 4 However, in most cases, with the slight variation in concentration, this factdr is of minor importance, just as Cg is. There must be another factor of greater importance. Composition and Balance of Milk Salts. Since electrolytes have a very marked effect upon the stability of colloids, we should expect that variations in the salt compo- sition would influence the stability of the casein in the milk. That the various salts exert an influence on the coagulating point was shown in a number of cases. 144 The Heat Coagulation of Milk The effect of an addition of ammonium oxalate to milk that previously coagulated is shown in Table VI. The removal of calcium by precipitation prevents coagulation in most cases and similarly in most cases the addition of small amounts of calcium salts lowers the coagulating point. This coagulation can again be balanced by means of sodium citrate or dipotassium phosphate (Tables VII, VIII, IX, and X). Coagu- lation caused by MgCl2 or BaCl2 can also be balanced by sodium citrate (Tables XI and XII). In most cases coagulation can be prevented by the addition of citrates or phosphates, the coagulation being due to an excess of calcium and magnesium. However, in a few cases the addition of citrates or phosphates did not prevent coagulation, but rather TABLE VI. Ammonium Oxalate Prevents Coagulation. 5 cc. of milk + 10 per cent (NH4)2 C2O4. Coagulation in 20 min. (NH4)2 C2O4 added. drops 1 2 3 4 + + + + + + hastened it. In these cases the addition of the proper amount of calcium salts prevents coagulation or at least raises the coagulating point (Tables XIII and XIV). From the data we see that the calcium and magnesium are balanced by the phosphates and citrates of the milk practicallj^ in gram-equivalent amounts. The sodium and potassium chlorides in the concentrations present do not have any marked influence on the coagulating point, so that the balance of the four constituents, calcium, magnesium, citrates, and phosphates, largely determines whether a milk will coagulate or not. If calcium and magnesium are in excess, the milk will coagulate on heating. If calcium and magnesium are properly balanced with the phosphates and citrates, the optimum stability obtains. If phosphates and citrates are in excess, coagulation will also result. H. H. Sommer and E. B. Hart 145 TABLE VII. Balance between Calcium and Citrates 25 cc. of milk plus. Coagulation time. m/2 Ca acetate. m/2 Na citrate. H2O cc. cc. cc. min. 0.0 0.0 1.3 3 0.3 0.0 1.0 i 2 0.3 0.1 0.9 2h 0.3 0.2 0.8 'A 0.3 0.3 0.7 2 0.3 0.4 0.6 n * The sodium citrate consisted of 25 cc. of sodium m/2 citrate plus 3 cc. of m/2 citric acid. This solution was distinctly acid, so that the balancing effect could not have been due to neutralization of acidity by means of the sodium citrate. TABLE VIII. Balance betiveen Calcium and Cilrates.* 25 cc. of milk plu3. Coagulation time. m/2 Ca acetate. m/2 Na citrate. H2O cc. cc. cc. min. 0.0 0.0 1.6 4 0.4 0.0 1.2 1 2 0.4 0.2 1.0 40- 0.4 0.4 0.8 40- 0.4 0.6 0.6 2i 0.4 0.8 0.4 2 * The sodium citrate consisted of 25 cc. of m/2 sodium citrate plus 1 cc. of m/2 citric acid. TABLE IX. Balance between Calcium and Citrates. 25 cc. of milk plus. Coagulation time. m/2 Ca acetate. m/2 Na citrate. H5O cc. cc. CC. min. 0.0 0.0 1.8 25- 0.8 0.0 1.0 i 0.8 0.4 0.6 1 8 0.8 0.6 0.4 25- 0.8 0.8 0.2 25- 0.8 1.0 0.0 ^ THE JOURNAL OP BIOLOGICAL CHEMISTRT, VOL. XL, NO. 1 146 The Heat Coagulation of Milk TABLE X. Balance between Calcium and Phosphates. 25 cc. of milk plus. Coagulation time. m/2 Ca acetate. m/2 K2HPO4 H2O cc. cc. CO. min. 0.0 0.0 1.2 20- 0.5 0.0 0.7 1 4 0.5 0.2 0.5 3 8 0.5 0.3 0.4 1 0.5 0.4 0.3 20- 0.5 0.5 0.2 20- 0.5 O.G 0.1 20- 0.5 0.7 0.0 6 TABLE XL Balance between Magnesium and Citrates. 25 cc. of milk plus. M/2 MgCh m/2 Na citrate. H2O cc. cc. cc. min. 0.0 0.0 0.7 20- 0.3 0.0 0.4 1 2 0.3 0.2 0.2 20- 0.3 0.3 0.1 20- 0.3 0.4 0.0 8 TABLE XII. Balance between Barium and Citrates. 25 cc. of milk plus. m/2 BaCli m/2 Na citrate. H2O cc. 0.0 (t.'J 0.2 cc. 0.0 0.0 0.2 cc. 0.4 0.2 0.0 min. 20- 1 8 20- Thus ilie coagulation of a milk sample on heating may be due either to an excess or a deficiency of calcium and magnesium. We may explain this in the following manner. The casein of the milk is most stahk' with regard to heat coagulation when it is in combination with a definite amount of calcium. If the calcium H. H. Sommer and E. B. Hart 147 combined with the casein is above or below this optimum, the casein is not in its most stable condition. The calcium in the milk distributes itself between the casein, citrates, and phosphates chiefly. If the milk is high in citrate and phosphate content, more calcium is necessary in order that the casein may retain its optimum calcium content after competing with the citrates and phosphates. If the milk is high in calcium, there may not be TABLE XIII. A Sample in Which Calcium Prevents Coagulation. 25 cc. of milk plus. Coagulation time. m/2 Ca acetate. m/2 Na citrate. H2O cc. cc. cc. min. 0.0 0.0 0.8 n 0.2 0.0 O.t) 20- 0.2 0.1 0.5 u 0.2 0.2 0.1 1 0.2 0.3 0.3 3 4 0.2 ■ 0.4 0.2 3 4 TABLE XIV. A Sample in Which Calcium Raises the Coagulating Point. 25 cc. of milk plus. Coagulation time. m/4 Ca acetate. HjO cc. cc. mm. 0.0 0.5 If 0.1 0.4 2 0.2 0.3 91 "4 0.3 0.2 3 0.4 0.1 6 0.5 0.0 2h sufficient citrate and phosphate to compete with the casein to lower its calcium content to the optimum. In such a. case the addition of citrates or phosphates makes the casein more stable by reducing its calcium content. The magnesium functions by replacing the calcium in the citrates and phosphates. In most cases the coagulation is due to an excess of calcium and magnesium. It is possible to balance this excess by citrates, 148 The Heat Coagulation of Milk Hi '-HC-li0050i00050':DrfCO(M'+ifOrt'-<050'-i(NCOt^'* (MiOCCOJiCiO'^COTt^'*-*-*!-^-^ M^CC CO (N ■5.E 6-2 HW Ml-* HlN fHIW H« rHiCS HN fhIN I 1 ■ 1 i-ii— i(N(NCCCO'#Ttl-^'^i£>COC0050000 r- 1 (M (N (N 2 1-2 2 02 OlMCOCOOcOCOCOC^IC^OCCiTtHC^O-^i-HO +. 1 + + + + + + 1 + + + + + + 1 1 1 S C§fl 1 .-hOOOOi-hOOOOOOOOOOOO + + + + + + + + + + + + + + + 1 + + o O o S rf ^ t- -' QJ '*00i-H05f000O05t^t^l:^-*(MCO'*(NO00 0505Ci'-iO(MOI>a5(MrHOOCD(N-*'OiO coconcoco'^-^cQco-^'^-^coco-^cciccico 3 01 «s o <£>Tj000505 COCOCO(MCOCOC0fOfO00COCOfOCOCC(NC<>^c^i'*< T-H(M<-Hi-HrHrH(M(MCI>-00050t-H(MCO'*iOCDI^CO H. H. Sommer and E. B. Hart 149 1-1 >o 1^ ^ 1—1 \ lO CO ^ CO 05 CO rt< 00 CO CO CO CO CO CO CO CO CM CO no r^ CO CO ^ o CO CO o 1—1 CO CO 00 lO CO o o r-l 00 00 CD CO CO CO C<1 CO CO CO CO CO CO CM CM CO T— 1 ■^ CO CM lO CM r^ lO '-H Oi lO CO CO Ti< li': lO CD lO lO If: •rt* lO o o o Q o o O o o o O o CO lO CO n> •^ CM 'ti CO Oi o fr> 00 00 "* CO »o 1—1 00 CO 05 l^ 00 ^ (N CO (M (M CO CM CM CM CM CM CO CM I— 1 ^ CO CO 00 CD -^ I^ t^ CM CM lO I— 1 (M 1— 1 l^ CM 1—1 t^ 05 CO O 00 (>> (N (N 1—1 CM CM CM 1—1 1— ( CM CM 1— 1 ^ 05 (N I— 1 CO r^ Oi r^ CO ^ ^ lO o O (N r-H CM 1—1 00 CM 1—1 CM t^ .— 1 t-H '-' 1—1 '-' ^ o T-l "-^ 1—1 O o I— 1 lO CO Ol 1—1 CM lO CO CM CM o o (N fN fN CN CM CM CM CM CM CM CM o O O o o O o O O O o o o O O o o o o o o o o O (-5 CO (N lO CO 00 00 '^t* CO CM 05 Oi I— ( 05 i-H CO 7-H l^ lij -* CO 1— ( CO l^ CO d d d d d d d d d d d d tJ* C5 o CO CO o ^/^ CO OJ ,_, ^ CO CO "Tt^ '^t^ o^ lO 'Tt^ n 05 a> CM Q