660
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y .
“Humin” iVitroge.it.-The compounds precipitated b y the excess of magnesia from the acid solution of the soils are for the most part the so-called humin substances or melanoidins. The term “humin” was first used . b y Berzelius’ in describing certain darkcolored constituents of mold. Mulder showed that on boiling protein matter with strong acid, that brown or black particles, which resemble the dark bodies of putrefying matter, separate. These bodies are also formed from many other organic compounds, including carbohydrates. However, they are uniformly high in carbon and low in hydrogen and nitrogen ; they are generally insoluble in water and in acids; in alkaline solutions they are somewhat soluble and are precipitated from them by acids; they do not give any particular reactions and their decomposition products are not well defined. Samuellyl holds that the indole, pyrrol, pyridine, and tyrosine groups of the albumin molecule contribute t o the formation of these compounds. I n the magnesia precipitate mentioned above there would also appear other compounds containing the pyridine ring, such as picoline carboxylic acid, first isolated from an Hawaiian soil b y Shoreys and later from other soils by Schreiner and Shorey.4 They conclude that such a compound might possibly arise in the soil as a secondary decomposition product of the ‘protein molecule. I t is generally considered that when the nitrogen is combined in the ring, as in pyridine compounds, that it is unavailable, and may even be harmful to plants. It would seem probable that a large part, if not nearly all of the nitrogen classed here as “humin” nitrogen, is combined in such a way as t o be unavailable t o plants. Table IV shows t h a t there is over 2 5 per cent, of “humin” nitrogen soluble in acid from four of these soils. Soil 2 1 is much lower in soluble humin nitrogen than the other soils, showing only about half as much. Since the nitrogen not soluble in acid map also be considered t o be “humin” nitrogen, the total amount of nitrogen in such combination is, in four soils, over 53 per cent., while in soil 2 1 , which has received dried blood, it is only 43 per cent. Amino Acid Nitrogen.-The two classes of amino acids have been thoroughly studied, and their nature and composition is fairly well understood. The nitrogen is in a form from which it can be readily changed into ammonia by various agents. The amino acids are resistant to the action of acids and alkalies, but the amino group may be removed b y the action of enzymes and bacteria, common both t o the plant and the animal. I n regard to the action of bacteria on amino acids, Czapek5 and Emmerlinge have shown that the a-amino acids are excellent nutritive materials. The, amino acids are acted on b y bacteria in two different ways: I. The amino acids are converted into simple acids with the elimination of ammonia. 2 . Carbon dioxide is split off, Poggendorff’s, A w . , 44, 375 (1838). Hofmeister’s Beifriige, 2, 355 (1902). a Hawaii Agr. Expt. S a . , Ann. R p t . , 1906, 55. 4 Bull. 53, p. 32. Bur. Soils, U.S. Dept. Agr. 6 Hofmeister’s Beifriige. 1, 538 (1902): 3, 47 (1902). 8 Ber.. 36, 11, 2289 (1902). 1
2
Sept., 1911
leading, as EmmeriingI has shown, to the formation of Brieger’s diamines, thus, lysine is converted into cadaverine. As a rule, the process does not stop here and the amino group is finally changed into ammonia. That plants may use amino acids directly in their building process does not seem a t all improbable. Different amino acids are present in the germinating plant seed and are used by young plants in their growth. Lefevrez has shown that in the absence of carbon dioxide plants are able t o use certain amino compounds as a source of their carbon. He states that this transfer was not simply osmotic, but that these compounds were taken up by the plant roots. Work with nitrogenous compounds of this nature in water culture work in these laboratories indicates quite strongly that plants are able to use amino compounds in their building process. It would seem then that either directly or through the agency of enzymes, bacteria, etc., amino acids become available t o plant use. These soils differ rather widely in the amounts of diamino and monoamino acids which they contain, shown in Table IV, but there seems to be no agreement between the amino acid nitrogen formed and the plat treatment. However, the amino acid nitrogen as a whole represents 2 2 to 3 4 per cent. of the total nitrogen in the decomposition products of the soils. These five samples of soil are really the same soil under long-continued treatment of different kinds, It is not improbable that work on widely different soils will show even much larger variations than those here noted. The work shows, however, that even in such cases there is a difference in the nitrogenous compounds in the soil and that different decomposition of the nitrogenous matter has and probably will continue to take place under the different conditions imposed upon the soils in the field. BUREAUOF SOILS, VASHINGTON, D. C.
-_____
[CONTRIBUTION FROM THE
TEXASEXPERIMENT STATION.]
AN IMPROVEMENT IN THE METHOD FOR ESTIMATING HUMUS IN SOILKa BY J. B. RATHER Received
M a y 3, 1911
The term “humus” in this article, is used t o signify the ammonia-soluble organic matter of the soil, the vnaterie noire of Grandeau.4 The Huston and McBrideS modification of the Grandeau method is the present one of the Association of Official Agricultural Chemists. The chief difficulty with this method is that with some soils considerable clay is brought into suspension by the ammonia. Most of this clay cannot be removed by filtration, and upon ignition i t loses water and thereby increases the quantity of organic matter which apChem., 29, 334 (1900). Rev. Gem. Bot., 18, 145, 205, 258, 302 (1906). 3 Full details of this work will be published in Bulletin 139 of the Texas Experiment Station. 4 Compf.rend., 1872, p. 988. 5 Bulls. 37 and 107 (Revised). Bukeau of Chem.. U. S. D . A. 12. physiol.
2
Sept., 1911
T H E J O U R N A L OF IAiDUSTRI,4L A N D E N G I N E 4 R I N G C H E M I S T R Y .
parently is present. With some soils the error is so large that i t is several times greater than the amount of organic matter actuallj- present. This fault with the method has been recognized for some time and several modifications have been suggested to remedy it. The most promising of these appears t o be that of Mooers and Hampton.1 Their modification consists in evaporating the solution to dryness on a water bath and heating on the water bath for several hours. The humus is taken up with 4 per cent. ammonia and filtered. The process is repeated and the filtrate evaporated, the residue dried, weighed, and ignited as usual. The Mooers and Hampton method has been used in this laboratory, and although a considerable amount of clay was removed, further improvement seemed desirable. The length of time necessary to complete a n analysis, when heavy clay soils are used, makes the method very long, if evaporation and solution are repeated a sufficient number of times t o give a clear solution. I t is well known that salts can coagulate clay, and throw i t out of suspension. Fraps and Hamner* used non-volatile salts to precipitate the, clay but call attention to the fact that the salt used might be decomposed or otherwise lost on ignition. I t occurred to us that a salt which volatilized or decomposed below I O O O C. would overcome this difficulty. Ammonium carbonate decomposes a t 8j0 C., and was accordingly tested for this purpose. EXPERIMENTAL.
The methods described below were compared on twelve soils, some high and some low in humus. The solutions were prepared by washing the soils with one per cent. hydrochloric acid to remove lime, with water to remove the acid, digestion with 4 per cent. ammonia and filtration, as prescribed in the Official Method. All comparisons were made on the same solutions. I. Humus was determined in I O O cc. by direct evaporation (Official method). The residue was dried 3 hours a t I O O O C., weighed, ignited and washed again. 2. One hundred cubic centimeters were evaporated to dryness on a steam bath, baked two hours, the humus redissolved in 4 per cent. ammonia and decanted from the clay. The process was repeated, the ammonia solution filtered, and the analysis completed as in the official method ( I ) (Mooers and Hampton method). 3. To 130 cc. of the humus solution in a glassstoppered cylinder was added 0.6jo gram ( j grams per liter) ammonium carbonate. The salt was allowed to dissolve and the solution shaken. The clay precipitate was allowed to settle over night, the clear, supernatant liquid decanted through a filter and I O O cc. evaporated, and the determination completed as in the official method. Numerous tests showed that 3 hours was a sufficient time to dry the residues. A test of the ammonium carbonate used showed that it left no residue in the above process. 2
-7. A m . Chem. Soc., 30, 800 (19081. Texas Experiment Station, Bull. 129.
661
RESULTS O F ‘ T H E WORK.
The results are shown in the following table: PERCENTAGE
OF
IIIJ>ICS AND
HUMUS
ASH
ESTIMATED BY DIFFERENT
METHODS.
Precipitation of Evap- clay with amoration and monium carsolution (2). bonate (3).
--Direct evaporation ( I ) .
Laboratory number. Humus. 114 Travis gravelly loam.. . ... 3 35 fine sandy 823 Orangeburg loam, subsoil., . . . , . . , , , 5 .00 829 Houston loam ... . , . . . . , . . 1 . 8 5 830 Laredo gravelly loam. . . . . . I . 7 0 896 Norfolk fine sandy l o a m . , . 1 . 2 8 978 Lufkinclay.. , . . . . . . . . . . . 2 . 1 5 1.75 982 Cameron ciay. subsoil 0 86 993 Orangeburg clay.. . . . . 947 Soil from alfalfa field, N. Dakota .... , . , . . . . , , . . . 6 . 2 0 941 Houston l o a m , , , , . . . . . . . . 3 . 1 4 1203 Houston clay. subsoil . . . . . 2 . 1 8 949 Soil from a n old field, Edge3.81 ly, North Dakota. Average, . . , . . . . . . 2.77 10 per cent. correction for water in ash... . . . . . . . . . 1 . 75
.
Ash. Humus. Ash. Humus. Ash.
19.43
0.78
0.95
0.58
0.21
33.45
0.86 1.51 1.11 1.32 1.29
0.78 1 .33 0 90 1.20 0.95 1.08 0 45
0.19 0.27 0.28 0.57 0.47 0.56 0.37
4.40
1.42
3.97
0.58
0.51 0.72 1.18 0.69 1.97 1 54 0.35
13.09 11.95 9.51
5.77 1.25 1 .OO
3.94 0.55 1 .53
5.Oi 1.07 0.83
0.i8 0.20 0.45
5.67 10.20
3.40 1.69
2.53 1.37
3.06 1.14
1.03 0.45
4.50 4.81 0.66 11.00
1 .55
1.39
The amount of humus “ a s h ” by the official method varied from 3.97 per cent. (soil 993) to 33.45 per cent. (soil 823) and averaged I O . 20 per cent. The humus by the Mooers and Hampton method ( 2 ) averaged two-thirds as much as b y the official method. The difference in some cases was very great. Humus in soil 823 by the official method was 5.00 per cent.; by the Mooers and Hampton method it was 0.86 per cent. The “ a s h ” by the Mooers and Hampton method varied from 0.35 per cent. (soil 993) t o 3.94 per cent. (soil 947), and averaged I . 37 per cent. This is about one-seventh as much as the average for ash by the official method. The humus by the ammonium carbonate method ( 3 ) was lower in all cases than by the Mooers and Hampton method ( 2 ) and the ash averaged onethird a s much as the ash by the latter method ( 2 ) . The average for humus b y the ammonium carbonate method was 1.44 per cent. and the average for ash was 0.45 per cent. This method gave very low reuslts for ash with soils that gave a large amount of clay into suspension. With soil 823 the ash was 33.45 per cent. of the official ( I ) method; it was 0 .j~ per cent. by the Mooers and Hampton ( 2 ) method, and only 0.19 per cent. b y the ammonium carbonate (3) method. With soil 114 the results for ash by methods I , 2 and 3 are, respectively, 19.43,0.95 and 0 . 2 I per cent. With soil 947 the results for ash are 13.09,3.94 and 0.78 per cent. An examination of the table shows similar differences with other soils. When ten per cent. of the ash’ is subtracted from the humus results, the difference between the results by the different methods are not so great. However, this correction is purely arbitrary and represent averages. Fraps and Hamner examined a large number of soils and found that the amount of water in the clay which is held in .suspension b y the ammonia varied from 8 to 2 0 per cent. A correction, therefore, is of little value. The above results show that the official method is (I)
Peter and Averitt. Kentucky Experiment Station. Bull. 126. p. 63.
662
T H E JOURNAL 0F.INDUSTRIAL A N D ENGINEERING CHEMISTRY.
Sept.,
1911
entirely misleading with some soils and that the Mooers to twelve hours. Such tests have been run day and and Hampton method does not remove all the sus- night for the past-two years to determine the efficiency pended clay. It is possible that continued evapora- of wet scrubbers and Theisen washers. The grains tion and solution would finally remove the clay from of dust per cu. ft. a t the entrance and exit, of any the soils we have used, but the amount of time re- cleaning device, obtained in this way, while not ultiquired renders this process impracticable. During mately correct, are comparative where the gas is evaporation the ammonia is liable t o absorb acid not too dirty, and serve to show what the cleaning fumes from the laboratory and the continued baking device is doing. is liable t o oxidize or decompose some of the conDuring the past year, however, there has been a n stituents of the humus. increasing demand for some method of making dust The advantages of the method proposed by us are determinations in dirty gas, which would give absosimplicity, speed, and the production of perfectly lutely correct ultimate results. Tests run near the clear solutions. blast furnace, where the dust per cu. f t . runs higher Proposed Modification of the Official Method.-On than 1.5 grains, demonstrated that under such condithe basis of the work just described the following tions the results obtained by merely allowing the modification of the official method is proposed: gas t o pass through the filter and deposit its dust Prepare the humus solution as described in the were absolutely worthless. After encountering inofficial method, and dissolve 0.65 gram ammonium numerable difficulties due to the irregularities of carbonate in 130 cc. of the solution. Allow to stand blast-furnace conditions the form of apparatus and over night in a glass-stoppered cylinder to allow the method of operating to be described have been clay to settle and decant the clear supernatant liquid worked out, and the results appear, theoretically and through a dry filter into a dry flask. Evaporate IOO practically, to be correct ultimate figures. By means cc. of the filtrate in a tared platinum dish, dry for of this apparatus and method of operating the same three hours a t IOO', weigh, ignite, and weigh again. the dust may be determined in any blast-furnace gas and a t any stage in its passage from the furnace Record loss on ignition as humus. to the stoves, boilers or cleaners. By means of two SUMMARY A N D CONCLUSIONS. sets of such apparatus the efficiency of any cleaning I . Clay in humus solutions may be precipitated by system may be accurately determined. The results ammonium carbonate and the precipitant disap- of many tests have shown that very slight changes pears on evaporating the solution and drying the in the construction of a dry gas cleaner may have a residue. wonderful effect on the efficiency of the same. The 2 . Evaporation and solution does not remove value of such efficiency dust tests may, therefore, the clay completely from humus solutions. readily be appreciated. 3. Precipitation of the clay with ammonium carBefore the apparatus was constructed the experibonate is more nearly complete than by evaporation ence of five years of preliminary work had made it and solution and is a much shorter method. clear that in working with dirty gas several important principles must be observed. These principles may be enumerated as follows: THE DETERMINATION OF DUST I N BLAST-FURNACE GAS. I. I n sampling gas containing solids in suspension By WILLIA1.I BR.4DY AND L. A. TOUZALIN. the sample shall be withdrawn from that region of The comparatively recent introduction of gas the main in which the mean velocity of the gas is engines for power development from blast-furnace found to exist. 2 . I n withdrawing said sample, means must be gas, and the general realization of modem blast-furnace operators of the desirability of more thorough employed to change the direction of flow gradually cleansing for gas used in stoves and boilers, has brought and to cause the flow in the entrance end of the samthe iron and steel chemist face to face with the rather pling pipe to be a t the same velocity as the passing perplexing determination of dust in a stream of cons- stream of gas. 3. Where possible, samples shall be taken from tantly flowing gas. a vertical main and a t least I j feet from any bend The writers have given this question considerable or obstruction. attention, and during the past five years various 4. The gas must be filtered above 2 I z O F. and cooled forms of apparatus have been constructed, many experiments have been made, and every conceivable below its dew point before being delivered t o the filtering medium has been tried. I t was early deter- wet test meter. In regard to Principle No. 2 , it had been considered mined that, with a curved sampling pipe facing the flow of gas, and an extremely porous filter, fairly an important point t o have equal velocities in the satisfactory comparative results could be obtained, main and sampling pipe, several years before where the gas contained less than I . j grains per cubic the correctness of the principle could be proved foot, by allowing the gas t o flow through the filter in practice. Within the last few months, the imand then through a wet test meter into the atmos- portance of observing this fundamental principhere. The static pressure of the gas alone was ple of dust determinations has been absolutely proved utilized to cause such flow, and the test was allowed by tests run on gas flowing through a supply pipe t o to take care of itself for periods varying from eight a stove. The gas was traveling a t an average velocity