THE DETERMINATION OF NITRATE NITROGEN IN ...

THE DETERMINATION BY ROBIN

OF NITRATE PLANTS.

C. BURRELL

AND

THOMAS

NITROGEN

IN

G. PHILLIPS.

(FTomthe Departmentof Agricdtura~

Chemistry, the Ohio &ate University, columbus.)

(Received for publication,

May 25, 1925.)

229

Downloaded from http://www.jbc.org/ by guest on September 9, 2017

The methods used for the determination of nitrate nitrogen in plant materials have been adapted from methods used in the analysis of soils and fertilizers. The one most used at present involves reduction by Devarda’s alloy, and the determination by titration of the ammonia formed. This method has been studied in detail for soils by E. R. Allen (l), Davisson (2), and Whiting, Richmond, and Schoonover (3). The peculiar difficulties met in applying it to plant extracts, however, have not been investigated sufficiently. The object of the first part of this work was to learn whether nitrate nitrogen can be determined by reduction by Devarda’s alloy in the presence of the other nitrogen compounds likely to occur in plant extracts. The materials used were sodium nitrate, made by neutralizing a standard solution of C.P. nitric acid with sodium hydroxide; ammonium sulfate; asparagine; and alanine. These were of the highest purity obtainable and their composition was checked by analysis. The distillations were made in 500 cc. Pyrex Kjeldahl flasks through Davisson scrubber bulbs and block tin condensers. Methyl red was used as indicator and the solutions of sodium hydroxide and sulfuric acid were 0.02 N. Ammonia-free water was used wherever necessary. Distilling from 1 gm. of Devarda’s alloy and a volume of 300 cc. of approximately 0.1 N sodium hydroxide for 1 hour, gave recoveries from 2 mg. of nitrate nitrogen varying between 97.1 and 99.7 per cent, with an average for eight determinations of 98.4 per cent.

Nitrate

Nitrogen

in Plants

Other

nitrogen

Recovery of nitrate nitrogen.

present.

1 mg. alanine nitrogen.. . . . . . . . . . . . . . . . . . . . . . . . 1 “ ammonia

“

.. .... . . ... . . ... .

2 “

‘I

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

“

1 “

amide nitrogen

1 “

amino acid nitrogen

as asparagine..

.. ....

. .. ..

.. .

99.6 99.7 91.1 88.2 89.9 85.7


Attempts were made to recover nitrate nitrogen from mixtures containing 1 mg. each of nitrate nitrogen, ammonia nitrogen, and amide nitrogen (as asparagine), and 2 mg. of amino nitrogen (1 mg. as asparagine and 1 mg. as alanine). To remove nitrogen other than nitrate that might be evolved as ammonia during the alkaline reduction the methods of Davisson (2) (preliminary boiling with sodium hydroxide), and that of Whiting, Richmond, and Schoonover (3) (preliminary boiling with sodium peroxide), were used. The results obtained were not consistent and were usually high. It was suspected that the amide nitrogen of asparagine was not completely removed by the preliminary treatment, and continued to be evolved during the reduction. The Davisson method was carried out with samples containing no nitrate nitrogen but 2 mg. of amide nitrogen as asparagine, with yields of “nitrate nitrogen” varying from 0.28 to 0.31 mg. These methods for nitrate nitrogen, then, are not applicable in the presence of asparagine. The comparison method as used by Strowd (4) was then tried. Two aliquots were distilled in the same way, except that Devarda’s alloy was used in one of them. The difference in the amount of ammonia recovered was taken as due to the nitrate (and nitrite) nitrogen reduced. The recoveries from 1 mg. of nitrate nitrogen amounted to 75.9, 75.9, and 85.3 per cent. In order to learn which constituent of the solution interfered with the recovery of nitrate nitrogen, the following mixtures were analyzed. Each contained 1 mg. of nitrate nitrogen.

R. C. Burrell

and T. G. Phillips

231


The comparison method, then, is not applicable in the presence of ammonia.nitrogen, and is especially faulty in the presence of amide nitrogen. Because of the failure of the reduction method, and because it is often desirable to determine amounts of nitrate nitrogen much less than 1 mg., it was decided to attempt the application of a calorimetric method. Two methods seemed to give promise, the phenol disulfonic acid method, especially as applied to soils, and the modified reduced strychnine method (5). The latter was soon discarded because of the off tints developed in the presence of nitrites and other compounds that are frequently present in plant extracts, and because of the rapid change of the color on exposure to light, which made quantitative comparisons difficult. In attempting the application of the phenol disulfonic acid method, three principal difliculties were met: (1) clearing the plant extracts; (2) removing sugars and other substances that char on the addition of the reagent; and (3) overcoming the effect of chlorides. Clearing the Plant Extracts.-The method is to be applied to the 80 per cent alcoholic extracts of fresh plant material. The alcohol is evaporated from an aliquot of such an extract and the residue taken up with water. Samples prepared in this way often are very highly colored. Preliminary tests of most of the usual protein precipitants and clarifiers showed charcoal and compounds of lead and copper to be most promising. These were studied further. The use of copper sulfate, calcium hydroxide, and magnesium carbonate, as applied to soils by Harper (6), gave filtrates that were noticeably colored except when the extracts themselves were very light colored. In the case of some soil extracts, Harper used a preliminary treatment with a special charcoal. This was tried with plant extracts, but if they were very highly colored, two treatments with the charcoal were necessary, and on evaporation of these seemingly very clear solutions, finely divided charcoal began to separate and had to be removed by a copper hydroxide treatment. The charcoal residue from the clarification was frequently gummy. On adding known amounts of nitrates (0.1 mg. of nitrate nitrogen to 50 cc. of the extract), variable but con-

232

Nitrate

Nitrogen

in Plants


siderable losses were observed, amounting to from 10 to 20 per cent for each charcoal treatment, even when the charcoal residue was washed three times with 10 cc. portions of hot water. Lead acetate solution alone would not give a clear filtrate. When 7.5 cc. of normal sodium hydroxide and 5 cc. of a 25 per cent lead acetate solution were added to 50 cc. of the extract prepared as described above, the desired result was obtained. The excess lead was removed by adding sulfuric acid. Removing Substances That Char on Addition of the Reagent.-On adding the reagent to the residue from the evaporation of the clear filtrate obtained by the above treatment, there was always a slight charring which gave an off color to the final piorate solution, making an accurate comparison with the standard impossible. This was found to be due largely to the presence of small quantities of organic substances not removed by the lead acetate clarification. A fraction of a mg. of glucose, for instance, is sufficient to give a distinctly off color, even though the apparent charring is relatively slight. These disturbing substances must be removed. Direct oxidation of the whole plant extract by sodium peroxide was tried after removal of the alcohol by evaporation. This required considerable time and the quantity of salts, especially carbonates, in the final residue was so great that there was serious nitrate loss when the phenol disulfonic acid reagent was applied. A much smaller quantity of sodium peroxide was sufficient to oxidize the disturbing substances in the filtrate from the lead acetate clarification; and this procedure proved satisfactory. Overcoming the Effects of Chlorides.-Enough chlorides are present in most plant material to prevent the quantitative determination of nitrates by the phenol disulfonic acid method. It is probable that the loss is due to the formation of apua regia (7). In attempting to avoid this difficulty, the Gericke modification (8) was tried, but it gave very inconsistent results. It seemed necessary to remove the chlorides. Precipitation by silver sulfate was tried. The precipitate was, in part, of such a colloidal nature that it could not be removed by ordinary means. However, the use of copper sulfate, calcium hydroxide, and magnesium carbonate carried down the colloidal silver chloride along with the excess silver and the copper and gave a perfectly clear filtrate, entirely free from interfering organic matter and chlorides.

R. C. Burrell

and T. G. Phillips RESULTS.

The method, as described below, was applied to the following solutions, with the results given in Table I. TABLE

I.

Nitric nitrogen added.

50 50 50 50 50 50 50

cc. Solution “ “ “ “ “ “ “ “ “ “ “ “

A. A. A. B. B. B. c.

Total amount of nitric nitrogen found.

7nQ.

VZQ.

0.00 0.10 0.50 0.00 0.10 0.50 0.00

0.00* 0.097 0.490 0.825 0.920 1.305 0.50

Average.

Range of recovery of added nitrate.

I 0

nitrate. per cent

"Q.

0.096-0.098 0.485-O ,495

97 98

0.092-0.098 0.47 -0.50

95 96 100

4 2 3 2 10 3 2

-

* Traces of nitrites. The Modijied

Method.

An aliquot of the alcoholic extract of the sample (usually 50 cc., representing 2.5 gm. of the fresh leaf material) is evaporated on the steam bath until the alcohol is removed. Water is added to make the volume about 50 cc.; 7.5 cc. of normal sodium hydroxide and 5 cc. of 25 per cent lead acetate solution are added; and the mixture is stirred. The heavy precipitate is removed best by centrifuging, and washed twice in the centrifuge with 20 cc. portions of hot water. Excess lead is removed by adding to the clear liquid 0.5 cc. of concentrated sulfuric acid. The lead sulfate is filtered or centrifuged out, and washed once with a small portion of hot water. If any difficulty is experienced in removing the precipitate because of its small amount, this is easily overcome by adding a drop or two of the lead acetate solution. to the centrifuge tube. The deleaded liquid is transferred to a Kjeldahl flask, the volume made up to about 150 cc., 2.0 gm. of sodium peroxide are added, and the mixture is boiled down to a volume of 10 to 15 cc. (until it starts to bump). It is cooled, rinsed into a flask, diluted to about


Solution A.-This contained 0.025 gm. of ssparagine, 0.1 gm. of glycine, 0.01 gm. of ammonium sulfate, and 0.5 gm. of glucose per liter of aqueous solution. Solution B.-The 80 per cent alcoholic extract of 50 gm. (fresh weight) pinnate leaves of 6 week old navy bean seedlings. Total volume, 1 liter. Solution C.-Standard nitric acid exactly neutralized with sodium hydroxide and diluted so that each cc. contained 0.01 mg. of nitric nitrogen.

Nitrate

Nitrogen

in Plants

SUMMARY.

The various modifications of the Devarda’s alloy method for the determination of nitrate nitrogen have been found to be inaccurate in the presence of amide nitrogen. A modification of the phenol disulfonic acid method has been developed which gives excellent results in the determination of nitrate nitrogen in plant extracts. BIBLIOGRAPHY.

1. Allen, E. R., .I. Id. and Eng. Chem., 1915, vii, 521. 2. Davisson, B. S., J. Ind. and Eng. Chem., 1918, x, 600. 3. Whiting, A. L., Richmond, T. E., and Schoonover, W. R., J. Ind. and Eng. Chem., 1920, xii, 982. 4. Strowd, W. H., Soil SC., 1920, x, 333. 5. Scales, F. M., and Harrison, A. P., Ind. and Eng. Chem., 1924, xvi, 571. 6. Harper, H. J., Ind. and Eng. Chem., 1924, xvi, 180. 7. Lombard, M., and LaFore, J., Bull. Sot. Aim., 1909, v, series 4, 321. 8. Gericke, W. F., J. Ind. and Eng. Chem., 1917, ix, 585. 9. Chamot, E. M., Pratt, D. S., and Redfield, H. W., .I. Am. Chem. Sot., 1911, xxxiii, 366. 10. Bear, F. E., and Salter, R. M., West Virginia Agric. Exp. Station Bull. 169, 1916, 23.


100 cc., and made neutral to litmus by adding sulfuric acid a drop at a time, with constant stirring. Next 10 cc. of saturated silver sulfate solution are added, followed by 1 cc. of normal copper sulfate solution, 0.2 gm. of calThe mixture cium hydroxide and about 0.5 gm. of magnesium carbonate. is shaken for 10 minutes. The precipitate should appear grayish; if it does not, more calcium hydroxide is added. The mixture is filtered and the The filtrate residue washed three times with small portions of hot water. is evaporated almost to dryness on the steam bath, 5 cc. of the phenol disulfonic acid reagent, prepared as recommended by Chamot, Pratt, and Redfield (9), are poured in the center of the dish where the nitrates are concentrated and allowed to react several minutes. The mixture is rubbed to a paste with a small pestle, enough water is added to dissolve the salts, and ammonium hydroxide is stirred in until present in excess. Sometimes The solution is it is necessary to filter out a slight flocculent precipitate. made up to some definite volume, usually 50 to 200 cc., and compared with freshly prepared standards as recommended by Bear and Salter (lo), using either Nessler tubes or a calorimeter.

THE DETERMINATION OF NITRATE NITROGEN IN PLANTS Robin C. Burrell and Thomas G. Phillips J. Biol. Chem. 1925, 65:229-234.

Alerts: • When this article is cited • When a correction for this article is posted Click here to choose from all of JBC's e-mail alerts This article cites 0 references, 0 of which can be accessed free at http://www.jbc.org/content/65/1/229.citation.full.ht ml#ref-list-1


Access the most updated version of this article at http://www.jbc.org/content/65/1/229.citation