INDUSTRIAL AND ENGINEERING CHEMISTRY
July 15, 1931
roller, 55, cooperating with the roller 56, to secure the test piece to the carriage. The speed of jaw separation used for testing is 30 inches per minute, but provision is made to operate a t 20 inches per min40 ute whenever required. Operation of Machine
T h e o p e r a t i o n of t h e machine is apparent from the preceding description. T h e t y p e of t e s t piece used is the ordinary dumbbell, 1 cm. being the width of t h e reduced p o r t i o n . Marks a t a definite distance Figure 9-Diagram of Follower are made on the narrower Mechanism part. The test piece is secured a t one end to the clamping device of the carriages, 33 (Figures 5 and 8), attached to the inclination balance. Bridging over the pointers of the elongation carriages, the other end is clamped to the sliding member 34. The plate is adjusted by means of the small drum in such a position that the tracing device moves on the ordinate, passing through the value zero of the abscissa, when the pendulum is in the vertical position. The carriages 31 and 32, drawn in contact by the weight of the member 30, are dimensioned in such a way that the corresponding pointers are a t the same distance as the distance between marks. During the initial elongation of the test piece these pointers ,7
239
are set to correspond with the marks, no further attention being required after the first few inches of stretch. Further, since the carriages 31 and 32 follow the separation of the marks, the tracing pen moves proportionally along the ordinates of the coordinate paper. In reference to the load, when the pendulum lever is moved in response to the stretching of the test piece, the weight is permitted to draw the plate along its support, whereby normal movement of the plate with respect to the tracing device results. A curve is thus traced which provides a record showing the relation between the load imposed and the resulting elongation. Acknowledgment
The author desires to express his appreciation for the valuable suggestions and criticism offered by W. W. Vogt during the development of the machine. Literature Cited (1) Beadle and Stevens, J. SOL.Chem. Ind., 28, 1111 (1909). (2) Breuil, Caoutchouc & gutta-percha, 4 (1907); “India Rubber and Gutta-percha,” pp. 250-257, Scott, Greenwood, 1910. (3) Burkley, Indza Rubber World, 67, 24 (1922). (4) CheneveauandHeim, J . phys., 2 (5), 535-50 (1912); Caoutchouc 6’ gutta-percha, 11, 8429-8432 (1914). (5) Depew, I n d i a Rubber World, 67,154 (1922). (6) Olsen, Proc. Am. SOL. Testzng Materials, 10, 588-91 (1910). (7) Schob, Gummi-Ztg., 37, 235 (1923). (8) Schopper and Dalen, “Lectures on India Rubber,” pp 289-293 (1908) (9) Schwartz, J. Insl. Elec. Eng., 44, 693-752 (1910). (10) Williams, Indza Rubber World, 72, 411 (1925).
Accurate and Adaptable Micro-Kjeldahl Method of Nitrogen Determination’ W. F. Allen KEDZIECHEMICAL LABORATORY, MICHIGAN STATECOLLEGE, EASTLANSING, MICH.
HE demand for an accurate, simple, and adaptable method for the micro-determination of nitrogen to be used in connection with caffeine analysis of decaffeinated coffee resulted in the development of the apparatus shown here. It is essentially a micro-Kjeldahl distillation apparatus similar to those of Pregl’s, and Parnas and Wagner’s (9). However, it was found to be easier to operate and capable of handling larger samples than the latter. The use of rubber connections was not eliminated, as was accomplished in the apparatus perfected by Kemmerer and Hallett ( I ) , but the apparatus gave very accurate results, and was found to be easily constructed by anyone capable of doing simple glass blowing. The entire apparatus was made of Pyrex glass. Parts I (23.5 cm. long) and J (21.0 cm. long) were made from 25 X 200 mm. test tubes. The inside condenser tubing extending over to the top of the safety trap J was made of 9-mm. tubing. The condenser is 60 cm, in length. The method of procedure is as follows: The sample was digested in the 250-cc. Florence flask A. This size flask was found capable of handling samples requiring not over 10 cc. of concentrated sulfuric acid in the digestion mixture. With samples containing a large amount of organic matter, it was found to be very convenient to add a few drops of 30 per cent hydrogen peroxide during the last part of the digestion. The cooled digested sample was then diluted with freshly boiled distilled water, care being taken that the flask was not over one-fourth full. The flask was placed in the asbestos-
T
1
Received January 20. 1931.
lined jacket B and clamped into position as shown. The jacket eliminated the use of a second burner by keeping condensation of steam down to a minimum. However, very satisfactory results were obtained without the use of the jacket by applying heat to flask A only during the last 3 or 4 minutes of the distillation. For samples containing up to 20 mg. of nitrogen, 15 cc. of 4 per cent boric acid (S) and 15 drops of 0.02 per cent methyl red indicator were measured into the 150-cc. evolution flask C, which was then placed in position under the condenser D with a wide cork resting on top of the flask. Freshly boiled distilled water was next added through funnel E into the 500cc. balloon flask F used as a steam generator. Heat was applied to flask F by a Bunsen burner G, having a shield around the top to prevent air currents from causing variation in the amount of heat applied. Immediately after starting to heat the steam generator, a slight excess of concentrated sodium hydroxide was added through funnel H and then washed down with a few cubic centimeters of water. This funnel was immediately closed with a pinch clamp. Distillation was continued a t such a rate that no bubbling of steam occurred in the receiving flask C. After about 125 cc. of liquid had distilled over, in about 12 minutes, the receiving flask was lowered so that the tip of the condenser was about 2 cm. above the surface, and the tip washed down with a small stream of water. Distillation was continued at an increased rate for one minute longer. The pinch clamp on funnel H was then released before removing the steam generator to prevent suction. A blank was always run using the same
240
ANALYTICAL EDlTlON
Vol. 3, No. 3
directly with dilute standard sulfuric acid from a microburet; and the end point was taken when the shade of color of the indicator exactly matched that of the blank. This proved to be a very faint pink a t a pH of about 5.7. By matching the blank it was unnecessary to subtract a blank correction from the amount of acid used. Table I-Comparative Accuracy of Micro-Kjeldahl M e t h o d SAMPLES NITROGEN ERROR WITH ERROR WITH ANAIN MACROMICROLYZED SAMPLE KJELDAHL KJELDAHL Mg. % % 8 21,114 0.62 0.29
Apparatus for Micro-Kjeldahl Method
reagents and procedure as with the samples, after the apparatus had been thoroughly steamed for a t least 15 minutes. carewas taken that the Same amountof liquid was contained in each flask to insure the same intensity of color. The &Ontents Of the receiving flask from a sample were titrated
c
6 4
1,000 0.583
4
0.100
1.85
1.20 0.24 3.00
It can be readily observed that this apparatus and procedure can be used with rapidity and accuracy on samples requiring as much as 10 cc. of concentrated sulfuric acid in the digestion mixture, as well as samples containing as little as 0.1 mg. of nitrogen. The comparative accuracy of this method with the macro-Kjeldahl method is shown in Table I. By running two sets of apparatus a t the same time it was found that 20 samples could easily be analyzed in a half day. Literature Cited (1) Kemmerer and Hallett, IND. E m . CHEM, l % 1295 (1927) (2) Preg!, “Quantitative Organic Microanalysis,” pp. 94-104, Blakiston, 1924.
(3) Scales and Harrison, J. IND
END.
CHEM.,12, 360 (1920).
Use of Boric Acid in Micro-Kjeldahl Determination of Nitrogen’ Norman M. Stover and Reuben B. Sandin DEPARTMENT OF CHEMISTRY, UNIVERSITY OF ALBERTA,EDMONTON, ALBERTA
The use of boric acid in absorbing ammonia in and thus a saving is made in H E use of a saturated nitrogen determinations by Pregl’s micro-Kjeldahl the use of standard reagents. solution of boric acid method has been found to give accurate results. Also, it is considered by some in place of a standard The distillate containing the ammonia does not need to beeasier toprepareastand.acid to absorb ammonia in a r d acid solution t h a n a to be boiled before titrating. pitrogen determinations by A mixed indicator containing methyl red and tetrastandard alkali solution, and $he K j e l d a h l m e t h o d has bromophenol blue has been found to give good results a carefully standardized acid been reported a number of solution-is less subject to &imes. I t was f i r s t proin boric acid solution. change in storage. posed by Winkler (8),has been mentioned by Scales and Harrison (G), and recomExperimental Procedure mended by Spears ( 7 ) . The authors ( 5 ) have also used boric acid to absorb ammonia in the determination of organic Attempts were made to find a suitable indicator for use in ,nitrogen in liquids by an alkaline fusion method. boric acid solution when absorbing small amounts of ammonia. The use of boric acid in nitrogen determinations is based on The following indicators were tried: bromophenol blue (0.1 the principle that ammonia is absorbed by the weak boric gram in 3 cc. of 0.05 N sodium hydroxide and diluted to 250 acid and is titrated in this solution with a strong acid, such as cc.), tetrabromophenol blue (0.1 gram in 20 cc. of warm alcosulfuric or hydrochloric. . Ammonium borate is probably hol and diluted to 100 cc.), sodium alizarin sulfonate (1 per formed, and in the presence of a strong acid the ammonia is cent in water), methyl red (0.1 per cent in 95 per cent alcohol), again released to form the salt of the strong acid. a mixed indicator containing methyl red and methylene blue The fact that boric acid gives results almost identical with (0.1250 gram of methyl red and 0.0825 gram of methylene $hose obtained when using a standard acid has led the authors blue in 100 cc. of 90 per cent alcohol), and a mixture of methyl to determine the possibility of using it in the micro-Kjeldahl red and tetrabromophenol blue (made by mixing equal method as outlined by Pregl (4). It has several advantages, volumes of solutions of the separate indicators). The mixed as stated by Scales and Harrison (G), and also a few other small indicator containing methyl red and methylene blue has been points in its favor. A minimum amount of the only standard recommended by Johnson and Green (3). The new indicator, Bolution (standard acid) required is used. There is no back tetrabromophenol blue,* was developed recently by Harden titration as the ammonia in the boric acid is titrated directly, 2 W. C . Harden of the laboratories of Hynson, Westcott, and Dunning
T
1 Received
January 30, 1931:
kindly supplied the sample of this indicator.
