Determination of Sodium in Biological Fluids - American Chemical

suggested a colorimetric modification of the Barber and Kolt- hoff (2) technique for the determination of sodium. Spectro- photometric examination (by...
1 downloads 0 Views 411KB Size
Determination of Sodium in Biological Fluids M. C. DARNELL, JR., ANDB. S. WALKER Evans Memorial, Massachusetts Memorial Hospitals, Department of Biochemistry, and Boston University, Boston, Mass.

T

HE observation of Muller (7) that sodium salts of pheno-

lic acids develop colors with solutions of uranyl salts has suggested a colorimetric modification of the Barber and Kolthoff (2) technique for the determination of sodium. Spectrophotometric examination (by the Color Measurements Laboratory, Massachusetts Institute of Technology) of the orange color formed b y sulfosalicylic acid and sodium acetate with uranyl zinc sodium acetate revealed a rather wide absorption band with a maximum below 400 m p and a slight irregularity at about 450 m p (Figure 1). Optical densities of a series of known dilutions of the color were observed with all the standard filters of the Evelyn (4) photoelectric colorimeter. The filter showing maximum transmission a t 440 m p (range 410 to 475 mp) gave the best proportionality of optical density to concentration. Further characteristics of the color were studied, and the color was used in the photometric determination of sodium. A standard solution was prepared containing uranyl acetate dihydrate, zinc acetate dihydrate, and sodium acetate trihydrate in the ratio in which they occur in (UO&ZnNa(C2H,O2)9.6H20, and equivalent to 0.0165 mg. of sodium per ml. T o varying amounts of this solution were added in order 4 ml. of 5 per cent sulfosalicylic acid, 4 ml. of 10 per cent sodium acetate trihydrate, and enough water to make 100 ml. Table I shows the observed values of the optical density of these solutions in the Evelyn macrocolorimeter with the 440 m p filter, and the calculated values of the ratio of optical density to concentration. The progressive decrease of this ratio in Table I indicates t h a t the color does not conform exactly with Beer's law. However, the data in Table I were repeatedly checked, with multiple dilutions of several independently prepared standard solutions, and were found to be constant. The ratio of density to concentration was expressed as a function of the density b y the method of least squares (8) i n the equation

K = 0.889

- 0.1449L

(1)

where K is the ratio of density to concentration and L is the density. Within the range of concentrations studied the color was observed to be stable for a period of 3 hours and over a temperature range of 5' C. The color was found to be sensitive to changes of acidity. The addition of mineral acid diminished

the color markedly. The dropwise addition of sodium hydroxide first increased the color to a density almost twice as great as that used in these experiments; further additions caused rapid fading of the color. The most intense color could be obtained by neutralizing the sulfosalicylic acid to a phenolphthalein end point before adding i t to the uranyl solution. This intense color was not used in the determination of sodium, however, because i t was found to be much more sensitive to changes of acidity than was the color with sulfosalicylic acid and sodium acetate, and because i t did not conform with Beer's law as closely as did the color used.

FIGURE 1. ABSORPTION SPECTRUM Solution contains 44 mg. of (UO?jaZnNa(C?HsOz)~.6H10,4 ml. of 5 per cent sulfosalicylic acid, and 4 ml. of 10 per cent NaCzH309.3H20 i n 100 ml.

Variations in the relative amounts of sulfosalicylic acid and sodium acetate used in the development of the color were found to cause variations in the intensity. Consequently, the concentrations of sulfosalicylic acid solutions were standardized by titration against standard sodium hydroxide; 5 ml. of sulfosalicylic acid solution required 17.25 ml. of 0,100 iV sodium hydroxide for neutralization to a phenolphthalein end point.

Reagents and Method

Trichloroacetic acid, 20 per cent solution. Ashless filter paper, Whatman No. 40. Uranyl zinc acetate reagent, prepared according t o Weinbach TABLE I. RELATIONSHIP OF OPTICALDEKSITY TO C O S C E S T R ~ - (11). Solution A: 77 grams of uranyl acetate, UOZ(C,H,OZ)Z.2H20, and 14 ml. of glacial acetic acid are dissolved by heating TIOS OF SODIUM and stirring in 400 ml. of water. Solution B: 231 grams of zinc L, K, acetate, Zn(C2H302)2.2H20, and 7 ml. of glacial acetic acid are Optical Density Density/Concentration dissolved by heating and stirring in 400 ml. of water. The two Sodium (Average) a (hverage)O solutions are mixed while still warm. After standing 24 hours, MQ. the solution is ready for use. I t is filtered immediately before 0.05so 0.879 0.066 using. 0.876 0.1156 0.132 0.1707 0.862 0.19s Ethyl alcohol, 95 per cent. 0.2264 0 . s5s 0.264 Ethyl acetate-acetic acid wash reagent: 300 ml. of c. P . ethyl 0,2403 0.857 0,2805 acetate diluted to 1 liter with glacial acetic acid. 0 . s54 0.2537 0.297 0.2656 0.847 0.3135 Ethyl ether, redistilled. 0.848 0.2798 0.330 Sulfosalicylic acid, 5 per cent solution, checked by titration, 0.2924 0.844 0.3465 as described above. 0.363 0.840 0.305 0,3795 0.319 0.841 Sodium acetate trihvdrate, 10 per cent solution: 50 grams of 0.838 0.396 0.332 c . P. sodium acetate, XaC2H302.3H201dissolved in water and di0.462 0,385 0.833 luted to 500 ml. 0.528 0.437 0.827 0.594 0.487 0.819 Sulfuric acid, 6 N . concentrated nitric acid, and 30 Der cent 0.660 0.536 0.812 hydrogen peroxide. ' a Equation 1 was derived from individual, not average, values of K and L. For the determination of sodium in blood serum, urine, or cerebrospinal fluid, 1 ml. of 1 to 10 trichloroacetic acid filtrate of the 242

APRIL 15, 1940

ANALYTICAL EDITION

material, or 0.1 ml. of material wet-ashed by the procedure of Hoffman and Osgood ( 6 ) , is analyzed. In the preparation of trichloroacetic acid filtrates, 1 ml. of biological material is delivered from a pipet (calibrated to deliver the material measured) into 7 ml. of distilled water in a test tube; while shaking the tube gently, 2 ml. of 20 per cent trichloroacetic acid are added. The contents of the tube are thoroughly mixed by stirring and allowed to stand for 10 minutes. After filtration through an ashless filter paper, 1 ml. of the clear filtrate is delivered from a pipet calibrated for delivery of water into a 15-ml. conical centrifuge tube for treatment with uranyl zinc acetate reagent. In the preparation of ashed samples, 0.1 ml. of the material to be analyzed is delivered from a pipet calibrated to deliver that material into a 15-ml. conical centrifuge tube of heat-resistant glass. After addition of 0.2 ml. of 6 N sulfuric acid and 0.1 ml. of concentrated nitric acid, the tube is placed in a beaker of boiling water for 10 minutes or longer. The tube is then removed from the water bath and the contents are charred over a free flame. By allowing a small flame from the microburner to strike just above the level of the liquid in the constantly shaken tube, very little danger of loss from spattering is encountered. The tube is allowed to cool while another is being similarly treated. Then one drop of 30 per cent hydrogen peroxide is added from a capillary pipet. The contents are again evaporated over a free flame until sulfur trioxide fumes fill the tube, and allowed to cool during the treatment of the second tube. The addition of hydrogen peroxide is continued until, after evaporation to sulfur trioxide fumes, only a colorless drop of liquid remains in the tube. (All tubes in any series are treated with equal amounts of hydrogen peroxide in order to ensure constancy of the blank; 6 drops or less will generally be enough to produce a clear solution.) After the final evaporation with hydrogen peroxide, the tube is allowed to cool for 10 minutes or more, and to it is added 0.9 ml. of water. After mixing the contents of the tube, it is ready for treatment with uranyl zinc acetate reagent. To 1 ml. of solution prepared by one of the methods described above, 5 ml. of freshly filtered uranyl zinc acetate reagent are added. A t 5-minute intervals are added seven 0.3-ml. portions of ethyl alcohol. These additions must occupy at least 0.5 hour, and may take longer. After each of the first five additions of alcohol the liquid in the tube is mixed first by tapping the bottom of the tube, producing a rotatory motion in the upper part of the liquid, then by rolling the tube back and forth between the palms of the hands, thus producing effect,ivemixing in the lowest part of the tube. The last two additions of alcohol serve to wash down the walls of the tube and are alloIved to remain layered on the solution. After the last addition of alcohol, the tube is centrifuged a t 2000 r. p. m. for 10 minutes, decanted, inverted, and allowed to drain for 5 minutes. The mouth of the tube is wiped dry and the precipitate is agitated by bloxing on it a fine stream of about 2 ml. of ethyl acetate-acetic acid wash liquid. (In the case of urines which contain much phosphate the precipitate must be agitated by stirring with a glass rod, which is washed off with the wash liquid after each use.) The walls of the tube are washed down with a small amount of the liquid. Centrifuging for 10 minutes, draining for 5 minutes, and wiping are carried out as before, and about 5 ml. of ether are used to wash the precipitate and the walls of the tube; precipitates from urines must be stirred with a glass rod. The tube is centrifuged for 5 minutes, decanted, and drained for 1 minute. (Longer draining may allow the flaky precipitate to become so dry that it crumbles and falls out of the tube.) The precipitate is washed a second time with 5 ml. of ether, centrifuged 5 minutes, decanted, and drained for 1 minute. The tube is put in a warm place for 5 minutes to evaporate the last traces of ether. To this point the procedure is identical for blood, urine, and cerebrospinal fluid, xith the exception of the agitation of precipitates, as noted. For blood and cerebrospinal fluid, the remainder of the procedure is as follows: The washed, dried precipitate is dissolved in 1 to 5 ml. of water and t,ransferred quantitatively to a 100-ml. volumetric flask. The solution in the flask is diluted to about 70 ml., and to it are added in order 4 ml. of 5 per cent sulfosalicylic acid, 4 ml. of 10 per cent sodium acetate trihydrate, and water t o make 100 ml. The contents of the flask are mixed thoroughly and a 15-ml. portion is transferred to an Evelyn colorimeter tube and read in the colorimeter, using the 440 mp filter. The instrument is set to read 100 Tvith a tube containing 4 ml. each of 5 per cent sulfosalicylic acid and 10 per cent sodium acetate in 100 ml. The optical density, L, is calculated from the formula L = 2 - log G, where G is the corrected galvanometer reading. The concentration of sodium in the sample tube is read from a plot of data in Table I, or is obtained by dividing L by the correct value of K from Equation 1. The sodium content of a similarly treated reagent blank must be subtracted.

243

TABLE 11. EFFICIEKCY OF PRECIPITATION AND WASHING IN SODIUM CHLORIDE SOLGTIONS Sodium Taken

Sodium Found

Jig.

.Mg.

%

0.168 0.168 0.li3 0.231 0.231 0.238 0.238 0,301 0.301 0.301 0.332 0.332 0.336 0.336 0.336 0.399 0.399 0.399 0.463 0.463 0.463 0.534 0.534

0.166 0.168 0.175

-1.2 0 f1.2 0

0.231

0.229 0.237 0.237 0.301 0.301 0.299 0.334 0.332 0.334 0.336 0.334 0.401 0.399 0.399 0.463 0.463 0.463 0.534 0.534

Error

-0.9

-0.4 -0.4 0 0 -0.7 f0.6 0 -0.6 0 -0.6 +0.5

0 0 0 0

0 0 0

If c is the concentration of sodium in the sample tube, b is the concentration in the reagent blank, and v is the volume of biological material taken for analysis, then 100 ( c - b ) / v = mg. of sodium per 100 ml. of

biological material

(2)

The triple salt precipitated from urine is dissolved by shaking Any uranyl phosphate present remains in the form of a gelatinous precipitate. The tube is centrifuged for 10 minutes a t 2000 r. p. m. Five milliliters of the clear supernatant fluid are pipetted into a 50-ml. volumetric flask, diluted to 35 ml. with water, treated with half-quantities of the color reagents, diluted to volume, and read in the Evelyn colorimeter. The result is calculated in the same manner as in the procedure for blood or spinal fluid, using Equation 2. it with 10.00 ml. of water.

The method as presented is modified from that of Weinbach (11)in three important respects: the use of ethyl acetate in acetic acid as the wash liquid, the removal of phosphate after the precipitation of sodium, as in the method of Hoffman and Osgood ( 6 ) , and photometric instead of titrimetric quantitation of the precipitated (U02)3ZnSa(C2H30r)o.6H20. Saturated solutions of the triple salt in alcohol (2j, glacial acetic acid ( I O ) , and acetone (11) have been used as wash liquids for the precipitated triple salt. Hoffman and Osgood (6) found that all three of these saturated solutions produced precipitates of uranium salts when brought in contact with uranyl zinc acetate reagent, while the pure liquids did not. By decreasing the solubility of the triple salt in acetic acid by the addition of ethyl acetate, it was possible to avoid saturation of the wash liquid with the triple salt. A solution of 30 volumes of ethyl acetate in 70 volumes of glacial acetic acid was found to give no color n-ith ferrocyanide after shaking with the triple salt, and to produce no precipitation when mixed with uranyl zinc acetate reagent. Although this solution has a very irritating odor, it can be handled without discomfort by delivering i t from a wash bottle with an extra, finger-stoppered hole in the stopper and a Bunsen valve on the inner end of the mouthpiece. The efficiency of this wash liquid, as used in conjunction with the precipitation method, was tested by analyzing varying amounts of standard sodium chloridcb solution b y the above method. Samples were treated immediately with the precipitating reagent, without pretreatment with the deproteinizing reagents. Results are shown in Table 11.

VOL. 12,NO. 4

INDUSTRIAL AND ENGINEERIKG CHEMISTRY

244

As a check on the validity of the method of phosphate removal, 1-ml. portions of a solution containing 0.324 mg. per ml. of sodium as disodium hydrogen phosphate were analyzed by the procedure given above for the analysis of urine. When compared with the results of similar analyses of portions from which phosphate had been removed by the method of Butler and Tuthill (S), no difference in reliability was observed.

Results Table I11 shows the results of analyses of representative biological fluids. The average value of the sodium concentration in 15 normal human blood sera, determined on ashed samples, was found to be 334 mg. per 100 ml.; the range was from 324 to 342 mg. per 100 ml. As determined on trichloroacetic filtrates of sera, the average was 341 mg. per 100 ml.; the range was from 333 to 349 mg. per 100 ml. The average value for ashed samples is in excellent agreement with the figure of 335 mg. per 100 ml., which Peters and Van Slyke (9) give as the normal concentration. The variation of the extremes was less than *4 per cent, which Hoffman and Os-

good (6) state as the normal variation of human serum sodium. Trichloroacetic acid filtrates gave values from 0.9 to 4.0 per cent higher than did ashed samples of the same serum. The average difference was 2.3 per cent. Hald (6) and Ball and Sadusk (1) also found that the values for sodium were higher when determined on trichloroacetic acid filtrates than on ashed samples. The values for the sodium content of two samples of cerebrospinal fluid were comparable with those of the blood sera analyzed. The accuracy of the method is demonstrated by the agreement between duplicate samples treated in the same way. As an absolute check on the accuracy of the method, known amounts of sodium chloride and mixtures of standard sodium chloride with analyzed blood sera were analyzed by the method given. Results are shown in Table IV.

TABLEIV. RECOVERY OF SODIUM FROM MIXTURES OF ANALYZED SERUM AXD STANDARD SODIUM CHLORIDE SOLUTION Treatment Ashed

TABLE111. SODIUM CONTENT OF BIOLOGICAL FLUIDS Material Serum

Spinal fluid

Urine

Sodium Found Ashed Preoipitated sample sample Mg. per IO0 ml.

Apparent Increase in Preoipitated Sample

%

Filtrate

Sodium Taken From From NaCl serum

Mg. 0

0.340 0.340 0.340 0.340 0.340 0.340

0 0.334 0.334

Mg.

0

0 0 0.336 0.336 0.342 0.342

0 0 0

Total Sodium Found

Sodium from NaCl (Total Serum Blank)

-

Mg.

lug. 0.005 0.344 0.347 0.681 0.681 0,688 0.688

0.339 0.342 0.340 0.340 0.341 0.341

0.016 0.350 0.352

0.334 0.336

....

....

-

Error

%

...

-0.3 +0.6 0 0 +0.3 f0.3

...

0 +0.6

337

34 1 341

1.2 1.2

331 331

334 334

0.9 0.9

33 1

342 342

3.3 3.3

Summary

334 331 337

345 342 342 342

3.8 2.9

342 342

349 349

2.0 2.0

340

346

1.5

337 337

345 345

2.4 2.4

326 324

336 33s

3.4 4.0

326 326

338 338

3.7 3.7

332 332

344 344

3.6 3.6

335 335

344

2.7

329 326

333 333

1.7 1.7

The color developed with uranyl zinc sodium acetate by sulfosalicylic acid and sodium acetate has been studied. While the color does not follow Beer’s law exactly, it has been found to be reproducible and stable to time and temperature. A new method, consisting of precipitation of uranyl zinc sodium acetate from alcoholic solution, removal of the excess of precipitant by washing with ethyl acetate in acetic acid and with ether, removal of phosphate as uranyl phosphate, and photoelectric measurement of the color developed by sulfosalicylic acid and sodium acetate, has been developed for the determination of sodium in biological fluids. Recovery of sodium from known amounts of sodium chloride, both in simple aqueous solution and added to blood sera, has indicated that the various steps, as well as the assembled procedure, have a maximum error of less than 1 per cent.

335

33s

0.9

335 335

34 1 338

1.8 0.9

335 335

341 341

1.8 1.8

332

341 338

2.7 1.8

...

387 387

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

(3) Butler, A. M., and Tuthill, E., J. B i d . Chem., 93, 1 7 1 (1931) (4) Evelyn, K. A., Ibid., 115,6 3 (1936). (5) Hald, P. M., Ibid., 103,4 7 1 (1933). (6) Hoffman, W. S., and Osgood, B., Ibid., 124, 347 (1938). (7) Muller, Chem.-Ztg., 43, 739 (1919). (8) Pearl, Raymond, “Introduction to Medical Biometry and Statistics”, p. 334, Philadelphia, W. B. Saunders and Co., 1923. (9) Peters, J. P., and Van Slyke, D. D., “Quantitative Clinical Chemistry”, Vol. 1, p. 753, Baltimore, Williams and Wilkins Co., 1931. (10) Salit, P. W., J . B i d . Chem., 96, 659 (1932). (11) Weinbach, A. P., Ibid., 110, 95 (1935).

... ...

PRESENTED before the Division of Biological Chemistry a t the 98th Meeting of the American Chemical Society, Boston, Mass. Abstracted from a dissertation submitted by Matthew C. Darnell, Jr., t o the Graduate School of Boston University in partial fulfillment of the requirements for the degree of doctor of philosophy.

...

.... ._ .... .. ... ... ... ...

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

...

493 490 162 162 120 120 190 192 203 20 1 215 212 186 186 190

1.5 1.5

... ... ...

Literature Cited (1) Ball, E. G., and Sadusk, J. F., Jr., (1936). (2) Barber, H. H.. and Kolthoff, I. M., J . (1928).

J . BioZ. Chem., 113, 6 6 1 A m . Chem. SOC.,50, 1625