ANALYTICAL CHEMISTRY
1944 (8),cannot interfere because it is cleaved by periodate to acetic acid. The presence of glycolic acid, a normal product of the periodate scission of ketoses, can cause high results when the chromotropic acid reaction is carried out at high sulfuric acid concentrations (Table VI). Apparently this effect obtains from sulfuric acid oxidation of glycolic acid to formaldehyde, and it may be completely eliminated without loss of sensitivity by employing less concentrated sulfuric acid for developing the dye (Table IV). (Note also the results for dihydroxyacetone in Table VII.) Effect of Acidity of Oxidation Mixture. Comparison of the data in Table V for the oxidation of glucose, xylose, and glyceraldehyde by unbuffered periodic acid with data obtained by oxidation of these substances by the recommended procedure (Table V I I ) indicates that scission of these reducing sugars to formaldehyde is slower under acidic conditions than it is in bicarbonate buffer. Hence, regulation or knowledge of the acidity is important in this kind of determination. The data in Table T' also demonstrate a n apparent unreliability in the Fleury and Lange method for following the progress of the periodate osidation-for example, in the oxidation of glucose the production of formaldehyde is only 63% of the theory a t the end of 1 hour, whereas a t this time periodate consumption measured by iodidearsenite reduction is 99% of theory. Results similar to these have been reported by Hughes and Nevell ( 4 ) for production of formic acid from glucose during periodate oxidation, and these
investigators have attributed the lag in the appearance of formic acid, as compared with apparent periodate reduction (measured by Fleury and Lange method), to formation of stable periodate complexes which are not reduced by iodide arsenite mixtures. The present data concerning forrnald?hj.de formation a t different acidities comprise a preliminary report of an examination of reducing sugar-periodate interaction which is still in progress. LITER.4TURE C I T E D
Bricker, C . E., and Vail, W..1., . ~ N . A LC . H m f . , 20, 647 (1948). Corcoran, A. C., and Page, J. H.. J . Biol. Chem., 170, 165 (1947). Fleury, P., and Lange. J., J . pharm. chim.,17,107 (1933). Hughes, G.,and Nevell, T. P., Trans. Faraday Soc., 44, 941 (1948). Lanibert, M., and Neish, h.C., Can. J . Research, 28 B, 83 (1950). AIitchell. 1%'.E. A, and Perrival, E., J . Chem. Soc., 1954,1423. Reeves, It. E., J . A m . Chent. Soc., 63, 1476 (1941). Speck, J. C . , ANhL. C H E X . . 20, 647 (1948). Witzemann, E. J., Evans, W. L., Hass, H., and Schroeder, E. F., "Organic Syntheses," Coll. Vol. 11, p. 305, New York, John Wiley & Sons, 1943. RECEIVEDfor review June 28, 1954. Accepted August 26, 1964. Presented before the Division of Carbohydrate Cheinistry a t the 124th Meeting of the . b i E R I C A N CHEmchL SOCIETY, Chicago, Ill., September 1953. Work supported by t h e Quartermaster Food and Container Institute for the Armed Forces. Views presented herein a r r the authors' and are not t o be construed as reflecting t h e views or endorrrment of the Derxwtnient of Defense.
Direct Ethylenediamine Tetraacetate Titration Methods for Magnesium and Calcium CHARLES W. GEHRKE, HAROLD E. AFFSPRUNG', and YUNG C. LEE2 Department o f Agricultural Chemistry, Missouri Agricultural Experiment Station, Columbia, M o .
A direct titration method for magnesium in 1imesLones employing ethylenediamine tetraacetate as the titrant and Eriochrome Black-T as the indicator is described. The calcium is precipitated as calcium sulfite at a pH of 5 to 7, and the calcium sulfite-RZ03 precipitate is remoied by filtration. The direct titration procedure is simple, accurate, and precise, and requires less time than the oxinate method or the indirect ethylenediamine tetraacetate titration method. The end point is excellent and the pure blue color formed is pronounced and stable, A n ethylenediamine tetraaceta te titration method is described for the routine analysis of calcium in all ty-pes of biological substances, such as plant materials, milk, experimental rations, grains, feces, and other substances such as fertilizers in which inorganic phosphate interferes with the titration. 3Iurexide is used as the indicator. .in anion exchange resin in the chloride form in a small column is used to remove interfering anions. The results obtained by the proposed procedure are equal in accuracy and precision to those of the classical oxalate permanganate method, and the method is much less time-consuming.
B
ECAUSE of the large number of analyses of limrstone samples and of agricultural materials made each year in control chemical laboratories, any procedure TI hich permits the rapid determination of calcium and magnesium in these substances will be of great value. The classical standard mcth1
Present address, New York State College for Teachers, Albany, N. Y .
a Present address, Sharp and Dohme Plant, Lansdale, Pa.
ods of analysis for thesf) elemeiits are slow and time-consuming. Shoitly after the introduction nf ethylenediaminetetraacetic acid (EI>Td) as a complexing agent and titrant, many ulip1ic:itions of this reagent m r e made in analytical chemistry. Generaliy, t,hese methods involve two titrations, a titration using Eriochrome Black-T as indicator for total calcium and magnesium, and a t,itration using murexide R S indicator for calcium. The magnesium is then calculated a s the difference between the tn-o titrations. Any errors inherent in either titration \vi11 he reflected in the values for magnesium. Thus, t,he major objrctives of this investigation were to develop reliable, accurate and precis?, simple, analytical methods for the determination of magnesium in limestone and calcium in :i variety of agricultural products :ind biological substances.
MAGNESIUM IN LIJIESTONE AIethods have been developed which are based on the origind work of Schwarxenbach and coworkers (5, 16, 16) for the determination of calcium and magnesium in water (3,4,9, I O ) , limestone and soils ( 2 , 6-8). Several investigators have presented direct methods for magnesium in limestone ( 2 ) and water (3, IO) using oxalate as the precipitating agent for calcium. Some difficulties were experienced in regard to the end point and completeness of separation. The present paper describes a simple direct disodium dihydrogen et,hylenediamine tetraacetate titration procedure for magnesium in limestone. The method is rapid and accurate. APPARATUS AND R E A G E N T S
Artificial light source, such as Precision Scientific Co. Titralite Catalog No. 9555.
V O L U M E 2 6 , NO. 12, D E C E M B E R 1 9 5 4
1945
Filter paper, Whatman S o . 5 or equivalent. pHydrion indicator paper. Buffer solution I, pH 10. Dissolve 67.5 grams of C.P. ammonium chloride in 200 ml. of distilled water, add 570 ml. of reagent grade concentrated ammonium hydroxide, dilute t o 1 liter (IO). Buffer solution 11, pEI 8. Dissolve 80 grams of C.P. ammonium chloride in 200 nil. of distilled water, add 60 ml. of reagent grade concentrated ammonium hydroxide, dilute to 1 liter. Eriochromc Black-T indicator (F-241). Dissolve 0.2 gram of the indicator powder (Eastman Kodak, 1'6361) in 50 nil. of analytical reagent grade methanol containing 2 grams of hydroxplamine hydrochloride (6). Potassium cyanide solution, 2% aqueous solution. Use reagent grade potassium cyanide. Standard solution of disodium dihydrogen ethylenediamine tetraacetate 0.4 and 0.1%. Dissolve 20 grams or 5 grams of the reagent in double distilled water and make t,o 5 liters. Standardize the solution against standard magnesium and calcium solutions. The titers are approximately 0.27 or 0.068 mg. of inagnesium per ml., i i n d 0.44 or 0.11 mg. of calrium per ml., rrspectively . Standard magnesium solution. 1)issolvc 1 gram of analytical reagent grade magnesium turnings in dilute hydrochloric acid, tlilute to 1 liter wit,h double distilled water. This solution contains 1 mg. of magnesium per ml. Sodium sulfite solution. Dissolve 20 grams of T.P. sodium or ammonium sulfite in 100 ml. of douhle distilled water. Pre])are this solution just before use.
Table 111. Recovery of Magnesium Added to Standard Limestone Samples IIac11 Chemical C o .
Recovered, Found
%
aarnlilt,h
2.50
52 S(i
52.01 52.01 52.01
98.95 98.95 98.95
1001
Ki.81
115 81
114 97 114 70 114 90
99.27 99.04 99 21
1002
2.44
.52.44
32 01 J2.01 52.01
99.18 99.18 99.18
1005
5,04
. i , 5 , 04
,i4 81 .i4. 9.5
>- t,he t\vo inethods aglwd closely. The data are presented iii Tut)le 11. A rcroverJ- study was made, the data of which are presented in Tal)!e 111. Ten standard samples of limestone obtained from thv Hwh Cliemical Co. and the National Bureau of Standards \wrcl \vrighcd i L i i d dissolved in 1 to 1 hydrochloric acid, xarmed gcantly, then evaporated to dryness. The following werc :~tl(ledto the rrsiduc-an aliquot of a standard magnesium chloride solution cont:iining 50 mg. of magnesium, 1 ml. of 1. to 1 hydrochl.tric acid, and 10 ml. of water. The sample solutions were then buffered to pH 5 t,o 7, t'he R20, precipitated, and the calcium separated a s calcium sulfite. The filtrates were diluted to 100 ml., and 10-ml. xliquots were t,aktsn for the magnesium titrations. The average recovery of magnesium was found to be 99.04%. The recovery of 99.04% of the magnesium added to
1946
ANALYTICAL CHEMISTRY
Table IV. Taken
Founda
1.00
1.00
2.00 5.00
LOO
10.00
15.00 20.00 25.00
30.00 40.00 a
Recovery of Magnesium from Standard Solution of Magnesium Chloride
2.00 10.01 15.00 20.01 25.09
30.14 40.24
Magnesium, M g . Difference
.. .. .. +'O
o: 1
i'O.01 +0.09 +0.14 +0.24
Each value represents an average of three independent analyses.
standard limestones is slightly low but is reasonably satisfactory. The recovery range was 98.07 to 99.84%. The per cent of magnesium oxide in the standard samples used in the recovery studies was from 0.85 to 21.82%. The amount of magnesium added to these samples ranged from a factor of 2 times the magnesium present to a factor of about 25. There was no correlation between low recovery values for magnesium and the chemical composition of the samples-that is, the content of calcium oxide, phosphorus pentoxide, magnesium oxide, silicon dioxide, or the Rz03 group. I n order to determine the titration error several samples of a standard solution of magnesium chloride were titrated using Eriochrome Black-T a s the indicator (Table IV). The end point and the stoichiometric point corresponded well when the amount of magnesium in the aliquot v a s from 1 to 20 mg.; at a higher concentration a positive error was introduced due to the color range of the indicator at the end point. However, in routine limestone analysis, the aliquot taken for the titration by the proposed direct method seldom contains more than 20 mg. of magnesium. A blank Kith the blue end point color was prepared for comparison purposes, although this as not necessary. It was found desirable to titrate using an artificial fluorescent light as a background light. I n this way all of the analytical work was conducted with the same background illumination. The color change approaching the end point is gradual, but definite, and a blank correction was not found to be necessary nhen the aliquot contained 20 mg. or less of magnesium. DISCUSSION
The direct titration of magnesium in limestones is simple, accurate, and rapid. Several hundred commercial limestone samples have been assayed by this method in the RIissouri Experiment Station Laboratories and the results have been very satisfactory. An average standard deviation of zkO.03 was obtained by the direct method on 16 samples of standard and commercial limestones ranging from low to high (0.89 to 21.54) in percentage of magnesium oxide. At least three independent determinations were made on each sample. These data show that the method has a good degree of precision. Most previously published methods for the analysis of magnesium in limestone or in plant and animal materials, using ethylenediamine tetraacetate as the titrant, involvr two titrations and the magnesium is calculated as a difference. Any errors inherent in these two determinations will be reflected in the calculated value for magnesium. The proposed direct method involves a single titration for magnesium onlv. The direct titration procedure for magnesium requires less time than the oxinate method or the indirect ethylenediamine tetraacetate titration method. The end point in the direct method is excellent. Also, the pure blue color is more pronounced and stable, and does not fade as is commonly experienced in the absence of sulfite. The end point color remains unchanged for 24 hours or more. This improvement in end point color is due to the excess of sulfite ion present in the solution, as in the absence of sulfite, the normally observed fading occurs. Bray's group (8) reported that if a sam-
ple contains more than 1%of iron, the solution may turn from a blue to a brown color in a few minutes. This change in color did not occur in the proposed procedure, even when the samples contained 4 or more % of R203. I n this method the Rz03 is removed when the solution is buffered to pH 5 to 7 previous to the removal of calcium by the addition of sodium sulfite. A rather large difference exists in the solubility of calcium and magnesium sulfites (CaS03.2H20, 0.0043, and MgS03.6H20, 1.25 parts per hundred parts of water). The separation of calcium from magnesium is based on this observation. The sulfite solution must be added sloaly and a t room temperature, for when the sulfite solution is added rapidly, or a t elevated temperatures, excessively low results are obtained for magnesium. A quantitative spectrographic study was made on commercial and standard limestone samples to determine the extent of coprecipitation of magnesium sulfite with the R203 group and the calcium sulfite residue. The magnesium content of the residues was of the order of 1 to 5000 to 1 to 2000. -4 satisfactory separation of the magnesium from the calcium was thus obtained by means of a single precipitation of the calcium as calcium sulfite. However, the recovery data seem to show considerable coprecipitation, but as these data vere obtained from samples to vhich rather large amounts of magnesium had been added and the rcsulting solutions were quite concentrated, it is felt that these results represent the maximum error to be expected. The solutions used for the spectrographic study were much less concentrated than those used for the recovery study, by a factor of 2 to 25, and these data show that excrssive coprecipitation does not occur unless the concentration of magnesium in the solutions used is high. CALCIUM IN BIOLOGICAL SUBSTANCES Attempts have been made to apply ethylenediamine tetraacetate methods for the determination of calcium in plant materials (6, If, 17). Difficulties are encountered in these titrations when orthophosphate ions are present. A direct method has been developed for calcium in milk and milk fractions ( I S ) . hlason (14) worked out a method in which calcium was separated from phosphate by a cation exchanger. The calcium was then eluted from the exchanger and titrated with ethylenediamine tetraacetate in the effluent. -4definite need was realized for a direct titration procedure for calcium in biological materials which would be free of interference due to phosphate ions. The present paper describes a rapid and accurate anion exchange separation-direct titration procedure. APPARATUS AND REAGENTS
Anion exchange resin. dmberlite IR-4B, an anion exchanger with a high capacity for phosphate was used. The resin particles are 0.4 to 0.6 mm. in size, and are approximately 20- to 50-
Table V.
Recovery of Calcium from Standard Solutions
Containing Phosphate Phosphate5 (PO&---) Added,
Added
Mg.
10.00 10.00 10.00 10.00 10.00 IO.00 10.00 10.00 10.00 10.00 10.00 10.00
6.00
10.00 16.00 20.00 26.00 36.00 40.00 46.00 50.00 56.00 60.00 80.00 70.00
Calcium, JIg. Found 10.00 10.02
10.02 I O . 02 I O . 02 9.99 9.99 9.97 9.98 9.99 10.00 9.98 5.00 5.00 4.99
5.00 5.00 5.00 Standard deviation 10.017 Phosphate added as potassium dihydrogen phosphate. 90.00 100.00
-
Difference 0.00
10.02 co.02
+0.02 +0.02
-0.01 -0.01 -0.03 -0.02 -0.01 0.00 -0.02 0.00 0.00 -0.01
V O L U M E 26, NO. 1 2 , D E C E M B E R 1 9 5 4
1947
mesh. This resin can be obtained from the Rohm and Haas mg. Thus, it is possible to pass a large number of aliquots through a column before regeneration of the resin is necessary. Co., Philadelphia, Pa. Anion exchange column. Regenerate the resin to the chloTitration of Calcium. Add 5 ml. of lOy0potassium hydroxide and about 50 mg. of murexide indicator to the effluent and washride form by the batch process. First exhaust it with three ings in the Erlenmeyer flask. Titrate the sample, with swirling, separate portions of 5% sodium carbonate or sodium hydroxide. using a standard solution of ethylenediamine tetraacetate of the Wash until all excess base is removed. Then treat the resin with a t least three se arate portions of 50/, hydrochloric acid desired titer. The end point is from salmon pink to a deep with stirring. Rinse t i e resin with distilled water until there is purple. .4t the end point no pink coloration is observed upon ' viewing the solution against a background of artificial light. no further color throw. Prepare a glass column approximately It is always preferable to titrate with an artificial light, and a 23 cm. long and 2 cm. in diameter. Divide the column 5 cm. from the bottom by sealing in a coarse porosity sintered glass plate. blank is used for comparison. Attach a 2-mm. two-way stopcock to the bottom of the column Calculations. The calcium content may be calculated as follows when a 10-gram sample is ashed, made to 250-ml. volume, for regulating the flow of solution through the column. Place and a 25-ml. aliquot is used for titration. 30 to 50 grams of IR-IB-C1 in the column. The column of resin is about 9 cm. in height. Stir the resin to remove air bubbles and maintain the water level above the resin surface. = per cent calcium Calcium indicator. Intimately mix 40 grams of C.P. potassium sulfate and 0.2 gram of murexide powder (Eastman Rodak A = milligrams of calcium pcr ml. of titrant Product) in a mortar (4). B = milliliters of titrant Potassium hydroxide solution, 10%. Dissolve 10 grams of reagent grade potassium hydroxide in 100 ml. of distilled water. RESULTS Standard calcium solution. Dissolve 2.4972 erams of reagent grade calcium carbonate, previously dried a t 170' C., in d h t e The results of a series of detci niinatinn~are presentcd in Table hydrochloric acid. Dilute to 1 liter a.ith double distilled water. V, in which increasing amounts of phosphate were addrd to This solution contains 1.00 mg. of calcium per ml. Standard phosphate solution. Dissolve 2.8658 grams of C.P. ____ uotassium dihvdroeen Dhowhate in double distillea 6ater: and Table 1-1. Determination of Calcium in Forages by Titration Methods make to 1 liter. This solution Calcium, % Deviation contains 2.00 mg. of phosphate (POa---)a Oxalate ETA Oxalate ETA per ml. Sample No. Phosphate method method Diff method method 0.80
ANALYTICAL PROCEDURE
Preparation of Sample. Ash a 10-gram sample of the plant material overnight in a quartz crucible at 550" C. Place the sample in a cold furnace and gradually increase the temperat,ure to 550" C. Add 75 ml. of 1 to 1 hydrochloric acid and 2 drops of concentrated nitric acid. Digest the sample on a hot plate or steam bath for 2 hours and finally take to dryness. Add about 3 ml. of 1 to 1 hydrochloric acid to the residue and approximately 100 ml. of distilled water and digest for about 1 hour. ,4110~the solution t,o cool and transfer quantitatively to a 250ml. volumetric flask. Make to the mark with distilled water and mix thoroughly. R e m o v a l of P h o s p h a t e . Transfer a 25-ml. aliquot, of the sample solution to a 250-ml. beaker. The aliquot should contain between 10 and 15 mg. of calcium. Xeutralize the solution to pH 3 to 4 with a 10% solution of potassium hydroxide using pHydrion indicator paper. Pass the sample solution through the column of anion exchanger in the chloride form, and collect the effluent in a 300-ml. Erlenmeyer flask. Control the rat'e of flow to about 2 t o 3 ml. per minute. Wash the resin column thoroughly with 150 ml. of distilled water in three separate portions. Pass the first portion of 50 ml. through a t t,he same rate as the sample. Pass the second portion through a t a rate of about 9 ml. per minute. Allow the final portion to pass through the column freely. Twelve samples can he handled a t one time using a bank of 12 columns. The exchange capacity for phosphate of one column containing 30 grams of resin is about 1800
0.200 0.227 0.268 0 213 0 222 0 209 0 207 0 234
Average Standard deviation 2 (forage)
240 240 244 240 242 244
0.242
0.847
0 836 0 818 0 826 0 827 0 820 0.822 0.836 0.827 0 833 0.827
0.833 0.847 0.861 0.864 n . 850 0 840 0 830
f0.011 a0.019
0.020
0.847
Average Standard deviation 4 (small animal ration)
1 16.i
1.103
1 693
1,693 1 644 1 793 1 814 1 809 1 783 1 784
1.787 1.787 1.778 1.747 1.747 1.742 1.744 1.747 1,747
1.752
1.758
0,812 0,840 0.848 0.898 0.902
0 879
0.000 -0.014 0.000 4-0.014 +0.017 4-0.003 -0.007 -0.017
-0,020
0.0111
+o
012 t O 003 +0.012 - 0 011 - 0 009 -0 009
6.48
1.47
0.883 0.888
Average Standard deviation G (forage)
+ O 004 + O 045 -0.010 -0.001 -0,014 -0.016
240 242
0,223 0.60
-0.023
244
Average Standard deviation 3 (forage)
Average Standard del iation .i (dehydrated spinach)
0.86
I
Average Standard deviation
Q
0 0 0 0 0 0 0 0 0
-4verage standard deviation Phosphate expressed a s per cent P o 4 - - - .
-0.062
-0,059 -0.108 +0.041
+0.062 +0.057 +0.031 +0.032 +0.006
0.880
1.280 1.280 1.267 1.276
1.251 1.253 1.251 1.252
0,000.5 -0.055
-0.027 -0.019 10.031
0.889 0.880 0.880 0.880 0.882 0.880 0.871 0 880
O.8G7
0.0100
-0,059
+0.035 +0.016 +0.021
+0.013
0.031R +0.004 +0.004 -0.009
f0.002
-0.002
0.000 -0.002 -0.002 +0.002 - 0 002 0.000 $0.002
0.0017
+ o . 009
-0,009 -0.001 0.000 -0.007 -0.005 +0.009 0.000
1-0.006
0.0063 +0.003 -0,001 +0.001 -0,001 +0.001 -0,003
0 0025 f O 029 f0.029 +0.020 -0.011 -0.011 -0.016 -0,014 -0,011 -0,011
0.0183 -0,001 + o , 009 0.000 0.000
0.000 +0.002 0.000 -0.009 0.000 0,0044 -0.001 + o . 001 -0.001
-0.024 0.0053
k0.023
0.001 10.0057
ANALYTICAL CHEMISTRY
1948 Table V I I .
Recovery of Calciuni from Standard Solution of Calcium Chloride (Titration eiior) Calcium, ME. -. I ounda Ilifference ~~
Added 1 00 2.00 5.00 10.00
1 00 2 00 5 02 10 04 15 04 20 40 35 61 30 70 40 80 :o 48
pink color was observed. With csperience and a good artificial light source one can become quite proficient in detecting the purple end point. I n order to avoid a titration error in routine application of the method the aliquots were a l ~ a y schosen so as to contain 15 mg. or less of calcium. DISCUSSION
+'0.'02
The anion exchange separation-ethylenediamine tetraacetate titration method is well suited to the routine analysis of all types +0.40 20.00 of biological materials. S t the Missouri Argicultural Esperi+O.Cil 26 00 ment Station during the past two years hundreds of analyses have +0.76: 30.00 been made by this method on plant materials, milk, urine, serum, + o . 80 40.00 +0.98 50.00 esperimental rations, grains, and other substances in which phosphate interferes with the titrat,ion. Each value is a n average of two indepcndent determinations The results obtained by the proposed method are equal in accuracy to those for the accepted classical oxalate method, and the procedure is much less time-consuming. I n this study the number of analyses made by each method on forages and rations known amounts of calcium. These knon-n solutioiis \vcre 1 1 i . t ~ was approximately 50. The average difference in per cent calpared so as to h a w ratios of ilhosphate to calcium from IPS th:m cium \vas found to be 0.024 for all the results by the proposed 1 to 1 to 20 t o 1. After the wlutiozis Lvere p:tPsed through t h e method as compared to the results by the oxalate method. I n an anion eschange column t,he tdcium K:IP titrated with vthylrneover-all evaluation the ethylenediamine tetraacetate values were slightly lower than the oxalate values. This average difference, diamine tetraacetate using murexide as the indicator. Thc 0.024, is nearly equal to the average st,andard deviation of the osaeffluents were also analyzed for phosphorus I I J ~ the volunietl.ic late method which was 10.023. These data show that the reA.O.A.C. method ( 1 ) and nonf, was fourid. The recovery of sults of the proposed method are in good agreement with those calcium was excellent for the solutions containing a sninll amount of the classical oxalat,e procedure. The small average standard deviation of =!=0.0057for the proposed procedure indicates that as n-ell a s a large amount of phosphate. The standard deviation the reproducibility is excellent. \vas found to be 1.0.017 for 12 samples containing 10 mg. of calIf only a small amount of sample solutioii is available the ion cium and different amounts of phosphate ranging from 6 t.o 100 nig. esrhange column can be made proportionately smaller. It is The data obtained upon analyzing various for:igP anti fecd important to keep the quantity of calcium below 20 mg. in the s:tniplr solution being titrated, as a significant titration error samples are presented in Table VI, The sampler were rhosen in o(a(urswhen attempts are made to titrate larger amounts. Large order to have a range in phosphate content, a s \ w I l ils of calcium. :mounts of potassium chloride were found to have no effect on The determinations were conducted by both t h r c l a 4 c a l oxalate thcx analysis for calcium. method ( I ) , and the proposed "ion exchange-eth~.lenecli:tmiri(~ Drj- ashing was preferred over the various wet ashings procedures for the removal of organic material in the forage samples. tetraacetate titration procedure." For samples 1, 2, 3, 4) iind -5 The dry ashing procedure is rapid and convenient. When wet three independent ashirlgs were made a t differrut times. For ashing is used the anion eschange columns become exhausted each aPhing, two or three replicate determinatiolls ~veremade 11:. more quickly as a result of the removal of sulfat,e, perchlorate, both methods. For sample 6 only one ashing \V;I.S made :ind ii :wid nitrate anions used in the n-et ashing method. As a result, more time is consumed in regenerating the resin and refilling the triplicate determination wts conducted by both mc,thods. Thcs columns. Several hundred samples of feces, foods, and urine phosphate (POa) content of the,-e samples rangrd from 0.54 to lvere \vet ashed using nitric, sulfuric, and perchloric acids. 6.48%. It was observed that the phosphate ions L V ~ I ' Peffectively FYhen aliquots of these solutions mere passed through the columns removed by the anion exchanger. The difference in the, overages for phosphate removal, a more frequent change of resin was needed as t,he aliquots contained large quantities of anions. of the per cent calcium between the oxalate and ethylene:iiamirie Such difficulties were not experienced when t'he same samples tetr:racetate methods ranged from -0.062 to +0.019%'0, \vhic.h \Yere dry ashed. When a column containing 30 grams of resin is is within the analytical tolerance usually allowed foi, thr t*I;issic:il used, a large number of aliquots can be passed before regeneration method of analysis for calcium. The precision of tlie ethylc~nt~- is necessary. diamine tetraacetat,e method was escellent and muc.tr Ilcttter LITERATURE CITED than for t.he oxalate method; however, the precision of thc os:ll:ltr Assoc. Offic. Agr. Chemists, ".\lethods of .\nalysis," 7th ed., method was also considered to be good (Table VI). 1950. Recovery experiments were made using vnrious types of forBanewie, J. J.. and Keniier. C , T., AXIL. CHEM.,24, 1186 ages, such as ladino clover, orchard grass, red clover, soyl)enns, (1952). and alfalfa. Kno\vn amounts of c:ilcium \vwe atlded as the tahloBets, J. D., and Soll, C . A . , J . A m . Wuler F o r k s Assoc.. 42, 49 (1950). ridc t.o aliquots of the original ash solution of thts aumplc~s. The Ihid., p. 749. final concentration of c:llciurn in the aliquot \viis betnecw 1 0 aiid Biedermann, W.,and Sch\v:arzent)ach. G., Chivriu ( P r a g u e ) , 2, 16 mg., a s this quantity gavr tlie smnllest titratioii t.i'i'or. The 58 (1948). solutions were adjusted t,o pH 3 to 5, t,hen passed through tlie (~heng,K. L., and Bray, I
