306
ANALYTICAL EDITION
specific gravity may serve with nearly equal accuracy as a measure of the total solids content, the use of the former proves much more convenient, as it may be determined with greater ease and speed and with only a drop or two of the sample. conc~usion,we wish to point out that it is fundam e n t a b unsound to apply this method to a study of egg yolk or whole egg, since these mixtures contain more than two major componentsa This condition makes possible an
Vol. 4, No. 3
indefinite series of mixtures all of which may have the same refractive index. LITERATURE CITED (1) Getman, “Outlines of Physical Chemistry,” 3rd ed., P. 113, Wiley, 1932. (2) Hoist and Nmquist, Hilgardiu, 6, 45 (1931).
RECEIVED January 9, 1932. Contribution 16, Division of Poultry Husbandry, Agricultural Experiment Station, University of California.
Rapid Method for Determination of Boron Suggested Procedure for Evaluation of Boron-Bearing Ores for Boron Content WILFREDW. SCOTTAND COLLEAGUES, University of Southern California, Los Angeles, Calif.
S
B y using sodium hydroxi& to precipitate the The s e n i o r a u t h o r finds that authorpublishedamethod iron and aluminum in boric acid solutions the use of ammonia, w i t h t h e f o r t h e e v a l u a t i o n of r e s u l t i n g f o r m a t i o n of amquantitatively, when using sofnol red or p-nitroboron-bearing ores in his work monium chloride, to high on“StandardMethodsof Chemiphenol indicator,at least 30 minutes are saved results and an u n c e r t a i n end cal Analysis.” The procedure over the method of using sodium carbonate. The point in the titration; and also results of approximately one hundred and J i f y that in t h e p r e s e n c e of free followed the generally accepted determinations show that the method givesvery b o r i c acid aluminum m a y be principles for the evaluation of boronatrocalcite, horocalcite, satisfactory results. The end points are good. boracite, and boron-bearing ores p r e c i p i thydroxide sodium a t e d by addition completely, of The error in Ihe determinationshas not been w i t h o u t f o r m a t i o n of the decomposed by hydrochloric acid, but was n o t a d a p t e d to greater than 0.25 per cent. The method is soluble sodium aluminate, a t the applicable to commercial boron-bearing ores such stage when approximately onehigher grades of ore now found in as crude borax, kernite, tincalconite, ulexite, third of the f r e e b o r i c a c i d California, or to concentrates. Considerable contrary opinions is neutralized. A suitable incolemanite, and boracite. exist on t h e details for a n dicator is used to ascertain the exact and rapid method for the end point of subsequent neucommercial evaluation of these ores, which induced one of tralization, before addition of the polyhydric alcohol. A the large shippers of the ore and ore concentrates to request method for the recovery of occluded boric acid has been inthe senior author to investigate the existing procedures more vestigated. generally accepted as standards. The matter being of perOF OCCLUDED BORICACID sonal interest, the senior author undertook the work and RECOVERY carried out a series of tests in his laboratory a t the University EFFECT OF IRON AND A positive error of Southern California and worked out details of the method was resulted in direct proportion to the amount of salts present. The general procedure recommended in this tested by colleagues-Arthur Zeismer, Russell G. Dressler, For gram Of (aqua) added H, W. olson, and L, K. Gates-and certain modifications required an additional titration of 25.3 ml. of 0.552 N sodium were made in the original method. Valuable information hydroxide* was obtained from commercial laboratories. We would OCCLUSION O F BORIC ACID BY I R O N AND ALUMINUM PREmention specially A. R Maas, Maas Chemical Laboratory, CIPITATES. I n a Sample Of “tinCa1” Containing 1.9 per Cent L0s Angeles; George A, Connell and K. Jacoby, Pacific Of iron and aluminum oxides and over 44 per cent Of boric Coast Borax Company, Los Angeles; George Tate, Analytical oxide, a loss of 1.15 Per cent of boric oxide occurred by OCClULaboratory, Liverpool, England; Dr. c. Ahrens and Dr, Ad. sion of the boric acid by the iron and aluminum hydroxides Gilbert, Offentlisches Chemisches Laboratorium, Hamburg, precipitated, by sodium carbonate solution washed six times; Germany; H. L. Payne, Baverstock and Payne, Los Angeles; and 1-53 Per cent by the Precipitate washed twice. It was A. Osgyani, Raymond G. Osborne Laboratory, Los Angeles; found that 02355 gram of aluminum hydroxide held 0.1922 Smith-Emery Company, Los Angeles: and Dr. K. J. Suckow, gram of boric oxide, washed twice. Further results appear Suckow Borax Mines, Calif. later. The presence of iron and aluminum in the ores necessitates GAIN OF MOISTUREI N BORATECONCENTRATES. The their removal from the solution titrated for boric acid, since necessity of keeping the representative sample in a closed these would also titrate and lead to high results. Provision container is shown by the following tests: A 5-gram sample is generally made for their removal in the accepted methods, taken from material as received was exposed to the air for but little attention is given to the recovery of occluded boric 1 hour. I n this time the sample gained 24.5 mg. of moisture. acid carried out by the precipitates. Wherry and Chapin I n 2.5 hours it had gained 66.1 mg., 01 over 1 per cent. (2) recommend the precipitation of the boric oxide group by WATER DISPLACEMENT OF CLAY RESIDUE. I n rapid addition of calcium carbonate; Connell and Jacoby use methods where the clay is allowed to settle and the solution barium carbonate; Gilbert and Tate suggest the use of is made to definite volume, an allowance is made for the sodium carbonate. Ammonium hydroxide, magnesium car- volume occupied by the residue in the flask. The senior bonate or oxide, and sodium hydroxide have been used. author found that 10.08 grams of this residue in the analysis OME years ago the senior
TABLE111. OCCLUSIONOF BORICACID BY ALUMINUM HYDROXIDEB
of ‘%tincal” occupied a space of 4.8 ml., or approximately 0.5 ml. per gram of residue.
PRECIPITATION OF AI~UMINUM HYDROXIDE BY SODIUM HYDROXIDE IN PRESENCE OF FREE BORICACID. The sample from the acid extraction of the borate concentrate was titrated in the presence of phenolphthalein indicator to a faint pink color (the polyhydric alcohol being omitted), aluminum hydroxide was completely precipitated, and no evidence of sodium aluminate in the filtrate was obtained. It was found that approximately one-third of the boric acid in the solution was neutralized a t this stage, and the free boric acid evidently prevented the formation of the soluble sodium aluminate.
RAPIDMETHODvs. CHAPIN’SDISTILLATION METHOD. Some doubt has been expressed regarding direct titration methods by the customary procedure. It is felt that the isolation of boric acid from interfering elements is necessary for accurate results. Any one who is familiar with the distillation method realizes the disadvantages of this long procedure in the evaluation of boron-bearing ores and the danger of error due to factors in the procedure and the small amounts of boron that can be determined. The comparisons given in Table I, however, are of value and interest. Better checks with the Chapin method are obtaine‘d by using mannitol and titration in the presence of sofnol indicator. TABLE I. DIRECTTITRATION METHOD vs. CHAPIN’S DISTILLATION METHOD“ DIRECTTITRATION Cram
CHAPIN’S Gram
0.2195 0.2189 0.2199
BORIC OXIDE, 0.6-GRAM SAMPLEb
PURE BORIC
ACID,
0.2221 0.2228 0.2213
With glycerpl and p-nitrophenol indicator 0.1012 0.1011 0.1011 With mannjtol and sofnol indicator 0.0996 0.0996
0.1005 0.1002
+
Data, by H. W. Olson. b In Chapin’s method, 0.5 gram of ore; in direct titration method, aliquot portion of sample equivalent to 0.5 gram. Q
The results of tests reported in Table I1 show the degree of precision in checks that can be expected by direct titration of the solution containing the borate. OF RESULTS OF DIRECTTITRATION TABLE11. COMPARISON OF SOLUTION COHTAINING BORATE
% 25.00 25.52
(0.6184-gram sample taken) SAMPLE
ALUMINUM ADDED
&Boa OBTAINED
a b
AVERAQE HnBOa OCCLUDED
%
%
%
100.00b 100.00b 99.75 99.75 99.25 99.50 98.55 98.55 97.34 97.10
0.00 0.00 0.25 0.25 0.75 0.50 1.45 1.45 2.66 2.90
0.00
% 1 None 2 None 3 0.2 4 0.2 5 0.5 6 0.5 1.0 7 1.0 8 2.0 9 2.0 10 Data by Arthur Zeismer. Standard solution.
HaBOa OCCLUDED
0.25 0.62 1.45 2.78
I n the analysis of ten synthetic samples, including two blanks, the occluded boric acid was recovered and the amount added for total percentage of boric acid. A maximum error of 0.25 per cent was obtained. Since aluminum hydroxide is a gelatinous substance, the amount present in a sample will determine the length of time required in filtering the solution in order to obtain all of the occluded boric acid. The following report of analysis of a California concentrate received at Liverpool, England, was received from the George Tate Laboratory: LotA Lot B:
SCOTT’S METHOD 42.40 42.50
7 BaOa
8 BaOs
METHOD BY TATE 42.50 42.25
An excess of acid used in the original digestion of the ore has practically no effect on results, this excess being taken care of in the procedure. The following results from the Raymond G. Osborne Laboratory supports the statement. Ten-gram samples were examined.
0.1-GRAM S A M P L E
0.0972 0.0966 0.0967
CRUDE CRUSHED ORE 30-gram samule
307
INDUSTRIAL AND ENGINEERING CHEMISTRY
July 15, 1932
AMT.OF N HCl
%
50 75 100
43.84 44.15 44.15
WATER-SOLUBLE BORATES.It may be desired to evaluate, the ore by determining its water-soluble content. The California deposits of sodium borates are largely watersoluble. The residue must be examined also if total borate is desired. The water extraction is made in a covered beaker, the reflux condenser being unnecessary.’ The following report was obtained from a local commercial laboratory. SCOTT’S
Water-soluble BaOa COARSE FINE CONCENTRATEBz01in residue Total BiOa DUMPINQS DUMPINGSCONCENTRATE AT MINE 10-gram 10-gram 10-gram 10-gram sam~le samole samDle samole
%
%
%
%
13.40 13.45
28.28 28.80
44,300 44.12 44.10 44.38
48.20 48.38
44.42b 44.54 44.25 a Data by Russell G. Dressler. b Data by H. W. Olson.
OCCLUSIONOF BORICACID. I n studying the occlusion of boric acid by aluminum hydroxide when precipitated with sodium hydroxide, which has been mentioned, it was found that definite amounts of the acid are occluded. The amount depends on the concentration of the solution and the per cent of aluminum present. From the data in Table I11 it will be seen that if the amount of aluminum present is less thqn 0.2 per cent, the occluded boric acid need not be run when the concentration of the sodium hydroxide reagent is above Q.5 normality.
BORICOXIDE
Cc.
GILBERT’S
AMETHoD B
OSBORNE’S
AMETHODB
%
%
%
46.35 1.32 47.70
46.24 1.65 47.89
.,.
...
%
...
...
46:04
48:08
48:t)S
4j:84
%
AMETHoD B
%
DETERMINATION OF TOTAL ACID-SOLUBLE BORATE The finely ground ore is boiled with hydrochloric acid in a flask connected to a reflux condenser. The solution is made up to a definite volume and portions of the filtrate taken for analysis. The iron and aluminum are first precipitated quantitatively with sodium hydroxide using p nitrophenol as the indicator. The iron and aluminum hydroxide precipitate is filtered off, and the boric acid is titrated with standard sodium hydroxide in the presence of glycerol, using phenolphthalein as the indicator. The iron and aluminum hydroxides are redissolved in 6 N hydrochloric acid and then reprecipitated with sodium hydroxide added very cautiously, and filtered to separate the occluded boric acid. The solution from the iron and alumina is now titrated with sodium hydroxide as above.
1 Even in the presence of acid practically no loss of boron occurs a t temperatures just below boiling. This confirms the results of Dodd and Scott (1).
ANALYTICAL EDITION
308
ACID EXTRACTION. The method .is applicable for the determination of total boric acid in borates of sodium, calcium, and magnesium in materials such as crude borax, tincalconite, kernite, boracite, ulexite, colemanite, etc. Silicoborates required a preliminary fusion with sodium carbonate. The acid residue should be examined for these. Iron, aluminum, ammonium salts, and substances other than boric acid should be absent from the solution if they react with sodium hydroxide. REAQENTS 1 or 0.5 N ~~-~ HCl 1 or 0.5 N NaOH (carbonate-free) 6 N HCl (acid of constant boiling point is satisfactory) ”. 50% solution of NaOH (carbonafeelfree) ~
Indicators:
T’trophenol (saturated water solution) f%iol red No. 1 Phenolphthalein (1% in 50% alcohol solution) Mannitol or glycerol Water, carbon dioxide-free
A five-gram sample, finely ground, is placed in a 250-cc. volumetric flask, 15 to 20 cc. of 6 N hydrochloric acid are added, together with an equal volume of water, and the flask connected to a reflux condenser. The mixture is heated to boiling and boiled for 20 to 25 minutes. After allowing to cool slightly, 75 cc. of water are poured into the flask through the condenser tube, the solution mixed, and then the heating and boiling repeated for 10 to 15 minutes. The flask is again allowed to cool slightly and about 50 cc. of water poured through the condenser tube into the flask. The condenser is disconnected, and the flask placed in a cold water bath and cooled.2 Water is now added to the 250-cc. mark, and if necessary an additional amount to allow for the volume occupied by the residue.8 After the residue has settled, the clear solution in the quantity desired is decanted through a dry filter into a dry, clean beaker, and portions of this filtrate taken for analysis. Fifty cubic centimeters are equivalent to 1 gram of the original sample. Of the filtrate, 50 or 100 cc. are taken, according to the strength of the reagents used, and placed in 400-cc. beakers. REMOVALOF IRONAND ALUMINUM.To the boric acid solution in a beaker are added 3 to 4 drops of sofnol red or p-nitrophenol indicator. Standard sodium hydroxide is added cautiously until the yellow color of the indicator remains. All the iron and aluminum will be precipitated a t this point. The solution is heated gently, then allowed to stand for several minutes. The iron and aluminum hydroxide are atered off and washed with hot water, and the filtrate and washings titrated for boric acid according to the procedure given under titration of boric acid below. RECOVERY OF OCCLUDED BORICACID. The iron and aluminum hydroxide carry out very appreciable quantities of boric acid when precipitated in the presence of a large excess of this acid. If the amount of these hydroxides is appreciable, the recovery of boric acid is essential for accurate results. The hydroxides are dissolved in the filter by addition of 6 N hydrochloric acid, added in sufficient excess. The acid solution is caught in a beaker together with hot washing of the filter. Three to four drops of sofnol red or p-nitrophenol are added to the solution, and sodium hydroxide (50 per cent solution) is now added dropwise until about neutral, then 0.5 N sodium hydroxide until the yellow color of tlie indicator remains. The solution is heated gently, then allowed to stand for several minutes. The iron and aluminum hy1 The addition of 2 to 5 grams of sodium chloride hastens settling of the suspended material. The salt does not interfere with the accuracy of the method. a One gram of the residue occupies a volume slightly leas than 0.5 ce. Generally the amount is EO small that this additional water is unnecessary.
Vol. 4, No. 3
droxides are filtered off, and the precipitate is washed with warm water. The filtrate is acidified with 0.5 N hydrochloric acid, the solution neutralized to sofnol red or p-nitrophenol indicator by addition of sodium hydroxide drop by drop, and the boric acid determined by titration with sodium hydroxide in the presence of mannitol or glycerol with phenolphthalein indicator. Time is saved by carrying out this titration separately from the main solution. TITRATION OF BORICACID. The solution from the precipitation of iron and aluminum is acidified with hydrochloric acid and then just neutralized cautiously with standard sodium hydroxide to the yellow color of sofnol red or p nitrophenol. About 10 drops of phenolphthalein indicator and 25 to 50 cc. of neutral glycerol or mannitol (according to the amount of boric acid titrated) are added, and then the standard alkali added until a distinct reddish pink color appears. The true end point for boric acid is p H 11. NOTE. The boric acid in the solution recovered from the iron and aluminum precipitate is determined also as directed above, and the amount added for total percentage of boric oxide.
-
1 cc. of N 0.03482 g. BSOa; or 0.9536 g . Na~BtO7.10H~O; or 0.0818 g. HaBOa 1 cc. of 0.6 fl NaOH = half above amounts
The acid extraction generally effects complete solution of the borates available in crude borate minerals. Should silicoborates be present, a fusion of the acid residue should be made with sodium carbonate flux, and the fusion examined for borates. The reflux condenser is used t o avoid loss of the boric acid by volatilization. (The 6 N hydrochloric acid is approximately acid of constant boiling point.) This is unnecessary in water extractions for water-soluble borate. The removal of iron and aluminum from the extract is necessary, as iron and aluminum salts lead t o high results for boric acid by the action of their combined acid with the standard base. Definite amounts of borate are invariably carried out by the iron and aluminum hydroxides, so that a recovery must be effected if the precipitates are present in appreciable amounts. The isoelectric point of aluminum is between pH 6 and 7.5, and it is at this point that the aluminum compound is least soluble. Some aluminum hydroxide will go into solution if too much sodium hydroxide is added to precipitate the aluminum. Therefore, care must be exercised in exactly neutralizing in the presence of a suitable indicator. The end point in the presence of phenolphthalein is a combination of the two colors yellow and pink, and is a distinct reddish pink color. The sofnol red or p-nitrophenol gives a distinct yellow color in alkaline solutions. Both indicators are colorless in acid solutions. A concentrated solution of sodium hydroxide is used t o neutralize the 6 N hydrochloric acid so as not t o increase the volume for filtering. Other strengths or concentrations may be used. Neutral glycerol is prepared as follows: To 1000 cc. of glycerol add 4 to 5 cc. of phenolphthalein and neutralize with regular sodium hydroxide reagent. Owing to the formation of acid, the color will fade, but addition of sodium hydroxide from time t o time will keep the color pink. A sharper end point is obtained with mannitol. If end points are run over, back titrations are possible with either indicator. If a carbonate is used in the precipitation of iron and aluminum, it should be removed by acidifying the solution and gently heating at a “simmering” temperature for 10 to 15 minutes in a covered beaker. Loss of boron occurs if the beaker is not covered. The presence of carbonic acid in the solution leads t o an error in the boric acid titration. The use of sodium hydroxide in place of sodium carbonate in the precipitation of iron and aluminum avoids the necessity of removing carbon dioxide from the solution. LITERATURE CITED (1) Dodd and Scott, Analyst, 54,715 (1929). (2) Wherry, E.T.,and Chapin, W. H., J . Am. Chem. Soc., 30, 1687 (1908). R ~ C E I VSeptember ~D 29, 1931.
