JOURNAL OF CHEMICAL EDUCATION
SEPARATION AND DETECTION OF THE ALUMINUM ION IN QUALITATIVE ANALYSIS1 STERLING B. SMITH and JOHN M. SHUTE Trinity College, Hartford, Connecticut
T H E use of "aluminon," the ammonium salt of aurin tricarboxylic acid, as a reagent in the qualitative test for aluminum is widespread. The purpose of this reagent is to distinguish between the gelatinous precipitate of aluminum hydroxide and a similar precipitate of hydrated silica.. When a few drops of the "aluminon" reagent are added to an acidic solution containing the aluminum ion, and the mixture made alkaline with ammonium hydroxide, a red "lake" result^.^ No such color is formed with hydrated silica. For the past 16 years we have been using in our courses in qualitative analysis in this laboratory the excellent textbook by Hogness and J ~ h n s o n ,now ~ in its fourth edition. Their confirmatory test for the
alumiuum ion is to dissolve the gelatinous precipitate suspected to be aluminum hydroxide in one milliliter of 3 M hydrochloric acid, washing the filter paper with one milliliter of water. Three drops of 3 M ammonium acetate and an equal quantity of a 0.1 per cent "aluminon" solution are then added. The solution is stirred, made slightly alkaline with 6 M ammonium hydroxide, and gently heated. A red flocculent precipitate indicates the presence of aluminum in the original soluti~n.~ The results using these directions have been most unreliable and unsatisfactory. Almost without fail if either iron or chromium ions, especially the latter, are present a false positive test for the aluminum ion is obtained. 1 The material for t,hk r naner~wan .t,aken ~from~a thesis ~ bv John This immediately suggests that the previous separaM. Shute, presented to the graduate committeeof Trinity College tion of aluminum from chromium and iron is not comin partial fulfillment of the requirements for the M.S. degree. HAMMEW, L. P., AND C. T. SOTTERY, J . Am. Chem. Soc., 47, plete. A study of this separation and the subsequent 1423 (1925). test for the aluminum ion was therefore undertaken. 8 HOGNESS AND JOHNSON, "Qualitative Analysis and Chemical 4 H~GNESS AND JOHNSON, Bid.,3rd ed., 1947, pp. 397-9. Equilibrium," Henry Holt and Ca., New Yark. ~~~~~~~
~
~
~
~
~~
~
~
~~
~~
.
~
JULY, 1955
381
Preliminary t o this study, a ample of aurin tricarboxylic acid was synthesized. This material and five This is the white precipitate that is tested in order to different commercial samples were used in carrying out distinguish aluminum hydroxide from hydrated silica. the "aluminon" test to determine if the unreliability This is done, as previously described, by dissolving the of the test might be due to any variance in the reagent precipitate in acid and adding "aluminon" reagent itself. No significant differences mere observed. and ammonium hydroxide. A thorough study was made to determine the optiAI(OH)B+ 3HC1- AICls + 3Hs0 mum conditions for the red lake formation between N&OH "aluminon" and aluminum ions. I t was concluded AlCL + "aluminan" A bright red lake that the best red lake is formed when an acidic solution of pH less than 3.5, containing both the dye and alumI t is apparent from the above that if the chromium inum ions, is reduced in acidity to pH from 5 to 7.2. were not completely oxidized by the hydrogen peroxide A study was made t o determine which of the common to the hexavalent state, it would revert to the cation metals react with "aluminon" to form colored lakes, upon acidification, interfering with the aluminum test and the minimum amount of each metal necessary for as shown in the following equations: precipitation was establiihed. Of these metals, lead, CrCb + 4NaOH NaCrOz 3NaC1 2H,O zinc, cobalt, manganese, iron, and chromium gave NsCrOz + 4HCI CrCL + NsCl + 2H20 tests similar to that of aluminum. These elements Cr(0H). + 3NH.CI CrCl, + 3NHtOH must therefore either he absent or rendered harmless before a solution is tested for aluminum. Cr(OH)*+ 3HCl- CrCL + 3H10 By increasing the amount of ammonium acetate N&OH from three drops, as directed by Hogness and Johnson, CrCla + "aluminon" A bright red lake to one milliliter, it was found that zinc, cobalt, and of the 1, the endeavor to secure complete manganese caused no interference unless the amounts chromium, excess peroxide was added to the alkapresent exceeded one millimole. Such amounts are line solution containing iron, aluminum, chromium, unlikely to be present when the "aluminon" test is No improvement ~ to Mellor6 ~ if was noted. ~ performed if the previous procedures have been per- excess peroxide is present, a partial reduction of the formed carefully. chromium is inevitable upon acidification. Small amounts of lead, chromium, and iron still satuwted bromine waterwas substituted as the oxcause interference even after the increase in added idizing agent, hi^ proved very satisfactory as far ammonium acetate. Lead, however, is extremely un- as the chromium was concerned, but iron still interlikely to be Present even in small amounts a t this fered. Experiments using varying amounts of sodium point. Small amounts of chromium and iron are hydroxide on a very dilute solution containing ferric likely to be present a t this point, and it is to remove ions showed t h t all the iron was precipitated and these interfering ions that an alternate procedure has that the more sodium hydroxide used, the greater was been devised. the amount of unprecipitated iron. According to work that exThe procedure in Hogness and Johnson is t o treat recorded in ~ ~ lG. C. l wittstein ~ ~ ,believed ~ a mixture of the chlorides of ferric iron, aluminum, cess sodium hydroxide dissolves a minute amount of and chromium with hydrogen peroxide and excess ferric hydroxide, but L. Schaffner claims that the ferric sodium hydroxide: hydroxide is in colloidal suspension. FeC1..+ 3NaOH + Fe(0H)r 3NaCI Since it is necessarv to have excess sodium hvdroxide . . .present to dissolve the amphoteric hydroxides, small MCI8 + 4NaOH + NaAlOl 3NaC1 2HsO amounts of iron must necessarily remain in the filtrate 3Hz02 2Na~CrOl 6NaCI 8HzO 2CrC13 + lONaOH containing supposedly only aluminum and chromium. Thus, insoluble ferric hydroxide is obtained which In order to prevent this iron from later precipitation may be filtered off and its separation effected. The along with the aluminum hydroxide, it seemed desirable amphoteric elements remain in solution in the excess to tie up the iron by complex-ion formation. The common complexing agents, such as oxalate and of strong base and a t the same time the chromium is oxidized to the hexavalent state. When this filtrate is tartrate, were of no value since they form complexes acidified, the aluminum returns to its cation state and with aluminum as well as with iron. A search for a chelating agent for iron which did not unite with aluthe chromium remains in the complex anion. minum led to the use of the monosodium salt of N, NeAIO, + 4HC1- MClr + NaCl 2HsO Ndihvdroxvethvl glvcine (sold in solution under the
--
+
-
+
+
+
+
+
+
+
+
NatCrO,
+ 2HC1-
HpCrO,
+ 2NaC1
Addition of slight excess of NH40H t o this solution produces the white gelatinous precipitate of aluminum hydroxide, leaving chromium in solution. ALCL
+ 3NH4OH
-
AI(OH),
+ N&CI
6 MELLOR, J. W., "A Comprehensive Treatise on Inorganic and Theoretical Chemistry," Longmans, Green & Co., New York, 1922, Vol. 1, p. 944. 6 MELLOR, J. W., ''A Comprehensive Treatise on Inorganic and Theoretical Chemistry," Longmans, Green & Co.. New York, 1934, Vol. 13. p. 372.
~
JOURNAL OF CHEMICAL EDUCATION
382
trade name of Versene Fe-3 Specific).' Experiments 3 M ammonium acetate and throe drops of "aluminon" reagent. using Versene Fe-3 Specific showed that relatively Mix thoroughly. Now make the solution slightly alkaline to with 6 M ammonium hydroxide. Heat the mixture in a large amounts of iron are chelated, whereas the amount litmus water bath for from two to three minutes. A bright red floecuof chelated aluminum is not appreciable. For effective lent preoipitate shows the presence of aluminum. chelation the pH should be below 12. Heat as well as AI(0H)s 3HC1- AlCls 3Hz0 time adversely affects chelation. NHlOA As a result of these experiments, the following proAlClr + "aluminon" bright red lake cedure is suggested for the separation of aluminum Experiments were conducted using the above profrom iron and chromium and the subsequent use of cedure to determine the maximum amounts of iron "aluminon" reagent. (alone), chromium (alone), and iron and chromium PROCEDURE together which can be present without interfering with Diasolve the precipitate containing the hydroxides of ferric the aluminum test. The minimum amount of alumiiron, aluminum, and chromium by distributing drop by drop one num which could be detected in their presence following milliliter of 3 M hydrochloric acid over the filter paper on whioh the above procedure was determined. I t was shown it wae colleoted. Wash the filter paper with one milliliter of that the concentration of aluminum ion normally water, adding the wash liquid to the filtrate. present in an unknown solution for semimicro qualiFe(OH)$ + 3HCI FeCls + 3Hz0 tative analysis far exceeds the minimum amount which AI(OH), 3HCl- AIC1, + 3HsO can be detected. At the same time the maximum amount of iron and chromium which can be present Cr(OH)r 3HC1- CrCls 3Hn0 without interference is considerably greater than that To this solution add two milliliters of 6 M sodium hydroxide and two milliliters of saturated bromine water. Stir well. Heat normally present in semimicro work. The amount of elements present in a solution taken in a beaker of boiling water for three to four minutes. Filter the solution, retaining any precipitate for the test for iron. for semimicro analysis is usually 6 X moles. This corresponds to three milliliters of a solution which FeCI. 3NaOH Fe(OH), + 3NaCI is 0.02 M in concentration of each ion present. HowAlC4 4NsOH NaAIOl 3NaCI + 2Hz0 ever, the concentrations may vary from one-fifth t o twice this amount and still be properly separated and detected. Carefully neutralize the filtrate (whioh must be clear or reDuring the course of the investigation, 121 separafiltered until it is so) with 12 M hydrochloric mid and add one tions were satisfactorily made following the above prodrop in excess. cedure. As a final check, 20 solutions, each three NaAlOl + 4HC1- AICla + NaCl + 2H,O milliliters in volume were prepared, all of which were 0.02 M in concentration of iron and chromium and nine Na,CrO, + 2HC1- HaCrO, + 2NaCI Add five drops of Versene Re-3 Specific (a 10 per cent solution of which were also 0.003 M in concentration of alumiby volume of the commercial solution) and stir thoroughly. num. These were given to an analyst who did not Make the solution alkaline with 6 M ammonium hydroxide; know which solutions or how many contained alumadd two drops in excess, and stir. Filter this solution, retaining inum. Following the above procedure he correctly the filtrate for the test for chromium. determined the presence of aluminum in the nine cases AI(OHh 3NHCI AICls 3NH,OH and its absence in the eleven. I n conclusion, it may be pointed out that when this HlCrO, 2NH40H (NH,),CrOl 2H,O procedure was followed by a class of 22 freshmen taking Wash the precipitate on the filter paper with two one-milliliter of water, these Dissolve the pre- qualitative analysis, only one man reported aluminum cipitate on the filter paper with one milliliter of 3 M hydrochloric to be present when the unknown contained chromium acid and wash the paper with one milliliter of water, adding the but no aluminum. This suggests that the test showing washings to the filtrate. To this solution add one milliliter of the presence or absence of following this Manufactured and sold b~ the Bersworth Chemical co.. procedure may be regarded to be as reliable as the Framinghsm, Mass. ' test for other cations.
+
+ +
+
-
+
+
-
+ +
--
+
+
+
-
+
