OXIDATION i'.îETHODS FOR THE VOLUMETRIC DETERMINATION OF HYP0PHÜ5PHATE, PHOSPHITE, AND HïPOFHOSPHITE By Stanley J. Carlyon A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. OXIDATION METHODS FOR THE V0LUI-4ETRIG DETERMINATION OF HYPOPHOSPHATE, PHOSPHITE, AI'D HYPOPHOSPHITE By Stanley J. Carlyon AN ABSTRACT Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry Year 1953 Approved Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stanley d. Carlyon THISIS IBSTrtAGT Volnmetrlc oxidation methods were developed for tlie determination of hypopnosphste J phosphite and hypophosptilte using standard solutions of sodium hypochlorite, potassium dlchromate, and cerlc ammonium sulfate under suitable conditions. The compounds used for the anal;yses were dlsodlum dlhydrogen liypophosphate hexaliydra.te which was prepared by the method of Lelnlnger and Chulskl,^ reagent grade phosphorous acid, and purified sodium hypophosphlte. The compositions of the phosphorous acid end sodium hypophosphlte were established by a gravimetric procedure In which the compounds were oxidized to orthophosphate by repeated evaporations with aqua, regia, and the orthophosphate converted to mag­ nesium pyrophosphate. iiypophosphate , phosphite, and hypophosphlte were quantitatively oxidized to phosphate by an excess of standard potassium dlchromate In 12 normal sulfuric acid at the temperature of a. boiling water bath. The oxidations were complete In one hour. The excess dlchromate was determined lodometrlcally after adjusting the sulfuric acid concentration to approximately three normal by the addition of sodium hydroxide solu­ tion. The excess dlclirornate was also determined by adding an excess of standard ferrous anmionlum sulfate solution and backtltratlng the excess ferrous Ion with standard dlchromate . It was found that there was a loss of 0.01 to 0.03 ml. of 0.1 normal dlchromate during the oxidations and corrections were applied to compensate for this loss. 1 Lelnlnger, Z ., end Ghulskl, T., J. Am. Cliem. Soc ., 7 1 , 238$ (I9k9). — 1 — Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stanley J. Carlyon iiypophosphate J phosphite ^ ano Irypophospliite were quantitatively to phosphate uno.er sue.tsole conditions by an excess of standard sodiuiri hypochlorite , Hypopliosphate was determined vjith an excess of standard sodium hypochlorite in a solution made neutral >ri.th sodium bicarbonate . The excess hypochlorite was determined after jO minutes by adding an excess of standard sodium arsenite and beckti.trating the excess arsenite with standard sodium hypochlorite using Bordeaux as indicator. Tiie excess hypochlorite was also determined by titrations with standard arsenite in wïrich the endpoint was determined either potentiometrically or by the deadstop technique. The iodometric de­ termination of hypochlorite in the presence of bicarbonate or buffers gave unsatisfactory results. Phosphite was determined by a sligiit excess of standard sodium hypochlorite in a solution containing bicarbonate and bromide, or essentially by hypobromite. The oxidation was sufficiently rapid under these conditions tha,t a direct titration of the phosphite with hypo­ chlorite was possible if the equivalence point in the titration was determined by the deadstop technique. The use of Bordeaux as indicator in this titration was not satisfactory. HypophospMte was only very slowly oxidized by an excess of sodium hypochlorite in neutral solution but was quantitatively oxidized to phosphate by excess hypochlorite in one normal sulfuric acid in ten hours. Phosphite and hypophospliite were quantitatively oxidized to phos­ phate by an excess of standard eerie ammonium sulfate in dilute sulfuric - 2 - Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stanley J. Carlyon acid solution in one hour at boiling temperature. incompletely oxidioed under these conditions, Hypopliosphate was à small loss of eerie suH-fate during the oxidations required the use of blank eorrections . -3Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 AGKNŒ'JLil.DŒ'iENT The writer wishes to express his sincere appreciation to Doctor Elmer Leininger for his most helpfxil counsel and guidance dur­ ing the course of this work. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ill T;aLE OF C01\iTi!;i\iTS PÎGE I . B3TR0DUCTI0N............................... II. III. IV. V. HISTORICAL................................................ 3 itiiFiGENTS AND STANDiUtD SOLUTIONS........................... 16 A , Preparation of Standard Solutions.................. B. Standardization of Solutions ...... .............. G. Indicators......................................... 16 Ifa 19 DETERiilNATION OF HYPOPHOSPHiVfE............................ 21 A. Oxidation of Hypophosphate by Excess Potassium Diciiromate in Sulfuric Acid Solution............ 3. Oxidation of Hypopliosphate By Excess Sodium Hypochlorite..................................... 0, Oxidation of Hypopliosphate With Sodium Hypochlorite in a Solution Containing Bromide....... D . Oxidation of iîÿpophosphate by Excess Ceric Sulfate in Sulfuric Acid Solution....................... Summary of Results of Hypophosphate Determination... 32 33 DETERHIiLlTION OF PHOSPHITE................................ 37 A. Oxidation of Phosphite By Excess Potassium Bichrom­ ate in 5u].furic Acid Solution................... B. Oxidation of Phospliite by Excess Hypochlorite in the Presence of Bromide.............. C. Oxidation of Phosphite vrith Sodium Hypochlorite D. Oxidation of Phosphite By Excess Ceric Sulfate in Sulfuric Acid Solution.......................... Summary of Volumetric Methods for the Determination of Phosphite..................................... VI. 1 DETERMINATION OF HYPOPHOSPHITE............................ A. Oxidation of Hypophosphlte by Excess Bichromate in Sulfuric Ac id Solution....................... B. Oxidation of Hypophospliite by Excess Sodium Hypochlorite in a BicarbonateMedium............. C. Oxidation of Hypophospliite by Excess Hypochlorite in the Presence of Bromide...................... D . Oxidation of Hypophospliite by Excess Hypochlorite in Sulfuric Acid Solution....................... Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21 32 U? 60 6U 69 76 60 62 83 8? 87 68 IV Tj-iBLE OF CONTENTS - Continued PAGE E. Oxidation of iiypophospliite by Excess Ceric Sulfate in Nitric i cid Solution........................... F , Oxidation of Hypophospliite By Excess Ceric Sulfate in Sulfuric Acid Solution......................... Summary of Volumetric Methods for the Determination of Hypophosphite ................................. VII. VIII. DISCUSSION................................................ SUfilARY................................................. 91 93 97 99 106 LITliltATURE CITED...................................................108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 I. INTRODUCTION Phosphorus forms a series of oxy scids and salts of these acids. The first tiiree members of this series include hypophosphorous ^ Ji, phosphorous acid, and hypophosphoric acid in wliich the oxidation numbers of phosphorus are positive one, positive tliree, and positive four respectively. Some of the properties of these acids and their salts such as their physical properties, molecular structure, stability, acid strength, and preparation have been fairly well established but accurate rapid methods for their determination are not available. Better analyti­ cal methods for the determination of the members of this series would assist greatly in further studies made of the series . Theoretically, phosphorous and hypophosphorous acids and their salts are powerful reducing agents; however, reactions of these acids and their salts with even the strongest oxidants are slow, and in many cases require several hours for completion. The slow oxidation of hypophos­ phate by strong oxidants has been explained by the presence of P-P bonds in its structure and in some cases by the time required for its hydrolysis to phosphite and phosphate before the oxidation can proceed. An equi­ librium between tautomeric forms of the compound has been suggested as the reason for the slow oxidation of phosphite and hypophosphite. An appreciable amount of work has been done on the development of analytical methods for hypophosphite largely because of its use in medicine. The literature describes the work of several investigators Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. who proposed methods for the determination of phosphite out very little has been published on the determination of 1ijpop ho sph ate . Much of what has been written on the determination of these acids, especially on the first two members of the series, is often misleading and has been subjected to criticism by other investigators. There has been much disagreement concerning the ejxperimental. conditions for carrying out many of the methods which have been proposed. Some of the methods have been based on materials which -were considered to be primary standards and in many cases it is very doubtful that the material could be con­ sidered as a pure compound ivith a definite composition. Chulski (12) made a limited study of the oxidation of hypophosphate to phosphate by sodium hypochlorite and by potassium die tiromate in sulfuric acid solution. The results indicated that ttie oxidations may be made quantitative under ttie proper conditions. It was considered that the use of these oxidants might be extended to the determination of the other reducing acids of phosphorus. The purpose of this work is to develop new volumetric methods for the determination of hypophosphate, phosphite, and hypophosphite with emphasis on the use of potassium diciiromate and sodium hypochlorite. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3 I I . iilSTORICAL In aqueous solution hypophosphoric acid, is tetrabasic. The most common salts are of the type iihgPgOg, but salts in which the other hydrogens are replaced by metals ere known. The thorium salt, ThPgOg, and normal silver salt, Ag^PgOg, are practically insoluble, the former even in strong mineral acids. The disodium salt is sparingly soluble and is the usual form in whicii hypophosphate is separated from other phosphorus salts . Aqueous solutions of the acid and its acidified salts disproportionate according to the following equation: H4P2O6 + HgO -- » HaPOa + HaPO^ That hypophosphoric acid is a dimer rather than the monomer, H PO , 3 3 follows from cryoscopic data on the acid and its salts, the.diama.gnetic properties of a number of its salts, and the crystal structure of the ammonium salt (iil) . Indications that the structure involves a P—P bond include the fact that the acid and its salts are formed only from materials containing this bond, and the remarkable resistance of hypopiiospliate to oxidation. Their chemical behavior characterizes hyp op ho sph at es a.s intermediate between phosphites and phosphates; they reduce only the strongest oxidizing agents but not the salts of the noble metals. They are not oxidized by halogens but are oxidized by permanganate and diciiromate and by these probably only £is the acid is liydrolyzed to piiospliorous and phosphoric acids (^9). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. h The couples (I|l) H4P 3O 6 + 2HaO ^ 2H 3PO4 + 2H"*" + 2e"; = ca. 0.8 V and SHaPOa ^ H4P 3O 6 + 2ii'^ + 2e“ ; E° = ca. - 0 .U V einphssize the fact that even powerful reducing agents do not convert phosphoric acid to hypophosphoric acid and that treatment of phosphorous acid with an oxidizing agent sufficiently strong to give hypophosphoric acid vjill give phosphoric acid instead. lost and Russell (72) state that hypophosphate is not oxidized by boiling dlchromate. Ho>/ever, this may mean that hypophosphate itself is not oxidized by hot diciiromate but that in acid solution it is oxidized to phosphate through hydrolysis and oxidation of the phosphorous acid. Solutions of the salts of hypophosphoric acid are much more stable to decomposition than are those of the acid, and the rate of decomposi­ tion of the acid alone is too slow to be measured at room temperature (72). Solutions of the acid and its salts are unstable at higher temperature. Methods have been proposed for the determination of hypophosphate based on oxidation to phosphate or on precipitation of slightly soluble salts. Salzer (53) determined hypophosphate by direct titration with per­ manganate at boiling temperature but satisfactory results could be obtained only if the permanganate solution was standardized against pure disodium dihydro gen hypophosphate. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mo If and Jtmg (70 ) determined hypophosphate in the presence of phosphite and phosphate by precipitating silver hypophosphate from a solution adjusted to a pH of one to two. The excess silver was de­ termined in the filtrate after tlie precipitate was removed. Probst (U?) dissolved the precipitated silver salt in ammonium hydroxide and con­ verted it to silver chloride. Treadwell and Schwarzenbach (66) precipitated hypophosphate as uranous hypophosphate, with a standard uranous sulfate solution in ah atmosphere of carbon dioxide. The endpoint was determined electrometric ally. Wolf and co-workers (?l) made a direct titration of hypophosphate i-Tith silver nitrate in a. solution buffered with disodium phosphate. Exclusion of air was not necessary. Grundmann and Hellmich (20) stated that the indirect determination of hypophosphoric acid as the silver salt is unsatisfactory in the presence ox phosphate but can be determined in the presence of phosphate and phosphite by a potentiometric titration in a solution buffered with sodium acetate using a silver iodide indicator electrode . Van Name and Huff (6?) hydrolyzed hypophosphoric acid to phos­ phorous and phosphoric acids by repeated evaporations with hydrochloric acid and determined the phosphorous acid with excess iodine in a solution buffered with disodium phosphate. Blaser and Halpern (S) found that sodium hypophosphate was oxidized to pyrophosphate by bromine: Na^PgOg + Br g + H gO N a.gHgp gO ^ + 2NaBr Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The oxidation was rapid and quantitative at a pH of 8 ^ >?hile outside the limits of pH 5 to 11 the reaction was scarcely detectable, Moeller and Quinty (U2) oxidized hypophosphate to phosphate with excess ceric ammonium nitrate . The cerium phosphates were kept in solution by boiling in nitric acid. The excess cerium was determined by a potentiometric titration with arsenite. They found that no blanks were necessary and also that the disodium dihydrogen hypophosphate hexaliydrate prepared by the method of Leininger and Chulski (37) was a primary standard, Meta and pyro phosphorous acids and their salts are obtainable but the ortho acid is the only one which is inport ant. The meta. and pyro acids hydrate rapidly to the ortho acid in aqueous solution (IfL) . The ortho acid behaves as a dibasic acid, wliich suggests that one hydrogen atom is covalently bonded to the phosphorus atom. Only tvjo series of salts are kno>m, MUsPOg and MgiiPOs, but two series of esters with structures P(0R )3 and RPO(OR)g suggest the tautomeric equilibrium: P(OH)a -- » HPO(OH)a Mitchell (Uo) assumed the above tautomeric forms of phosphorous acid in proposing a mechanism for the oxidation of the acid by iodine. The couples (Ip.) H 3PO3 + HgO H3PO4 + + 2e"; = 0.20 V in acidic solution, and HPOa"^ in alkaline + 30H“ + 2H 2O + 2e“ ; E° = 1.0^ V solution, indicate that phosphorous acid and phosphites are Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. strong reducing agents. However, reactions with oxidizing agents are of ten slow, particularly at ordinary temperature , for example those with the halogens (UO) and dlchromate. Phosphoric a.cid or phosphate is the oxidation product, Hypophosphorous acid, H3PO2 , behaves as a monobasic acid in aqueous solution and ttiis suggests that only one hydrogen atom isattached an oxygen atom. the acid In aqueous solution The solutions decompose aoove lUO^' C. is to moderatelystrong(72). Jenkens and Jones (23) found that a 0.05 molar solution of C . P , sodium hypophosphite stored at room temperature at a pH of five decomposed to the extent of 65 per cent in one year and 0.8 per cent in one year at a pH of 1.5. As shown by the couples (Ul) HaPOs + HgO HgPOs + + 2e” ; E° = 0.5? V and P + 2H 3O H3PO2 + + e” ; = 0 .29 V in acidic solution, and the couples HgPOa" + 30H“ ^2^ HPOa"^ + 2HgO + 2e“ ; E° = 1.65 V . and P + 20H" H 2PO3" in alkaline solution, reducing agents. + e” ; = 1.82 V hypophosphorous acid and its salts are strong However, as in thecase of phosphites, many oxidizing agents is very slow. oxidation by The reactions with the halogens proceed at a measurable rate (39,19) . Because of the reducing power of phosphorous acid, strong oxidants convert hypophosphites to phosphates rather than to phosphites. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 Methods have been-proposed for the determination of phosphorous and hypophospiiorous acids and their salts based on their oxidation to higher oxidation states. Two gravimetric methods (65) which have been used are (a) oxidizing to orthophosphate by repeated evaporations with nitric acid and determining the phosphate as magnesium pyrophosphate and (b) oxidizing to phosphate by excess mercuric chloride and weighing the mercurous ctxloride formed, Sandri (5L), and Schwicker (5?) modified the mercuric cliloride procedure by determining the mercurous ciiloride formed with a standard potassium bromate solution using methyl orange as indicator. Bond ( ) stated that the mercuric chloride method gave 6 results wlrLch were only 8 0 per cent of theoretical but that the determination as magnesium pyrophosphate is satisfactory, Rosenlieim and Pinsker ( 5 0 ) determined phosphite and hypophospliite by excess potassium permanganate in the presence of sulfuric acid heated to 80 to 90° C. Zivy i l h) recommended the addition of the phos­ phite or hypophospliite sample to a mixture of sulfuric acid^ manganous sulfate and excess standard permanganate. for 2 5 The mixture was then refluxed minutes, an excess of st^dard oxalic acid added, and the excess oxalate ba.cktitra.ted with permanganate. Kolthoff (28) pointed out the errors from the instability of permanganate in hot solution and proposed oxidation in cold sulfuric acid solution with blank experiments to compensate for permanganate decomposition. He recommended 2li hours standing for the oxidation of hypophosphite (28), and two hours standing for phosphite (2?) . Koszegi (36) has indicated that errors exceeding one to four per cent are possible in this method when used for hypophos­ phite. He recommended (35) the use of permanganate in neutral solution Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9 at boiling temperature followed, by an iodometric determination of the excess permanganate after filtering. Schweiclcer (5?) used an excess permanganate in sulfuric acid solution and a 30 minute heating period. He stated that the rate of oxidation is greatly increased by the addition of a few drops of a five per cent ammonium molybdate solution. Pound (h6) reported that the oxidation of hypophosphite in neutral or acid solution by direct titration iirith permanganate is impossible and that even after several days standing vriLth excess permanganate in acid solution oxidation does not exceed 96 per cent. He stated that a small amount of potassium bromide has a cataiytic effect, possibly due to the forrriation of free bromine, and tha.t the oxidation is complete, in a. few hours at room temperature if the bromide is present. Gall and Ditt (16) determined phosphite and hy%;op ho sphi te vjith an excess of standard manganate solution at elevated temperature in alkaline solution. Stamm (63) suggested the use of a.n excess of alkaline permanganate for the oxidation of phosphite and hypophosphite. The permanganate was reduced to manganate and precipitated by barium ion. The oxidation was reported to be complete in five minutes. The quantitative oxidation of phosphite and hypophosphite by the halogens under different conditions has been proposed by many investi­ gators , Rupp and Finck (5l) appear to be the first to suggest the use of iodine for the determination of phosphite . The oxidation was carried out in a bicarbonate medium and the excess iodine titrated with sodium Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10 thiosulfate after the mixture had been standing for two hours . Boyer and Bauzil (6) used a similar procedure for phosphite and determined h^-pophospliite by oxidizing it to phosphite by excess iodine in sulfuric acid solution. The oxidation was completed to phosphate by excess iodine in neutral solution. The procedure required eight to ten hours. Wolf and Jung (70) pointed out inconsistencies in the method of Hupp and Finck (5l) and developed a method of determining phosphite and hypophosphite separately or in the presence of each other . They determined phosphite by an excess of standard iodine solution in a bicarbonate medium saturated %\rith carbon dioxide. The excess iodine was determined by a titration with standard sodium arsenite after minutes. Hypophosphite was determined by treating it >7ith an excess of standard iodine solution in dilute sulfuric acid, if ter 10 to 12 hours the solution was made neutral witli sodium bicarbonate and the oxidation completed to phosphate. Haguet and Pinte (U8) used a wann borax solu­ tion in place of the bicarbonate. Bond (?) stated that this method gave results for hypophosphite which were 20 per cent low. Schwicker (56) recommended a solution of ammonium borate for the buffer. Kamecki (25) found that the rate of the oxidation of hypophosphite to phosphite by excess iodine was increased by increasing sulfuric acid concentration but was practically constant for one to four normal acid. He recommended the oxidation of hypophosphite by iodine in one to two normal sulfuric acid, and that the excess iodine be determined after three hours. He claimed that phosphite did not interfere. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 Jones and Swift C2U) modified the method of Wolf and Jung (70) and proposed a method for the determination of hypophosphorous acid and phosphorous acid alone or in the presence of each other. They pointed out that the data of Wolf and Jung showed results for the determination of hypophospliite wliich were O.U per cent lower than the corresponding gravimetric results and also that phosphite interferes in the method proposed by Kamecki. They developed the method on the assumptions that (a) phosphorous acid can be oxidized quantitatively by excess iodine in a neutral solution in one hour and that the oxidation of h^piophosphorous acid under these conditions is negligible and (b) hypophosphorous acid can be oxidized quantitatively by excess iodine in one to two normal hydrochloric acid. They found the assumptions to be valid if the oxida­ tion of the phosphorous acid was carried out in a solution buffered to a, pH of approximately 7.3 by a phosphate buffer and if the oxidation of hypophosphorous to phosphorous acid was carried out in about 1.5 molar hydrochloric acid. The results compared favorably with analyses made from neutralization titrations on the acids using a glass, electrode pH meter. Vermeil (68) suggested a micro determination of phosphite by excess iodine in ammonium borate buffer, and determined the excess iodine spectrophotometrically. Manchot and Steiniia.user (36) determined phosphite with an excess of standard bromine solution in a bicarbonate medium. Hypophosphite was determined in a similar manner using sodium acetate in place of the bicarbonate and heating the mixture . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12 Brukl and Be hr (lO) determined phosphite and hypophosphite byoxidation to phosphate with an excess of standard iodic acid solution. The procedure was completed by boiling off the iodine which was liberated in the oxidations and determining the excess iodic acid. An alternate procedure was carried out by collecting the iodine wliich was distilled off and determining i t . Rupp and Kroll (^2), and Schwicker (^7) recommended the use of an excess standard bromate-bromide solution for the determination of hypophosphite. Komarowski and co-workers (3U) used an excess chloramine-T in dilute sulfuric acid for the oxidation of hypophosphite to phosphite. The oxidation required 2li hours and the results were somewhat lo-wer than the results obtained by the method of Rupp and Finck (^l). Benrath and Ruland (3) reported that hypophosphorous acid is oxidized to phosphorous acid at boiling temperature idien titrated -wi'th ceric sulfate. No other conditions and no data were given. Cocking and Kettle (13) separated phosphite from hypophosphite by removing the former with lead acetate and determined the hypophosphite on an aliquot portion of the supernatant solution with excess one normal potassium diciiromate in approximately five normal sulfuric acid. The oxidation with diciiromate was carried out by lieating the mixture in a water bath for one hour. metrically. The excess dichrornate was determined iodo- Pelizza and Risso (U3) used an excess of 0.1 normal di­ chr ornate in a similar procedure carried out in five to six r^ormal sulfuric acid in which the-mixture was refluxed for one hour. This procedure Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13 and the one described by Cocking and Kettle are someidiat limited in that only very small samples of the hypophosphite solution were used. Kirson (26) found that the oxidation of phosphorous acid to phosphoric acid by chromate ion is about twice as fast in the presence of perchloric acid as in the presence of sulfuric acid and that the reaction is faster with increasing acidity. Kolthoff (27) determined phosphite VTith excess standard liypobromite in bicarbonate medium. The mixture was allowed to stand for one-half to one hour and the excess hypobromite determined iodometrically. Schwicker (^7) oxidized phosphite by an excess of standard sodium hypoctilorite in a, solution made neutral with sodium bicarbonate. After the mixture had been standing for ten minutes, an excess of sodium bicarbonate was added and the excess hypochlorite determined by a titration vri-th standard potassium iodide solution. Sodium hypochlorite is a highly reactive substance but many investi­ gators have found that hypochlorite solutions are quite stable if they ere properly stored. It has many advantages as a titrimetric reagent but its use has been somewhat limited in comparison to many of the standard oxidants. Jellinek and Kresteff (2l) described the use of a sodium hypochlorite solution as a valuable volumetric aid. They found that a 0.1351 normal solution changed to 0.1330 normal in 17 days. They attributed this loss to the fact that they made no attempt to shield the solution from sunlight, Jellinek and Kulm (22) prepared a standard sodium hypochlorite solution and found that its concentration remained practically constant for several weeks. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lit Belcher (l) found that a sodium hypochlorite solution prepared from commercial, sodium liypoclilorite was stable for a week, Goldstone and Jacobs (18) recommended the use of Clorox for the preparation of a standard sodium hypochlorite solution. They added sufficient sodium hydroxide to the solution to bring the pH to about 12.5, the pH of greatest stability. They found that a 0.00^000 normal solution stored in a brown bottle at room temperature changed to O.OOÜ4.977 normal in S6 days . Chapin (ll) studied the effect of hydrogen ion concentration on the decomposition of hypohal.ites and concluded that aside from notable stability in strongly acid solution in the absence of halide ions the stability of dilute sodium hypochlorite solutions was greatest at a pH of 13.1 and least at a pH of 6.7. Chopin also found that the decomposi­ tion of sodium hypochlorite solutions, at least over certain ranges of pH, was notably accelerated by acetate, borate, and carbonate and that even phosphate seemed to exert some effect. It seems to be well established that the stability of hypochlorite solutions is greater if they are stored away from sunlight (U,17,lU). Kolthoff and Stenger (33) proposed the use of a calcium hypochlorite solution as a standard oxidizing agent and found that the solution could be kept in dark bottles for a long time without appreciable change in titer, A disadvantage in the use of calcium hypochlorite as compared to sodium hypochlorite is the formation of a turbidity on standing as the result of the precipitation of calcium carbonate. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15 in many cases hypobromite reacts faster than hypochlorite but hyipobromite solutions are less stable. A hypobromite solution of knoivn concentration can be prepared extemporaneously- by adding an excess of alkali bromide to a standard hypochlorite solution: OCl" + Br" --> OBr” 4- Cl“ The reagent is particularly useful for oxidations which require a. neutral or alkaline medium. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16 III. REAGENTS AND Sl'AND/iRD SOLUTIONS ivll reagents used in this work were, unless otherwise stated, analytical reagent grade. The volumetric equipment used was calibrated Kimble Normax and corrections were applied vdiere necessary. All standard solutions which were prepared from weighed amounts of reagents were adjusted to volume in a Kimble N'oxmiaoc volumetric flask wliich was certified by the National Bureau of Standards. All weighings were made with calibrated weights. A. Preparation of Standard Solutions Potassium diciiromate, 0.1000 normal, was prepared from i'-iallinckrodt, Primary Standard Analytical Reagent which had been dried at 160° C. for three hours. The solution contained U.9035 grams of the reagent per liter of solution. It was found to be equivalent to a 0.1000 normal solution prepared from National Bureau of Standards reagent (N.B.S. number 136) in a titration of a ferrous ammonium sulfate solution. Sodium arsenite, 0.1000 normal, was prepared from Baker and Adamson, Primary Standard Analytical Reagent arsenious oxide which had been dried at 110° C. for two hours. To grams of the dry reagent was added approximately eight grams of sodium hydroxide pellets and 20 ml, of water. "When the mixture was dissolved it was diluted with water to 100 m l ., made sli^itly acid to litmus with sulfuric acid, approximately Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17 three grams of sodium bicarbonate added, and diluted to one liter. It was found to be equivalent to a 0.1000 normal solution prepared from liational Bureau of Standards reagent arsenious oxide . Sodium hypochlorite, 0.1 normal, was prepared by diluting approxi­ mately ml. of Clorox (Clorox Chemical Company), 5.2^ per cent sodium lijypoclilorite by weight, to one liter. The solution was stored in an amber bottle and kept out of direct light as much as possible, Ceric ammonium sulfate, 0.1 normal, was prepared by dissolving ?0 grams of reagent grade ammonium tetrasulfato cerate (G. Frederick Smith Chemical Company) in 1^00 milliliters of water containing 28 ml. of concentrated sulfuric acid and slowly diluting the solution to one liter, The solution was allowed to stand for at least 2li hours and was filtered m t h a sintered glass crucible before it was standardized. Iodine solution, 0.1 normal, was prepared by dissolving 6.5 grams of reagent grade iodine and 20 grams of reagent grade potassium iodide in 30 m l . of water and diluting to one liter. Sodium tliiosulf ate, 0,1 normal, was prepared by dissolving 25 grams of reagent grade sodium thiosulfate pentahydrate and 0.5 gram of sodium carbonate in water and diluting to one liter . The solution was allowed to stand for at least 2h hours before it was standardized. Ferrous ammonium sulfate, 0.1 normal, was prepared by dissolving 39 grams of reagent grade ferrous ammonium sulfate hexaliydrate in 200 ml. of water containing UO ml. of six normal sulfuric acid and diluting to one liter. M Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18 D. Standardization of Solutions The solutions used in the experimental work described in this thesis were standardized according to the following procedures unless the procedure was changed for a specific purpose in which case the procedure is described in the section where the solution was used. Sodium thiosulfate, 0.1 normal^ was standardized by the following procedure (bh). To 25.00 ml. of 0.1000 normal potassium bichromate solu­ tion in a 500 ml. Srlenmeyer flask was added 100 ml. of two normal sulfuric acid . Ipproximately two grams of powdered sodium carbonate was added in small portions with constant swirling, and then about six grams of potassium iodide dissolved in ten ml. of water was added. The flask was stoppered, allowed to stand for about eight minutes, the contents diluted to about 350 ml., end the liberated iodine titrated vbLth the sodium thiosulfate solution. Four lul. of 0,5 per cent freshly prepared starch solution was added near the equivalence point. Ferrous ammonium sulfate, 0.1 normal, was standardized by the follow­ ing procedure (30) . To 25.00 ml. of the ferrous ammonium sulfate solution in a 500 ml. Erlenrneyer flask were added 20 ml. of 1:5 sulfuric acid, five rnl. of 85 per cent phosphoric acid, and six drops of 0.01 molar sodium diphenylamine sulfonate indicator. The solution was titrated with 0.1000 normal potassium bichromate to the appearance of the first tinge of purple or violet-blue. /n indicator blank of 0 .05 ml. of 0,1000 normal potassium dictirornate was subtracted from the volume of bichromate used (55). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19 Csric aiTilîloiii-um sulfate, ü .1 normal, was standardized by the follow­ ing procedure (62). To 25.00 ml. of 0.1000 normal sodium arsenite solution in a 300 ml. Erlenrneyer fla.sk were added 65 ml. of water, 20 ml. of six normal sulfuric acid, one drop of ferroin indicator, and tlree drops of 0.01 molar osmium tetroxide. The solution was titrated at room temperature with the eerie ammonium sulfate solution to the elimination of the red tint. Sodium hypochlorite, 0.1 normal, was standardized against sodium arsenite by the following procedure (32). To 25.00 ml. of 0.1000 normal sodium arsenite solution in a 200 ml. Erlenrneyer flask were added one gram of potassium bromide and 0.5 gram of sodium bicarbonate. The solu­ tion was titrated rvith the sodium lij,’poch.lorite solution until vjithin a few ]iil. of the expected endpoint, one drop of Bordeaux indicator added, and the titration continued until the red color of the indicator faded. One more drop of indicator was added and trie titration continued dropwise until the color of the solution fleshed from pink to colorless or light yellow green. An indicator blank of 0.03 m l . of 0.1 normal sodium hypo­ chlorite was subtracted from the volume of liypochlorite used. G. Bordeaux. Indicators A 0.2 per cent aqueous solution was prepared from Bordeaux, British Color Index 88, G. Frederick Smith Chemical Company. Diphenylamine sodium sulfonate, 0.01 molar, was prepared (69) by dissolving 0.32 gram of diphenylamine barium sulfonate in 100 ml. of water and adding 0.5 gram of sodium sulfate. After standing over night Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 the clear solution was decanted from the precipitated barium sulfate. Ferroin. i prepared solution obtained from the G. Frederick Smith Chemical Company was used. Quinoline Yellow. A 0 .2 per cent aqueous solution was prepared from quinoline yellow, technical, British Color Index 801, Eastman Kodak Company. isrtrazine. A 0.05 per cent aqueous solution was prepared from tartrazine, practical, British Color Index 6J4O , Eastman Kodak Company. Cresyl violet. A 0.2 per cent aqueous solution was prepared from cresyl violet. Allied Chemical and Dye Corporation. Starch. A 0.5 per cent aqueous solution of soluble starch, freshly prepared. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21 IV. DiiTi’iiitMli'iATION OF lliPOPHOSFuATÏÏ It was desirable to use as a, basis for a study of methods for the .10termination of hypophosphato a liypophosphe.te which is a primary standard, reasonably stable in solution, and one wliich does not contain cations wliich may interfere in the methods used. Disodium dihydrogen typophosphate hexaiiydra.te prepared by the method of Leininger and Chulski (37) meets these requirements (37,12,^2). Oisodium hypopliosphate was, prepared by the method of Leininger and C'lulski. The product was recrystallized twice from water; the crystals were allowed to form at room temperature. They were finally washed five times with five ml. of ice-cold water and air dried e,t room temperature until they tumbled freely vtien stirred. They were stored in a tightly stoppered bottle. Standard 0.02500 molar solutions of the preparation were made by dissolving 7.651^ grams of the crystals in hOO ml. of warm water and diluting to one liter after adjusting to temperature. i . Oxidation of Hypophosphate by Excess Potassium Bichromate in Sulfuric Acid Solution Determination of excess dichromate with ferrous ion. Ghulslci (12) found that disodiuiri dihydrogen hypiopliosphate is oxidized by an excess of potassium dichromate in sulfuric acid solution. He found that the excess potassium dichromate could be determined by adding an excess of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 j'.u’rous ion and titrating the excess with potassium dichromate. This titration could, be done in mucli stronger acid than recommended (55) if there was present ttiree ml. of Ü5 per cent phosphoric acid for each 50 m l . volume of solution. The volume of phosphoric acid present had to be equal to at least one half the volume of concentrated sulfuric acid present. The reverse titration of potassium dichromate xvLth ferrous ion was not possible . To measured- samples of O.050OO normal disodium phosphate in 500 .ii-l. brlenmeyer flasks which had been thoroughly cleaned with hot chromic acid cleaning solution and thoroughly rinsed ifith water were added a measured excess of 0.1000 normal potassium dichromate and sufficient concentrated sulfuric acid to majce the resulting mixture 12 normal in acid. The flasks were covered vrLth small beakers and hea,ted in a boiling water bath. The contents of the flasks were cooled to room temperature^ diluted vjith water to 250 ml., 15 ml. of 05 per cent phos­ phoric acid added^ and then a measured excess of standard ferrous ammonium sulfate solution. The excess ferrous ammonium sulfate was immediately titrated with the standard dicliroma.te using six drops of diphenylamine sodium sulfonate as indicator. An indicator blank of 0.05 ml. of 0.1 normal dichromate was subtracted from the volume of dicliroma.te used. Blanks were determined in the same manner, substituting water for the hypophospha.te solution. A minimum of ferrous ion was added before the backtitration to keep the cliromic ion concentration low and provide a sharper endpoint. The results are shox-jn in Table I. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23 TABLE I UETERi-iIi'JiiTION OF HYPOPHOSPE&TE WITH EXCESS P0TASSIU14 DIGHROI-IivTE (BACK'riTRATIOW WITH FERROUS lOw) Conditions; hi. O.OS N AagiigjP2O 0 9.99 19.98 L9.96 5.00 9.99 9.99 19.98 19.96 19.96 25.00 25.00 25.00 19.98 1:9.96 0.0 0.0 0.0 0.0 12 W Sulfuric Acidj Heating Time One Hour Ml. 0.1 h KgOr 2O7 25.00 25.00 L9.96 5.00 9.99 25.00 19.98 19.96 19.96 25.00 25.00 25.00 L9.96 L9.96 25.00 25.00 25.00 25.00 Meq. Ha^HgP gOg T aken 0.500 0.999 2.1:99 0.250 0.500 0 .500 0.999 0.999 0.999 1.250 1.250 1.250 2 .1:99 2.L99 — — — — • wagHgPgOg Found Neq. Percent 0.1:9L 0.995 2.L75 0.251 0.503 0.506 0.999 1.001 1.002 1.252 1.250 1.248 2 .487 2.494 —— — — — 99.6 (a) 99.6 (a) 99.0 (a) 100.4 100.6 101.2 100.0 100.1 100.2 100.2 100.0 99.6 99.5 99.6 , 0.09: 0 .05; — 0 .02 (a) ^ Ml. 0.1 H KgCraO? lost (a) heated I4O minutes. An indicator blank of 0.03 ml. of dichromate was used in the calcu­ lations of the results in Table 1 but the blanks resulting from the small loss of dichromate were disregarded. The data in Table 1 show that heating for W minutes in 12 normal sulfuric acid at the temperature of a boiling water bath is not quite Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2h lon(^ onougb to complote the oxidation. ii^Toophosphate hydrolyzes to phosphite and phosphate in the hot acid solution. It is quite possible that the rate of the oxidation of hypopliosphate by dichromate is de­ termined by the rate of the hydrolysis and that the phosphite, rather than the hypophospha.te, is oxidized. Hovrever, the reacting ratio is the same if hydrolysis does or does not take place . In both cases the equivalent weight of the disodium diliydrogen hypophosphate is one-half of the molecular weight and the equivalent weight of the potassium ciichromate is one-sixth of the molecular imight. The accuracy of the method using a hot acid solution of potassium dichromate depends upon the stability of potassium dielirornate in hot acid solution and upon the accuracy of the procedure used to determine the excess dichromate. In order to study the stability of the diclirornate in hot acid solution and determine the accuracy of the procedure used to determine the excess dicliroma.te a series of blanks was determined by the method used for the determination of hypophosphate substituting water for the hypophosphate solution. Each blank contained sufficient sulfuric acid to make the resulting mixture 12 normal in acid and all were heated for one hour in a boiling water bath. A slight excess of standard ferrous ammonium sulfate was added to each sample before ba.cktitration with standard dielirornate . Table II shows that potassium dichromate is stable in 12 normal sulfuric acid at the temperature of a boiling water bath. The small loss of dicliromate may result from the oxida.tion of small quantities of reducing substances in the sulfuric acid. If the blanks are to be kept Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 TABLE II Si'ABILITY OF POTASSIUM DICmOILATE IN HOT SULFURIC ACID SOLUTION ml. 0.1 N KgCr 3O 7 25.00 25.00 25.00 25.00 9.99 9.99 9.99 9.99 M l . Concentrated H 2SO4 Ml. Water 25 25 25 25 10 10 10 10 25 25 25 25 10 10 10 10 HI. 0.1 N ^sCrgO? Lost 0.02 0.03 0.02 0.01 0.00 0.00 0.01 0 .00 small, great caxe must be taken to use flasks which are tiioroughly cleaned because of the drastic oxidizing conditions used. In order to determine the effect of making the calculations on the basis ox blanks, a series of liypophosphate samples was determined oy the same procedure as used in the previous determination of hypophospnate by excess dicliromate with the exception that the results were calculated on the basis of blanks determined in the same manner as the samples. The data in Table III show that slightly better results are obtained if the small blanks are taken into consideration in the calculations. The best results are obtained if tion taken for analysis is limited to the volume of hypophosphate solu­ 20 to 30 ml. If the volume of sam ple taken is very large, a large amount of sulfuric acid is required and this makes the endpoint in the back—titration less sharp. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Also, the 26 TABLE III Dii.TERi-IIlNÎATIOK OF HYPOPHOSPHATE WITH EXCESS DICHR02-I/iTE MAKIWG CALCULATIONS ON THE BASIS OF BLANKS ill. 0.05 w i-iagHgPaOe 19.98 19.98 19.98 19.98 25.00 25.00 25.00 25.00 0.0 0.0 Ml. 0.1 N K^Cr gO? 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25 .00 Meq. NasHaPaOs Taken 0.999 0.999 0.999 0.999 1.250 1.250 1.250 1.250 —— Me q . Found Percent 0.998 1.001 0.998 0.999 1.2L6 1.2U9 1.2L9 1.2U7 —— 99.9 100.2 99.9 100.0 99.7 99.9 99.9 99.8* 0.04_ 0.02" * Ml. 0.1 H KsCrzO? lost. ftjiiount of excess ferrous ion added should be kept at s. minimum in order that 3. smaller amount of chromic ion x-ri.ll be present at the endpoint. The color of a high concentration of cliromic ion makes the endpoint less sharp. If the oxidation of hypophosphate by excess dicliromate is to be complete in approximately one hour, it must be carried out in at least 12 normal sulfuric acid. In order to determine the effect of added phosphate on the determin— otion of hypophosphate xsrith excess potassium dicliromate in sulfuric acid solution, samples and blanks were determined by the procedure described x-rith the exception that phosphate was added before the heating period. The phosphate was added in the form of reagent grade phosphoric acid and monopotassium phosphate. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27 It, was found that a small amount of dichromate was used by the phosphate. However, the amount of dicliromate used was not constant for a constant amount of added phosphate and varied with the source of the phospliate. It even varied tvith the same amount of phosphoric acid taken from different bottles of the reagent grade phosphoric acid. The amount of dicliromate used was smaller for the same amount of phosphate added in the form of potassium phosphate than in the form of phosphoric acid. The phosphoric acid used was marked as meeting A.O.S. specifications which means that its reducing properties are such that ten ml, of the 0$ per cent acid should not decolorize 0.20 ml. of 0.1 normal potassium permanganate (h9) . It was found that the addition of ten ml. of the phosphoric acid resulted in a blanlc of approximately 0,13 m l . of 0.1 normal dicliromate wiiich corresponds to a reducing property which is well ■vrithin the A.O.S. specifications. It was concluded that phosphate has no effect on the determination of hypophosphate with potassiiam dichromate in hot sulfuric acid solution except the effect due to reducing substances present in the added phosphate . Determination of the excess potassium dicliromate iodometrically. because dicliromate liberates iodine from an acid solution of potassium iodide it should be convenient to determine an excess potassium di­ chromate iodometrically. However, the acid concentration present in the determination of hypophosphate by excess dichromate in 12 normal Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o'jjric a.cid is mucîi liighsr* tlian uhs scid. GonGsriirsirion r'scorriimended the iodometric determination of dichromate (31,ù5). in order to determine the effect of a much stronger arid solution uiian recommended on the iocometric determination of dicliromate e series oi dicliromate samples was determined iodometrically in solutions of different sulfuric acid concentration. To 19.9d ml. samples of 0,1000 normal potassium dicliromate in 500 ml. drlenmeyer flasks were added various amounts of water and sulfuric scid so that the total volume of the solution was 100 ml. To each flasl was slowly added tiro grams of powdered sodium carbonate and six grams of potassium iodide dissolved in 15 m l . of water, /tter standing for eight minutes the contents of the flasks were diluted i-rith water to scout 350 m l . and the liberated iodine was titrated with a sodium Ohiosulfate solution. The results are shown in Table IV. TABLS IV EFFBOT OF ACID CONCEI'ITRATIGK Oh THE IODŒIKFRIG DEI’ERMU'lATIOh OF POTASSIUM DICHROMATE Sample No . 1 2 3 h Approximate Concentration of Acid 2 I'r 3.6 N 7.2 N Ml. 0.1 N nagSgOg 19.91 19.92 19.97 20.03 * Recommended concentration, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29 Samples 1 and 2 in Table IV were determined by the recommended procedure for the iodometric determination of dicliromate. The data in Table IV show that a larger volume of thiosulfate is used when the acid concentration is higher than recommended. The larger amount of thio­ sulfate used in a solution of high acidity is probably due to air oxidation of the iodide. In order to study the effect of determining the excess dichromate iodometrically in the 12 normal acid used in the determination of hypo­ phosphate ^ a series of hypophosphate samples was oxidized by excess potassium dicliromate in 12 normal sulfuric acid by the procedure previously described and the excess dichromate was determined iodoraetricslly in the 12 normol acid. The volume of each solution was adjusted so that each flask contained about 75 ml. of 12 normal sulfuric acid during the oxidation. All samples were heated for one hour. The results sre shovm in Table V. TABLE V DJSTERMIMTION OF HYPOPHOSPHATE BY EXCESS POTASSIUM DICHROMATE, BAGKTITRATING THE EXCESS DICHR0I4ATE IODOMETRICALLY IN STRONG ACID SOLUTION HI. 0.05 N ha.gHgPgpg Ml. O.IN KgCr 2O7 Meq. NagHgPgOg Taken 9.99 19.96 19.98 25.00 9.99 19.98 19.98 25.00 0.500 0.999 0.999 1.250 KS3H 2P 2OG Found Meq , Percent o.hio 0.968 0.965 1.201 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94.0 96.9 9 6 .6 96.1 30 I'soleV shows that low results would be obtained if the excess bichromate were determined iodometric ally in the 12 normal sulfuric acid . It seemed reasonable to expect that dichromate could be accurately determined iodometricelly in a highly .acid solution if part of the acid were first neutralized so tliat the acidity was the same as required by the recommended procedure. In order to determine if such were the case, a series of dichromate samples was treated with water and con­ centrated sulfuric acid so that the resulting mixtures each had a volume of approximately 75 m l . and were 12 normal in sulfuric acid concentration. The mixtures were cooled in an ice water oath and different weighed amounts of sodium ir>tlroxide dissolved in a constant volume of water added. The mixtures were cooled to remove the heat given off in the neutralization. After the addition of t'wo grams of sodium carbonate to each mixture, the dichromate was determined iodometrically. The results obtained were compared to results obtained by the recommended procedure. Table VI shows that potassium dichromate can be determined iodo­ metrically in 12 normal sulfuric acid if the acid is partially neutralized so that the détermination is completed in t>ro to three normal acid. If the determination is carried out in this manner, the resists obtained agree with the results obtained by the recommended procedure. The date in Table VI also indicate that it should be possible to carry out an iodometric determination of the excess dichromate in an oxidation using excess dichromate in 12 normal sulfuric acid. This procedure was used to determine the excess dicliromate in the oxidation of hypophosphate i-Jith dichromate . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31 TA6LÜ VI lODOHLTKIC DjiTMINAÏION OP DICHROMATP IN riIGNLï ACID SOLUTION BY NEUTRALIZING PiüT OF THE ACID Ml. 0.1 N IvgCr 2O7 25.00 25.00 25.00 25.00 25.00 25.00 25.00 M l . Cone. H3SÜ4 (a) (a) 25 25 25 25 25 Grams NaOH Added 0 0 26 20 20 17 17 Approximate Acid Concentration" Ml. NBgS 2O 3 2 N 2 N ,1.6 N 2.5 w 2.5 w 3 M 3 M 24.23 24.24 24.18 24.21 24.23 24.25 24.23 (a) 100 m l . of 2 N HgSO^ (recommended procedure) ■K Before the addition of N 82003. Measured samples of hypophosphate %'jere treated with sulfuric acid 8nd 8 measured excess of standard potassium dichromate. After heating for one hour in a boiling water bath the samples were cooled in ice water and sodium'hydroxide solution sdded until the resulting mixtures had a, volume of approximately 100 ml. and i-rere two to three normal in sulfuric acid concentration. The excess dichromate was then deteirnined iodometrically using a standard sodium thiosulfate solution. Table VII shows that the excess dichromate in the oxidation of hypophosphate with dichromate can be determined iodometrically. The results obtained are essentially the same as the results obtained when the excess dicliromate was determined with a ferrous ammonj.um sulfate solution. Table VII also shows that increasing the time of heating to one and one—half hours has no effect on the results. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32 ’i'/iBLE VII OF h y p o p h o s p h a t e b y e x c e s s DICHROMATE IN 12 H SULFURIC ACID DEIEtmiNING THE EXCESS DICHRCE'iATE IODOMETRIC ALLY AFTER NEUTRALIZING PART OF THE ACID (Heating Time One Hour) HI. 0,0$ H - agH gP gOg Ml. 0.1 N K gCr 2O7 Meq, i\lagligPgOg T aken 9.99 9.99 19.98 19.96 19.98 19.98 19.98 2^.00 25.00 25.00 25.00 25.00 L9.98 9.99 9.99 19.98 19.98 19.98 25.00 25.00 25.00 25.00 25.00 25.00 25.00 h9.98 Ü.50O 0.500 0.999 0.999 0.999 0.999 0.999 1.250 1.250 1.250 1.250 1.250 2.L99 NasH^PaOg Found (a) Meq , Percent 0.L98 0 .500 1.003 0.996 0.999 0.99S 0.996 1 .22:7 1.250 I.2U 6 1.2L8 1.2147 2.1:96 99.6 100 .0^ 100.l4_ 99. 100 .0 99.6 99.7_ 99.6T 100.0 99.7 99.8 99.8 99.9 Heated for one and one-haif hours . (a) Calculated on the basis of blanks. The iodometric determination of the excess dichromate has no particular advantage over the procedure using a ferrous ammonium sulfate solution except that the former is a direct determination. B , Oxidation of Hypophosphate By Excess Sodium Hypochlorite Chulski (12) made a limited study of the oxidation of hypophosphate by sodium hypoctilorite and concluded that liypophospliate is oxidized by excess sodium hypoclilorite in approximately neutral solution. He de­ termined the excess hypochlorite iodometrically in the presence of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33 buffers and the results obtained were somewhat irregular in that better results were obtained if calculations were made on the basis of a. direct standardization of the hypochlorite solution than if blanks , treated in the same manner as the hypophosphate samples, were used as the basis for the calculations. The same irregularities in the iodometric determination of hypo­ chlorite were experienced during the investigation described in this thesis. Tills is described more fully in the section of this thesis dealing >7itli tiie determination of piiosphite witii hypochlorite. Arsenite reacts rapidly with hypochlorite in a bicarbonate solution, however, no indicator is available for determining the equivalence point in a titration of hypochlorite %-n.th arsenite because the indicators usually employed in oxldation-reduction systems are destroyed by hypo­ chlorite . It seemed reasonable to expect that hypoclilorite could be determined by adding to it an excess of standard arsenite solution and titrating the excess arsenite with a standard oxidant. Arsenite can be titrated with hypochlorite in bicarbonate solution using Bordeaux indicator to determine the endpoint (32) . It is necessary to have bromide present at the endpoint because Bordeaux is not decolorized by a slight excess of hypochlorite but it is decolorized by a slight excess of hypobromite. The hypobromite is f o m e d by the reaction between the bromide and hypochlorite. Determination of hypophosphate with excess hypochlorite, determin­ ing the excess with excess arsenite. To measured samples of 0 .0^000 normal hypophosphate in 2^0 ml. iodine flasks were added approximately Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3U one gram of sodium bicarbonate and a measured excess of standard sodium hypoclilorite solution. The flasks were allowed to stand in the dark for various time intervals, a measured excess of standard sodium arsenite added, and then approximately one gram of potassium bromide. The excess sodium arsenite was titrated with the standard sodium hypo­ chlorite solution using Bordeaux indicator as in the procedure for the standardization of hypochlorite against sodium arsenite. An indicator blank of 0.03 m l . of 0.1 normal hypochlorite was subtracted from the volume of hypochlorite used. Blanks were determined in the same manner, substituting water for the hypophosphate solution. The hypophosphate is oxidized to phosphate: haClO + hSgHaPsOg + H2O --- > 2ilaH2P04 + haCl The data in Table VIII show that hypophosphate is oxidized com­ pletely by excess sodium hypoclilorite in bicarbonate medium. The data also show that the oxidation is complete in pO minutes, incomplete in 13 minutes, and that increasing the standing time up to tvro hours has no effect on the results . The blanks due to the loss of hypochlorite during the standing are negligibly small. Several series of hypophosphate samples were determined by excess sodium hypoclilorite using the same procedure . The results of the analy­ sis of a typical series in which the hypophosphate samples were allowed to stand with excess standard hypochlorite for various time intervals from 30 to 90 minutes are shown in Table IX. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35 TABLE VIII DETERMINATION OF HYPOPHOSPHATE WITH EXCESS SODIUM HYPOCHLORITE AS A FUNCTION OF T L Œ M l. nl. 0.05 N nahgP gOg 0.1 N NaClO Added With Sample 2 3 .00 25.00 25.00 25.00 25.00 0.0 0.0 19.98 19.98 19.98 19.98 19.98 19.98 19.98 * Meq. NaghgP gOg Taken Standing Time Min. 1.250 1.250 1.250 1.250 1.250 — —— — 15 30 55 60 120 30 60 NSgHgP pOg Found Meq. Percent 1.250 1.250 1.252 1.258 1.251 99.2 100.0 100.2 99.8 100.1 0 .01% 0 .00% — — MI. 0.1 N hypochlorite los t. Table IX shows that the amount of excess hypoclilorite used can be varied from two to four times: the amount required for the oxidation of the hypophosphate ivith no effect on the results TABLE IX DETERÎ4INATI0N OF HYPOPHOSPHATE WITH EXCESS HYPOCHLORITE (ARSENITE-HYPOCHLORITE MODIFICATION) Ml, 0,05 N N a^H 3P gO g Ml. 0.1 N NaClO Ad.d.ed With Sample 9.99 9.99 9.99 19.98 19.98 25 .00 25.00 25.00 19.98 19.98 19.98 19.98 19.98 19.98 19.98 25.00 Meq. NagHgPgOg Taken 0.500 0.500 0.500 0 .999 0.999 NagHg>PpOg F cund Percent Meq . 1.250 0.597 0 .501 0.599 0.999 1.000 1 .251 1.259 1.250 1.250 1.250 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99.5 100.2 99.6 100 .0 100.1 100 .1 99.9 100 .0 36 Determination of excess hypochlorite by a potentiornetrie titration with standard sodium arsenite. Preliminary ex^^eriments were carried out to determine if it vrere feasible to determine the excess hypochlorite by a direct potentiometric titration with standard sodium arsenite rather than using an excess arsenite . Potentiometric titrations of hypochlorite with arsenite and of arsenite with hypoclilorite were made and the re­ sults compared with the results of titrations of arsenite with hypochlorite using Bordeaux indicator. Previous oxidations using hypochlorite had been carried out in stoppered iodine flasks. Because it would be incon­ venient to carry out potentiometric titrations in flasks, it would be necessary to use beakers. It was then necessary to determine if there was a measurable loss of hypochlorite from a covered beaker during the time required for the oxidation of hypophosphate with excess hypochlorite. In order to determine this, hypophosphate samples were determined by excess hypochlorite in beakers wliich were covered with a watch glass, and in stoppered iodine flasks . The standing time varied from 30 to 60 minutes, The excess hypochlorite was determined by adding an excess of standard arsenite and titrating the excess arsenite ivith hypochlorite using Bordeaux indicator. The results are shown in Table X. The data in Table X show that there is no measurable loss of hypo­ chlorite from a covered beaker during the time required to carry out the oxidation of hypophosphate with excess liypochlorite. In order to compare potentiometric titrations of arsenite with h;^poclilorite with titrations using E'ordeaux indicator, a 0.1000 normal sodium arsenite solution was titrated vxith a hypochlorite solution Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37 TABLE X DEl'ElmlNATION OF HïPOPHOSPiiJ.TL IN BEAKEitS AS COMPARED TO DETERi^IWATION IN IODINE FLASKS Sample" Meq. uagHgfaÜG Taken Meq. N a . ^ H ^0s Found 3i 1.2^0 1.250' 1.250 1.259 Fg Bp 1 .2 5 0 1 .2 5 0 1.250 1 .2 5 0 Ba 1.250 1.259 B refers to determination made in beaker, F in flask; listed in order of standing: time . using; Bordeaux indicator and it was found that 19.96 ral. of the arsenite solution required 19.26 and 19.26 ml. of the hypochlorite solution in 'inro titrations with no correction for the indicator blank. The arsenite solution was then titrateo with the hypochlorite solution and the equi­ valence point determined potentiometric ally by the folloi-jing procedure . To 19,98 ml. samples of the arsenite solution in 150 ml. beakers were added 35 ml. of water, 0.5 gram of sodium bicarbonate, and one gram of potassium bromide . The solution was titrated T-ûth the hypochlorite solution and the equivalence point determined with a Sargent potentiometer with platinum and saturated calomel electrodes. Data, for a typical titration are shoi-m in Table XI. Other 19.98 ml. samples of the sodium arsenite solution titrated potentiometric ally in the same mannei-..required 19.25 and 19.25 ml. of tile hypociilorite solution. A 19.98 ml, sample of tiie arsenite solution Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. à 38 TABLE XI PÛTEOTIOMETHIC TITRATION OF A R S m i T E WITH HYPOCHLORITE IN THE PRESENCE OF BROMIDE Ml. 0.1 N NaClO 18.57 1 8 .66 18.76 18.83 19.17 19.19 19.23 19.26 19.27 19.29 19.32 End point E Volts A S Volts 0.132 0.139 0.115 0.150 0.157 0.159 0.309 0.661 0.737 0.753 0.758 19.25 ml. 0.007 0.006 0 .005 0.007 0 .002 0 .150 0.352 0.076 0.016 0.005 A V HI . 0.09 0.10 0.07 0 .3a 0.02 O.OL 0.03 0.01 0.02 0.03 a e / a v 0.0 78 0.060 0.071 0.021 0.10 3.6 1 1 .0 7 .6 O.oO 0.17 V Ml. 18.62 18.71 18 .60 19.00 19.18 19.21 19.25 19.28 19.28 19.32 Has titrated, ’t-n.th the hypochlorite by the seme procedure except tiiat no potassium oromide was added to the srsenixo sample. The results are shown in Table X I I . The data in Tables XI and XII show that sodium arsenite solution in sodium bicarbonate medium can. be titrated with sodium hypochlorite and the equivalence point determined potentiometrically. ihe data also show that the presence oT bromide does not influence the equivalence point in the titration. 00 The. average volume of h^ipochlorite required oxidize 19.98 ml, of the arsenite solution was found to be 19.28 ml. (before the indicator correction) in two titrations using Bordeaux as indicator. Tlie a.verage volume of hypochlorite required by 19.96 ml. of the arsenite in four potentiometric titrations was found to be 19.25ml Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39 TiiBLii XII "OTENÏIOIIETRIC TITRATION OF hi. 0.1 N NaClO E Volts Volts 0.176 19.05 0.162 0.006 19.15 0.220 19.22 Ü.O38 0 .110 0.330 19.23 0.168 1 9 .2L 0.678 19.26 0.652 0.176 0.661 0.009 19.27 19.32 0.036 0.695 0.020 19.61 0.715 End point 19.26 ml. A R S Z N IT E WITH H Y P O C H L O R IT E AV hi. AS/ AV V Ml. 0.10 0.07 0 ,01 0.01 0.02 0 .01 0.05 0.09 0 .60 0.56 11.0 16.6 6.7 0.90 0.68 0.22 19.10 19.19 19.23 19.26 19.25 19.27 19.30 19.37 The 0.03 ml. Indicator blank used in titrations using Bordeaux indicator is therefore well substantiated. iixoex’iments were carried out to compare the results obtained oy potentiometric titrations of hyrjoclilorite t*n_th arsenite wiun the results obtained by titrations of arsenite with liypochlorite using Bordeaux as ind.icator. The normality of the hypochlorite was found to be 0,09789 and 0.09769 b y two titrations using Bordeaux as indicator; the normality was calculated on the basis of the normality of the standard arsenite solution. metrically. The normality of the hypochlorite was also determined potentio­ To 19.96 ml. samples of the hypochlorite solution in 150 ml. beakers were added 35 m l . of water and 0.5 gram of sodium oi carbon ate The solution was titrated with the standard arsenite solution and the equivalence point determined by means of a Fisher Titrimeter with Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Uo Mlst/ri.nuin and s a/buna “bed. calomel electrodes. to a null point at 0.00 volts. The instrument vrss adjusted Data obtained in a t;ypical titration arc siiQ\-m in Table XIII. TiæiE XIII POÏENTIOhiETRIC TITRATION OF HIPOCHLORITE WITH i\RSENITE hi. 0.1 N Arsenite E Volts 0.703 19.45 19.46 0.691 0.669 19.46 19.51 0.645 0.550 19.53 0.231 19.54 0.130 19.55 0.121 19.57 0.104 19.65 End point 19.54 AE Volts AV Ml. 0.012 0.022 0.024 0.095 0.319 0.101 0.009 0.017 0.01 0.02 0.03 0.02 0.01 0.01 0.02 0.08 A E/ A V 1.2 1.1 0.60 4.8 32.0 10.0 0.45 0.20 V Ml. 19.46 19.47 19.50 19.52 19.54 19.55 19.56 19.61 Other 19.98 ml. samples of the liypochlorite solution determined by a potentiometric titration with the standard arsenite using the same procedure, required 19.55^ 19.5^ and 19-55 m l . of the 0.1000 normal arsenite. The average normality of the hypochlorite on the basis of the potentiometric titrations was then calculated to be 0.09755, i-diich a.grees well with the normality of 0.09789 found by the method using Bordeaux as indicator. The preliminary expei’iments indicated that the excess hypochlorite in the detenmination of hypophosphate by excess hypochlorite could be determined b y a, direct potentiometric titration with standard sodium Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. arsenite. This procedure was used to determine a series of hypophos­ phate samples. A Fisher Titrimeter was used to determine the end point in the arsenite titration. The hypochlorite solution was standardized against standard sodium arsenite using Bordeaux as indicator. The h}l)ophosphate samples were allowed to stand with excess hypochlorite for 35 to 90 minutes before the determination of the excess hypochlorite The results are shown in Table XIV. ï/iîLÏÏ XIV Ddfdm^IHATION OF HYPOPHOSPHATE WITH EXCESS HYPOCHLORITE, DEl'ERMINING THE EXCESS HYPOCHLORITE BY A POTENTIOMETRIC TITRATION WITH STAND/Jm SODIUM ARSENITE ml. 0.05 H Na.gH gPgOe 25.00 25.00 25.00 25.00 19.98 19.98 HI. 0,1 h KaClO Added With Sample i'leq . N a gH gP gO g Taken 19.98 19.98 19.98 19.98 19.98 19.98 . 1.250 1.250 1.250 1.250 0.999 0.999 i\a aHgPgOg Found Percent Meq . 1.2li9 1.250 1.252 1.250 0 .999 1.001 99.9 100.0 100.2 100 .0 100.0 100.2 Determination of excess hypochlorite by a titration with arsenite using the dead stop endpoint. The dead stop endpoint is often a con­ venient means of detecting the equivalence point in a titration, and is especially valuable if a good indicator for the titration is not avail­ able. Experiments were carried out to deteimiine if the equivalence point in the titration of hypochlorite with arsenite could be detected by the dead stop technique. It was found by experiment that if a small Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. U2 potential was applied across platinxim electrodes immersed in a hypo­ chlorite solution containing bicarbonate a current flowed between the electrodes. Also^ if arsenite was added to this solution, the current decreased to zero and then increased as the volume of arsenite added was increased. Experiments were also carried out with measured volumes of hypo­ ciilorite and arsenite. A sodium hypochlorite solution was standardized against a standard sodium arsenite solution by the procedure in which Bordeaux is used as the indicator. The normality of the hypochlorite solution was found to be O.IO5I4; this value was the average of four determinations . The normality of the hypochlorite was also determined by the dead stop procedure. To 19.96 ml. samples of the hypochlorite solution were added 35 m l . of water and 0.9 gram of sodium bicarbonate. Platinum electrodes were immersed in the solution and connected to a Fisher Elecdropode. electrodes. A potential of U90 millivolts was applied to the The standard arsenite solution was added from a burette until close to the equivalence point and current readings taken as the arsenite solution was added in small increments. stirred with a magnetic stirrer. The solution was The equivalence point was taken as the point where the current was at a minimum. Data for a typical titra­ tion are shown i n Table XV. Table XV shows that there is a large change in current before and after the equivalence point and the end point is easily detected. Ooher 19.98 ml. samples of the liypochlorite solution were titrated with the 0.1000 normal arsenite solution by the same procedure. The average of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. U3 TABLE XV DEADSTOP TITRATION OF SODIUE HYPOCHLORITE WITH SODIUII ARSENITE Ml, 0.1000 N Arsenite Current. Reading 20.7 U 20.87 20.98 21.00 21 .02 21.03 21 .05 21.07 21,08 21.11 305 220 12U 102 78 58 18 2 21 51 End point 21.07 ml. six titrations was 21.0? ml. of 0.1000 normal arsenite used for a 19.98 ml. sample of arsenite with a maximum deviation of 0,03 ml. of arsenite, The normality of the hypoclilorite solution wa.s then calculated to oe 0.1053 normal which agrees well with the normality of O.IO^U determined by titrations using Bordeaux as indicator. The results indicateo that in an oxida.tion using excess hypochlorite, the excess can be determined by a titration with standard sodium arsenite, detecting the endpoint with the deadstop technique unless some other component in the mixture interferes. • Samples of hypophosphate were determined by excess hypochlorite in sodium bicarbonate medium and the excess hypochlorite determined bj a titration with standard arsenite determining the endpoint by the dead­ stop procedure. The samples were allowed to stand for 35 to 80 minutes, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jiii voltage of h$0 millivolts was applied to the electrodes in the dead­ stop ba,cktitration. The results are shown in Table XVI. TABLE XVI HYPOPHOSPHATE BY EXCESS HYPOCHLORITE AND DEADSTOP BACKTITRATIQN WITH ARSENITE determination of 141. 0.05 w i'ia.gllaP 9.99 9.99 9.99 19.98 19.96 19.96 19.96 19.98 25.00 25.00 25.00 Ml. O.IN NaClO 19.98 19.98 19.98 19.98 19.98 19.98 19.96 19.98 19.98 19.96 19.98 M e q . i'4a.gH ^P ^0 g T a.ken 0.500 0.500 o .500 0.999 0.999 0.999 0.999 0.999 1.250 1.250 1.250 Me q . aOe Found Percent 0.501 0.501 0.502 0.999 1.001 0.999 0.999 1.000 1 .2L9 1.250 1 .2U9 100.2 100.2 100. U 100.0 100.2 100.0 100.0 100.1 99.9 100.0 99.9 Table XVI shows that the excess hypochlorite in the oxidation of hypophosphate can be determined by a deed-stop titration with standard arsenite. It would be very convenient if an indicator were available which could be used in a titration of hypochlorite with arsenite. Indicators have been proposed for the titration of arsenite inLth hypochlorite. Some of these indicators have been reported to be somewhat reversible in tiiat if the indicator is bleached by hypochlorite the addition of arsenite will restore the color of the indicator. These indicators were investigated to determine if any were sufficiently reversible to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. U5 use in a titration of hypochlorite with arsenite and to determine if any of them were more convenient to use than Bordeaux in the reverse titration, Sinn (60) recommended the use of a, 0.2 per cent aqueous solution ox quinoline yellow for detecting tlie equivalence point in titrations of arsenite with hypochlorite. potassium bromide. The arsenite solution contained Young and Gupta (73) used quinoline yellow for titrations of hypochlorite witii arsenite but did not state that bromide v/as present. Belcher (2) reviewed the use of this indicator and found that it was slightly reversible in the presence of bromide . A 0.2 per cent aqueous solution of quinoline yellow was prepared. Titrations of. arsenite with hypochlorite and the reverse titrations were tried using ttiis indicator witii and ïfithout bromide present. It was found that several drops of 0.1 normal hypochlorite were required to decolorize the indicator if bromide was not present ^ but the color was bleached from yellow to colorless by a trace of hypochlorite in the presence of bromide. The indicator was found to be slightly re­ versible , but when tlie color was bleached by a small amount of liypochlorite in the presence of bromide and an excess of arsenite was added the return of the color required about one minute. The indicator could not be used for a titration of hypochlorite with arsenite . Sheentsis (58) recommended the use of cresyl violet for titrations of arsenite with hypochlorite and claimed that the results for the determination of available chlorine agreed with those determined iodometrically. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. k 6 Experiments showed that the red color of cresyl violet was bleached to colorless by a trace of hypociilorite vjithout bromide present but the indicator wa.s found to be completely irreversible. It was found to be somewhat more sensitive to local excesses of hypo­ chlorite than is Bordeaux ajid its use does not present any advantage over the use of Bordeaux in titrations of arsenite with lypiochlorite except that the presence of bromide is not required when cresyl violet is used. Belcher (l) used an aqueous solution of tartrazine as indicator in titrations of arsenite with hypochlorite in tlie presence of bromide. Studies on the use of this indicator showed that the yellow color of the indicator is decolorized only by a large excess of hypochlorite if bromide is not present but is decolorized b y a trace of hypochlorite in the presence of brom i d e . The indicator was found to be slightly reversible but the re turn of the color upon addition of excess arsenite was not rapid enough that the indicator could be used for titrations of hypochlorite with arsenite even if the indicator was added close to the equivalence point. It was found that none of the indicators studied is better than Bordeaux for detecting the endpoint in titrations of arsenite with hypochlorite and none was sufficiently reversible to oe used for titra­ tions of }y»poclilorite inLth arsenite , Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. h .l C, Oxida.tion of Hypophospliate '■Jith Sodinjn Hypochlorite in a Solution Containing Bromide It was found by experiment that phosphite is rapidly oxidized by 8 slight excess of hypochlorite in a solution containing sod.ium bicarbonate and potassium bromide , This is discussed more fully in the section of this thesis dealing with the determination of pliospliite by hypochlorite in a bromide solution. hypophosphate under the same conditions. An attempt was made to oxidize It was found that the reaction between hypophosphate and hypochlorite in the presence of bromide, or essentially the reaction between liypophosphate and hypobromits, is much slower than the reaction of phosphite with hypobromite. The reaction is so slow that the yellow color indicating and excess hypociilorite appeared on the first addition of hypociilorite to a solution containing Ir/pophosphate and bromide and disa.ppeared very slowly especially near the equivalence point. It was therefore not possible to determine hypophosphate by a slight excess of hypochlorite in a bromide solution as was done in the case of pliosphite^ because the slow reaction did not readily allow the detection of a. slight excess of hypochlorite. A n attempt was made to determine hypophosphate in the presence of bromide by a measured excess of hypochlorite. To a series of 0.0^000 normal hypophosphate samples were added one gram of sodium bicarbonate and one gram of potassium bromide. Measured amounts of standard sodium hypochlorite were added from a burette. The samples were allowed to stand for various time intervals ^ an excess of standard sodium arsenite solution addedj and the excess arsenite titrated with hypochlorite from Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lib :1:3 same burette using Bordeaux to determine the endpoint. blanks VCre determined in the same manner, substituting water for the h}-po)hosphate solution. The results are shown in Table XVII. TABLE XVII DETERI-lIMnTION OF HYPOPHOSPHATE l'ATTH EXCESS HYPOCHLOklTE Ih THE PRESENCE OF BROMIDE AS A FUNCTION OF TIME OF STANDIFjG AND EXCESS HYPOCHLORITE il. 0.05 N 2^-2^ 2^6 19.96 19.96 19.96 19.98 19.98 19.98 19.96 25.00 25.00 25.00 25.00 25.00 25.00 0 0 Approx. M l . O.IN NaClO Before Standing 11 (a) 11 (a) 12 (a) lit 16 17 18 13 13 16 20 20 20 10 10 Meq . NagHgPgOe Taken Standing Time Min. 0.999 0.999 0.999 0.999 0.999 0 .999 0.999 1.250 1.250 1.250 1.250 1.250 1.250 marmm —— it 6 10 3 3 3 1 5 5 3 3 5 1 2 ba^riaPaOs Found -Meq. Percent 0.996 0.995 0.999 i.ook 1.007 1.007 0.996 1.2^0 1.229 1.2b^ 1.250 1.267 1.254 -- e 99.7 » 99.6 100.0 100.5 100.6 100.6 99.9 99.2 96.3 99.6 100.0 101.4 100.5 0.16" 0.26T -& Ml. 0.1 N NaClO lost. (a) The hypochlorite was added slowly with constant swirling until a yellow color persisted then allowed to stand the time indicated. The data in Table XVII show that hypophosphate is oxidized by excess hypoctilorite in bromide solution. The data also shoifs that a. large excess of hypobromite in the presence of bromide is unstable, or forms products on standing which do not react with arsenite in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. k bicarbonate m e d i u m . 9 This accounts for the large loss of hypochlorite when blanlcs were determined. Quantitative results for the determination of liypophosphate were not obtained, for the results depend on the amount of excess hypochlorite used and on the time of standing. F ark as and le win (l^) made a study of the reaction between sodium hypochlorite and potassium bromide and found that the rate of the re­ action (1) CIO" + Br" decreases with increasing pH, --- > BrO" + Cl” At lower pH values other reactions take place ; + BrO" -----> 2Cl” (2) 2010“ (3) CIO” + 2HC10 > 21"^ (U) BrO" + 2HC10 ^ BrOg" + BrOg” + 2Gl" + 2h'^ + CIO3" + 2Gl” and the rate of these reactions increases with decreasing pH. They found that in the pH range of 9.0 to 9-U an excess of hypochlorite quant it a.tively oxidized bromide ion according to equation (l) , and that the oxidation was complete in five minutes . If the reaction were allowed to proceed for a longer time, the effect of the other reactions was apparent and loss of hypochlorite occurred. Because the results of the determination of hypophosphate by excess hypociilorite in a solution containing bromide and bicarbonate ivere erratic, tlie determination was tried in a solution buffered to a pH of approximately 9.2. It was considered that under these conditions the oxidation of the hypophosphate migiit be confined to a reaction between Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50 in.popliosphate and hypobromite without loss of hypochlorite due to the foi-miation of br ornate and clilorate . * iv buffer of pH 9,U was prepared from boric acid and sodium hydroxide. The pH of a mixture of 25.00 ml. of O.050OO normal h;^T?o- phosphate, 19.98 ml. of 0.1 normal sodium hypochlorite, one gram of potassium bromide, and 50 ml. of the buffer of pH 9 ,h was found to be 9.2 as determined with a Beckman Mo del H-2 pH meter. To measured samples of hypophosphate were added one gram of potassium bromide and 50 ml,, of the pH 9 ,h buffer. A measured excess of hypochlorite was added from a burette, the mixture allowed to stand for 15 minutes, an excess of standard arsenite added, and the excess arsenite titrated 1-ri.th standard hypochlorite from the same burette using Bordeaux indicator. The hypochlorite was standardized against the standaz'd arsenite solution in the presence of SO ml. of the pH 9 .h buffer and it was found that the results obtained were the same as those found when bicarbonate was used in place of the buffer. Table XV111 shows that the oxidation of hypophosphate by excess h;>p)Ochlorite in the presence of bromide at a pH of approximately 9.2 is complete in 15 minutes if the excess hypochlorite is 60 per cent or above. However,' the results obtained are sligkitly high indicating some loss of liypoclfLorite but are not as erratic as the results found in a bicarbonate medium (Table XVIl) . The determination was repeated using approximately 60 per cent excess hypociilorite and varying the time of standing. The results are shown in Table XIX. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51 TABLE X V III DLTjiRMINivTION OF HïPOPHOSPlIi-.TE B Y EXCESS HYPOCHLORITE 1IÎ THE PRESENCE OF BROMIDE AT A pH OF APPROXIMATELY 9,2 AS A F m C T I O N OF EXCESS HYPOCHLORITE Ml. 0.05 N N a^H gP gO 0 9.99 19.98 9.99 19.98 25.00 25.00 19.98 19.98 29.97 25.00 19.98 39.96 25.00 Approx. Ml. 0.1 H NaClO 20 25 10 20 25 25 16.5 17.5 25 20 15 25 15 i'-.eq . HaHgPgOg T aken Approx. Percent Excess NaClO 0.500 0.999 0.500 0 .999 1.250 1.250 0.999 0.999 1.^99 1.250 0.999 1.998 1.250 300 150 100 100 100 100 85 75 66 60 50 25 20 Me q . 0.502 1 .00k 0.501 1.003 1.252 1.254 0.996 0.996 1.505 1.252 0.992 1.996 1.227 6 Found Percent 100.k 100.5 100.2 1 0 0 .k 100 .2 100,3 99.7 99.7 1 0 0 .k 100.2 99.3 99.9 98.2 Table XIX shows that if a 60 per cent excess hypochlorite is used, the oxidation of hypophosphate is complete in ten minutes end that a longer standing period up to 30 minutes has no appreciable effect on the results. The method using hypociilorite in the presence of bromide has no particular advantage over the method using hypochlorite without bromide except that the oxidation of hypophosphate is complete in s. shorter time if bromide is present. This advantage is minimized by the necessity of using a buffer and a critical excess of approximabely oO %)ercent hypochlorite when bromide is present. The latter requirement makes the method somewhat empirical. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52 Ï/03LE XIX ÛüTERI-lINATION OF HYPOPHOSPHATE BY EXCESS HYPOGHLORITE IN THE PRESENCE OF BROMIDE AT A pH OF APPROXIï^'IATELY 9.2 AS A FUNCTION OF TN^IE OF STANDING i-il. 0.05 N iMa.^igPgOg 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25 .00 D. Ml. 0.1 N i^IaClO 19.98 19.98 19.98 19.98 19.98 19.96 19.98 19.98 me q . N a^H gP gO g Taken 1.250 1.250 1.250 1.250 1.250 1.250 1.250 1.250 Standing Time Min. 5 8 10 10 12 15 20 30 NasHgP pOg Found Meq. Percent 1.2U5 1.2L6 1.251 1.2ii9 1.2L9 1.2lt8 1.251 1.252 99.6 99.7 100.1 99.9 99.9 99.6 100.1 100.2 Oxidation of Hypophosphate by Excess Ceric Sulfate in Sulfuric Acid Solution It was found by experiment that hypophosphite is completely oxidized by excess ceric sulfate in sulfuric acid solution at boiling temperature. It was also found that the excess ceric sulfate could be determined oy a titration with a standard ferrous aniinoriium sulfate solution or with a standard arsenite solution. A small amount of ceric sulfate wa.s lost during the boiling period and it was necessary to apply corrections for tliis loss. Determinations by excess ceric sulfate are discussed more fully in the section of this thesis dealing with the determination of hypophosphite with ceric sulfate. Experiments were carried out to de­ termine if hypophosphate could be determined by excess ceric sulfate in sulfuric acid solution. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53 To measured samples of 0 .OpOOO normal hypophosphate were added a measured excess of standard ceric ammonium sulfate and ten ml. of concentrated sulfuric acid. The mixtures were refluxed for various time intervals in 500 ml. flasks attached by glass joints to watercooled condensers , The precipitate of cerium hypophosphate wliich formed when the ceric sulfate was added to the hypophosphate dissolved when the flask was heated and remained in solution during the détermin­ ation. Each condenser was rinsed vjith about 30 ml. of water which drained into the attached flask. The contents of the flasks were cooled to room temperature ^ two drops of ferroin added^ and the excess ceric sulfate titrated >rith a standard ferrous ammonium sulfate solu­ tion. The excess ceric sulfate in some of the samples was determined, by adding two drops of ferroin, three drops of osmium tetroxide solution and titrating with a standard arsenite solution. Blanlcs were also determined in the same manner, substituting water for the hypophosphate solution. The results were calculated on the basis of the blanks and are shown in Table XX. Table XX shows that hypophosphate is not completely oxidized to phosphate by excess ceric sulfate in approximately six normal sulfuric acid in one hour at boiling temperature. The oxidation is fairly rapid when the reacting mixture is first heated but increasing the heating time from to 60 minutes has little effect on the oxidation. It is quite possible that the oxidation takes place tlirough a hydrolysis of the hypophosphate and oxidation of the phosphite formed and that in the relatively low concentration of acid used the hydrolysis is not completed Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. s 5U TABLE XX DLTEHIXIN/iTION OF HYPOPHOSPHATE BY EXCESS CiiEIG SULFATE IN SULFURIC ACID SOLUTION AS A FUNCTION OF TIME OF HEATING Note: Excess ceric sulfate determined with standard ferrous ammonium sulfate except where indicated. Ml. O.OS N Na^tigP gOg Ml. 0.1 N Ceric Sulfate 25\oo 23.00 19.96 19.98 19.98 25.00 25.00 25.00 25.00 25.00 25.00 25.00 0.0 0.0 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25 .00 25.00 25.00 25.00 25.00 25.00 25.00 Me q . I'Ia.2H gP gO g I aken Refluxing Time Min. 1.250 1.250 0.999 0.999 0.999 1.250 1.250 1.250 1.250 1.250 1.250 1.250 15 25 30 US 50 50 60 60 60 60 60 60 60 60 —— NagHgP gOg Found Me q . Percent 1.220 1.229 0.98U 0.965 0 .963 1.233 1.236 1.23U 1.233 1.232 1.236 1.23U —— 97.6 98.3 98.5 98.6 96 .U 98.7 98.9 98.7 98.6 9 8 .6 (a 96.9 (a 98.7 (a 0.07 e 0.09 - -K i-il. 0.1 N ceric sulfate lost. (a) Excess ceric sulfate determined with arsenite. in one hour. Also, it was found in another part of tiiis work that phosphite is not readily oxidized by ceric sulfate and this may oe a, factor in the oxidation of the hypophosphate. The oxidation of hypophosphate may possibly be completed if carried out in stronger acid solution or if heated for a, period longer than one hour, but the investigation was discontinued because other methods for the determination of hypophosphate developed during this work are much more satisfactory. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55 Svuranary of Results of Hypophosphste Determination The average results obtained by the methods developed for the determination of hypophosphate are as follows: Average Percent h agH 3P 3O e Method ____________________________________ bxcess Dicliromate j Ferrous Sulfate Procedure (Mo blank corrections).................... .. ................. Ixcess bichromate, Ferrous Sulfate Procedure (Blank corrections).................................. 99.9 Ixccss Diehronate ^ lodometric Procedure (Slanlc corrections) Ixcess h;,pochlorite, backtitration withExcessArsenite.. Excess xfvpochlorite , Potentiometric Backtitration T-jitli Arsenite ................... .................... Ixcoss h;vpochlorite , DeadstopBacktitration with. Arsenite 1 0 0 , 1 1 0 . 2 Z 0 .1 ^ 99 .6 _ C .2 100.0 1 0 .1 100 .0 _ 0 .1 100.0 Z 0.1 Tiie results listed are the average results of analyses made on samples containing from 10 to 50 ral. of O.Op normal hypophosphate solu­ tion. The average deviations are also given. The data, show that the method in wïiich an excess of hypociilorite is used is the most accurate and precise and that the procedure used to determine the excess hypochlorite has no influence on the results. The liypoclilorite method is much better tiisn tlie dicliromate method for several reasons: it is more accurate and precise, it is faster and carried out at room temperature, no blank otlier tlian an indicator blank is required, and the oxidant is less expensive, Tlie hypochlorite method is well adapted to the de­ termination of a large number of hypophosphate samples for the standing time and amount of excess hypochlorite can be varied within wide limits. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56 Several methods of equal accuracy s.nd precision are available for determining the excess hypochlorite. If the amount of excess hypo­ chlorite is approximately knovm it can be determined quite rapidly by a dea(3stop titration with standard arsenite for current readings need to be taken only near the endpoint. Plotting the changes in current is unnecessary even if tlie approximate endpoint is not known because tlie current changes before and after the endpoint are very pronounced. The dichrornate method is more difficult to use because of the heating required. It is not well adapted to samples of large volume. It may be used to advantage for the determination of hyp op ho spiiate s which are insoluble in the bicarbonate medium used in the hypochlorite method but which are soluble in acid solution. For routine determinations the blanks can be disregarded if precautions are taken to eliminate any extraneous reducing material from the reaction flask. The methods which have been developed for the determination are far superior to the gravimetric procedures which are commonly used, at least as far as the time required for the analysis is concerned. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57 V. DiiTEitMIî^ATION OF PHOSPHITE A phosphite was required as s. basis for the comparison of several methods for Its determination. A survey of the literature Indicated that a primary standard phosphita had not been prepared. Therefore, It wa.s necessary to prepare a phosphite which couJLd be made homogeneous and contain no substance which would Interfere with the methods used for t}ie determination of phosphite. For this purpose approxâmetely sO grams of reagent grade phosphorous acid was ground in a mortar and placed In a desiccator over phosphorus pentoxlde. The grinding and crying process was repeated several times until the small crystals could be poured freely. The phosphorous acid was weighed In a closed welgtilng bottle and allowed to stand In the open weighing bottle over phosphorus pentoxlde for a week. The 50 grams lost about 16 milligrams and It was considered that further drying was not necessary. The weighing bottle was then closed, the acid mixed thoroughly by shaking, and the closed bottle keist in a desiccator over phosphorus pentoxlde , Samples of the phosphorous acid were weighed as required to make sample solutions by pouring approximately the required amount Into a small dry weighing bottle which had been previously weighed, and weigh­ ing again to find the weight of the phosphorous acid. The transfer was made with as little exposure to the air as possible, for the acid was very hygroscopic . The vjeighed samples were dissolved In water and diluted to one liter to malce solutions of approximately 0 .025 molar or Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5b 0.05 normal in terms of oxidation of the phosphite to phosphate. The purity of the phosphorous acid was established by oxidizing the acid to orthophosphate and determining the phosphate as magnesium p^TTophosphate ^ a method which has been used by others to check volu­ metric results. One hundred ml. samples of a phosphorous acid solution containing 2 .2U21 grams of the acid per liter were evaporated to dryness three times with 30 ral, of concentrated hydrochloric acid and ten ral. of concentrated nitric acid after adding to each sample sufficient ms.gnesium cliloride equivalent to the formation of magnesium orthophos­ phate. The residue was dissolved in water, diluted to 100 ml., magnesia, mixture added, and magnesium ammonium phosphate slowly pre­ cipitated with ammonium hydroxide. The precipitate was allowed to stand overnight, filtered, washed with cold ten per cent ammonium hydroxide, dissolved in tr/drochloric acid and reprecipitated, ffter standing overnight the precipitate was filtered by means of a. sintered porcelain crucible, washed with cold ten per cent ammonium hydroxide and ignited to magnesium pyrophosphate at 1000° C . in a muffle furnace to constant weight. The per cent purity of the phosphorous acid was calculated from the weight of pyrophosphate obtained. The per cent purity was found to be 99.57, 99.51, 99.LG, 99.51 for four samples; the average was then calculated to be 99.52 per cent. It was assumed that the remainder of the phosphorous acid was water for it was not dried to an anliydrous state. Tests for other substances were not made . All results for the determination of phosphite reported in this thesis are expressed on the basis of the average of the gra.vimetric results. In each table. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59 the iTiilliequivalents of H3PO3 taken are the xnilliequivalents of H 3PO3 •wiiich were present in the sample taken for analysis. fhe composition of the phosphorous acid was further established by the method of Wolf and Jung (70) . Tliis method was revievmd by Jones and Swift (21:) who stated that it is accurate for the determination of phosphite. To measured samples of approximately 0,05 normal phosphorous acid in 250 iml. iodine flasks was added 50 ml. of 0,2 molar sodium bi­ carbonate solution wiiich was saturated with carbon dioxide. A measured excess of at least ten rnl. of 0.1 normal iodine solution was added to eacli flask. The flasks were allowed to stand in tiie dark for I|.5 to 60 minutes, and the excess iodine titrated with standard sodium arsenite solution, Tlie results were calculated on the basis of blanks determined in tiie same m ^ n e r as tiie samples, substituting water for tlie phospliorous acid solution. Tlie results are shoxm in Table XXI. TiibLE XXI DETEIMIN a TION Ml. 0.05 N HaPOa 25.00 25.00 25,00 25.00 25.00 of PHOSPHITE BY METHOD OF WOLF AND JUNG Ml. 0,1 N Iodine 25.00 25.00 25.00 25.00 25.00 Meq. H 3PO3 Taken 1.360 1.360 1.360 1.360 1.360 . H 3PO3 Found Percent Meq . 1.359 1,360 1.360 1.359 1.359 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99.9 100.0 100 ,0 99.9 99.9 60 The average per cent found was calculated as 99-9 per cent from the data in Table X X i . The results by the Wolf and Jung method there­ fore agree well with the results of the gravimetric method. 1. Oxidation of Phosphite By Excess Potassium Dichromate in Sulfuric Acid Solution Preliminary experiments were conducted to determine if phosphorous acid is oxidized by excess potassium dichromate in sulfuric acid solu­ tion and to determine the effects of time of heating and concentration of sulfuric acid, A solution of approximately 0.025 molar phosphorous acid was prepared from reagent grade crystals of the acid wliich were not dried and -viiich were moist on the surface . The undried reagent was used only in these preliminary experimentsj the dried reagent was used in all other procedures dealing viith the determination of phosphite. measured samples (19.98 ml.) of the 0.025 molar solution were treated with 19 .96 ml. of 0.1 normal potassium dichromate and sufficient con­ centrated sulfuric acid to make the resulting mixture 12 normal in sulfuric acid. The mixtures were heated in a boiling water bath and the excess diclirom ate determined iodometrically after neutralizing part of the acid as previously described in the section of this thesis dealing with the determination of hypophosphate by excess dieIrornate. The results are shoifn in Table XXII. The procedure was repeated using different concentrations of sul­ furic acid and various heating times. Each sample contained 19.98 ml. of the 0.025 molar phosphorous acid and 19.98 ml. of 0,1 normal di­ chromate. The results are shown in Table XXIII. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61 TABLE XXII OXIDATION OF PHOSPHITE BY EXCESS DICHROMATE IN 12 NORMAL SuLFURIC ACID AS A FUNCTION OF TIME OF HEATING Time of Heating Min. Meq. H 3PO3 Found 5 10 15 30 60 90 120 0.913 0.966 0.98U 0 .985 0 .986 0 .986 0.986 TiHLE XXIII OXIDATION OF PHOSPHITE BY EXCESS DICHROMATE AS A FUNCTION OF SULFURIC ACID CONCENTRATION AND TIME OF HEATING Approximate Normality H 28O4 12 12 6 6 6 3.6 3.6 Time Heating Hours 1 1 0.5 1 1.5 0.5 1 Meg. H 3PO3 Found 0.985 0.989 O.8U1 0.9h6 0.980 0 .61L 0 .8L6 Tables XXII and XXIII indicate that phosphorous acid is completely oxidized by excess diclirom ate in 12 normal sulfuric acid in 30 to minutes and-that if the oxidation is to oe complete in one hour, the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62 sulfuric acid concentration must be at least above six normal. The preliminary experiments were carried out to determine the conditions necessary to oxidize phosphorous acid with excess dietirornate . The composition of the phosphorous acid was not accurately known, but it was assumed that the oxidation was complete if no further oxidation took place with a longer heating period using the same acid concentration. Solutions of phosphorous acid were prepared from the dried acid. To measured samples of the solutions were added a measured excess of standard potassium dichromate and sufficient sulfuric acid to make the resulting mixture 12 nonnal in sulfuric acid. The mixtures were lieate d for one hour in a boiling water bath and the excess dichromate was determined by a standard ferrous ammonium sulfate solution as previously described in the section of tliis thesis dealing with the determination of hypophosphate i-xith excess diclirom ate. Table XXIV shows that if the blanks are taken into consideration, the results are in better agreement with the gravimetric results than if the blanks are neglected. The determination was repeated by the same procedure except that the excess diclirom ate was determined iodometrically after neutralizing part of the sulfuric acid as described in the section of tliis thesis dealing with the determination of liypophosphate with dicliromate. The results are shown in Table XXV. Table XXV shows that phosphite is completely oxidized by excess dichromate in 12 normal sulfuric acid in minutes at the temperature of a boiling water bath. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63 TABLü XXIV D EI'ELIM IN ATIO N OF PHOSPHITE BY EXCESS DICrlROMATE IN 12 NOKM/..L 3U IF U R IC A C ID (Ferrous Sulfate Backtitration) Conditions : 1:1 . 0.05 W ii3PO2 Ml. 0.1 N K gCr aO«p 19.98 19.98 19.98 25.00 25.00 25.00 25.00 19.98 19.98 19.98 25.00 25.00 25.00 25.00 Time of Heating One Hour Meq. H 3PO3 Taken 1.171 1,171 1.171 1.376 1.376 1.376 1.376 _ HaPOa., Found Me q . Percent 1.172 1.171 1.171 1.380 1.381 1.377 1.377 100 .1-ÏÏ 100 .0* 100 .0* 100.2 100.3 100.2 100.2 Results calculated on the basis of blanks. Ti3LE XXV DEC ELIMINATION OF PHOSPHITE BY EXC ESS DICHROMATE IN 12 NOHHAI SUIi'"uRIC ACID (lodometric Backtitration Ifter Neutralizing Part of The Acid) Conditions : Time of Heating 1+5 Minutes Ml. 0.05 N H3PO3 H I . 0.1 N KgCr gO-7 Meq. H 3PO3 T aken 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 1.376 1.376 1.376 1.376 1.376 1.360 1.360 1.360 1.360 1.360 Na?P.3 Found (a) Percent Me q . 1.377 1.376 1.373 1.377 1 .375 1.361 1.360 1.361 1.360 1.357 * Heated one hour. (a) Correction for blanks. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100.1 100 .1 99.8 100 .1 99.9 100.1 100.0 100.1 100 .0 99.8 6h B. Oxidation of Phosphite by Excess Hypochlorite In The Presence of Bromide hltrlte and arsenite are oxidized rapidly b y hypochlorite in a solution containing bromide and bicarbonate (29). It seemed reasonable to expect that phosphite may be oxidized under the same conditions. To measured samples of 0.05 normal phosphorous a.cid in 200 ml. Erlenmeyer flasks were added one gram of sodium bicarbonate and one gram of potassium bromide . Standard sodiuia hypochlorite was added from a burette until a sliglit yellow color appeared, indicating a slight excess of hypochlorite. The hypochlorite could be added to the solution as rapidly as the burette would deliver it, and the yellow color would not persist until there was an excess of hypochlorite . The flasks were alloi-ied to stand for two to three minutes, a small measured excess of standard arsenite added, and the excess arsenite titrated id.th the standard hypochlorite from the same burette using Bordeaux as indicator. The volume of the solution was adjusted to 50 to 75 ml. at the endpoint. In indicator blank of 0,03 ml. of 0.1 normal h^^pochlorite was sub­ tracted from the volume of hypochlorite used. An excess of about 0.5 ml, of 0.1 normal hypochlorite was used with each sample. The data in Table XXVI were obtained using two different phos­ phorous acid solutions of slightly different concentration but prepared from the same reagent. Table XXVI shows that phosphite is completely oxidized by a slight excess of hypociilorite in the presence of bromide and the results obtained agree fa.irly well with tlie results obtained by the gravimetric and Wolf and Jung methods, The data also show bhat Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i 65 TABLE XXVI DETERiuINATION OF PHOSPHITE BY A Sî^li'-LL EXCESS OF HYPOCHLORITE IH A SOLUTION CONTAINING BROMIDE AND BIC/HBONATE Ml. Ü .05 N H 3P03 9.99 9.99 19.98 19.98 19.98 19.98 19.98 19.98 19.96 19.98 25.00 25.00 I49.98 a9.98 (a) (b) (c) (d) Meq . H 3P O 3 T aken 0.616 0.616 1.231 1.231 1.231 1.166 1.168 1.166 1.168 1.166 1.5L0 1.5U0 3.060 3.060 ... Meq . Found Percent 0.617 0.616 1.230 1.230 1.229 1.168 1.168 1.152 1.168 1.155 1.538 1.538 3.076 3.076 100.1 100.0 99.9 99.9 99.8 100.0 100.0 97.8 100.0 98.9 99.8 99.8 99.9 99.9 (a) Cb) Cc) Cd) Allowed to stand 10 m i n . before adding arsenite, Direct titration, endpoint indefinite. 30 ml. buffer pH 7 -U used in place of bicarbonate. KBr added after arsenite instead of with sample. phosphite is not completely oxidized in two to tiiree minutes by a small excess of hypochlorite unless bromide is present. In order to determine the stability of the phosphorous acid used for this work, the acid was determined by the hypochlorite-bromide method after the acid had been standing for four months in a closed weighing bottle which was kept in a desiccator over phosphorus pentoxide The results are shown in Table XXVII. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66 TABLE XXVII D jJ T Æ I W A T I O N o f stability of phosphorous acid o n STANDING^ USING HYPOCHLORITE-BROMIDE METHOD OF ANALYSIS Ml. 0.05 N H 3PO3 Meq. HaPOs Taken 25.00 25.00 25.00 25.00 25.00 1.376 1.376 1.376 1.376 1.376 „ HaPOa, Found Meq . Percent 1.37U 1.376 1.376 1.375 1.375 99.9 100.0 100 .0 99.9 99.9 Table XXVII shows that there was no apparent change In the composi­ tion of the phosphorous acid during four months, for the results agree with the results in Table XXVI which were obtained four months earlier. Direct titration of phosphite in a bromide—bicarbonate medium with hyrpochlorite . Attempts were made to titrate phosphite directly in a solution containing sodium bicarbonate and potassium bromide using a. solution of Bordeaux to determine the endpoint. The reaction near the equivalence point is so slow that the indicator was destroyed before a true endpoint was reached. Several drops of indicator were added and decolorized before an excess of hypochlorite could be detected. If a reversible indicator were available for titrations with liypochlorite a direct titration may be possible. The possibility of using the deadstop technique to determine the endpoint in a direct titration of phosphite in a bromide-bicarbonate medium with hypochlorite was investigated. It was found that when a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67 small potential was applied, to platinum electrodes immersed in a phosphorous acid solution containing sodium bicarbonate and potassium bromide, no cui'rent flowed as determined by a Fisher Elecdropode. However, one drop of 0,1 normal sodium hypochlorite in a. bicarbonatebromide medium, produced a current under the same conditions. It was then considered that a very slight excess of hypochlorite may be detected by the deadstop procedure. To measured samples of 0.05 normal phosphorous acid in 150 ml, beakers vrere added one gram of sodium bicarbonate and one gram of potassium bromide. Platinum electrodes were immersed in the solution and a. potential of h$0 millivolts applied by means of a Fisher Elecdro­ pode . The solution was stirred with a. magnetic stirrer. upon the addition of standard sodium hypochlorite from a burette , a current flowed which decreased to zero in a. few seconds, bear the equivalence point in the titration the addition of a drop of hypochlorite produced a large increase in current which rapidly decreased to zero when the addition of hypochlorite was stopped. The titration was continued until a fraction of a drop of hypochlorite caused a steady current to flow. This point was taken as the equivalence point. If the Irypochlorite was added until within 0.5 ml. of the endpoint, the titration could be com­ pleted in two to tliree minutes, The full sensitivity of the instrument was used near the endpoint, The hypochlorite solution was standardized against standard sodium arsenite using Bordeaux as indicator, Table XXVIII shows that phosphite can be determined by a direct titration with hypochlorite in a solution containing bromide and bi­ carbonate , Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 68 TABLE XXVIII DIRECT TITRATION OF PHOSPHITE WITH HYPOCHLORITE HI A SOLUTION CONTAINING BROMIDE USING A DEADSTOP ENDPOINT Ml. 0.05 M H 3PO3 . Meq. HgPOg Taken 19.98 19.98 19.98 19.98 19.98 19.98 19.98 19.98 1.101 1.101 1.101 1.101 1.101 1.101 1.101 1.101 Meq. H3PO3 Found Percent 1.099 1.097 1.100 1.099 1.099 1.100 1.100 1.098 99.8 99.6 99.9 99.8 99.8 99.9 99.9 99.7 In order to have a. deadstop endpoint in which a current flows between the electrodes before and after the endpoint with zero current at the endpoint it is necessary that there be a system capable of under­ going electrolysis before and after the endpoint (6U). The current changes observed on the addition of sodium arsenite to sodium hypo­ chlorite are typical of this type of endpoint. Sufficient information is not available to predict just what the electrolysis before the end­ point involves because of the variable oxidation states of ciilorine but after the endpoint the electrolysis presumably involves the arseniteersenate system. Another type of deadstop endpoint, such as that encountered in the titration of phosphite with hypochlorite in the presence of bromide, is characterized by no current flowing between the electrodes before the endpoint and a current after the endpoint. The requirements for Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69 this type of endpoint are that s system incapable of electrolysis imder the conditions used must be present before the endpoint and one vrtiich can be electrolysed after the endpoint. The phosphite and phos­ phate do not undergo electrolysis, and as soon as there is a permanent slight excess of hypoclilorite present, a system is available for electrolysis and a current flows. The electrolysis after the endpoint may involve oxidation states of bromine because of the hypobromite present from the reaction of the first trace of excess hypochlorite viith the bromide present. C. Oxidation of Phosphite with Sodium Hypochlorite In order to determine if there is an optimum pH at wiiich phosphite is oxidized by hypochlorite, a series of buffers was prepared and the oxidation carried out with excess hypochlorite in the presence of each buffer. pH meter. The pH of each buffer was determined with a Beckman Model H-2 Because in some cases the pH of the mixture of the buffer and tiie other components of the reacting mixture was slightly different than the pH of the buffer itself, the pH of each reacting mixture was also measured. To 19.98 ml. samples of 0.0$ normal phosphorous acid solution in 2$0 ml. iodine flasks wpre added $0 ml. of buffer solution and 25.00 ml. of 0.1 normal sodium hypochlorite. The gutters were filled with saturated potassium iodide, solution and the flasks were allowed to stand in the dark for 20 minutes. The potassium iodide solution was then allowed to flow into the flask, 20 ml. of glacial acetic acid Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70 added, and the liberated iodine titrated with standard sodinm thiosulfate after about one minute. The hypochlorite solution was standardized against the standard thiosulfate in two ways. In one, a measured volume of the hypochlorite was added to $0 ml. of water, 10 ml. of saturated potassium iodide solution added, and then 20 m l . of glacial acetic acid. After about one minute the liberated iodine was titrated vrith thiosulfate. In the other method, $0 ml. of buffer of pH 7.0 was used in place of the $0 ml. of water. The comparison of the hypochlorite with the thiosulfate was made several times using different buffers and in every case the amount of thiosulfate required for a certain volume of hypoclilorite solution was smaller when a buffer was used than was the case for a comparison made without the buffer. Also, if the hypochlorite was allowed to stand tcLth a certain volume of buffer, the time of standing had no measurable effect on the volume of thiosulfate required. Table XXIX .shows that the oxidation is more complete in the pH range of 7.0 to 7,6 and that the oxidation takes place only to a very small extent at a pH above 12 . It also shows the large difference in results obtained if the hypochlorite is standardized by the different methods discussed. It would be expected that better results would be obtained if the comparison of the hypochlorite and thiosulfate was made in the presence of a buffer as in the determination of the excess iiypochlorite used with the phosphite samples, or in other words, by a blank. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71 TABLE XaiX EFFECT OF pH OK THE OXIDATION OF PHOSPHITE BY EXCESS HYPOCHLORITE Conditions: 20 minute standing. 150^ excess KaClO pH Percent Oxidized (a) 5.9 6.1 6.6 7 .0 7 .3 7.6 7.6 8.9 9.2 12.1 100.2 100.0 100.1 100.6 100.5 100.5 100.6 56.7 37.2 1.0 Percent Oxidized (b) 99.If 99.2 99.2 99.7 99.7 99.7 99.7 55.9 3 6 .il l.ii * Buffers of pH range S .9 to 6.6 were prepared from EsHgPO^ and I'isoHPO^, pH 7.0 to 8.9 from KH^PO^ and HaDH, pH 9.2 from H 3BO 3 and HaOH, and pri 12 .1 from NaOH and acetic acid. (a) Basis of standardization of NaClO in the a.Dsence of a, buffer I'o) Basis of standardization of NaClO in buffer pH 7.0. In order to determine more closely the optimum pH at which phosphite is oxidized by liypochlorite, another series of buffers was prepared and the determinations repeated by the same procedure. The results are shox-m in Table XXX. Table XXX shoxfs that the oxidation of phosphite by excess hypo­ chlorite is most complete over a pH range of ( .3 to 7,8. ihe results in this table cori’espond more closely to the results obtained oy the gravimetric method if they are calculated on tiie basis of the normality Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72 EFFijCT ÜF pii ON TNF 0XID7:.TI0W OF PHOSPHITE BY EXCESS HYPOCHLORITE Conditions r Time of standing 20 min. , 130^ excess hypochlorite. pH^" Percent Oxidized (a) 6.9 7.3 7.5 7.8 8.1 99.6 100.0 100.0 100.0 96.0 Percent Oxidized (b) 99.0 99.lt 99.lt 99.lt 95.lt Bulfer pH 6.9 was prepared, from NaHgPO^ and Fa^HPO^ pH 7.3 to b.l from rjaHgFO^ and NaOH. (a) Basis of standardization of NaClO in the absence of a buffer (b) Basis of standardization of HeClO in buffer pH 7.5. of the ir^'pochlorite determined in the absence of s. buffer whereas the opposite was true in Table XXIX. The only difference was in the reagents from which the buffers were prepared. A series of phosphite samples was determined by the same procedure using buffer 7.8; it was the same buffer that was used for the data in Table XXIX. The results obtained agreed with the results in Table XXIX, that is, better results were ob­ tained on the basis of the normality of the hypochlorite determined in the presence of the buffer than in the absence of the buffer, This would seem to indicate that the extent of the oxidation depends not only on the pH of the solution but also on the buffer u sed. The accuracy of the method depends upon an accurate iadometric determination of hypochlorite in the presence of a ouffer. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ihe hypochlorite 73 used canno’b be considered as a solution of a pure compound for it may contain variable amounts of chlorate and chlorite depending on tlie age of the solution and on the age of the Clorox from idcich the solu­ tion was prepa,red, hypochlorite and chlorite are determined, iodo- metrically in acetic acid solution however Kolthoff and Sandell ( 3 0 ) state that chlorite liberates iodine from an iodide solution slowly in acetic acid solution. Chlorate liberates iodine from iodide only from a solution made highly acidic with a strong acid. In a determination of available chlorine in a bleaching solution using arsenite in a bi­ carbonate m e d i u m t h e r e is no interference from chlorite or cilorate and only the cllorine from the hypochlorite is determined. The iodometric determination of available chlorine i-ra.s carried out under various conditions . It was found tha.t the volume of thio­ sulfate solution used for the same volume of h;ypiochlorite varied with the a.cid used to acidify the potassium iodide solution. A smaller volume of thiosulfate was used i-rith acetic acid than tcLth sulfuric acid and if the concentration of sulfuric acid was increased, a slightly larger volume of tillosulfate was used. The effect of the presence of buffers on the iodometric determina­ tion of hypochlorite was determined. It was found that the volume of tliiosulfate solution required for a constant volume of hypochlorite was smaller if a buffer of pH 5.7 to 6.5 was present than if it was not present. This effect was found if the solution was acidified with either acetic acid or sulfuric acid before the liberation of the iodine. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tit The effect of buffers is shown by the following . P. solution of three grams of sodium bicarbonate in $0 m l . of water was added to each of several iodine flasks. Different weighed amounts of reagent grade monopotassium phosphate were added followed by 19.93 ml. of 0.1 normal hvpochlorite . Ten ml. of saturated potassium iodide solution was added to each flask and then 25 ml. of glacial acetic acid carefully added. The liberated iodine was titrated with 0.1 normal sodium thiosulfate solution. The results are shown in Table XXXI. TiiBLE XXXI EFFECT OF BUFFERS Oh THE IODOMETRIC DETERMINATION OF HYPOCHLORITE Grams XH2PO4 Ml. 0.1 h I'ia.^SgO 2 0 1 2 3 k h 2C, .82 20.82 20.80 20.79 20.75 20.75 Table XXXI shows that smaller amounts of iodine are liberated from a potassium iodide solution by a constant amount of hypochlorite in a bicarbonate medium if increasing amounts of monopotassium phosphate are added. It means that the buffer either interferes with the liberation of iodine or affects the titration of iodine with thiosulfate. The results are opposite to what would be expected as the pH of the solution Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73' decreases with increasing sirionnts of the phosphate and if the effect is one of pH only, it would not be expected that a smaller amount of iodine would be liberated at a. lower p H . /. possible explanation of the effect is that the carbon dioxide in solution from the reaction between the bicarbonate and phosphate liberates hypochlorous acid which decomposes, Hypochlorite was determined iodometrically in the presence of in­ creasing amounts of haHgPO^ and it was found that the amount of phosphate present, at least up to four grams, had no measurable effect on the results. The same was K2HPÜ4 or NagHPO^ were found to be the case when increasing amounts of used. However, when a buffer prepared from mono and disodium phosphates was used, the amount of iodine liberated was less than was the case if the buffer were not present. This suggests tiiat the amount of iodine liberated depends upon the pH of the solution. The effect of using different acids in the iodometric determination of hypochlorite was determined. To 50 ml. of various buffers were added 19.98 ml. of a fresiily prepared 0.1 normal hypoclilorite solution, two grams of potassium iodide, and 20 ml. of either glacial acetic acid or six normal sulfuric acid, The liberated iodine was then titrated with a thiosulfate solution. in place of the buffer. In one sample 50 ml. of water was used The results' are shown in Table XXXII. Table XXXII shows tha.t when the iodometric determination of hypo­ clilorite is carried out in the presence of a buffer s smaller amount of thiosulfate is used than if the buffer is not present and also that when a buffer is present the type of acid used has little effect on the results. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76 ÏVæLE XXXII EFFECT OF ACID USED ON IODOMETRIC DETER14INATI0N OF HYPOCHLORITE IN THE PRESENCE OF BUFFERS pH of Buffer Acidified With Acetic Acid Ml. NazSgOs Acidified With Sulfuric Acid M l . NagSgOa 18.80 ■ 18.70 18.71 18.72 16.81; 18.70 18.70 18.69 Water 5.7 6.6 6.5 It also shows that when a buffer is not present, a larger amoiint of thiosulfate is used when the solution is acidified with sulfuric acid rather than acetic acid. Because of the erratic results found in the determination of phos­ phite by excess hypochlorite in the presence of buffers, the determina­ tion was attempted in a bicarbonate medium. The results in Dicarbonate medium were not as erratic as the results found using buffers but the presence of the bicarbonate had an effect on the iodometric determination of hypochlorite similar to that found when a buffer was present. The results for the determination of phosphite were in better agreement i-irith the results obtained by other methods if the comparison of the hypo­ clilorite and thiosulfate solutions wa.s made in the absence of bicarbonate, Experiments were conducted to study the oxidation of phosphite by hypochlorite with no buffer or bicarbonate present. To measured samples of 0.05 normal phosphorous acid solution in 250 ml. iodine flasks was Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77 added a measured excess of standard sodium hypochlorite. The flasks were allowed to stand in the dark for various time intervals, ten ml. of saturated potassium iodide solution and 20 ml. of glacial acetic acid added and the liberated iodine titrated with a. standard sodium thiosulfate solution. Blanks were determined in the same manner sub­ stituting water for the phosphorous acid solution and calculations were made on the basis of the blanks. The results are shown in Table XXXIII, TidLE XXXIII DETEkMINATIGN OF PHOSPHITE B£ EXCESS HYPOCHLOklTE IN AN UNBHFFEHED SOLUTION hi. 0.05 N H3PO3 hi. 0.1 h NaClO 19.98 19.98 19.98 19.98 25.00 25.00 25.00 25.00 Meq. H3PO3 Taken Meq. H3PO3 Found 1.177 1.177 1.177 1.177 Tine Standing Min. 1 .17U 1.176 1.175 1.176 3 5 10 Uo Percent Oxidized 99.7 99.9 99.8 99.9 Table XXXIII shows that phosphite is rapidly oxidized by sodium hypoclilorite in an unbuffered solution. These data also support the conclusion that the iodometric determination of hypochlorite in the presence of buffers is erratic. A large number of phosphite samples were determined by excess hypochlorite in the presence of buffers and sodium bicarbonate and the results were inconsistent. Chulski (12) experienced the same irregularities in determining the excess hypo­ chlorite used in oxidations of hypophosphate in the presence of buffers and sodium bicarbonate. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78 The determination of lypochlorite by a. comparison against standard sodinm arsenite gave consistent results if the determination was made in the presence of bicarbonate or buffers of pH approximately six to ten, and gave results wiiich agreed with the iodometric determination made without a buffer present and acidified with acetic acid. For this reason, during this work, an excess liypochlorite was determined by the arsenite method rather than by the iodometric method. The iodometric determination of hypochlorite in the presence of buffers requires more s tudy. The determination of phospliite by excess hypochlorite has no ad­ vantage over the determination using a slight excess of hypochlorite in the presence of bromide . The method in wiiich bromide is used is somevrtiat shorter and is easier to carry out. D. Oxidation of Phosphite By Excess Ceric Sulfate in Sulfuric Acid Solution The details of oxidations using excess ceric sulfate in sulfuric acid solution are more fully discussed in the section of this thesis dealing with the determination of hypophosphite by ceric sulfate. Experiments were conducted to determine if phosphite were oxidized by excess ceric sulfate in sulfuric acid solution. To measured samples of 0,05 normal phosphorous acid were added a measured excess of 0.1 normal ceric sulfate and ten ml. of concentrated sulfuric acid. The mixtures were heated under reflux for one hour and the excess ceric sulfate determined by a titration with standard ferrous ammonium sulfate Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79 BS described in the section dealing >d.tij the determination of hypophospliite by ceric sulfate. Blanks were determined by the same procedure, substituting water for the phosphorous acid solution, and calculations made on the basis of the blanks. The results are siio>m in Table ZXXIV . TABLE XXXIV DEI’EmahATION OF PHOSPHITE WITH EXCESS CERIC SULFATE IH SULFURIC ACID SOLUTION Conditions; 6 h 13804, boiling one hour. hi. 0.0$ N HsPOa Ml. 0.1 N Ceric Sulfate 19.98 19.98 19.98 19.98 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25 .oo 25.00 25.00 Meq. HsPOg T aken 1.101 1.101 1.101 1.101 1.376 1.376 1.376 1.376 HsfOs. Found Percent Meq. .. . 1.095 1.099 1.092 1.096 99.6 99.9 99.5 99.6 1.376 100.0 1.372 99.7 99.9 99.3 1.37U 1.367 * % 3 normal sulfuric acid used. Table XXXIV shows that phosphite is oxidized by excess ceric sulfate in sulfuric acid solution, but as a quantitative method, the results are slightly louer and less precise than the results obtained by other methods for the determination of phosphite . Preliminary experiments showed that the oxidation requires at least one hour at boiling tempera­ ture in six normal sulfuric acid. It is possible that the oxidation Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 would be faster and more complete in a. stronger acid solution than six normal but the data in Table XXXIV show that there is little change in results on increasing the silLfuric acid concentration from tlaree normal to six norinal. The method offers no advantage over the method using an excess hypochlorite in the presence of bromide and is much more inconvenient to use. Summary of Volumetric Methods For The Determination of Phospliite In Table XXXV is shown a summary of results obtained by the volu­ metric methods developed for the determination of phosphite . The results are based on the gravimetric procedure in vrtiicli the phosphite is determined as magnesium pyrophosphate . Results obtained by the method of Wolf and Jung are shoxne in the same table so that the volumetric methods developed may be compared xfith a reported volumetric method. The average results listed are averages of results obtained using the methods on several different solutions of phosphorous acid prepared from the same reagent. The average deviation for each method is also shoi^m in Table XXXV . Table XXXV shows that the results obtained by the volumetric methods are in slightly better agreement i^th the results obtained by the Wolf and Jung method than with the results of the gravimetric method and are in general slightly lower than the results of the gravimetric method. The reason for this may be that the phosphorous acid contained a small amount of pliosphate. Table XXXV shows that for accurate results, the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81 TA B LE XXXV SLlliiulY OF RESULTS OF VOLUMETRIC METHODS FOR THE DETERMINATION OF PHOSPHITE Method Average Percent __________________________________________________________________________ Wolf and Jung 99.9 « 0 . 1 Excess Bichromate, Ferrous Sulfate Baclctitration (ho Blank Corrections)............................100.2 1 0.1 Excess Diclrrornate , Ferrous Sulfate Backtitration (Blank Corrections)...............................100.0 + 0 . 1 Excess Diclu’omate, Iodometric Backtitration (Blank Corrections)...... ....................... 100 .0 _ 0 .1 Slight Excess NaClO in Presence ofBromide 99.9 2 O .1 Direct Titration with NaClO,De ads topEndpoint 99.8 i 0.1 Excess Ceric Sulfate in H 3SO4 Solution (Blank corrections).............................. 99.7 1 0.2 blanks should be taken into consideration in the dichromate method. The results obtained by the ceric sulfate method are lower and less precise than the results by other methods , As in the determination of hypophosphate, the use of a standard hypochlorite solution is most convenient for the determination of phosphite . The oxidation of phos­ phite by hypochlorite in the presence of bromide, or essentially by hypobromite, is rapid and opxantitative . The Wolf and Jung method for determining phosphite appears to be accurate and precise but the method is not as convenient to use as tiie hypochlorite-bromide method. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82 V I . DErERi4lKATI0N OF HYPOPHOSPHITE A hypophosphite was required as a. basis for the comparison of several methods for its determination, A survey of the literature indicated that a primary standard iiypophosphite has not been prepared. Therefore it was necessary to use a hypophospliite which could be made homogeneous and contain no substance wliich would interfere vrith the methods used for its determination. Sodium hypophosphite monohydrate, haHgPO^"H^O ^ was used for this purpose. The reagent was prepared by reerystallizing the C.P. reagent tliree times from redistilled ethyl alcohol and air drying the product. The purified reagent was stored in a tightly stoppered bottle. Brooks (9) found that the reagent prepared in tiiis manner did not contain phosphite and that it could not be completely changed to the anhydrous salt without decomposition by oven drying. Standard solutions of the purified reagent were prepared by dissolving weighed amounts of the salt in freshly boiled water and diluting to one liter with the boiled water. The composition of the sodium hypophosphite was established by oxidizing aliquots of a solution of the hypophosphite to orthophosphate by repeated evaporations to dryness with aqua, regia. The orthophosphate was then precipitated twice as magnesium ammonium phosphate and ignited to magnesium p^rrophosphate as outlined in the procedure for the gravi­ metric determination of phosphite . The composition of the sodium hypo­ phosphite was determined from the weight of pyrophosphate obtained. The results are shown in Table XXXVI. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83 T/BLE ]OŒVI GRAVIMETRIC DETEIR-IINATIOH OF HYPOPHOSPHITE 1 Grams Hypophosphite Taken Grams Mg^P^O^ Found Per cent NaHsPOg Per cent NaHgPOg.HaO 0.3000 0.3225 85.00 102 .U Sampl e Humber 2 3 O .3000 0.3226 85.03 102 .U il 0 .3000 0.3221 84.89 102.3 0,3000 0.3222 8L.92 102.3 Table XXXVI shows that the sodiinn hypophosphite was partially dehydrated and could not be considered as a mono hydrate. The average per cent HaHgPOg was calculated from the data, in Table XXXVI to be 6ii.96jo, and the average per cent haH^POg’HgO was 102 Data for the volumetric determination of hypophosphite listed in the following tables are based on the average of the gravimetric results. In each table the milliequivalents of haHgPOg taken are the milliequivalents of NaHgPOg wiiich were present in the sample taken for analysis. The gravimetric results were compared with those obtained by the method of Wolf and Jung C?0) . This method was recently reviewed by Jones and Swift (2U) who pointed out that the method, as used by Wolf and Jung, gave results which were 0 .U per cent lower than the corres­ ponding gravimetric results. Tne data of Wolf and Jung show that the oxidation of hypophosphite to phosphite in the first part of the de­ termination is very slow and that it is impossible to find the time in wliich oxidation is complete to phosphite because of the simultaneous oxidation of the phospMte to phosphate. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8U The sodium hypophosphite was determined according to the procedure described by Wolf and Jung. To measured samples of 0.07 normal hypo­ phosphite in 250 m l . iodine flasks were added ten ml. of six normal sulfuric acid and a measured excess of at least ten ml. of 0,1 normal iodine solution. The flasks were tightly stoppered, the gutters filled with water, and the flasks allowed to stand for 12 hours. The flasks were then opened and a slurry of sodium bicarbonate carefully added until the evolution of carbon dioxide ceased. Fifty ml. of 0.2 molar sodium bicarbonate solution saturated with carbon dioxide was added to each flask, the flasks tightly stoppered and allovred to stand for one hour. The excess iodine was then titrated with standard sodium arsenite using freslily prepared starch solution as ind.icator. Calculations were made on the basis of blanks treated in the same manner as the samples, substituting water for the hypophosphite solution. Table XXXVII shows that the results obtained by the method of Wolf and Jung are approximately 1.3 per cent lower than the results of the gravimetric method. A disadvantage of the Wolf and Jung method is the loss of iodine wiien the acid solution of iodine is treated with sodium bicarbonate . This loss depends on the amount of iodine present and on the temperature. The only way that the method could be used accurately would be to have the same amount of iodine present in the blanks that would be present in the sample after the oxidation of the hypophosphite to phosphite. For an unknown sample, this would be rather difficult to arrange . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85 t;j3Le XXXVII DiSrm^mvJATION Ml. 0.07 N NaHgPOg 19.98 19.98 19.98 19.98 19.98 19.98 19.98 of hypophosphite by the method of wolf and JUivG Ml. 0.1 h .Iodine Meq. WaHgPOg Taken 25.00 25.00 25.00 25.00 25.00 25.00 25.00 1.5L5 1.5L5 1.5U5 1.545 1.545 1.545 1.545 NaHgPOa Found Me q . Percent 1.530 1.524 1.528 1.525 1.525 1.522 1.524 99.0 98.6 98.9 98.7 98.7 98.5 98.6 Volumetric methods for the determination of hypophosphite developed during this work and described in the following pages of this thesis support the conclusion that the Wolf and Jung method for determining hj^ophosphite gives low results. A. Oxidation of Hypophospliite by Excess Dichromate In Sulfuric Acid Solution Experiments were conducted to determine if hypophosphite is oxidized by excess potassium dichromate and to determine the time required to complete the oxidation in 12 normal sulfuric acid at the temperature of a boiling water bath. To measured samples of 0.05 normal sodium hypophosphite were added a measured excess of standard potassium dichromate and sufficient con­ centrated sulfuric acid to make the resulting mixture 12 normal in acid. The mixrbures were heated in a boiling water bath for various time Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86 intervals and. the excess dichromate determined iodometric ally after neutralizing part of the sulfuric acid present as described in the section of this thesis dealing with the determination of hypophosphate by excess dicliromate . Blanks were determined in the same manner, sub­ stituting water for the hypophosphite solution. The results are shown in Table JOOCVIII. TABLE XXXVIII DETERITINATION OF HYPOPHOSPHITE BY EXCESS DIGHROi-lATE IN 12 NOIMAL SULI’UHIC ACID AS A FUNCTION OF TIME OF.HhATING Ml. 0.05 H NaHgPOo (a) 25-00 25.00 25.00 25.00 25.00 19.98 19.98 19.98 19.98 Ml. 0.1 Ai KaCr gO^ 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 Meq. NaHgPOg T aken 1.282 1.262 1.282 ' 1.282 1.282 1.5L5 1 .5U5 1.5L5 I.5b5 Heating Time Min. 10 15 25 ho 70 60 60 60 60 N aH gPO Me q . i.lUo 1.269 1.261 1.278 1.283 1.5^1 1.5L3 1.5U2 l.5b3 Found Percent 88.9 99.0 99.9 99.6 100 .1 99.7 99.9 99.7 99.9 (b) (b) (b) (b) (a) The 19.98 ml. samples were from1 a. different hypophosphite solution than the 25.00 ml. samples; the solutions were prepared from the same salt. (b) Calculations made on the basis of blanks. Table XXXVIII shows that hypophospliite is completely oxidized by excess dicliromate in one hour a.t the temperature of a boiling water ba.th and that the oxidation is not complete in 15 minutes . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87 B. Oxidation of Hypophosphite by Excess Sodium Hypochlorite in a Bicarbonate Medium Scln-d-cker (5?) found that phosphite could be oxidized to phosphate in a bicarbonate medium with no interference from h^'pophosphite . To measured samples of a. sodium hypophosphite solution were added one gram of sodium bicarbonate and a measured excess of sodium hypocllorite solution. After standing for different time intervals an excess of standard sodium arsenite was added, and then one gram of potassium bromide . The excess arsenite was titrated with the standard hypochlorite solution using Bordeaux indicator. It was found that after two and one-half hours tlie Ig/pophosphite was oxidized about 16 per cent and after 13 hours about $0 per cent assuming that it was oxidized to phosphate , It may be possible to determine phosphite with excess hypochlorite in the presence of a small amount of hypophospilte with little interference from the latter, out it seems unlikely that the determination could be made in the presence of a. large amount of hypophosphite . C. Oxidation of Hypophosphite by Excess Hypochlorite in the Presence of Bromide Experiments were carried out to study the oxidation of hypophos­ phite by a slight excess of hypocllorite in a. solution containing potassium bromide and nodi urn bicarbonate. It was found that like hypo­ phosphate, but unlike phosphite, the oxidation was very slow and the procedure used for the determination of phosphite was not successful. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88 The oxidation was found to be more rapid when a. large excess of hypo­ chlorite was used in the presence of bromide ^ but the large loss of hypochlorite under these conditions made the method unreliable, D. Oxidation of Hypophosphite by Excess Hypochlorite in Sulfuric Acid Solution In order to use hypoclilorite in a sulfuric acid solution, it was first necessary to determine the stability, or change in oxidizing capacity, of hypoclilorite in an acid solution. To 19.98 ml. samples of 0,1 normal sodium hypoclilorite in 2^0 ml. iodine flasks were added 25 ml. of water and 10 ml. of six normal sulfuric acid. The flasks were tightly stoppered, the gutters filled with saturated potassium iodide solution, and the flasks allo-vjed to stand in tlie dark for various time intervals. The flasks were then cooled in ice-water to reduce the pressure in the flasks in order that when they were carefully opened the potassium iodide solution was drawn into the flasks. The flasks were allowed to stand for about one minute and the liberated iodine titrated with standard sodium tliiosulfate . .The results are shown in Table XXXIX. Table XXXIX shows that there is a loss of approximately 0.05 ml. of 0,1 normal hypoclilorite in approximately two hours in one normal sulfuric acid. This does not necessarily mean that the hypochlorite remains as hypochlorite, or as hypochlorous acid, but that its oxidising capacity toward potassiun] iodide in acid solution is only slightly Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89 TABLE XXXIX OXIDIZING CAPACITY OF SODIUI-i HYPOCHLORITE IM OWE HOIMAL SULFURIC ACID Standing Time ITinutes Ml. 0.1 h i'j3.28303 0 0 0 30 20.29 20.27 20.29 20.27 20.26 20.26 20.2k 20.23 60 90 125 changed on standing. Its decomposition products probably liberate iodine under the same conditions, Experiments were conducted to determine the extent of the oxidation of hypophosphite by excess sodiuiii hypochlorite in dilute sulfuric acid solution. To measured samples of 0.06 normal hypophosphite in ^50 ml. iodine flasks were added a measured excess of standard sodium hypochlorite and ten ml. of six normal sulfuric acid. The flasks were tightly stoppered and the gutters filled t'dth saturated potassium iodide solution. After standing in the dark for various time intervals, the flasks were cooled in ice water to reduce the pressure which had built up in the flasks during the reaction. The flasks were then carefully opened and the potassium iodide solution was draxijn into the flasks by the reduced pressure . The flasks were allowed to remain stoppered for about one Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90 rainute and the liberated iodine titrated x-ri.th standard sodi-urn thio­ sulfate solution. Blanks were determined in the same manner ^ substitut­ ing water for the hypophosphite solution, and calculations made on the basis of the blanks. The blanks were allowed to stand for approximately the same time as the samples. The hypochlorite solution was standardized iodometrically against the standard thiosulfate solution by pipetting 25.00 ml. of the hypo­ clilorite solution into an iodine flask, adding 20 ml. of water, 10 ml. of six normal sulfuric acid, and 10 ml. of saturated potassium iodide solution. The liberated iodine was then titrated with the standard thiosulfate solution. Table XL shox-xs tliat hypophosphite is not completely oxidized to phosphate in six hours by excess sodium hypochlorite in approximately one normal sulfuric acid solution. Tlie data show that the oxidation is complete in ten hours, and that there is no appreciable change in the results if the time of standing is increased to 19 hovu's . The data, also show that the oxidation is somewhat slower in hydrochloric acid tha.n it is in sulfuric acid of the same concentration. The use of hypochlorite in acid solution for the determination of hypopliosphite gives fairly accxxrate and precise results as compared to the gravimetric procedure, but the long time required for the oxidation makes the method inconvenient to use. Also, the flasks must be opened very carefully after the oxidation because of the pressure built up during the reaction. Even with the tightly fitting stoppers of the iodine flasks, a fexf of the samples showed evidence of leakage during Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91 1V3LE XL DETERMINATION OF HYPOPHOSPHITE WITH EJCGESE HYPOCHLORITE IN DILUTE SULFURIC ACID AS A FUNCTION OF TIME OF STANDING Ml. 0.06 N NaHgPOg (a) M 1. 0.1 N NaClO M e q . NaHgPOg Talcen Standing Time 19.98 19.98 19.98 19.96 19.98 19.98 19.98 19.98 19.98 19.98 19.98 19.98 19.98 25.00 25.00 25.00 19.98 19.98 19.98 19.98 25.00 19.98 25.00 19.98 19.98 25.00 25.00 25.00 25.00 25.00 25.00 25.00 1.026 1.319 1.026 1.026 1.319 1.026 l.51t5 1.026 1.026 1.51:5 1.169 1.169 1.169 1.1:63 1.1:63 1.1:63 50 min. 60 min. 70 min. Go rain. 90 m i n . 100 m i n . 6 hr . 10 h r . 10 tir-. lU hr. 17 hr. 17.5 hr. 18 hr . 18.5 hr. 19 hr. 19 5-ir. HaHgPO a Found (b) Percent Meq. 0.951: 1.215 0.977 0.979 1.238 0.9L2 1.523 1.027 1.025 1.535 1.169 1.166 1.168 1.1:59 1.1:61) 1.1:61 93.0 92.1 95.2 95.1: 93.8 91.8 (c: 98.5 100.1 99.9 (c 99.3 100.0 99.7 99.9 99.7 100.1 99.9 (a) Several solutions prepared from the same salt were used. (b) Calculated on the basis of blanks which amounted to a loss of 0.01). to 0 .07 ml. of 0.1 N NaClO. (c) 10 ml. of 6 K HCl used in place of 10 m l . of 6 N H 2SO4 . standing as indicated by coloration of the potassium iodide solution in the gutters of the flasks; these samples were discarded. E. Oxidation of Hypophosphite by Excess Ceric Sulfate in Nitric Acid Solution Moeller and Quinty (1:2) determined hypophosphate by excess ceric nitrate in nitric acid solution at boiling temperature, Ihe excess Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92 ceric nitrate was determined by a potentiometric titration with standard arsenite . Experiments were conducted to determine if hypophosphite could be determined with excess ceric sulfate in nitric a,cid solution. To measured samples of 0,0^ normal hypophosphite in 1^0 ml. be alters were added a measured excess of standard ceric ammonium sulfate and ten ml. of concentrated nitric acid. The beakers were allored to stand for various time intervals and heated to boiling for various times. The excess ceric sulfate was determined by a potentiometric titration i2) . 63. Treadwell, F ., and Hall, W., "Analytical Chemistry," Vol. II, p. 3i40, 8th Ed., 1935, John Wiley and Sons, Inc., New York. 66. Treadwell, W. D . , and Schwarzenbach, G., Helv, Chim. Acta,., 11, 1,05 (l928). — 67. Van Name, R. Ü., and Hnff, W. J., Am.. J. Sci., 68. Vermeil, C ., Anal. 91 (1918). Chim . Acta., _7j> 191 (1932). 69. Willard, H . H ., and Furman, H . H., "Elementary Quantitative Analysis," Third Ed., p. 2I4.8 , D. Van ho strand Co., hew York, 19^0. 7 0 . Wolf, L., and Jung, W. , Z. anorg. u. allgam. Chem., 201, 337 (1931). 7 1 . Wolf, L., Jung, W . , and Uspenskaja, L., Ibid ., 206 , 125 (1932). 7 2 . Yost, D ., and Russell, H ., "Systematic Inorganic Chemistry," Prentice Hall, Inc., hew York, 1943. 73. Young, J., and Gupta , A., Analyst , TU, 367 (1939). 73. Zivy, L., Bull. soc. chirn., 39, 396 (1926); C. A ., Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. 3661 (1926).