Fertilizers - Analytical Chemistry (ACS Publications)

K. D. Jacob. Anal. Chem. , 1950, 22 (2), pp 215–221. DOI: 10.1021/ac60038a004. Publication Date: February 1950. ACS Legacy Archive. Cite this:Anal. ...
0 downloads 0 Views 1MB Size
L U M E 22, NO. 2, F E B R U A R Y 1 9 5 0

215

Opfer-Schaum, R., and Pitisti, M., Fette i(. Seifen. 51, 133 (1944). Parftimerie und Kosmetik, Frankfort-am-Main, Gel many. Petit, G., Bull. soc. chim. France, [SI, 15 141 (1948). Petrova, L. N., Zhur. Priklad Khim., 22,122 (1949). Pronina, M. V., Zavodskaya Lab., 14,1493 (1948). Ramsey, L. L., and Patterson, W. I., ,J. Assoc. Ofic. A g r . C'hemists, 31,441 (1948). Rao, P. R., J . Sci. Ind. Research India. 7B,166 (1948). Rees, H. L., and Anderson, D. H., ASAL. CHEM.,21,989 (1949). Rutten, A. M. G., Chem. en Pharm. Tech. (Dordrecht) Holland, 4, 125 (1948). Schimmel & Co., Inc., New York, "Annual Report, 1946," 1949. Schimmel & Co., Ber. Schimmel & Co., 1939. Ibid., 1940. Ibid., 1941. Ibid., 1942-43. Ibid.. 1944-47.

183) \

/

(84) (85) (86) (87) (88)

(89) (90)

(91) (92) (93)

Seaman. W..11cComas K. H.. and Allen. G. A,. .I\\L. CHEM.. 21, 510 (1949). Siggia, S., and Hanna, J. G., Ibid., 20, 1084 (1948). Taylor, -i.N., Perfumeru Essent. Oil Record, 40,135 (1949). T e k t ' e v , 4.P., and Shor, K. I., Zhur. Obshchei Khim., 17, 2075 (1947). Trujillo, V. A. A , Rev. facidfad farm. y hioquim. C-niv. rtacl. mayor Sun Marco (Lima)Peru, 9, No. 36-36, 136 (1947). Vega, F. de la, and Capillas, A , , Galenica Acta Madrid, 2,No. 1, 47 (1949). Wearn, R. B., Murray, W. h i . , Jr., Ramsey, hi. P., and C'hanrller, N., ANAL.CHEM.,20, 922 (1948). White, J. W., Jr., Ibid., 20,726 (1948). White, J. W., Jr., and Dryden, E. C., Ibid., 20,853 (1948). Wurtzschmitt, B., 2. anal. Chem., 128,549 (1948). Zaugg, H. E., and Horrom, B. IT., .%N.~L. CHmi., 20, 1 0 6 (1948).

RECEIF-ED January 4 , 19:O.

FERTILIZERS K. D . JkCOB Bitrenu of P l a n t I n d u s t r y , Soils, and .4gricultural Engineering. Z-. S. D e p a r t m e n t of Agriculture, Beltsville, Md.

ETHODS of fertilizer analysis published in the period 1943 to 1948are reviewed in the February 1949 issue of A N ~ L Y T I CAL CHEMISTRY ('7'8). The present paper summarizes the subsequent developments in this field and the recent literaturechiefly for 1948-not previously covered. The survev relates not only t o procedures that are directly applicable in fertilizer analysis but also t o those that are of possible usefulness for this puipose. SAMPLIhG AND SAMPLE PREPARATIOV

Several phases of the problem have been studied by Allen (W), who reports t h a t analyses of mised-fertilizer samples taken n-ith single-tube and double-tube samplers, respectively, were in good agreement. As regards changes in the moisture content of the sample, it appears that double-walled asphalt-impregnated bags made of heavy paper are as satisfactory as glass jars for shipping fertilizer samples from the field to the laboratory. Evidence that there is no need for passing the sample through a 10-mesh screen before grinding the portion for analysis, as specified in the official procedure (4,p. 20))is presented. WATER

Methods for free water in fertilizers have been studied collaboratively by Hill et al. ( 7 d ) , with special reference to the airflow and vacuum drying procedures. These procedures are of wider application in fertilizer analysis than is oven drying a t 100" C. Caley and Gordon (25)describe an improved trap for determination of water by distillation with organic liquids. YITROGEN

The one hundredth anniversary of the birth of Johan Kjeldahl, who developed the Kjeldahl method for nitrogen which, with its several modifications, has served so well for the determination of this element in fertilizers, was celebrated in 1949 (1'7'3,186). The merit of selenium as a catalyst for the Kjeldahl reaction continues to receive attention. Rauen and Buchka (134) present data which indicate that a digesting period of 1 hour after the reaction mixture becomes clear is sufficient when selenium-sulfuric acid mixtures containing 0.7% of selenium dioxide are used, even when lysine and lysine-containing piotein are present. According t o Mallol (107) digestion of organic nitrogen materials t o complete decolorization of the solution can be effected i n less than 4

minutes by heating the sample with sulfuric acid containing 0.05 gram of selenium per 20 ml. and agitating R-hile concentrated perchloric acid is added dropwise. Bradstreet (23) reports that met,allic selenium, sodium selenite, and sodium selenate are more effective than the corresponding tellurium products as catalysts for the digestion of aromatic compounds. On the other hand, Willits et al. (181)show that there is danger of loss of nitrogen when selenium, alone or with ot,her catalysts, is present in the Kjeldahl digestion of heterocyclic nitrogen ring compounds. The only catalysts necessary with such compounds are mercuric oside and potassium sulfate, used in the ratio of 0.6 to 15 grams, with 25 ml. of sulfuric acid, and a digestion time of 3 hours. Hiller et al. (73) also state that mercury is the only cat,alyst that gives nitrogen values for proteins as high as those obtained by the Dumas combustion method. Detailed procedures given by the latter workers, involving the use of mercury catalysts, include macro- and micromethods with distillation and titration of the ammonia and a micromethod with gasometric determination of the ammonia by the hypobromite reaction. The results of a comprehensive collaborative study of numerous micro-Kjeldahl procedures for nitrogen are reported by Willits and Ogg (182). Methods for determination of total, water-soluble, and waterinsoluble nitrogen and of nitrate and ammoniacal nitrogen in fertilizers are outlined by Kascimento (118). The recommended method for total nitrogen involves treatment of the sample with phenolsulfonic acid and sulfuric acid with subsequent' addition of powdered zinc, anhydrous sodium sulfate, and metallic mercury, and digestion and distillation in the usual n a y . Post'nikov and Lapshin (152) state that aniline, dimethylaniline, and o-toluidine can be used for determining calcium nitride in the presence of calcium cyanamide. Although decomposition of calcium nitride by hot anhydrous ethyl alcohol is only 80% complete in 16 hours, this reaction can be used as the basis of an approximate method for calcium nitride in calcium cyanamide. T h a t loss of nitrogen may occur when fertilizers containing uitrate and chloride are analyzed by the official Iijeldahl method as modified to include nitrate nitrogen (4, pp. 27-28) is further confirmed by White and Ford (1'7'9)who show that appreciable loss can be expected if the ratio of nitrat.e ( S a K 0 3 )to chloride (HC1) is less than 5 to 2. The loss occurs in the first step of the procedure -the addition of t,he sulfuric-salicylic acid mixture. In a collaborat,ive study reported by Etheredge ( 4 7 ) better results for total nitrogen including nitrate were usually obtained by the Ford

ANALYTICAL CHEMISTRY

216 modification of the official method than by the Shuey procedure (152)or by substituting fuming sulfuric acid for concentrat,ed acid in the official method. For analysis of sodium nitrate the Devarda method (4, p. 28) is more rapid and appears to be more accurate than the Ford modification. Iingane and Pecsok (101) describe a method for the detcrmination of nit,rate which is based on reduction to animonium ion by chromous ion in dilute sulfuric acid solution. With amounts of nitrate ion of the order of 20 to 50 mg. the method is precise and is accurate to +0.2$7; wit,h 2 to 5 mg. of nitrate ion the accuracy is about +29/,. Large amounts of chloride do not interfere. S i trite undergoes the same reduction. A method for nitrate, described by Holler and Huch (75), involves nitrating 3,4-xylenol (3,4-dimethylphenol) in SO'% sulfuric acid and steam-distilling the 6-nitro-3,4-xylenol into dilute sodium hydroxide, forming t,he deeply colored sodium salt which is determined colorimetrically. The recommended concentration range is 0.10 to 0.35 mg. of nitrate nitrogen in 100 ml. of solution, using a cell depth of 1 cm. Interference by chloride and nit,rite can be easily avoided by suitable preliminary treatments. Confirming previous studies (46), Etheredge again reports good results by the formaldehyde procedure in a collaborative investigation of methods for nitrogen in ammonium nitrate (47). Pieters (126) points out that results by this procedure are influenced by the concentration of formaldehyde, the pH, and the duration of the reaction, but the determination can be made under uniform conditions and the titrating solution can be standardized against an ammonium salt of known purity. Determination of the ammonium ion in fertilizers by conductometric titration with sodium hydroxide is proposed by Jander et al. (81). Taras (162) recommends a 0.1% solution of disodium 4,4'-bis(m-tolyltriazeno)-2,2'-stilbenedisulfonat~ein acetone as an indicator in titrating the ammonia recovered from Kjeldahl distillations by absorption in boric acid. D a y and co-workers (34) describe an improvement in the microaeration technique of Sobel et al. (154)for absorbing Kjeldahl ammonia in boric acid. Developments in equipment for micro-Kjeldahl determinations include apparatus in which the digestion and distillation take place in the same vessel (11, l @ ) ,a distillation apparabus (83), arid B vibrator t o prevent bumping in digestions ( 7 0 ) . PHOSPHORUS

Mixtures of perchloric and hydrofluoric acids are frequently used to decompose siliceous phosphates, especially products made by high-temperature processes. I t is recognized that some loss of phosphorus may occur when phosphatic materials are heated at high temperatures with these acids. Chapman et al. (26) show that other elements may also be lost. Thus, among those of possible interest to the fertilizer chemist, complete volatilization of boron, silicon, and arsenic and appreciable loss of manganese and chromium occur at 200 C., whereas potassium, sodium, calcium, magnesium, copper, zinc, vanadium, molybdenum, and cobalt are not affected. For determination of phosphorus in protein-containing substances, decomposition of the sample with concentrated sulfuric acid and Perhydrol is recommrnded by Hahn (64). Tschirch (170) recommends weighing the phosphorus as magnesium iimmonium phosphate hexahydrate after treating with acetone and vacuum drying. With the aid of the Chevenard t.hei,mobalance (29), Duval (40) found that complete conversion of magnesium ammonium phosphate to magnesium pyrophosphate-the form in which phosphorus is usually weighed i n its be accomplished a t 285 C., as gravimetric determination-can compared with the previously reported temperature of 477 C. (411. Sarudi (144)outliiies an improved method for precipitating and weighing phosphorus as ammonium phosphomolybdate, (PI"&P04.1211003. I t is recommended that the precipitate stand at least 2 hours before filtering on a Gooch crucible with a sintered-

glass bottom (Xo. 2A) and that the precipitate be protected from atmospheric moisture during drying and weighing. Addition of ammonium nitrate or ammonium carbonate t o the wash solution prevents formation of a green color when the precipitate is dried at 160" to 180" C. The phosphorus pentoxide conversion factor is 0.0377, which is within 0.3y0 of the theoret,ical value. Ammonium phosphomolybdate is stable a t 410" C., but ammonia is lost at higher temperatures (39, 40). According to Bourdon and Cotte (22) the frartion of molybdic oxide fixed by animoriium phosphomolybdate is constant at high acidity ( 5 S)and excess molybdate (4 to 7 % solution in 31003). Scheffer's method for phosphorus (145) is entirely satisfactory under these conditions of precipitation. In the titration of phosphoric acid \vit,h sodium hydroxide Dijksman ( 3 7 ) reports that the pH a t the end point is 4.25 for titration to H2P04- and '3.1to 9.2 for titration to H P 0 4 - - . He suggests an alkalimetric, method for phosphorus in phosphat,es containing calcium, aluminuni, and iron, which involves removal of the metal ions by means of a cation exchanger (Dusarit). Determination of phosphorus in monoammonium phosphate by electrometric tit,ratiori wit,h sodium hydroxide is proposed by Jander el al. (81). According t o Toropova and Yakovleva (168), the phosphate ion can be determined successfully under certain conditions by electrometric titration with lead acetate in an atmosphere of hydrogen. Recently studied colorimetric methods for phosphorus include the well-known molybdenum blue procedure with reduction by stannous chloride (64, 110, l 4 O ) , amidol (2,4-diaminophenolhydrochloride) (43), or 1-amino-2-naphthol-4-sulforiicacid ( 7 ) ; and the molybdivanadophosphate method (103). Studies by Boltz et al. (19) show t,hat the polarographic waves of molybdiphosphoric acid are considerably influenced by the nat'ure of the supporting electrolyte; two stages of reduction exist, but the halfwave potentials are not well established. Improvements in the isobutyl alcohol procedure for extracting inorganic phosphate from colored solutions (12, 129), with its subsequent determination by the molybdenum blue method, are described by Martin and Doty (110). The precipitate formed with molybdivanadophosphate by hydroxyquinoline has the composition 6CpH7OK.P10j.V~Oj.2211003.12H20 (105); the complex is soluble in isoamyl alcohol and ethyl alcohol. Substitution of a pyridine reagent (equal volumes of pyridine and nitric acid) for st,rychnine in the strychnine-molybd_ate nephelometric method for phosphorus is proposed by Cupr and Hemala ( W ) , who describe a special nephelometer based on measuring hhe voltage rat,her than the current from bhe photoelectric, cells. Guthrie and Nance (63) propose a rapid method for direct determinat,ion of free phosphoric acid and water-soluble phosphorus in superphosphates, which involves stepwise titration of an aqueous ext,ract of the sample Lvith sodium hydroxide with the aid of suitable indic:itors. Cheritat, and Yignau (28) outline it volumetric molybdate method for total phosphorus in basic slag, Ivhich involves decomposing the sample with a mixture of nitric and sulfuric acids, precipitating the annnoniuni phosphomolybdate without prior removal of the sulfate, and heating on the steam bath. On the basis of previous studies (79, 238, 1 3 3 ) one uould expect, this procedure to give high results, owing to interference by the sulfate ion. Collaborative studies reported by Jacob et a l . (80) show that the present official neutral ammonium citrate method (4, pp. 2324) is a suitable procedure for the evaluation of the phosphorus in basic slag. On the basis of this work the neutral ammonium citrate method was adopted in October 1949 by the Association of Official Agricultural Chemists as the official procedure for "available" phosphorus in basic slag, thus replacing the 2% citric acid method (4, p. 25) which had long been used for this purpose. In agreement with othw investigators (71, 74, 80), Barbier and

217

V O L U M E 22, NO. 2, F E B R U A R Y 1 9 5 0 Trocm6 (8)show that within certain limits the solubility of basic slag phosphorus in 2% citric acid increases with the fineness of grinding. Finer grinding of the analytical sample, as with high-speed hammer mills, appears to have only a very small effect on the results for ammonium citrate-insoluble phosphorus in mixed fertilizers (52). Martens (109) discusses the solubility of phosphates in water, aIiimonium citrate, and mineral and organic acids. He concludes that present analytical methods do not permit, a quantitative separation and deterniinat,ion of the different calcium phosphates-a fact that is not well recognized in some quarters. Thus far, synt,hetic organic phosphorus compounds have found 110 place in the fertilizer industry. However, such compounds are of potential interest to the fertilizer analyst because of the recent developments in t.heir use as pesticides and the possibility of their inclusion in fertilizer preparations. According t o Wreath (183), decomposition of the sample by means of t,he semimicro Parr bomb method yields excellent result,s on all types of organic phosphorus compounds. Digestion with nitric and perchloric acids, with proper precautions against explosions, is also suitable for such compounds as hexaethyl tetraphosphate, tetraethyl pyrophosphate, and t,ributyl, tritolyl, and triphenyl phosphate. The difficulty in hydrolyzing such phosphates by means of aqua regia is pointed out by MacIntire et at. (104), who show that digestion periods of a t least 36 hours may he necessary for complete hydrolysis of some of these compounds. Jlethods for the determination of tetraethyl pyrophosphate are outlined by Hall (6.5) and Wreath and Zickefoose (184). The behavior of synthetic organic phosphates in the soil and their effect on plant growth cannot be foretold by the conventional analytical procedures used for the evaluation of phosphate fertilizers (104). POTASSIUM

Perrin (124)has developed a modification of thr chloroplatinate method for potassium in fertilizers, which is reported to have a number of advantages over the official procedure (4, pp. 31-32). It involves an improved wet combustion with aqua regia to destroy ammonium salts and organic matter, n.ith siIdultaneous precipitation of the potassium chloroplatinate. A modification of the perchlorate method, which does not, require preliminary removal of sodium, magnesium, ammonium, sulfate, and chloride ions, is descrilied by Lejeune (W). The procedure is rapid and is wid to be accurate within 2%. The hexanitrodiphenylamine method of Sheintsis (150) has been adapted t o the determination of potassium in fertilizers (56). Prior removal of sulfate, phosphate, and ammonium ions is necessary. The accuracy of the procedure is reported to tie as good as that of the chloroplatinate method. h voluinetric method for determination of potassium in mistures of magnesium and potassium sulfates is based on the use of Gaspar's reagent (2.5 grams of calcium ferrocyanide dodecahydrate dissolved in 500 ml. of water plus 500 ml. of ethyl alcohol, 141). Potassium in aqueous solution can be determined by precipitating and weighing as the salt of 2-chloro-3-nitrotoluene-5-sulfonic acid (13.5% potassium) (20); barium and ammonium salts interfere. Ford (52)reports that finer grinding of the analytical sample, as with high-speed hammer mills, does not generally result in higher values for soluble potassium in mixed fertilizers. On the contrary, somewhat lower values may often be obtained For the determination of potassium in siliceous materials Elving and Chao (45)have combined hydrofluoric acid decomposition of the sample tvith a convenient procedure for the removal of calcium and magnesium. The method is said to be accurate and much simpler in manipulative details than the J. LaLvrence Smith method. Gaudin and Pannell (58)describe a method for the radioactive determination of potassium in minerals, which is applicable to concentrations of potassium above 1%.

Several modifications of the cobaltinitrite method have been developed for the micro- and semimicrodetermination of potassium. They comprise titration of a potassium dichromate sdution of the cobaltinitrite precipitate with standard iron solution ( 2 1 ) ; photometric determination of the nitrite as nitrosoindole (84); and turbidimetric determination of the cloud formed by a reagent composed of sodium cobaltinitrite, sodium nitrite, methanol, and methylene glycol (166). A rapid and sensitive micromethod for pot'assium is based on the polarography of the periodate (156). Potassium can be completely separated from sodium by means of an organic ion exchanger (Amberlite IR-100 resin) with the aid of perchloric acid (87). Fixation of the ammonia with f o r m a l d ~ hyde eliminat,es its interference in the sodium lead iodide test for potassium (48). Investigations of the flame photometer met'hod for potassium include development of improved apparatus and techniques (96, 125, 178) and studies of sources of errors and their reduction (17, 122, 148, 168, 169). It appears that satisfactory application of the methotl to mixed fertilizers has not been achieved. CALCIUM

In a critical study of the oxalatc. method for caicium Holth ( 7 6 ) found that double precipit,ation is necessary for accurat,e analysis, a large excess of ammonium oxalate must be used to avoid precipitation of magnesium from supersaturated solutions, .and the solution must' be kept 4 hours at room temperature before filtering. For the analysis of phosphate-containing solutions Edwards and h1coc.k ( 4 2 ) remove copper and zinc and precipitat.e the osalate in the presence of tartaric acid, with suhsequent ignition to the oxide. Tschirch (fro) gives a procedure for determining calrium hy weighing as the oxalate after treating with acetone and vacuum drying. When calcium oxalate is heated, calciuni carbonate is the stable product a t 420" t,o 660" C., while conversion to the oxide is complete at, 840' C. (40). Methods for the determination of calcium by permanganate titration of the oxalate are outlirird by several investigators (77, 91,161,185). According to Ingols and Murray ( 7 7 ) ,the use of the hydrolysis of urea for raising the pFI of i t solution containing cnlcium and oxalate ions permits formation of large, readily filtered crystals of c:ilciuni oxalate. These crystals are less contaminated t,han the small crystals formed by the standard method, and the t8imerequired for the determination is reduccd as much as joyo. Hirnlxium and Shchigol (18) propose prr,cipitation of the calcium with standard sodium oxalate and potentiometric titrat,ion of the excess ox:tl:it,e with silver nitrate i n :~mmoniac-alsolution. Also, calcium c:in be determined by potentiometric titration of the oxygen-free solution with sodium os:date i n t,he presence of cadmium chloride and ethyl alcohol (1%). Shvedov (1%) report,s that separation of cidcium from magnesium by precipitation as calcium sulfate in the presence of acetone has no advantage over the usual oxalate separation. Gravimetric and titrimetric methods for determining calciuni wit,h the ferrocyanide ion are described by Flaschka and Spitzy (61). .I volumetric method for calcium involves formation of :t red solution \%-ithferric chloride and ammonium thiocyanate, and either visual or amperonietric titration with sodium fluoride in the presence of ethyl alcohol (136, 137). Small quantities of calcium can he determined colorimetrically with the aid of picidonic acid and methylene blue ( 8 9 ) or by precipitation as CaKaXi(S02)s and estimation of the nickel by the dimethyglyoxime reaction ( 1 7 1 ) . The chloranilic acid method ha7 been studied by Gammon and Forbes (54) with special reference to the interference of iron and magnesium. Paper chromatography has been ext'ended to the separation, detection, and determination of calcium ( 3 ) . Other methods for calcium include its determination by the flame photometer (169) and by subjective spectrophotometry (62). Murexide (146) and 5,6-benzoquinaldic acid (106) have

ANALYTICAL CHEMISTRY

218 been investigated as analytical reagents for calcium. In the absence of iron and aluminum, calcium can be separated from magnesium by precipitating the latter as magnesium hydroxide ( 6 1 ) . MAGNESIUM

Several studies of the analytical precipitation of magnesium ammonium phosphate and its subsequent treatment are summarized in the section on phosphorus (40, 41, 170). To determine magnesium in filtrates from calcium oxalate precipitation, by